Patent — Revocation — Genetic engineering — Recombinant DNA technology — Plasmids — Human tissue plasminogen activator — Discovery of DNA and protein sequences — Claims for product however produced — New route to known end. Validity — Whether patent could be revoked because claim not supported by description — Whether what was claimed was an invention — Whether claim related to discovery as such. Sufficiency of description — Whether mistake in description a ground of invalidity — Whether deposit required of micro-organism transformed by plasmid. Obviousness — Person skilled in the art — The Simpkin List — Matter available to the public. Amendment — Possible objection that amended claim not supported by description. Convention court — Status of rulings. Patent invalid but not for reason stated in decision appealed from — Broad claims not for an invention and related to discovery as such — All claims invalid for obviousness — Other objections to validity not made out. Patents Act 1977, sections 1(1), (2), 2(2), 3, 14(5), 60(1), 72(1)(a), (c), 91(1)(c), 125(1), and (3). European Patent Convention, Article 56, Protocol on the Interpretation of Article 69: GENENTECH INC’S PATENT COURT OF APPEAL (CIVIL DIVISION)  RPC 147
GENENTECH INC’S PATENT
COURT OF APPEAL (CIVIL DIVISION)
 RPC 147
HEARING-DATES: 27, 28, 29, 30 June, 1, 4, 5, 6, 7, 11, 12, 13, 14, 15, 18, 20, 21, 22 July, 31 October 1988
31 October 1988
Patent — Revocation — Genetic engineering — Recombinant DNA technology — Plasmids — Human tissue plasminogen activator — Discovery of DNA and protein sequences — Claims for product however produced — New route to known end.
Validity — Whether patent could be revoked because claim not supported by description — Whether what was claimed was an invention — Whether claim related to discovery as such.
Sufficiency of description — Whether mistake in description a ground of invalidity — Whether deposit required of micro-organism transformed by plasmid.
Obviousness — Person skilled in the art — The Simpkin List — Matter available to the public.
Amendment — Possible objection that amended claim not supported by description.
Convention court — Status of rulings.
Patent invalid but not for reason stated in decision appealed from — Broad claims not for an invention and related to discovery as such — All claims invalid for obviousness — Other objections to validity not made out.
Patents Act 1977, sections 1(1), (2), 2(2), 3, 14(5), 60(1), 72(1)(a), (c), 91(1)(c), 125(1), and (3).
European Patent Convention, Article 56, Protocol on the Interpretation of Article 69.
Principal claims of the patent in suit related to human tissue plasminogen activator (t-PA). t-PA, a protein, was known to occur in very small amounts in human tissue. Its function was to activate the conversion of existing precursor plasminogen into plasmin, an enzyme capable of dissolving fibrin in blood clots. t-PA was obviously desirable as a therapeutic agent if it could be produced in quantity. Cells of Bowes melanoma, a human tumour cell line, were known to secrete t-PA. By means of genetic engineering, using a particular route of recombinant DNA technology, the patentees took the relevant genetic information from the cell line and expressed it in microorganisms capable of producing t-PA as a therapeutically acceptable product. The background was briefly as follows.
A protein such as t-PA was known to consist of certain amino acids arranged in a particular sequence, and techniques for establishing the sequence were known although the whole sequence of t-PA itself was not. The sequence of amino acids was known to correspond to a sequence of nucleotides in the essential genetic material (DNA) of the cell in which the protein was made, according to a known code associating each triplet of nucleotides with an amino acid. It was known that, in forming protein, the information in the DNA was first transcribed into messenger RNA (mRNA) of a complementary nucleotide sequence and then translated from the mRNA into the amino acid sequence of the newly formed protein. It was also known that complementary DNA (cDNA) could be formed by reverse transcription from the mRNA, that this cDNA could be inserted into plasmids, that the plasmids could be used to transform suitable micro-organisms or other hosts, and that growth of the latter could replicate the cDNA and produce the protein free of other proteins of the cell from which the genetic information originated.
Reverse transcription of a population of mRNAs from Bowes melanoma cells formed a corresponding population of cDNAs. The technique used by the patentees to identify and isolate the individual cDNA corresponding to t-PA involved making short lengths of synthetic DNA which contained the nucleotide sequences corresponding to known fragments of the amino acid sequence of t-PA. Synthetic DNA having a nucleotide sequence identical with part of the sequence in the desired c-DNA would be expected to bind to it and could therefore be used to identify it.
On the evidence, at least five teams embarked independently on research in this direction. It involved laborious and costly effort and success was not certain. The patentees had been the first to discover the nucleotide sequence of the cDNA corresponding to t-PA (and hence the amino acid sequence of t-PA itself), and this discovery formed the basis of the alleged invention of the patent in suit.
Claims 1-9 and 16-20 of the patent were as follows.
1. Recombinant human tissue plasminogen activator essentially free of other protein of human origin.
2. Human tissue plasminogen activator unaccompanied by associated native glycosylation.
3. Human tissue plasminogen activator as produced by recombinant DNA technology.
4. Biologically active human tissue plasminogen activator in essentially pure form, unaccompanied by protein with which it is ordinarily associated.
5. Human tissue plasminogen activator per se, of the kind produced by expression of a recombinant DNA sequence coding therfor in a mammalian cell line.
6. Human tissue plasminogen activator according to claims 1-5 containing a sequence of polypeptide preceding the N-terminus of the first amino acid of said human tissue plasminogen activator.
7. A recombinant cloning vector comprising a DNA sequence encoding human tissue plasminogen activator.
8. A replicable expression vector capable, in a transforming microorganism or cell culture, of expressing a DNA sequence according to claim 7.
9. The plasmid p$99RIPA or pt-PAtrp12.
16. A process which comprises expressing DNA encoding human tissue plasminogen activator in a recomibant host cell.
17. A process for producing human t-PA, which process comprises:
a. preparing a replicable expression vector capable of expressing the DNA sequence encoding human t-PA in a host cell;
b. transforming a host cell culture to obtain a recombinant host cell;
c. culturing said recombinant host cells under conditions permitting expression of said t-PA encoding DNA sequence to produce human t-PA;
d. recovering said human t-PA.
18. A process according to claim 16 or claim 17 wherein the host cell is a mammalian cell line.
19. A process for producing human tissue plasminogen activator, substantially as described herein.
20. Human tissue plasminogen activator produced by a process according to any one of claims 16 to 19.
Petitioners for revocation alleged under section 72(1)(a) of the Patents Act 1977 that the claims did not relate to a patentable invention because there was no inventive step and also because the patent related to a discovery as such. The petitioners alleged under section 72(1)(c) that the specification of the patent did not disclose the invention clearly enough and completely enough. In particular, in respect of claim 9, they alleged insufficiency on the ground that there had been no deposit of the microorganism transformed by the plasmid. They also alleged insufficiency on the ground of a mistake in the sequence of the cDNA for t-PA as disclosed in the specification.
The patentees contended that certain of the objections amounted to an allegation that section 14(5) had not been complied with, and that such an allegation could not be raised under section 72(1).
In the Patents Court, Whitford J held that the broad product claims were invalid and that the patent must be revoked, but he dismissed other grounds of objection. He found that a specification containing claims which failed to comply with section 14(5)(a) to (c) was not directed to a patentable invention.
The patentees appealed to the Court of Appeal.
Held, dismissing the appeal:-
Lack of support dy description
(1) The Patents Act 1977 must be construed primarily in the context of the European Patent Convention. The authority and usage of previous patent law should be approached with caution and not applied without reserve. The Act had displaced any residual element of common law in the application for and grant of a patent, the protection of the monopoly against infringement, and the revocation of a patent (pp 196, 197, 235, 258).
(2) A patent could not be revoked under section 72(1) on the ground that it had not complied with section 14(5) and should never have been granted (pp 198, 236, 260).
Naimer’s Patent (Decision T23/86)  Official Journal EPO 316, followed. British United Shoe Machinery Co Ltd v Simon Collier Ltd (1909) 26 RPC 21, Mullard Radio Valve Co v Philco Radio & Television Corp of Great Britain Ltd (1936) 53 RPC 323, Swift & Co’s Application  RPC 37, and Mobile Oil Corp’s patent (Decision G’r01/84)  Official Journal EPO 299, referred to.
(3) (per Mustill LJ) An objection that a claim in a patent included products which the proprietor had not invented or embodiments which he had not practised amounted to an objection that the claim was not supported by the description for the purposes of section 14(5)(c) and could not be introduced into revocation proceedings (p 271)
(4) (per Dillon LJ obiter) Although section 14(5) did not provide grounds for the revocation of a patent once granted, an application by the patentee to amend his claim would let in objections that the claim did not comply with section 14(5) (p 248).
Naimer’s Patent (Decision T23/86)  Official Journal EPO 316, applied. Mobil Oil Corp’s patent (Decision Gr01/84)  Official Journal EPO 299, referred to.
Whether patent had been granted for an invention
(5) (per Purchas and Mustill LJJ) The question of compliance with paragraphs (a) to (d) of section 1(1) did not become relevant unless the claim disclosed something which could be described as an invention. An invention must be claimed by stating the essential characteristics of a product or a process (cf section 60(1)). In the type of invention concerned in the appeal, the claim must describe the method which embraced the idea (pp 224, 226, 262, 263, 266).
Mullard Radio Valve Co v Philco Radio & Television Corp of Great Britain Ltd (1936) 53 RPC 323, and Crabtree & Sons Ltd v Hoe & Co Ltd (1936) 53 RPC 443, referred to by Purchas LJ.
Chinoin’s Application  RPC 39, John Wyeth & Brother Ltd’s Application  RPC 545, and American Cyanamid Co (Dann’s) patent  RPC 425, referred to by Mustill LJ.
(6) (per Mustill LJ obiter: Dillon LJ contra) It was difficult to see how the ascription of a hitherto unknown property to a known substance could be either the invention of that “product” or the invention of a “process” (pp 263, 248).
(7) (per Purchas and Mustill LJJ) If something was not an invention, section 125(1) did not convert it into an invention by its inclusion in a claim (pp 224, 264).
(8) (per Purchas and Mustill LJJ) Claims 1, 2, 3 and 4 (and per Purchas LJ claims 5 and 6) were not claims to an invention and should be revoked for that reason (pp 228, 264, 269).
Whether patent related to a discovery as such
(9) (per Purchas and Dillon LJJ: Mustill LJ dubitante) A patent which claimed the practical application of a discovery did not relate to the discovery as such and patentability was not excluded by section 1(2) even if the practical application might be obvious once the discovery had been made (pp 208, 239).
Hickton’s Patent Syndicate v atents & Machine Improvements Co Ltd (1909) 26 RPC 339, and Vicom’s Application (Decision T208/84)  Official Journal EPO 119, followed. Chamberlain & Hookham Ltd v Mayor, etc, of Bradford (1903) 20 RPC 673, referred to. Merrill Lynch Inc’s Application  RPC 1, disapproved.
(10) (per Purchas LJ) A claim to a method embracing a discovery which was speculative as to the method by which the discovery was to be embraced or referred to developments in the art which were neither direct nor immediate would be a claim for protection of the discovery as such and therefore unpatentable by reason of section 1(2) (pp 227, 228, 233).
Mullard Radio Valve Co Ltd v Philco Radio & Television Corp of Great Britain Ltd (1936) 53 RPC 323, applied. Genentech Inc’s Aplication (Decision T292/85) 27 January 1988 (EPO Technical Board of Appeal), considered.
(11) (per Mustill LJ) Whether the combination of a new idea conceived by an inventor and an obvious application of it could lead to a patentable new article or process was not germane to the patentability of an invention founded upon the ascertainment of an existing fact of nature, ie a discovery. If the identification of a discovery as a foundation of a patent was not to be fatal to its validity, the only available means was reliance on the words “which consists of” in the opening passage of section 1(2) (pp 268, 269).
Lane-Fox v Kensington & Knightsbridge Electric Lighting Co Ltd (1892) 9 RPC 413, Reynolds v Herbert Smith & Co (1903) 20 RPC 123, Hickton’s Patent Syndicate v Patents & Machines Improvements Co Ltd (1909) 26 RPC 359, and Diamond v Diehr & Lutton (1981) 450 US 175, distinguished. Merrill Lynch Inc’s Application  RPC 1, Vicom’s Application (Decision T208/84),  Official Journal EPO 119, and Koch & Sterzel (Decision T26/86)  Official Journal EPO 19, referred to.
(12) Claims 2 and 4 were not claims for the practical application of the discovery of the sequences and fell foul of section 1(2) of the Act (pp 228, 239, 269).
(13) (per Purchas LJ: Dillon LJ contra) All the other claims except claims 9 and 19 could not be said to be claims to a method embracing a discovery rather than claims to the discovery as such. By virtue of section 1(2) they were not claims for inventions for the purpose of the Act (pp 211, 230, 233, 234, 240).
Whether invention sufficiently disclosed
(14) Claims 2 and 4 (and per Purchas and Mustill LJJ all the other claims except claims 9 and 19) could be revoked under section 71(1)(c) of the Act for lack of disclosure enabling a person skilled in the art to perform the invention as claimed (pp 227, 247, 271).
(15) (per Purchas and Dillon LJJ) The objection under section 72(1)(c) in relation to the extra “G” in figure 5 was not made out on the facts (pp 172, 236).
(16) (per Purchas LJ) The deposit of a cell line was not necessary for the claims to comply with section 72(1)(c). The specification gave full details of the use of certain expression vectors leading to the expression and replication of t-PA in both prokaryotic and eukaryotic cells (p 228).
Person skilled in the art
(17) (per Purchas and Mustill LJJ: Dillon LJ contra) Where the art by its nature involved intellectual gifts and ingenuity of approach, the person skilled in the art for the purpose of section 3 must have a degree of inventiveness (pp 214, 240, 280).
Valensi v British Radio Corp  RPC 337, and Technograph Printed Circuits Ltd v Mills & Rockley (Electronics) Ltd  RPC 346, distinguished. Samuel Parkes & Co Ltd v Cocker Bros Ltd, (1929) 46 RPC 241, referred to.
(18) (per Purchas LJ: Mustill LJ contra) There was no mandate for applying a different meaning to the words “skilled in the art” in section 72(1)(c) from that in section 3 (pp 215, 278).
Valensi v British Radio Corp  RPC 337, and King, Brown & Co v Anglo-American Brush Corp (1892) 9 RPC 313, referred to.
(19) (per Dillon LJ) The ability of a person skilled in the art to perform known techniques had nothing to do with the question of obviousness or inventive step under section 3 (p 243).
Valensi v British Radio Corp  RPC 337, considered.
(20) (per Mustill LJ) Where the person skilled in the art was represented by a team of individuals practising a variety of arts through a variety of individuals, it made no difference whether they worked together as a single unit or whether individuals were regarded as sub-contractors. The standard for skills was the same. The gifts to be ascribed to each depended on the contribution which he was to make to the work of the team by the exercise of his own particular art (pp 278, 280).
Badische Anilin und Soda Fabrik v La Societe Chimique des Usines du Rhone (1897) 14 RPC 875, referred to.
(21) (per Dillon and Mustill LJJ) The hypothetical team of persons skilled in the art should be credited with sufficient time and the best available equipment to carry out the work (pp 246, 278).
Matter available to the public
(22) (per Purchas LJ) Matter was not available to the public if it was merely known to one or two individual research workers pursuing their own experiments in private. The “public” was that community of research workers skilled in the art in general (p 204).
(23) (per Dillon and Mustill LJJ) The question whether an invention would or would not have been obvious to a person skilled in the art, having regard to matter which had been made available to the public at the relevant date, was a jury question to be considered broadly as a touchstone of the originality of the supposed invention. The judge must form his own mental picture of the art and of the skilled practitioner and see how it measured up against the tasks which the latter was assumed to be tackling (pp 246, 281).
General Tire & Rubber Co v Firestone Tyre & Rubber Co Ltd,  RPC 457, and Johns-Manville Corp’s Patent,  RPC 479, referred to by Mustill LJ.
(24) (per Purchas and Mustill LJJ) To assess inventiveness one must look not at individual achievements but at the total work and the result. There was a risk in seizing on a few aspects of the case and drawing too close to them to achieve a proper perspective (pp 222, 285).
British Westinghouse Electric & Manufacturing Co Ltd v Braulik (1910) 27 RPC 209, referred to.
(25) (per Mustill LJ) To fasten on the word “step” in the expression “inventive step” and look at the steps actually taken by the inventor was unsound. The history of the route by which he arrived at the invention was not material to the validity of the patent. The word step must carry the meaning of “activity”, as the other two texts of the European Patent Convention (Article 56) made plain (p 285).
(26) A novel idea for a project carried through to a successful conclusion, in a field where the necessary facts could not be discovered by ready reference to an available source, might involve the necessary inventive step to merit patent protection (pp 221, 243, 280).
American Cyanamid Co (Dann’s) Patent  RPC 425, IG Farbenindustrie AG’s Patents (1930) 47 RPC 289, Beecham Group Ltd v Bristol Laboratories Ltd,  RPC 153, Johns-Manville Corp’s Patent,  RPC 479, Olin Mathieson Chemical Corp v Biorex Laboratories Ltd  RPC 157, Tetra Molectric Ltd v Japan Imports Ltd,  RPC 541, and American Cyanamid Co v Ethicon Ltd,  RPC 215, considered.
(27) (per Dillon and Mustill LJJ) Being the first to succeed was not enough; but if no one else had set out to produce the invention this might be evidence that it was not obvious (pp 242, 277).
(28) (per Dillon and Mustill LJJ: Purchas LJ contra) Laborious and costly effort did not necessarily involve an inventive step even if amounting to more than the exercise of proficiency (pp 214, 221, 242, 280).
American Cyanamid Co (Dann’s) Patent,  RPC 425, considered, Valensi v British Radio Corp,  RPC 337, also referred to by Purchas LJ.
(29) (per Dillon LJ) An idea was obvious if at the relevant date a man skilled in the art would have thought the idea well worth trying out to see whether it would have beneficial results; or a research group would have been directly led to try the idea in the expectation that it might well produce a useful result; or the material in question were “lying in the road” and ready for the research worker to use (p 242).
Johns-Manville Corp’s Patent,  RPC 479, Olin Mathieson Chemical Corp v Biorex Laboratories Ltd,  RPC 157, and Philips (Bosgra’s) Application,  RPC 241, applied.
(30) (per Mustill LJ) Section 3 required the court:-
(a) to imagine the hypothetical skilled man, up to date with the prior art, looking toward the goal (whether or not precisely identifiable in advance) which would become the claimed invention if and when realised;
(b) to ask by what routes it would have been possible to proceed to the goal from the starting point;
(c) to see what obstacles he would have faced on those routes and to enquire how he could have overcome them; and
(d) to ask whether they could have been overcome by no more than pertinacity, sound technique or trial and error, or whether a spark of imagination would have been required beyond the imagination properly attributable to him as a man skilled in the art.
The requirement of section 1(1)(b) were fulfilled only if the final question was answered in the latter sense (p 275).
(31) (per Mustill LJ obiter: Dillon LJ contra) A patentable invention was not reached by the mere process of pursuing obvious lines of enquiry by trial and error until a favourable result chanced to turn up (pp 275, 276, 277, 240, 248).
May & Baker Ltd v Boots Pure Drug Co Ltd (1950) 67 RPC 23, American Cyanamid Co (Dann’s) Patent  RPC 425, and IG Farbenindustrie AG’s Patents (1930) 47 RPC 289, considered.
(32) (per Dillon and Mustill LJJ: Purchas LJ contra) Genentech were not operating in an entirely empirical field. The goal was known, and sufficient of the theory and practice was known for them to know what they were doing. It was obvious to the person skilled in the art to set out to produce human t-PA by recombinant DNA technology. All the steps taken by Genentech in finding out the composition of the sequences and applying that knowledge to produce human t-PA were applications of known technology without any original step. All the claims were invalid for obviousness and want of inventive step (pp 240, 242, 243, 247, 248, 276, 286).
May & Baker Ltd v Boots Pure Drug Co Ltd (1950) 67 RPC 23, distinguished.
(33) (per Purchas LJ, dissentiente) Having regard to the Protocol on the Interpretation of Article 69, fairness would not be achieved as between the patentee and the rest of those involved in the practice of the art if it was to be held that there was no inventive step involved in Genentech’s discovery of the sequences (p 221).
American Cyanamid Co (Dann’s) Patent,  RPC 425, Valensi v British Radio Corp,  RPC 337, and Genentech Inc’s Application (Decision T292/85), 27 January 1988 (EPO Technical Board of Appeal), considered.
Construction of Claim
(34) (per Purchas LJ) Under section 125(1), reference to the specification was for the purposes of interpretation only and was not a means of restricting a claim which was otherwise too wide or lacked substance. But if an interpretation of the claim within the classic dictum of Lord Russell in Electrical & Musical Industries Ltd v Lissen ((1939) 56 RPC 39) was inconsistent with an interpretation which achieved a position between the two extremes referred to in the Protocol on the Interpretation of Article 69, the latter should prevail (pp 197, 228).
Rulings of other courts
(35) (per Purchas and Dillon LJJ) The judges in the Court of Appeal had to evaluate the evidence afresh and form their own opinions (pp 228, 240).
Benmax v Austin Motor Co Ltd, (1955) 72 RPC 39, followed.
(36) (per Mustill LJ) The requirement in section 91(1)(c) that the court should take judicial notice of a decision of the relevant Convention court was directed to evidentiary matters, ie the proof of matters which might otherwise have had to be proved as foreign law. It did not give to rulings of other courts any greater status than they otherwise would have possessed (p 265).
Observed (per Dillon LJ): In examining an application to see whether the claims satisfied the requirements of section 14(5), the Patent Office ought to have very clearly in mind that it was undesirable to allow claims the object of which was to cover a wide and unexplored field or where there was no disclosure in the specification coterminous with the monopoly indicated in the claims (p 236).
Dr Sidney Brenner, CH, FRS sat with the court as a scientific adviser.
American Cyanamid Co (Dann’s) Patent  RPC 425 (HL).
American Cyanamid Co v Ethicon Ltd  RPC 215.
Badische Anilin und Soda Fabrik v La Societe Chimique des Usines du Rhone, (1897) 14 RPC 875.
Beecham Group Ltd v Bristol Laboratories Ltd  RPC 153 (HL).
Benmax v Austin Motor Co Ltd (1955) 72 RPC 39(HL).
British United Shoe Machinery Co Ltd v Simon Collier Ltd (1909) 26 RPC 21.
British Westinghouse Electric & Manufacturing Co Ltd v Braulik (1910) 27 RPC 209 (CA).
Chamberlain & Hookman Ltd v Mayor, etc, of Bradford, (1903) 20 RPC 673 (HL).
Chinoin’s Application,  RPC 39.
Crabtree (RW) & Sons Ltd v Hoe & Co Ltd, (1936) 53 RPC 443 (CA).
Diamond v Diehr & Lutton, (1981) 450 US 175 (USA Supreme Court).
Electrical & Musical Industries Ltd v Lissen, (1939) 56 RPC 39 (HL).
Genentech Inc’s Application (Decision T292/85), 27 January 1988 (EPO Technical Board of Appeal).
General Tire & Rubber Co v Firestone Tyre & Rubber Co Ltd  RPC 457 (CA).
Hickton’ Patent Syndicate v Patents & Machine Improvements Co Ltd (1909) 26 RPC 339 (CA).
IG Farbenindustrie AG’s Patents, (1930) 47 RPC 289.
John-Manville Corp’s Patent  RPC 479 (CA).
King, Brown & Co v Anglo-American Brush Corp (1892) 9 RPC 313 (HL).
Koch & Sterzel (Decision T26/86),  Official Journal EPO 19 (EPO Technical Board of Appeal).
Lane-Fox v Kensington & Knightsbridge Electric Lighting Co Ltd (1892) 9 RPC 413 (CA).
May & Baker Ltd v Boots Pure Drug Co Ltd, (1950) 67 RPC 23 (HL).
Merril Lynch Inc’s Application,  RPC 1.
Mobil Oil Corp’s Patent (Decision Gr01/84),  Official Journal EPO 299 (EPO Enlarged Board of Appeal).
Mullard Radio Valve Co v Philco Radio & Television Corp of Great Britain Ltd, (1936) 53 RPC 323 (HL).
Naimer’s Patent (Decision T23/86),  Official Journal EPO 316 (EPO Technical Board of Appeal).
Olin Mathieson Chemical Corp v Biorex Laboratories Ltd,  RPC 157.
Parkes (Samuel) & Co Ltd v Cocker Bros Ltd, (1929) 46 RPC 241.
Philips (Bosgra’s) Application,  RPC 241.
Reynolds v Herbert Smith & Co, (1903) 20 RPC 123.
Swift & Co’s Application,  RPC 37.
Technograph Printed Circuits Ltd v Mills & Rockley (Electronics ) Ltd,  RPC 346 (HL).
Tetra Molectric Ltd v Japan Imports Ltd,  RPC 541 (CA).
Valensi v British Radio Corp,  RPC 337 (CA).
Vicom’s Application (Decision T208/84),  Official Journal EPO 119 (EPO Technical Board of Appeal).
Wyeth (John) & Brother Ltd’s Application,  RPC 545.
This was an appeal by the patentees from a decision of Whitford J. In the Patents Court (reported  RPC 553) in a consolidated proceeding of a petition by The Wellcome Foundation Ltd for the revocation of patent No 2119804 of Genentech Inc, and an action by Genentech Inc and Boehringer Ingelheim Ltd against The Wellcome Foundation Ltd for infringement of the same patent. It stood admitted that if the patent was valid there was a threat to infringe the majority of the claims, and accordingly the only issue was validity.
Stephen Gratwick QC and Richard Hacon appeared for the appellants; Robin Jacob QC, Peter Prescott and Martin Howe appeared for the respondents.
PANEL: PURCHAS, DILLON, MUSTILL LJJ
JUDGMENTBY-1: PURCHAS LJ
PURCHAS LJ. In the judgments which we are about to hand down the court is unanimous about the order to be made. The judgments are necessarily of some length in view of the number of different issues raised during the argument and the desire of the court to deal in as helpful a way with as many of them as we properly could, notwithstanding differing views held on some of the topics involved. It would be difficult for the parties and their advisers to assimilate and consider the three judgments in the short time normally allowed between making advance copies available and the formal handing down of the judgments. Additionally in this case the court understands that there is a degree of general commercial interest over the result which would make giving advance notice of the result undesirable. We have been conscious of the importance of expedition with security. The fact that today’s date has been met is due in no small part to the efforts of our personal clerks who have achieved a result which would have been difficult using normal scrivenery services.
The court proposes to make the order resulting from the judgments, but not immediately to invite submissions as to consequential orders. Instead we propose to indicate the nature of the consequential orders which we would be minded to make without the assistance of further submissions. We shall then adjourn. The order will not be drawn up for seven days — and if before noon on Monday, 7 November, either party gives notice to the court that they wish to make submissions then a date for further argument will be arranged.
For the reasons given in the judgments about to be handed down the appeal will be dismissed. Subject to further argument costs will follow the event; if the appellant applies, the court is minded to grant leave to both parties to appeal to the House of Lords.
The court has before it an appeal from a judgment given by Whitford J on 7 July 1987 in two consolidated proceedings. One was an action brought by Genentech Inc (“Genentech”) against the Wellcome Foundation Ltd (“Welcome”) for infringement of letters patent No 2119804 (“the patent”) by writ dated 5 March 1986. The other was a petition for revocation of that patent lodged by Wellcome on the day the patent was granted, namely 26 February 1986. The patent is concerned with the synthesis of a protein, human tissue plasminogen activator referred to throughout these proceedings as “t-PA”. The appeal is concerned with the science and development of genetic engineering, otherwise referred to as recombinant DNA technology. During the last two decades this area of science had been the subject of intensive and widespread research which has resulted in rapid development and important advances in knowledge and technology. The appeal also raises, we are told for the first time at an appellate level, issues relating to the meaning of certain sections of the Patents Act 1977 (“the 1977 Act”) in the context of the Convention on the Grant of European Patents as presented to Parliament in November 1974 (“the European Patent Convention”). Before proceeding further with this judgment I wish to record the court’s gratitude to Dr Sydney Brenner, CH, FRS, Director of the Molecular Genetics Unit of the Medical Research Council, for the assistance we have received in relation to the intricate and complicated scientific questions arising out of the appeal and also to the assistance that the court has received from Mr Gratwick QC, who appeared for Genentech, and from Mr Jacob QC and Mr Prescott who appeared for Wellcome.
1.01. Although by the turn of the century it was known that the nuclei of the cells of plants and animals contained long structures called chromosomes which acted as carriers of hereditary characteristics, little was known about their structure and function and that of other constituent parts of cells. Early work in genetics established that the chromosomes contained components known as genes and that these controlled the development and function of the organism. The second half of the present century has, however, seen dramatic developments and further discoveries relating to the function of cellular components and in particular to the part played by two nucleic acids, namely deoxyribonucleic acid (“DNA”) and ribonucleic acid (“RNA”). Genes are made of DNA and serve as the repository of instructional information governing the organism, its composition and structure and to a large extent how it will grow as well as its behaviour and lifespan. In particular they specify the structure of the proteins of the cell, by sending the instructional information in the form of a RNA copy to the protein synthesising machinery of the cell. t-PA is one such protein, and in order to instruct a cell to make t-PA it is necessary to introduce into that cell a suitable DNA containing the appropriate coding message. This proposition requires further expansion in the context of this appeal.
1.02. The wound-healing mechanism of the human body involves two systems which act successively. The flow of blood is stopped by the coagulation system which deposits a clot composed of a protein called fibrin. The fibrin acts as a matrix and must be later removed and replaced by tissue. This is done by the fibrinolytic system which involves an enzyme called plasmin that digests fibrin. Plasmin is generated from an inactive protein precursor called plasminogen. Plasminogen is converted to plasmin by an activator. t-PA is one such activator.
1.03. The fibrinolytic process, when properly directed, was realised to be a powerful weapon in the armoury with which to combat circulatory conditions associated with blood clots causing partial or, in severe cases, total occlusion of a blood vessel. The most serious of these conditions is coronary thrombosis, that is, a clot in the blood vessels serving the heart which, if not quickly dealt with, leads to irreversible destruction of the heart muscle. Traditional anti-coagulant therapy did nothing to remove blood clots when once formed; this only fibrinolysis could achieve. t-PA has a further important advantage, namely it has a specific and high affinity for fibrin so that the curculatory plasminogen is activated at the target site rather than generally throughout the circulation system as in the case of other non-specific activators. Other plasminogen activators were known, eg streptokinase and urokinase which, at the time of the patent, were commercially available and in use for treatment of acute thrombosis. For reasons which it is not necessary to examine, each of these activators had serious disadvantages although urokinase is still being researched and developed. t-PA was originally found in human uterine tissue. In 1980, a female cancer cell line, know as Bowes melanoma cells, was found to produce t-PA, bt only in small quantities.
1.04. During the 1970/1980’s it was appreciated that the production of quantities of t-PA sufficient to treat patients was not only desirable but could also be a highly profitable operation. In the early 1980’s, several research teams were engaged upon the search for a method of synthesising t-PA in commercially viable quantities. The possibility considered by these teams was to produce t-PA by genetic engineering (recombinant DNA) methods which had already been realised for other human proteins, such as the interferons, growth hormone and insulin. In essence this had been done by finding or making a gene with a sequence specifying the protein and introducing it into an appropriate cell in which it directed the synthesis of its protein product. Besides the Genentech team there were teams at the Cold Spring Harbour Laboratory, New York, USA (“CSH”); the Genetics Institute, New York, USA (“GI”); the Univeristy of Umea, Sweden (“Umea”) and two teams working in parallel at Celltech Ltd in the United Kingdom and at the University of Leuven in Belgium (“Celltech/Leuven”). Genentech had identified the structure of t-PA and of the DNA which coded for it by December 1981. None of the other teams, for one reason or another, had succeeded in identifying the full structure of t-PA or of the DNA coding for it by the priority date of the patent, which was in May 1982. It will be necessary to consider, however, at a later stage the progress made by each of these teams along the path towards the objective.
1.05. Recombinant DNA technology is a very new development and arose from scientific discoveries in molecular biology and genetics. The first breakthrough in this field came in 1953 when two research workers at Cambridge, DR FHC Crick and Dr JD Watson, identified the basic structure of the DNA molecule — work for which they were awarded a Nobel Prize. This structure gave immediate insight into how the molecule might be replicated by specific base-pairing mechanisms and also suggested how the sequence of its components might carry the information to specify the proteins of the cell. The science advanced very rapidly, so that, within ten years, the basic features of the biological systems became known.
2.01. In order to deal with the legal issues raised in the appeals I must, however inadequately, embark on a layman’s exposition of some of the elementary principles of genetic engineering. I am particularly indebted to Dr Brenner in preparing this and the next ensuing technical sections of this judgment.
2.02. The constituent units of proteins are amino acids; they are simple organic molecules sharing the same overall structure and for present purposes there are only 20 of them. They consist of an amino group (NH$99) and a carboxylic group (COOH) attached to a central carbon atom (the alpha carbon); making up the remaining two links to the alpha carbon are a hydrogen atom and a fourth group known as the side chain. [for diagram see original]
It is in the nature of the side chain that amino acids differ one from another and it is upon these side chains that the various properties of the amino acids depend. In the construction of proteins each amino acid is joined to the next one by a peptide bond forming a polypeptide chain which folds to give the three dimensional structure of the protein.
2.03. Proteins vary immensely in size and their polypeptide chain may contain anything between 50 and 2000 amino acids or more. However long they are, there will be at one end an amino group exposed and at the other a carboxylic terminus exposed (the “N terminus” and the “C terminus”). [for diagram see original]
In the above tetrapeptide four amino acids are shown, namely methionine, aspartic acid, glutamine and glycine (MDQG). In nature proteins consist of much longer chains (polypeptides) and contain varying sequences of the twenty different amino acids. A full length of mature molecule of t-PA contains 527 amino acids, amongst which all twenty are represented in various numbers and positions in a defined sequence.
2.04. Returning to the nucleic acids, DNA and RNA, these have similar but different molecular structures. The component units are organic molecules called nucleotides. These consist of a pentose sugar and a phosphate group which is common, and a base which is individual, to each necleotide. The information and “operating instructions” reside in the linear sequence of the bases aligned along the length of the nucleic acid molecule. [for diagram see original]
2.05. Both DNA and RNA use combinations of only four bases each to perform their coding function. These are adenine, guanine, cytosine and thymine in the case of DNA. RNA also uses adenine, guanine and cytosine but in the place of thymine a chemically similar base called uracil is used. These bases are known respectively by their first letters, namely A, G, C, T and U. They have an important characteristic, namely that C always pairs, by means of a hydrogen bond, with G but not with A or T, and A pairs with T or, in the case of RNA, U, but not with G or C. This propensity to form paired structures (“base pairs”) is an important feature underlying the function of nucleic acids. Advantage is taken of this in the techniques of genetic engineering, as in the detection of sequences of hybridisation with a radioactive probe with the complementary sequences, or in the ligation of restriction enzyme fragments (see paragraphs 3.10, 3.12 and 3.13 (infra)).
2.06. In its normal state DNA consists of two complementary strands of nucleotides running in opposed directions. The two strands are held together by hydrogen bonds between the base pairs such that G pairs with C, and A with T, as represented in the following figure. [for figure see original]
2.07. When a cell replicates by division, the whole genetic complement of the cell must be faithfully copied so that it is inherited by each daughter cell. In this process the double helix unwinds rupturing the hydrogen-bond pairing of the bases from the two strands and allowing each strand to expose the linear sequence of its unpaired bases. An enzyme (DNA polymerase) promotes the creation of a new strand complementary to each of the parent strands with the appropriate base being joined on the new opposite strand, namely C acquires a G, T an A, etc. In other processes, eg transcription (see below), the helix is similarly locally denatured to expose the bases.
3. Transcription, translation and reverse transcription
3.01. DNA molecules are of a relatively stable nature and provide the permanent information bank. RNA molecules are short-lived and exist in various forms which serve different functions. This appeal is concerned mainly with messenger RNA (mRNA) but also to a lesser extent with transfer RNA (tRNA) and ribosomal RNA (rRNA).
3.02. When DNA is expressed in a cell, the two strands are separated locally and an enzyme, RNA polymerase, copies or, as it is said, transcribes the sequence from one strand into a sequence of bases of mRNA. Within every cell there are, for each of the many different genes, varying numbers of mRNA copies. Each mRNA will reflect the nucleotide sequences of the DNA for which it is the messenger and will have a similar composition with the exception that it has only one strand. [for diagram see original]
3.03. mRNA molecules are a small fraction of the RNA of the cell, usually less than 5% of the total RNA content. A given target mRNA will be only one amongst a large number of other irrelevant mRNAs. The proportion of a given mRNA can be very small indeed, as was eventually discovered in the case of t-PA mRNA. In the cell’s function of creating a protein the mRNA is then translated into a polypeptide chain. Thus if the gene is a blueprint for a protein its mRNA is a working copy. Ribosomes (rRNA) bind to an initiation signal in the mRNA and the sequence is then decoded, triplet by triplet, by the transfer RNAs carrying in their amino acids in the correct reading frame. When a termination signal in the mRNA is reached the ribosome will break off translation producing a completely free polypeptide chain:- [for diagram see original]
3.04. The correspondence between the nucleotide sequence of the mRNA and the amino acid sequence of the polypeptide chain is given by the genetic code. The bases are read in groups of three, and each triplet, or codon, codes for one specific amino acid. Because there are 64 (4$99) triplets but only 20 amino acids, in most cases more than one codon is used to code for any particular amino acid. The genetic code is conveniently displayed in the form of the table shown below:- [for table see original]
AUG is an initiation signal and also codes for methionine, while UAA, UAG and UGA do not code for any amino acid and are termination signals. While it is possible to translate a nucleotide sequence uniquely into an amino acid sequence, this cannot be done in the reverse direction. The redundancy or degeneracy of the code means that, with very few exceptions, it is possible to have several nucleotide sequences which correspond to a given amino acid sequence.
3.05. In recombinant DNA technology a viral enzyme called reverse transciptase is used to copy mRNA into DNA copies (cDNAs). The cDNA produced in this way can be made double-stranded and can then be incorporated into a vector DNA molecule, called a cloning vector. The cloning vector can then be introduced into an appropriate host cell where it is replicated and stably maintained. In this way the entire complement of mRNA extracted from a cell can be converted to cloned cDNA. These assemblages are referred to as “clone libraries”. A number of ways are available for incorporating lengths of cDNA into a cloning vector; but it is not necessary to give details of these at this stage.
3.06. By screening the clone libraries the genetic engineer hopes to isolate clones of the target DNA which can then be further manipulated and incorporated into another vector known as an expression vector. Suitable expression vectors exist for bacterial, yeast or mammalian cells, the commonly used hosts, in which the new construct of expression vector together with its DNA insert is not only replicated and stably maintained but also produces the appropriate protein encoded by that DNA.
3.07. The purpose of using the recombinant DNA process is to obtain enhanced production of a given protein by inserting a DNA coding for that protein into a host cell which, by having multiple copies of the gene and enhanced expression, produces more than a natural cell line. For example, t-PA is naturally found in human tissue and is also produced by the Bowes melanoma cell but in wholly inadequate amounts for therapeutic or clinical purposes. Recombinant DNA technology uses the limited quantities of the natural protein available in the human cell and its mRNA to create, clone and identify a cDNA which can be inserted into special host cells for efficient reproduction.
3.08. The steps involved in the recombinant DNA technology, can for the present purposes, be summarised as follows:-
1. The provision of an enriched sample of mRNA obtained from the natural source.
2. Reverse transcription of the enriched mRNA into cDNA using oligo-dT priming and conversion of the single-stranded cDNA into the double-stranded form using DNA polymerase with random priming.
3. Sizing by a fractionating gel to eliminate fragments less than a certain size (eg 350 base pairs) and selecting for fragments of the appropriate length.
4. The insertion by a process known as ligation of the cDNA fragments into specially designed vectors. In the present case C-tailed cDNA fragments were annealed and ligated with G-tailed plasmid vector pBR322. (C-bases combine with G-bases.)
5. Transformation in bacterial cells, eg E coli, with ensuing replication and the production of a cDNA library.
6. The screening of the library by one of a number of methods for the purpose of identifying the target DNA clone or clones if they are present.
7. By a second recombinant cloning process the insertion of the target clones so identified into an appropriate vector known as an expression vector to produce the target protein.
3.09. Vectors are DNA molecules usually less than 10 kilobases in size and will generally be either plasmids or bacteriophage chromosomes. For present purposes it is only necessary to consider the former variety. Plasmids are circular molecules of DNA which lead an independent existence in the bacterial cell. They almost always carry one or more genes and often these are responsible for useful characteristics displayed by the host bacteria. They are equipped with the necessary DNA sequences for separate replication and are able to multiply within the cell quite independently of the main bacterial chromosome. By May 1982 there were available to the genetic engineer a number of plasmids with specially designed characteristics and carrying appropriate genes. An example which played an important part in the history of the recombinant DNA technology in relation to t-PA was plasmid pBR322 which was 4.362 kilobases in length. It contained two particular genes endowing resistance to ampicillin (amp$99) and to tetracycline (tet$99) at various positions in its circular molecule in which there were also a number of restriction sites. These are of significance when it is required to insert further lengths of DNA into the plasmid in the recombinant DNA process. (See the figures at the end of paragraph 3.10 (infra)).
3.10. The restriction sites consist of particular sequences of nucleotide bases, and are recognised by a corresponding restriction endonuclease, which will cleave the DNA molecule at the recognition sequence and nowhere else. Restriction endonucleases are enzymes made by particular organisms, eg BamHI is made by a strain of Bacillus amyloliquefaciens and cleaves the DNA molecule at positions shown in the recognition sequences: [for diagram see original]
Other endonucleases cut at different but specifically individual recognition sites. Enzymes with staggered cutting positions as BamHI (above) leave fragments with single stranded ends which are complementary to themselves. Thus fragments of DNA cut with one such enzyme may be easily annealed and ligated to a vector cut with the same enzyme as shown below for the enzyme BamHI: [for diagram see original]
3.11. By arranging the cleavage and ligation so that one of the two genes amp$99 or tet$99 is replaced by the recombinant DNA insert, plasmids showing resistance to only one rather than both antibiotics can be separated. The plasmids in fact used by Genentech were more complex and specialised derivate of pBR322 or similar plasmids but the above example will suffice for this judgment. No inventive step is claimed by Genentech in the selection of the particular vectors into which part or whole t-PA clones were inserted in order to construct the two plasmids named in claim 9, p$99RIPA and pt-PAtrp12.
3.12. In normal circumstances the clone library will contain several thousand transformed bacterial colonies only a few, if any, of which are likely to contain the target cDNA sequences. There were prior to 1982 a number of screening methods which relied on the ability to recognise the gene product itself using either a specific antiserum or biological activity. These required a high abundance of the target clone. Other methods relied on the ability of a specific cDNA isolate to hybridise to its cognate mRNA (the affinity of complementary base pairs). This process was highly labour-intensive but was successfully used prior to 1982. Expression screening using antibodies to recognise the protein being produced by a recombinant clone could be used to screen large numbers of clones but was a newly emerging technique in 1982.
3.13. In the case of a low-abundance cDNA the process which was eventually proved to be successful was the use of oligonucleotide probes. In order to use this screening technique it is first necessary to isolate a sample of pure protein from a natural source and to obtain partial amino acid sequences of some peptide fragments. Favourable peptide sequences include amino acids with a low coding degeneracy. This means identifying a short sequence preferably with methionine or tryptophan which have unique coding sequences or, failing this, one of the amino acids for which there are only two alternative encoding triplets. Once the oligonucleotide sequences have been determined the appropriate probe can be chemically synthesised by sequential stepwise addition after covalent attachment of the first component to a solid support such as glass beads, cellulose paper or synthetic resin. The probes themselves are further provided with a radioactive atom, normally phosphorus atom $99P introduced at the 5′ end of the probe using a phosphorylating enzyme known as polynucleotide kinase. When the probe has attached by hybridisation to the target DNA the latter can then be identified by examination with X-ray sensitive film.
3.14. The interpretation of the results was a skilful and painstaking process. The successful hybridisation with the target cDNA had to be successfully contrasted with effects derived from irrelevant combinations. These effects are described as false positives or, alternatively, as “background” or “noise”. In particular the complementary base pairing of the G and C bases which have a strong affinity was liable to give false positives if there was a false combination towards either end of the obligonucleotide probe and an incorrect cDNA. The efficiency of the probing process was never 100% and depended upon three factors:-
1. The length of the probes used.
2. The G-C content of the probe.
3. The stringency of the hybridisation and the washes which were used to eliminate the false colonies.
Background arises, for example, if the probe wrongly sticks to the nitrocellulose filter or to debris from the bacterial colonies themselves. Hybridisation of all kinds is encouraged by conditions of low stringency. On the other hand, if the conditions are made too favourable then too many false positives will appear. The skill and judgment comes in deciding the exact degree of stringency which will reduce the number of false positives and the degree of washing which will remove the uncovenanted colonies and yet leave the target colonies. With strong uncovenanted G-C hybridisation forming false positives there was a danger of washing away the target cDNA colonies and leaving only false positives.
3.15. Any particular cDNA isolated is unlikely to represent the full length of the corresponding mRNA. This is not a major problem since the cloned cDNA can itself then be used as a unique hybridisation probe to search amongst the population of cDNAs for other isolates whose sequence content overlaps that of the original recombinant. In this way an overlapping series of sequences can be assembled such that between them they represent the entire sequence required to encode the protein.
3.16. Vectors play a second part in the recombinant process. This is in the expression of the isolated target cDNA which has been inserted into it for the purposes of the expression and replication of the target protein. The full-length cDNA sequence now contains the genetic information required to specify the complete amino acid sequence of the corresponding polypeptide chain. This genetic information remains incomplete, however, because the genetic signals required to potentiate transcription of the coding sequence are absent. Genetic transcription requires the recognition by the host cell RNA polymerase of a special sequence within the DNA known as a promotor. When the RNA polymerase binds to the promotor sequence it causes a local melting of the DNA duplex and initiates the transcription process already described in this judgment. In this second stage the expression vector will include a suitable promoter together with other features important in promoting the reception and replication of the exogeneous DNA by the host cell. Again, the selection and management of this process calls for a high degree of skill and technique in order to achieve optimum expression of the target protein. Two expression vectors, both plasmids, are of particular concern, namely p$99RIPA and pt-PAtrp12. The host cell may be either bacterial, yeast or mammalian.
3.17. The properties of the final protein product depend on the kind of cell which is used for its expression as this affects the special features of its amino acid sequence and molecular structure. Thus, in a mammalian cell, t-PA is extruded through the membrane, “trimmed” of superfluous amino acids and also glycosylated; that is, chains of sugar molecules are added at certain positions of the polypeptide chain. It is also folded and internally cross-linked to give a defined three-dimensional structure. The important fibrinolytic properties shown by t-PA depend upon it achieving its correct three-dimensional structure. When expressed in a bacterial cell t-PA protein is not glycosylated nor does it fold properly. It therefore lacks these special properties. Cells of any kind will not necessarily be receptive to the vector with its foreign cDNA nor will they necessarily process the protein correctly. The selection and construction of the expression system is a skilled and specialised operation. For t-PA it was found that one appropriate system to produce the correctly folded and glycosylated protein was a mammalian cell line, namely Chinese hamster ovary cells (“CHO”).
4. The synthesis of t-PA by Genentech
4.01. References in this section to “Stage 1”, etc, refer to the “Genentech Strategy Diagram” at paragraph 4.17 (infra (p 174)). Recombinant DNA technology had been successfully employed in synthesising a number of substances before May 1982 — eg human insulin, alpha and gamma interferon and human growth hormone. However, no protein having both an mRNA abundance as low as t-PA together with as long and as complicated a molecule as t-PA had ever been synthesised before Genentech succeeded in early 1982.
4.02. This briefest of outlines of the steps involved must do less than justice to the uncertainties of the procedures used and to the highly skilled, meticulous and detailed operations required, the time for which is measured in months and years rather than in days and weeks. It is also important to remember that in no case did one individual scientist carry out the entire operation. The work calls for the coordination of a team of skilled workers in each of the specialities concerned. Thus in the case of Genentech the natural t-PA was extracted from the Bowes melanoma cells by Dr Collen at the Centre for Thrombosis and Vascular Research of the University of Leuven in Belgium. He acted as consultant to Genentech and provided samples of purified t-PA. Dr Goeddel is an extremely experienced and highly qualified biochemist and a specialist in recombinant DNA procedures. He was in charge of a team of scientists including Drs Pennica and Holmes who specialised in the manipulation of mRNA and the production of cDNA libraries. In addition, the team was assisted by two chemists, William Kohr, who was a protein sequencer, and Gordon Vehar, who was a protein chemist. These scientists worked under the general direction of Dr Goeddel and the product of their efforts were brought together in the Genentech laboratory under Dr Goeddel’s supervision.
4.03. A purified sample of t-PA weighing 2 mg was provided by Dr Collen on 3 October 1980 together with a t-PA antiserum. Later in July 1981 this was augmented by a further amount of 8 mg of t-PA. Preliminary investigations on a part of the sample were carried out by Dr Pennica in order to confirm the presence of t-PA DNA (Stage 1). In March 1981 a further sample was sent to Dr Shiveley of the City of Hope Research Institute in California, who was a leader in the field of protein sequencing, and who was asked to sequence the N-terminus of the t-PA. Dr Shiveley wrongly concluded that the N-terminal amino acid was blocked and could not be sequenced and reported to this effect.
4.04. Between October 1980 and March 1981 Dr Pennica had been attempting unsuccessfully by various methods to confirm the presence of mRNA coding for t-PA in the RNA sample she had obtained from the t-PA received from Dr Collen. Dr Holmes was preparing total cDNA clone libraries from the same sample. In March 1981 the failure to identify the target mRNA was realised to be due to the extremely low abundance of mRNA for t-PA. Drs Goeddel and Pennica then decided to take steps to increase the concentration of the target mRNA (Stages 6-8).
4.05. The proportions of mRNA in the total RNA sample was first enriched using oligo-dT columns (Stage 7). This involved exploiting a characteristic of mRNA known as a poly(A)tail, which is a string of A-based nucleotides at the 3′ end. The columns consist of an insoluble resin with attached nucleotide chains of T-bases. These bind the poly(A)tails by base pairing and remove the mRNA from the remaining bulk of RNAs. The mRNAs are then detached from the resins by mild chemical means to give a sample enriched in mRNAs. This sample was then tested for the presence of mRNA coding for t-PA by injecting the mRNA into frog (Xenopus laevis) oocyte cells to test for expression of t-PA. The isolated mRNAs were further enriched by size fractionation using urea-agarose electrophoresis (Stage 8). Two slices were then separately tested for t-PAmRNA by translation with and without dog pancreas microsomes followed by immunoprecipitation with t-PA antiserum. In the event one of the two slices proved positive and this was used by Dr Holmes to produce a new cDNA library consisting of 4600 clones. All of these steps were difficult and, because mRNA is chemically labile, losses of mRNA were inevitable. Moreover, the longer mRNAs, such as that for t-PA, are most susceptible to chemical scission and are differentially lost. t-PAmRNA was larger than the low-abundance mRNAs that had been previously cloned.
4.06. Dr Holmes created the new cDNA library by copying the enriched mRNA into DNA using reverse transcriptase and an oligonucleotide composed of T’s to prime the synthesis from the poly(A)tails (Stages 9 and 11). The DNA was made double-stranded by copying it with DNA polymerase using random priming and it was then inserted into a cloning vector. The library was propagated by introduction of the vector into E coli (Stages 10 to 12).
4.07. After Dr Shiveley’s report in April 1981, Dr Goeddel halted attempts to sequence the N-terminal region and directed attention towards the C-terminal of the A-chain. Several months were spent in considering various methods to be adopted in sequencing the t-PA polypeptides. All involved the cleavage and purification of the polypeptides (Stages 1-2). After very considerable pursuasion, and by proving the method by reference to his work in connection with human growth hormone, it was agreed in July 1981 that William Kohr should use a process which involved digestion with an enzyme called trypsin followed by high-pressure liquid chromatography (HPLC) fractionisation to separate the sample into individual peptide fractions (Stage 3). These fractions were subsequently inserted individually into a device known as a Beckman spinning cup sequenator. This performed the sequential degradations by removing the reagents and their products by centrifrugal force. Each separated amino acid was reprocessed in the HPDC column and, by comparing the retention time of the peak produced on the trace with a calibration chart produced from testing known amino acids, the amino acid analysis could be identified (Stages 4 and 5).
4.08. Once some amino acid sequences had been determined it was possible to define and then synthesise lengths of nucleotides which coded for short sequences of amino acids (Stage 5). These oligonucleotide probes were then introduced to the clone library in order to bind to any cDNA with a complementary sequence (Stage 12). It was, therefore, important in selecting the short nucleotide probes to use amino acids with the minimum number of equivalent codons so as to reduce the number of probes needed.
4.09. The colonies showing most radioactivity above background, ie those which were most likely to contain the target DNA, were then checked. Even with a successful identification of a clone it was unlikely that a full-length cDNA would be obtained. However, the clones obtained would be of considerably greater length than the probe by which they had been detected and could then be used as long unique probes to go back either to the same or to a different library to probe for longer cDNA clones or extra fragments to allow the construction of a complete cDNA molecule. This procedure was well recognised before 1982.
4.10. About the end of August 1981, William Kohr identified a favourable amino acid sequence WEYCD. Dr Goeddel ordered eight 14-mer probes in two pools of four to cover all possible codons for WEYCD. There were produced in the Genentech laboratory. Using the synthesised pool of probes the new cDNA library prepared by Dr Holmes was probed, and as a result 96 potential colonies were identified. However, none proved to contain a target clone. Dr Goeddel ordered re-probing of the libraries. Four candidates were selected for further investigation and the DNA from each was digested with restriction enzymes to prepare suitable fragments for nucleotide sequencing. The clone showing the most positive features was 25-E-10. On 20 October Dr Seeburg reported that the restriction fragments of the 25-E-10 clone did not correlate with any of the t-PA peptide sequences found by William Kohr. This was a setback. But on rechecking the data Dr Pennica recognised that an amino acid sequence, EEEQKF, in the computer printout correlated with information which had been given by William Kohr during preceding weeks. This had not been incorporated into Dr Seeburg’s computer. Out of this oversight eventually emerged the one sequence which almost certainly formed part of a clone encoding for t-PA (Stage 13). From this Dr Pennica and Dr Goeddel were able to get a foot on the ladder from which they could progress in stages to larger sequences of cDNA and finally the synthesis of the full cDNA clone for t-PA.
4.11. Dr Goeddel directed that the cDNA library of 4600 clones be rescreened employing a large restriction fragment obtained from 25-E-10 as a probe (Stages 16-17). This rescreening identified only one further putative t-PA clone (25-B-2) which proved shorter than 25-E-10. Both clones lacked the 5′ sequences corresponding to the N-terminus of t-PA and the translation start signal, signal sequence, etc. The incomplete 25-E-10 clone was included in an expression vector, a plasmid p$99RIPA. Extracts were prepared from E coli transformed with p$99RIPA and tested for t-PA activity in a fibrin agar plate assay. On 8 February 1982, the DNA was shown to direct the synthesis of fibrinolytic activity which could be neutralised by antibodies to t-PA. This provided unequivocal confirmation that cDNA clone 25-E-10 was a substantial fragment of the target clone for t-PA.
4.12. Efforts were then made to isolate the missing N-terminal region by a technique known as primer extension. This involved hybridising a DNA fragment representing a known sequence to mRNA under conditions in which the primer sequence could be enzymatically extended by reverse transcription along the mRNA template. This was an intricate technique which, although reported in previous literature, had been only rarely used.
4.13. Dr Goeddel, wrongly as it later turned out, thought that this attempt to reach the 5′ region was frustrated by a loop in the RNA which was inhibiting the progress of reverse transcription. Accordingly, he adjusted the position of the primers relative to the site of the suspected loop thus getting closer to the N-terminal. After a great deal of work, described by Dr Pennica as taking over four months, a clone was eventually discovered which covered the N-terminal region. The work involved is described in paragraphs E1H (pages 34 to 37) of the patent in suit, and need not be repeated in detail here (Stages 18-21). It involved preparing a labelled DNA probe prepared from the 5′ end of the cDNA insert pPA25-E-10, which was a plasmid including the identified clone 25-E-10. This was then used to probe a human genome library as a result of which one clone was identified — pPA17, which was shown to contain the correct 5′ N-terminal region of t-PA, a signal leader sequence and an 84 base pair 5′ untranslated region.
4.14. From the two clones pPA25-E-10 and pPA17 the complete nucleotide sequence and restriction pattern of full-length t-PA clone were determined and are described in figures 5A to 5C (p 172) together with the deduced amino acid sequence. Figure 5 (infra) shows a string of untransferred bases leading to a pre-N-terminal sequence of codons leading to the start codon ATG at -35, the N-terminal at 1. The mature DNA sequence ends at the stop codon TGA at 528: [for table see original]
Two bases before the start codon, figure 5 as incorporated in the specification contains, in error, an additional G base interposed between AT and C and shown above in brackets. This demonstrated a fake start codon AT(G) which would have dictated the reading frame for the encoding process and negated the invention — it was an error which would have been obvious to a person skilled in the art. Mr Jacob argued that this error invalidated the patent ab initio, but I do not think it would do an injustice to him to say that he pressed this point without any great enthusiasm. Figure 4 of the specification showed the restrictions pattern for the t-PA gene:- [for table see original]
4.15. Plasmids pPA17 and pPA25-E-10 each contained an “overlap” restriction endonuclease site (HhaI). Making use of this and a third plasmid pLeIFAtrp103, a plasmid pt-PAtrp12 incorporating a full-length t-PA gene was prepared. This process is described in full detail in paragraph E1I of the specification and demonstrated in figure 9. It is of interest to note the relative positions of the DNA sequences included in the two plasmids in the whole sequence of the gene for t-PA:- [for diagram see original]
4.16. E coli transformed with this plasmid expressed full length t-PA. As I have already mentioned in this judgment, t-PA expressed in E coli was not glycosylated nor correctly processed and did not have the necessary activator characteristics. Further vectors containing the full-length cDNA were constructed for expression in mammalian cells in order to achieve the three-dimensional structure and glycosylation of the protein molecule which was thought to be necessary for the specific fibrinolytic properties of t-PA.
4.17. This short summary of the work involved, which extended over more than twelve months from the time that the Bowes melanoma cell t-PA was first provided by Dr Collen, shows that many man months of meticulous and skilful work by numerous experts in their own field had to be expended before the objective of determining the full sequence for t-PA gene and the deduced amino acid sequence for t-PA could be ascertained. The following diagram demonstrates the co-ordinated activities involved in the final solution of the problem:- [for diagram see original]
4.18. Mr Gratwick referred to the discovery of the full sequence as “a break-through” which effectively dislodged the logjam which hitherto had been preventing the synthesis of t-PA by recombinant DNA technology. This, he submitted, was the discovery upon which the invention depended which was properly protected by the patent in suit. Mr Jacob, on the other hand, by reference to the progress made by the competing teams, to which reference has already been made, asserted that there was no “logjam” and that the discovery of the precise information contained in the sequence was either something which was expected to be discovered in any event in the ordinary process of development or, alternatively, was not necessary to synthesise t-PA by recombinant DNA technology at all. This conflict which lies at the heart of this appeal leads me to consider shortly the progress and operations of the other teams upon which Mr Jacob based much of his case.
5. Umea University
5.01. Dr Per Wallen and others had been working with t-PA since the early 1970s in the Department of Physiological Chemistry at Umea. They had purified the protein to homogeneity from human uterine tissues and developed antibodies to t-PA. In August 1980 the department obtained from a gynaecologist at the University of Lund, Sweden a sample of the human melanoma cell (Bowes) which secreted a plasminogen activator. At this time researchers at Leuven had disclosed that this plasminogen activator was very similar, or identical, to t-PA. Dr Wallen, together with Dr Mats Ranby, a student in Dr Wallen’s group, had developed a particularly sensitive assay for t-PA. It was against this background that Dr Edlund and Dr Tor Ny embarked upon their research during 1980.
5.02. Quoting from the affidavit sworn by Dr Ny:-
“9. The first part of our stategy involved making a cDNA library which would contain one or more fragments of the t-PAcDNA. We extracted cytoplasmic RNA from Bowes melanoma cells and purified this in two stages. First we isolated polyadenylated mRNA by means of the oligo(dT) cellulose column. This technique separates the plyadenylated mRNA from the other types of RNA occurring in the cell. Secondly we used the technique of sucrose gradient centrifugation to size fractionate the mRNA and we then attempted to determine which fraction contained the t-PAmRNA by using an in vitro translation system.”
Drs Edlund and Ny, therefore, travelled the same path as Genentech through Stages 6, 7, 8, 9.
5.03. For two months Drs Edmund and Ny attempted translation in vitro using a rabbit reticuloocyte lysate system without success. After this they switched to an in vivo system and started to work with the frog Xenopus laevis oocyte system. This involved considerable practical skill in micro-injecting mRNA into the oocytes. In order to do this Drs Edlund and Ny had obtained tuition in the practical skills involved from one Lennart Nelsson, a student at the zoological physiology department at Umea University. They also had assistance from an American professor who was visiting the Karolinska Institute in Stockholm and who had a particular experience in this field.
5.04. t-PA activity in the translation products from the oocyte cells was quickly observed but further complications arose because the cells themselves produced a plasminogen activator of their own. This required inactivation by using a specially developed technique. By the latter part of December 1980 Drs Edlund and Ny were able to identify the mRNA fraction which contained the t-PAmRNA. With this they proceeded to prepare a cDNA library containing double-stranded cDNA from the mRNA. They then size-fractionated the cDNA again using a sucrose gradient centifugation so as to isolate cDNA molecules longer than about 400 base pairs. These were inserted into the PstI site of the plasmid pBR322 using poly DC/poly(dG) linkers. The resulting recombinant plasmids were then used to transform E coli strain resulting in a cDNA library of 4200 recombinant clones. This library was completed by the end of January 1981. They had, therefore, carried out stages 10 and 11.
5.05. Not having any amino acid sequence information Drs Edlund and Ny started screening their library using a hybrid selection technique. This involved preparing plasmid DNA from each recombinant clone in the library bound to a nitrocellulose filter and treating it with an mRNA preparation similar to that used as a basis for making the cDNA library. mRNA in the preparation corresponding to the cDNA insert in the plasmid should then hybridise to the bound plasmid DNA and following washing to remove non-bound mRNA the bound mRNA can be eluted. The eluted mRNA can then be investigated to detect mRNA coding for the protein of interest. This method was very tedious and time-consuming with the result that Drs Edlund and Ny managed only to screen about 500 clones from their library. They did not succeed in identifying a clone containing cDNA for t-PA. However, in about March 1981 some amino acids sequence data from the N-terminus of the B chain of t-PA became available from Dr Wallen. Oligonucleotide probes were synthesised for Drs Edlund and Ny by a firm, Kabigen AB, in Stockholm. This was not a novel process, although the techniques were relatively specialised and took some time to carry out.
5.06. Using two pools of oligonucleotides, which partially overlapped in terms of the encoded amino acid sequences, and a considerable amount of further experimentation and work in order to establish the appropriate hybridisation conditions, Drs Edlund and Ny identified a plasmid containing the t-PAcDNA fragment which was designated pPA1. Describing this work (see Evidence Day 4, page 46) Dr Ny said that he and Dr Edlund relied upon papers published by the Bruce Wallace Laboratory and discovered the optimum stringency conditions more or less by trial and error:-
“We increased our temperature in hybridisation and in washing until we got fewer (clones) that gave signal. We also had negative controls.”
In cross-examination Dr Ny described “negative controls” as E coli containing cloning vector pBR322 or an actin clone which was known not to contain a plasmid incorporating the target cDNA. As long as these gave “false positive” results, it indicated that the hybridisation conditions were not correct. Mr Gratwick relied upon this work as being inventive in character; but Dr Ny, upon whom Mr Jacob relied, clearly had a more modest opinion of his work.
5.07. The plasmid pPA1 was digested with PstI. It was shown that it carried a cDNA insert with 370 base pairs. This fragment was subcloned and part of the nucleotide sequence determined in the laboratory at Umea. In April 1982 it was established that the fragment contained a nucleotide sequence corresponding to amino acids 1-19 at the N-terminus of the B chain of t-PA which had been determined by Dr Wallen. The sequencing of the whole insert was carried out for Drs Edlund and Ny at Kabigen. Thus, they had continued to carry out Stages 12 and 13 although in relation to a different part of the t-PA-DNA gene and in this way arrived at their clone PA1: [for diagram see original]
Drs Edlund and Ny never took the further step of identifying the complete t-PA coding sequence. The former had moved to the United States of America whilst Dr Ny became interested in cloning the t-PA from the human genome. Their work was eventually published in January 1983.
5.08. Mr Jacob submitted that, had they continued, there is no doubt that Drs Edlund and Ny would have succeeded in obtaining a full-length cDNA clone. Mr Gratwick submitted that certain aspects of the work carried out at Umea were beyond the man ordinarily skilled in the art. He submitted, first, that such a man would not have access to the specialised knowledge of Dr Wallen, nor would he have had the inventive skill to overcome the difficulties in the oocyte translation. He would not have had the amino acid sequence information which was needed to construct the oligonucleotide probes, which was not published until much later. Mr Gratwick also submitted that using the hybridisation technique in the presence of the exceptionally low abundance of t-PAmRNA he would not have succeeded in isolating a clone at all unless he had adopted the approach by Dr Ny including the use of negative controls. Mr Gratwick also pointed to the specialised assay procedure devised by Dr Ranby and submitted that this would have involved inventive techniques and capacities not available to a person skilled in the art. Mr Jacob, on the other hand, submitted that the sequencing information produced by Dr Wallen had in fact been obtained by a graduate student called Gunnar Pohl working at the Karolinska Institute using normal sequencing equipment (spinning cup) and that this was routine work. He submitted that there was no relevance in the extra sensitivity of the Ranby assay since it merely afforded a saving of time and effort required when compared with the normal assay techniques. However, it was accepted that in conditions of extremely low abundance the Ranby assay technique might detect a clone which other methods might miss. In general Mr Jacob submitted that there was nothing in the work carried out at Umea University which passed from the arena of skilled and meticulous work into the arena of inventiveness and in this he included the use of negative controls.
6.01. GI is a company organised to develop the products of genetic engineering commercially. One of its founders was Professor Maniatis who gave evidence before Whitford J at first instance. The CSH Laboratory was set up with its primary functions to provide scientists with an environment where they could carry out their own research for which they had been able to obtain outside funding. As its Assistant Director, Dr Sambrook was charged with general overall supervision of the various scientists and their projects and considered himself similar to a chairman of an academic department rather than the research director of a commercial organisation. Dr Sambrook was involved in organising a course on the molecular cloning of eukaryotic and viral genes which was held in July 1980. This course included Professor Maniatis amongst the teachers. The notes for the various courses prepared by the course teachers were expanded and updated in July 1981. These notes formed the basis of a book subsequently published which was in evidence as indicating the state of the art. This was called “Molecular Cloning: A Laboratory Manual”. It was in fact published in 1982 shortly after the priority date but clearly the contents of the book which must have been in draft before the priority date, are evidence of the state of the art at the relevant time.
6.02. In late June or early July 1980 Dr Sambrook learned for the first time about t-PA and became aware of its commercial potential. Most of the scientists at CSH were working on their own projects but Dr Sambrook was able to start the t-PA project with a Dr Anderson who was visiting from Australia. The early stages of this work did not prove successful for a number of reasons. Amongst these was the lack of facilities for obtaining amino acid sequence information. This was mainly due to lack of availability of the sequencing facilities and persons to use them and a shortage of finance necessary to obtain outside help. Dr Sambrook hoped that the abundancy of the RNA would be sufficient to allow the use of hybrid selection procedures without oligonucleotide screening being necessary. This, of course, proved to be incorrect.
6.03. In November 1980 CSH incoporated a company known as Cell Biology Corporation (“CB”) as a commercial body to exploit discoveries and to attract funding for further research. Early in 1981 CB established contact with a company, Baxter Travenol (“BT”), as a result of which BT became interested in the t-PA project and eventually provided funds for the work to be continued by a joint team formed by an amalgamation of CSH and GI.
6.04. In the spring of 1981 t-PA and its antiserum were obtained by CSH. During the summer of 1981 Dr Sambrook, who was on an extended visit to London, and Dr Gething, who was on the staff there, had started working on t-PA at the Imperial Cancer Research Fund laboratories. As a result of this work Dr Sambrook came to the conclusion that the level of t-PAmRNA in Bowes melanoma cells might only be of the order of 0.01% of the total cellular mRNA, ie 50-fold lower than he had previously thought. Dr Gething started work on the construction of a cDNA library using the mRNA from the Bowes melanoma cells, and achieved a library of about 1000 clones. Late in 1981 Dr Sambrook decided to use oligonucleotide screening in order to meet his commitments to BT. There were no facilities for carrying out the amono acid sequencing at CSH at this time for use on the t-PA project. In 1982 Dr Sambrook approached Professor Maniatis who was at GI. During the early part of 1982 there were further negotiations which led to an agreement under which there would be co-operation between CSH and GI on an equal sharing basis. In April 1982 an advanced protein sequenator, which had been invented by Dr Hewick who had also joined GI, became available. This was known as a gas-phase sequenator and was not available elsewhere.
6.05. From late 1981 to early 1982 Dr Kay, who was a senior scientist at GI, was attempting to obtain support for work on t-PA. He presented a number of proposals at this time relating to the t-PA project. In a draft proposal prepared by Dr Kay for presentation to BT the project was described in the following terms: “to use the techniques of recombinant DNA to construct a strain of E coli that expresses large quantities of plasminogen activator . . .” The details of the scope of the project include all the steps leading to the preparation of cDNA libraries with alternative methods of screening these libraries including in the event of scarcity of the target clone the preparation of oligonucleotide probes. In March to April 1982 Dr Kay extracted mRNA from Bowes melanoma cells and examined it applying sucrose gradients and oocyte translation. The amino acid sequencing at GI was undertaken by Dr Hewick who had joined them in July 1981 as a senior scientist with particular responsibility for protein sequencing. On 1st March 1982 he received two samples from Professor Rifkin of New York University Medical School which he understood to be purified t-PA from Bowes melanoma cells. The total protein mixture was analysed by him using gel electrophoresis with a fibrin plate overlay. This technique separates the different proteins which migrate to a different extent on the gel and then the reaction with the fibrin plate overlay affords a sensitive means of detection of the plasminogen activator activity. The results of this analysis was that the samples were impure and considered unsuitable.
6.06. Accordingly Dr Larsen of GI was, in May 1982, sent to work in Professor Rifkin’s laboratory in order to obtain a purer sample of t-PA. This became available to Dr Hewick on 2 June 1982. By this time Dr Hewick had his gas-phase sequenator available and by 8 June had obtained sequence data covering the first 24 amino acid positions starting from the serine N-terminus of the protein, the sequence of residues 2-18 being without gap. This sequence contained a region of low degeneracy which allowed an oligonucleotide probe to be designed (QMIYQQ):- [see original]
6.07. Further work between August and November 1982 enabled GI to identify two further overlapping segments, ED1 and ED2, with which they were able to bracket the whole of the mRNA sequence coding for the full t-PA amino acid sequence. In December 1982 a complete cDNA was assembled in vectors for bacteria and yeast expression and on 13 January 1983 Dr Larsen observed the first expression of t-PA in E coli. The history of this achievement was as follows. Dr Larsen at GI had had two libraries available to him, namely, the 1000 clone library which had been brought by Dr Gething from London and a library of 8000 which had been prepared by a Dr Hanahan in America. Dr Larsen had screened both libraries and on 10 August 1982 he had identified one clone as a putative positive. The cDNA insert of this clone was sequenced and it was established on 18 August that it contained an 850 base pair fragment of the t-PAcDNA. It was found to contain the sequence corresponding to the 5′ end of the coding sequence of the t-PAmRNA and the 5′ untranslated region. This fragment was designated Sam 1. In February 1983 expression in COS cells was obtained which was followed in the ensuing month by expression in CHO cells. The combined achievements of GI/CSH are shown in the following figure:- [see original]
6.08. Mr Gratwick referred to the reports of meetings during July 1982 at CSH/GI. On 12 July the sequencing of 20 continuous amino acids from the N-terminus of mature t-PA by Dr Hewick was reported together with the production of specific oligonucleotides, 24 in number. There were long discussions concerning the manner in which the oligonucleotide pools would be screened to identify the one that contained the exact sequence of the t-PA gene in relation to the complication caused by the G+C content of different oligonucleotide ranges from 24% to 47%. It was appreciated quite clearly that the G+C content was a critical factor in the problem of suppressing “background” by choosing the level of stringency in hybridisation. Mr Gratwick submitted that these considerations were of an inventive nature and not within the general knowledge of a person skilled in the art. On 23 July 1982 the meeting in Switzerland, at which Dr Pennica disclosed some of the details of the work at Genentech, took place. It is to be noted that at the same meeting Dr Wallen himself gave details of a 20 amino acid sequence at the N-terminus of the single chain form of t-PA. These matters, together with Dr Pennica’s disclosures, formed the subject matter of prolonged discussions at GI on 30 July 1982. It was recorded in the minutes that Dr Hewick’s sequence from the N-terminus of the A-chain agreed with that presented by Genentech but there must be doubt about the veracity of this note and it is more probable that reference was in fact being made to the data disclosed by Dr Wallen.
6.09. On 27 August 1982 Dr Kay reported the achievement by GI of their first clone. It was a fragment 800 bases long. It was unknown at that time how far towards or beyond the protein start, that is the N-terminus, the clone extended. Dr Kay gave an estimate of the time required to produce a full-length clone under favourable conditions as one to two months; but if conditions were unfavourable then a much longer and harder path was anticipated.
6.10. the GI team, who were carrying out the joint work of CSH/GT, consisted of a Dr Brown who made the oligonucleotides, Dr Fritsch who made the clone libraries, Dr Hewick the protein sequencer, Dr Kaufman who did the work on expression of t-PA in mammalian cells, Dr Larsen who worked on cloning of t-PA, all of whom worked under the project leader Dr Kay, besides Professor Maniatis who was also available. Mr Jacob pointed to this team operating on the ground with every expectation of success, an expectation which was in the event justified without any substantial assistance from the information imparted by Genentech, as evidence in support of his case on obviousness and/or lack of inventiveness.
6.11. Mr Jacob relied in particular on the attitude of Dr Kay at GI at the outset of his work as disclosed in paragraph 14 of his affidavit:-
“Based on the level of t-PA activity secreted by the cells I thought that the t-PAmRNA would probably represent somewhere between 0.01% and 0.1% of total mRNA in Bowes melanoma cells but I recognise that this was little more than a guess.”
In this context Dr Kay was of the opinion that the successful completion of the exercise, namely the production of t-PA by recombinant DNA technology, was well within the capacity of GI. Dr Larsen observed first expression of the t-PA cloning sequence on 13 January 1983 using the construct of the molecule from the available fragments. This was, therefore, before GI had the full knowledge of the sequence data. In December 1982 Dr Kaufman, who was an expert in the mammalian expression, received the full sequence from Dr Rogers who had been assembling the groups of fragments of cDNA which he had obtained. He transferred the cDNA to a vector suitable for mammalian cells. The t-PA expression plasmid which was constructed was based on a vector very similar to the expression vector that Dr Kay had helped to develop during 1980 and 1981. In January 1983 Dr Kaufman transformed COS cells with the plasmid pLDSG but it was as long as January 1984 before a satisfactory cell line was eventually established.
6.12. Mr Jacob’s submission was that where there are a number of persons striving in the field towards the same objective and one of them succeeds and the others are just about to succeed, then it is not appropriate to grant a patent to the person who takes less time to succeed than his rivals all of whom have the knowledge, skill and art to achieve the final objective in varying times.
7.01. Dr Opdenakker at the Rega Institute extracted mRNA from Bowes cells and, by applying sucrose gradient fractionation and expression in frog oocytes, identified the presence of t-PA between June 1980 and January 1981 (Stages 6-8 followed by expression).
7.02. Dr Volckaert, who later held a fellowship at the University of Leuven, when applying for a post at the University in February 1981 made a proposal for:-
“Molecular cloning of proteins of clinical potential.
a) Cloning, characterisation and expression of a tissue plasminogen-activator (pa) gene in bacteria.
b) Construction of new vectors for use in cDNA Cloning.”
and in support stated:-
“In principle each cellular protein, whose mRNA can be demonstrated in some cell types of cell line, may now be produced by molecular cloning.”
7.03. Between January and June 1982 Dr Opdenakker was working under Dr Volckaert at Leuven using the induction of t-PAmRNA by phorbol esters. In this way they produced a cDNA library of 10000 clones. In the first half of 1982, Dr Volckaert realised that oligonucleotide screening was the method most likely to succeed, but no protein sequence information was available. In about the middle of 1982 tryptic peptide sequence information was made available by Professor Rombauts of Leuven’s Biochemistry Department. This data included the sequence WEYCD:- [see original]
Although Dr Cullen at Leuven was working for Genentech, the identification by both Genentech and Leuven Rega Institute of the same amino acid sequence for the preparation of oligonucleotide probes is not said to be other than coincidental. The sequence is conveniently placed for peptide fragmentation and has a low redundancy.
7.04. At Leuven’s request and using the amino acid information from Professor Rombauts, Celltech made the appropriate oligonucleotide pools. Dr Opdenakker started screening his library on 22 December 1982 using individually eight synthesised 14-mer probes. On 15 December 1982 Dr Harris, having produced probes, gave them to the Rega Institute, receiving Bowes cells in return. Using these probes on 22 December 1982 Dr Opdenakker screened for t-PA. In early January 1983 Dr Opdenakker had identified two clones, which were designated pL1 and pL2, to which three of the probes hybridised strongly. The two fragments were between 500 and 1,000 base pairs long. After receiving a copy of Genentech’s paper in Nature, which was published on 20 January 1983, Dr Opdenakker was able to confirm that one of the three probes which hybridised strongly to pL1 and pL2 corresponded with the published sequence WEYCD. It is true to record that at the time Dr Volckaert did not believe the pL1 was a t-PA clone, and did not trouble to sequence it. Mr Gratwick submitted that had not the information been received from Genentech’s paper in Nature, no one can tell what would have happened at Leuven. With regard to the work done by Professor Rombauts in arriving at the WEYCD sequence, Mr Gratwick pointed out that there was no evidence that the ordinary man skilled in the art could have done this.
7.05. In November 1982 at Celltech, Dr Harris was concerned with the use of automated equipment for synthesising the oligonucleotide probes based on the amino acid sequences from t-PA which had been given to Leuven by Dr Rombauts. On receiving from Leuven a sample of Bowes melanoma cells in December 1982 he used this to extract total RNA from which he isolated a total mRNA. From this he prepared cDNA by reverse transcription and produced a cDNA library of over 40000 clones from unenriched RNA and cDNA (Stages 6-9). Dr Harris intended to use pools of oligonucleotide probes but his plan was overtaken by the publication by Genentech of their paper. Dr Harris, therefore, used a single oligonucleotide having the correct sequence.
7.06. By means of this he identified sequence PTPA-A3. In his evidence Dr Harris testified to the fact that had he continued he would have expected to arrive at the full amino acid sequence for t-PA. The position reached by Dr Harris’ subsequent work at Celltech/Leuven finally produced full coverage of the whole sequence but in considering the work done by Dr Harris attention should probably be restricted to the line PA-A3:- [for diagram see original]
8. State of the art
8.01. The state of the art is taken to comprise all matter (whether a product, a process, information about either, or anything else) which has been made available to the public in any way (see section 2(2) of the 1977 Act (infra)). For present purposes much of the work described in the preceding paragraphs of this judgment was not made available to the public; but some of it was. In particular an article was published in Nature over the signature of Opdenakker and Collen and others at the Rega Institute at Leuven in the February edition of 1982 which had been received for publication on 22 June/23 September 1981. It was entitled “Messenger RNA for Human Tissue Plasminogen Activator”. It referred to a publication over the signature of Rijken and Collen in the July issue of Journal of Biological Chemistry 1981 entitled “Purification and Characterization of the Plasminogen Activator Secreted by Human Melanoma Cells in Culture.” This had been received for publication on 14 October 1980 and revised on 25 February 1981.
8.02. The earlier publication demonstrated the similarity between plasminogen activator derived from melanoma cells and the extrinsic plasminogen activator derived from human uterine tissue. In particular both enjoyed the advantage over urokinase already mentioned in this judgment. The melanoma cells had the further advantage of being capable of culture. These were being produced by Dr Rifkin at the Rockefeller University, New York. The article refers to the work carried out by Dr Collen in growing the cells to confluent monolayers in plastic tissue culture flasks. This was in fact work being carried out by Dr Collen for Genentech but was part of the article published describing the work at Leuven. It is necessary only to refer to the closing discussion where mention is made of the possibility of obtaining increased sources.:-
“Since plasminogen activator most likely represents the same enzyme, tissue extracts can be used as an alternative source. Although this offers the possibility of obtaining larger quantities of purified activator, still about 100 human uteri were necessary for 1 mg of purified protein. Out finding that the melanoma plasminogen activator is identical, or nearly identical, with the uterine tissue plasminogen activator makes it possible to produce substantial amounts of activator independent of human donors. This has allowed us to initiate studies on the molecular interactions which regulate fibrinolysis and on the thrombolytic properties of tissues plasminogen activator in animal models.”
8.03. There is, however, no mention at this stage of adopting the process of recombinant DNA technology although this had previously been used in the synthesis of interferon, human growth hormone and insulin. However, another publication with which Dr Collen and a Professor Billiau were concerned in the Lancet for November 1981 moved closer to the suggestion of recombinant DNA technology. This article described the treatment of a patient using what was called human extrinsic plasminogen activator (HEPA). In the discussion reference was made to another severe limitation on the use of extrinsic human tissue plasminogen activator:
“A severe limitation on this type of therapeutic experiment is that sufficient amounts of HEPA can be obtained only from a cell line originating from a malignant melanoma. Despite the apparent effectiveness and selectivity, and the absence of toxic side-effects therapeutic experiments must remain confined to patients in whom thromboembolism is life-threatening and for whom other forms of therapy have met with failure.
One possibility to obviate this objection would be to obtain HEPA from non-malignant cell cultures — eg, primary cultures of human vascular endothelia. The prospects of achieving mass production from such cultures are, however, small. Another possibility would be biosynthesis by bacteria through incorporation of the HEPA gene in an expression plasmid.”
8.04. Returning to Dr Opdenakker’s publication, although it did not refer to the article in the Lancet, it developed the programme suggested in the latter article to a considerable extent (Stages 6-11):-
“A human melanoma cell line (Bowes), which secretes extrinsic (tissue-type) plasminogen activator, was used as a source for the preparation of mRNA for extrinsic plasminogen activator. The cells were lysed and total RNA extracted with the phenol method. A poly(A)-rich RNA fraction was isolated by affinity chromatography on oligo(dT)-cellulose. This preparation was translated by oocytes in protein(s) that had biological activity in the assay for extrinsic plasminogen activator. On sucrose gradient centrifugation the translatable fraction of the RNA migrated with a sedimentation coefficient of approximately 19 S, a value compatible with the molecular weight of extrinsic plasminogen activator (70000). The translation product was characterised as being similar to or identical with authentic extrinsic plasminogen activator by the following criteria: (a) serological cross-reactivity with purified extrinsic plasminogen activator in neutralisation and immunoprecipitation reactions; (b) plasminogen dependency of fibrinolytic activity and (c) apparent molecular weight of 70000 on sodium dodecyl sulphate/polyacrylamide gel electrophoresis.”
In discussing the future programme in relation to the extrinsic plasminogen activator obtained the article says:-
“One possibility for achieving large-scale production at relatively low cost is the insertion of the extrinsic plasminogen activator gene into an expression plasmid of Escherichia coli (E coli). As a first step in this direction we have isolated the messenger RNA for extrinsic plasminogen activator in an active form from Bowes melanoma cells.”
The ultimate point of the work described in this article (although, of course, subsequently those at Leuven went considerably further) is as follows:-
“The purpose of our study was to isolate and purify the RNA for extrinsic plasminogen activator as a starting material and probe for cloning the extrinsic plasminogen activator gene into a prokaryotic vector. Our results indicate that the classic procedures of phenol extraction, affinity chromatography on oligo (dT)-cellulose and sucrose gradient centrifugation are suitable to obtain active mRNA for extrinsic plasminogen activator. However, the potency of individual preparations was highly variable and only the best preparations were quantitatively adequate for DNA-recombinant research. One can imagine possible explanations for this variability, eg degradation of mRNA during and after extraction or variations in cellular mRNA content depending on the physiological state of the melanoma cells at the time of extraction.”
8.05. The publication, therefore, stops at the point of preparing total mRNA and does not achieve the preparation of a cDNA clone library. It does, however, point very firmly in that direction. these were the only publications to which the court was referred on this aspect of the case. It is, however, not possible to part with this topic without mentioning a feature of the description contained in the patent in suit, namely the repeated reference at very many stages of the text to pre-existing publications. There was no analysis of these numerous publications which are to be found starting on page 53 under the heading “Bibliography” and it would, therefore, not be appropriate in this judgment to do more than comment that the patent itself acknowledges at very many stages the existence of the particular experiment involved to knowledge already existing in the state of the art.
8.06. Before trial, and in accordance with an order of the court, experts for each party agreed certain statements relating to the state of the art. These statements are of critical importance and the complete document must he rehearsed here:-
“Points of agreement between W Brammer and G Stark 5 August 1986*
1. The production of quantities of tissue plasminogen activator (t-PA) adequate to treat patients suffering from blood clots was known to be a desirable objective in early May 1982.
2. The Bowes melanoma cell line was available in early May 1982 and was known to be a source of t-PA and of the corresponding messenger RNA (mRNA).
3. Workers in early May 1982 knew how to prepare antibodies against t-PA derived from Bowes melanoma cells.
4. It was known that a population of mRNAs could be reverse transcribed, using reverse transcriptase and other enzymes, to form a corresponding population of complementary DNAs (cDNAs) which could in turn be made into a library. Such a library could be screened in various ways in an attempt to isolate a particular cDNA encoding, for example, all part of t-PA.
5. The probability of obtaining a particular cDNA clone was uncertain and depended upon many factors. For example,
(a) The relative abundance of the desired mRNA.
(b) Increase in relative abundance through purification of the mRNA, by virtue of its specific length, for example.
(c) Whether or not the relative abundance of the mRNA could be increased by induction.
(d) The skill of the workers involved in dealing with a population of mRNA, which is intrinsically less stable than DNA.
(e) Several different enzymes are needed to prepare the library. Skill is involved in assessing the quality of particular batches of such enzymes, for example with respect to freedom from contaminating activities such as nucleases.
(f) The yield in transfecting the library into host bacteria, so that enough clones will be obtained to contain the desired cDNA within the population.
(g) The availability and sensitivity of assays for the desired protein.
6. In early May 1982, skilled workers were familiar with many different techniques for manipulating RNA and DNA and in particular for preparing and screening DNA libraries. Shortly following the filing of Genentech’s first patent application, most of these techniques were collected in T Maniatis, EF Fritsch and J Sambrook, Molecular Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory (1982) (‘Maniatis’).
7. t-PA is larger and more complex than most other proteins that had been produced by recombinent DNA technology before 1982. It is a secreted protein with signal sequences, it is a glycoprotein and it is rich in cysteine residues.
8. The mRNA for t-PA is larger than most mRNAs and that had been successfully subjected to cDNA cloning efforts before early May 1982.
9. Before mid-1982, most proteins successfully produced by cDNA cloning were coded by abundant mRNAs, present intially in relative amounts greater than 1/200.
10. cDNAs corresponding to less abundant mRNAs (less than 1/200) had been cloned earlier. These mRNAs were usually smaller than that for t-PA, the cDNAs were cloned only incompletely, or they were derived from mRNAs whose relative abundance in the total population could be increased by induction. An example of the latter is the cloning of a cDNA for human alpha interferon by Weissman et al.
11. It was unknown in 1982 whether or not the relative abundance of the mRNA for t-PA could be increased by induction.
12. According to ‘Maniatis’, the ‘method of choice’ for cloning a low abundance mRNA involved chemical synthesis of a short DNA which hybridises to some part of the desired mRNA. One must know the amino acid sequence of a corresponding region of the target protein before this can be done.
13. Workers attempting to obtain sequence information for t-PA would have had access to and would have been familiar with the operation of automated liquid phase sequenators, in which amino acids are serially cleaved from a protein and analysed. Choice of sequencing strategies and the degree of care needed in sequencing depends on the quantity of material available. In early May 1982, a successful sequencing run might have yielded 10 to 20 amino acid residues and might have required a 1-2 mg sample of protein.
14. In the early 1980’s, workers would most commonly have approached the task of sequencing a part of t-PA by attempting in order:
(a) N-terminal sequencing, on the possibility that the N-terminus was not blocked.
(b) Cleavage at Methionine residues, which are relatively rare, by cyanogen bromide, to give a manageably small number of peptide fragments. Amino acid compositions available for t-PA would suggest that this was a reasonable approach.
(c) Cleavage of t-PA with a proteolytic enzyme, most likely trypsin, followed by separation and sequencing of the peptides.
15. In 14(b) and (c), each fragment must be separated before it can be sequenced. The difficulty of such separation is related to the number of peptide fragments produced and to the quantity of t-PA available.
16. In early May 1982, knowledge of the genetic code permitted workers to deduce a limited set of possible DNA sequences corresponding to a particular amino acid sequence. The exact number depends on the particular amino acid sequence, since the genetic code is ‘degenerate’ ie, most amino acids are encoded by more than one triplet codon.
17. Workers in 1982 could readily synthesise a mixture of all possible DNA sequences corresponding to a short sequence of ca 5 amino acids. In order to bind well to the target cDNA, each probe had to be at about 14 nucleotides long and preferably longer. The number of different sequences in the mixture is a function of the degeneracy of the code for a particular amino acid sequence.
18. Binding of some members of the mixed population of synthetic DNA sequences to unwanted cDNAs in the library will occur by random chance. The level of this background is a function of the complexity of the probe.
19. The larger the cDNA library, the more possibility for hybridisation of probes to unwanted cDNA clones, but the greater the likelihood that it will contain a cDNA corresponding to a low abundance mRNA.
20. Before May 1982, it was not unusual to isolate a cDNA clone identified by hybridisation to a short synthetic DNA probe and to determine its DNA sequence.
*Revised 20 January 1987.”
9. Patent in suit
9.01. Genentech filed their application on 4 May 1983 with a priority date 5 May 1982. The application was published, as already stated, on 26 February 1986. By their patent Genentech made 20 claims all of which were considered on the appeal. Reference must be made to the patent in suit but the claims particularly relevant to this appeal are set out here for the sake of convenience:-
3. Human tissue plasminogen activator as produced by recombinant DNA technology.
7. A recombinant cloning vector comprising a DNA sequence encoding human tissue plasminogen activator.
8. A replicable expression vector capable, in a transformant microorganism or cell culture, of expressing a DNA sequence according to claim 7.
9. The plasmid p$99RIPA or pt-PAtrp12.
16. A process which comprises expressing DNA encoding human tissue plasminogen activator in a recombinant host cell.
17. A process for producing human t-PA, which process comprises: a. preparing a replicable expression vector capable of expressing the DNA sequence encoding human t-PA in a host cell;
b. transforming a host cell culture to obtain a recombinant host cell;
c. culturing said recombinant host cells under conditions permitting expression of said t-PA encoding DNA sequence to produce human t-PA;
d. recovering said human t-PA.
19. A process for producing human tissue plasminogen activator, substantially as described herein.
20. Human tissue plasminogen activator produced by a process according to any one of claims 16 to 19.
9.02. The specification gives complete details of the processes and work carried out on the path to achieving the synthesis of t-PA. For the purposes of the appeal it is, I hope, necessary only to refer to certain extracts bearing in mind the comprehensive nature, however, of the specification from which they have been taken. These are as follows:
“The present invention arises in part from the discovery of the DNA sequence and deduced amino acid sequence of human plasminogen activator. This discovery enabled the production of human plasminogen activator via the application of recombinant DNA technology, in turn, enabling the production of sufficient quality and quantity of material to initiate and conduct animal and clinical testing as prerequisites to market approval, unimpeded by the restrictions necessarily inherent in the isolation methods hitherto employed involving production and extraction from existing cell cultures. This invention is directed to these associated embodiments in all respects.”
“The present invention is also directed to replicable DNA expression vehicles harboring gene sequences encoding human tissue plasminogen activator in expressible form, to microorganism strains or cell cultures transformed with them and to microbial or cell cultures of such transformed strains or cultures, capable of producing human tissue plasminogen activator. In still further aspects, the present invention is directed to various processes useful for preparing said gene cultures, and specific embodiments thereof. Still further, this invention is directed to the preparation of fermentation cultures of said microorganisms and cell cultures.”
As used herein, ‘human tissue plasminogen acitivator’ or ‘human’ t-PA or ‘t-PA’ denotes human extrinsic (tissue-type) plasminogen activator, produced by microbial or cell culture systems, in bioactive forms comprising a protease portion and corresponding to those tissue plasminogen activators otherwise native to human tissue. The human tissue plasminogen activator protein produced herein has been defined by means of determined DNA gene and deductive amino acid sequencing. It will be understood that natural allelic variations exist and occur from individual to individual. These variations may be demonstrated by (an) amino acid difference(s) in the overall sequence or by deletions, substitutions, insertions, inversions or additions of (an) amino acid(s) in said sequence. In addition, the location of and degree of glycosylation will depend on the nature of the host cellular environment.
“The potential exists, in the use of recombinant DNA technology, for the preparation of various human tissue plasminogen activator derivatives, variously modified by resultant single or multiple amino acid substitutions, deletions, additions or replacements, for example, by means of site directed mutagenesis of the underlying DNA. Included would be the preparation of derivatives retaining the essential kringle region and serine protease region characteristic generally of the human tissue plasminogen activator described specifically herein, but otherwise modified as described above. All such allelic variations and modifications resulting in derivatives of human tissue plasminogen activator are included within the scope of this invention, as well as other related human extrinsic (tissue-type) plasminogen activators, similar physically and biologically, so long as the essential characteristic human tissue plasminogen activator activity remains unaffected in kind.”
“‘Expression vector’ includes vectors which are capable of expressing DNA sequences contained therein, where such sequences are operably linked to other sequences capable of effecting their expression. It is implied, although not always explicitly stated, that these expression vectors must be replicable in the host organisms either as episomes or as an integral part of the chromosomal DNA. Clearly a lack of replicability would render them effectively inoperable. In sum, ‘expression vector’ is given a functional definition, and any DNA sequence which is capable of effecting expression of a specified DNA code disposed therein is included in this term as it is applied to the specified sequence. In general, expression vectors of utility in recombinant DNA techniques are often in the form of ‘plasmids’ which refer to circular double stranded DNA loops which, in their vector form are not bound to the chromosome. In the present specification, ‘plasmid’ and ‘vector’ are used interchangeably as the plasmid is the most commonly used form of vector. However, the invention is intended to include such other forms of expression vectors which serve equivalent functions and which become known in the art subseqently hereto.
‘Recombinant host cells’ refers to cells which have been transformed with vectors constructed using recombinant DNA techniques. As defined herein, t-PA is produced in the amounts achieved by virtue of this transformation, rather than in such lesser amounts, or, more commonly, in such less than detectable amounts, as might be produced by the untransformed host. t-PA produced by such cells can be referred to as ‘recombinant t-PA’.
B Host Cell Cultures and Vectors
The vectors and methods disclosed herein are suitable for use in host cells over a wide range of prokaryotic and eukaryotic organisms.
In general, of course, prokaryotes are preferred for cloning of DNA sequences in constructing the vectors useful in the invention. For example, E coli K12 strain 294 (ATCC No 31446) is particularly useful. . . .”
“Prokaryotes may also be used for expression. . .”
“In addition to prokaryotes, eukaryotic microbes, such as yeast culture may also be used. . . .”
“In addition to microorganisms, cultures of cells derived from multicellular organisms may also be used as hosts. In principle, any such cell culture is workable, whether from vertebrate or invertebrate culture. However interest has been greatest in vertebrate cells, and propogation of vertebrate cells in culture (tissue culture) has become a routine procedure in recent years [Tissue Culture, Academic Press, Kruse and Patterson, editors (1973)]. Examples of such useful host cell lines are VERO and HeLa cells, Chinese hamster overy (CHO) cell lines, and W138, BHK, COS-7 and MDCK cell lines. Expression vectors for such cells ordinarily include (if necessary) an origin of replication, a promoter located in front of the gene to be expressed, along with any necessary ribosome binding sites, RNA splice sites, polyadenylation site, and transcriptional terminator sequences.”
“Examples which are set forth hereinbelow describe use of E coli using the lac and trp promoter system and use of CHO cells as host cells, and expression vectors which include the SV40 origin of replication as a promotor. However, it would be well within the skill of the art to use analogous techniques to construct expression vectors for expression of desired protein sequences in alternative prokaryotic or eukaryotic host cell cultures.
“Satisfactory amounts of human t-PA are produced by cell cultures, however, later refinements using a secondary coding sequence serve to enhance production levels even further. The secondary coding sequence comprises dihydrofolate reductase (DHFR) which is affected by an external controlled parameter, such as methotrexate, thus permitting control of expression by control of the methotrexate (MTX) concentration.”
“E.1.E. Preparation of DNA Probe
Purified human tissue plasminogen acitvator was obtained according to the procedure of disclosed reference (19, 20).
The molecule was scanned in order to locate regions best suited for making synthetic probes, as follows: . . .
“. . . After sequencing about 25 of the best possible peptide peaks, all the sequence data that could be aligned was pooled to obtain a preliminary model of the primary structure of tissue plasminogen activator. From this data and model, several possible probes were located.”
“Plasmid DNA was isolated by a rapid method (71) from all colonies showing a positive hybridisation reaction. The cDNA inserts from these clones were then sequenced after subcloning fragments into the M13 vector mp 7(73) and by the Maxam Gilbert chemical procedure (74). Figure 3 displays filter number 25 showing in the hybridisation pattern a positive tissue plasminogen activator clone. The cDNA insert in clone 25E10 was demonstrated to be the DNA coding for tissue plasminogen activator by comparing its amino acid sequence with peptide sequence (see supra) obtained from purified tissue plasminogen activator and by its expression product produced in E coli as described in more detail infra. The cDNA insert of clone 25E10 (plasmid pPA25E10) was 2304 base pairs in length with the longest open reading frame encoding a protein of 508 amino acids (MW of 56756) and containing a 722 bp 3′ untranslated region. This cDNA clone lacked the N-terminal coding sequences.”
“E.1.I. Direct Expression of Full Length Tissue Plasminogen Activator cDNA Clone in E coli
A reconstruction of the entire coding sequence was possible employing the common HhaI restriction endonuclease site shared by both partial clones pPA17 and pPA 25E10. . . .”
The following passages in the specification traced the development step by step construction of plasmid pt-PAtrp12 and p$99RIPA the two plasmids in Claim 9.
10. Judgment of Whitford J
10.01. After rehearsing much of the technical detail and referring to the patent in suit Whitford J posed the question ( RPC 553 at 577, line 10):
“The question is, are these claims valid? In my view the majority of them are plainly invalid.”
After referring to the 1977 Act and the European Patent Convention, to both of which I must subsequently refer in more detail, the learned judge reviewed the evidence, including the agreement reached by the experts, and came to the following conclusions, at page 592:
“In my view, a specification which contains claims which are not supported by the description or which are neither clear nor concise and which fail to define the matter for which the applicant seeks protection is not a specification directed to a patentable invention. It is one in respect of which a patent should never have been granted and I am of opinion that these objections are open and available under the Act of 1977, as they were under the previous Act.”
Applying this approach, Whitford J came to the following conclusions in relation to what he described as the broad claims. He held that there was no basis in the specification for claim 2. In relation to claim 3 he said at page 594:-
“Claim 3 is a claim to a known product, t-PA and indeed its constituent amino acids were known, if not their sequence, made by any process involving recombinant DNA technology. Only one such route is particularly dealt with in the patent. Two other routes at the probing stage, though rejected by Genentech on the evidence as being unworkable, are still covered by claim 3.
“As a claim to a product, t-PA produced by any known or hereafter discovered route in the field of recombinant DNA technology, is too wide and is bad. There is no basis for it and again, if that were not an objection that is open, the directions outside the particular route followed in the patent must be wholly insufficient. It is a claim to an obviously desirable and potentially possible end reached by routes on which only limited guidance is given.
“The objections to these broad claims, claims 1 and 3, follow through so far as all the broad claims of the patent are concerned and, in consequence, I have reached the conclusion that none of the claims, other than claim 9 and the claim dependent thereon, can be supported. So far as claim 9 is concerned, I must say a little about that but, before I deal with this and indeed certain other remaining grounds of objection, there are two authorities to which I must make brief reference.”
10.02. Whitford J rejected two objections made by Mr Jacob against claim 9 and came to a conclusion that had the claim stood alone it would have been sustainable. In view of the conclusion which the judge reached in relation to the other claims, however, it was clear that the patent would have to be revoked and that, therefore, a more sustained consideration of the status of claim 9 was unnecessary. Whitford J accepted that there was a discovery which was capable of being usefully employed in the form of a patentable invention as disclosed in claims 9 and 19. It is fair to say, however, that the main thrust of Wellcome’s attack on the patent was based on obviousness and lack of inventiveness.
11. Statutory provisions
11.01. It is convenient at this point to refer to some of the provisions of the 1977 Act:-
“Preamble:- An act to establish a new law of patents applicable to future patents and applications for patents; to amend the law of patents applicable to existing patents and applications for patents; to give effect to certain international conventions on patent; and for connected purposes. [29th July 1977]
. . .
NEW DOMESTIC LAW
“1. — (1) A patent may be granted only for an invention in respect of which the following conditions are satisfied, that is to say —
(a) the invention is new;
(b) it involves an inventive step;
(c) it is capable of industrial application;
(d) the grant of a patent for it is not excluded by subsections (2) and (3) below;
and any references in this Act to a patentable invention shall be construed accordingly.
(2) It is hereby declared that the following (amongst other things) are not inventions for the purposes of this Act, that is to say, anything which consists of —
(a) a discovery, scientific theory or mathematical method;
(b) a literary, dramatic, musical or artistic work or any other aesthetic creation whatsoever;
(c) a schem, rule or method for performing a mental act, playing a game or doing business, or program for a computer;
(d) the presentation of information;
but the foregoing provision shall prevent anything from being treated as an invention for the purposes of this Act only to the extent that a patent or application for a patent relates to that thing as such.
(3) A patent shall not be granted —
(a) for an invention the publications or exploitation of which would be generally expected to encourage offensive, immoral or anti-social behaviour;
(b) for any variety of animal or plant or any essentially biological process for the production of animals or plants, not being a microbiological process or the product of such a process.”
. . .
“2. — (1) An invention shall be taken to be new if it does not form part of the state of the art.
(2) The state of the art in the case of an invention shall be taken to comprise all matter (whether a product, a process, information about either, or anything else) which has at any time before the priority date of that invention been made available to the public (whether in the United Kingdom or elsewhere) by written or oral description, by use or in any other way.
“(3) The state of the art in the case of an invention to which an application for a patent or a patent relates shall be taken also to comprise matter contained in an application for another patent which was published on or after the priority date of that invention, if the following conditions are satisfied, that is to say —
(a) that matter was contained in the application for that other patent both as filed and as published; and
(b) the priority date of that matter is earlier than that of the invention.”
. . .
“(6) In the case of an invention consisting of a substance or composition for use in a method of treatment of the human or animal body by surgery or therapy or of diagnosis practised on the human or animal body, the fact that the substance or composition forms part of the state of the art shall not prevent the invention from being taken to be new if the use of the substance or composition in any such method does not form part of the state of the art.
3. An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the state of the art by virtue only of section 2(2) above (and disregarding section 2(3) above).”
14. — (1) . . .
(2) . . .
(3) The specification of an application shall disclose the invention in a manner which is clear enough and complete enough for the invention to be performed by a person skilled in the art.”
. . .
“(5) The claim or claims shall —
(a) define the matter for which the applicant seeks protection;
(b) be clear and concise;
(c) be supported by the description, and
(d) relate to one invention or to a group of inventions which are so linked as to form a single inventive concept.”
“Revocation of patents
“72. — (1) Subject to the following provisions of this Act, the court or the comptroller may on the application of any person by order revoke a patent for an invention on (but only on) any of the following grounds, that is to say —
(a) the invention is not a patentable invention;
(b) the patent was granted to a person who was not the only person entitled under section 7(2) above to be granted that patent or to two or more persons who were not the only persons so entitled;
(c) the specification of the patent does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art;
(d) the matter disclosed in the specification of the patent extends beyond that disclosed int he application for the patent, as filed, or, if the patent was granted on a new application filed under section 8(3), 12 or 37(4) above or as mentioned in section 15(4) above, in the earlier application, as filed;
(e) the protection conferred by the patent has been extended by an amendment which should not have been allowed.”
. . .
“(5) A decision of the comptroller or an appeal from the comptroller shall not estop any party to civil proceedings in which infringement of a patent is in issue from alleging invalidity of the patent on any of the grounds referred to in subsection (1) above, whether or not any of the issues involved were decided in the said decision.”
. . .
“125. — (1) For the purposes of this Act an invention for a patent for which an application has been made or for which a patent has been granted shall, unless the context otherwise requires, be taken to be that specified in a claim of the specification of the application or patent, as the case may be, as interpreted by the description and any drawings contained in that specification, and the extent of the protection conferred by a patent or application for a patent or application for a patent shall be determined accordingly.”
. . .
(3) The Protocol of the Interpretation of Article 69 of the European Patent Convention (which Article contains a provision corresponding to subsection (1) above) shall, as for the time being in force, apply for the purposes of subsection (1) above as it applies for the purposes of that Article.”
. . .
130. — (1) In this Act, except so far as the context otherwise requires” —
. . .
(7) Whereas by a resolution made on the signature of the Community Patent Convention the governments of the member states of the European Economic Community resolved to adjust their laws relating to patents so as (among other things) to bring those laws into conformity with the corresponding provisions of the European Patent Convention, the Community Patent Convention and the Patent Co-operation Treaty, it is hereby declared that the following provisions of this Act, that is to say, sections 1(1) to (4), 2 to 6, 14(3), (5) and (6), 37(5), 54, 60, 69, 72(1) and (2), 74(4), 82, 83, 88(6) and (7), 100 and 125, are so framed as to have, as nearly as practicable, the same effects in the United Kingdom as the corresponding provisions of the European Patent Convention, the Community Patent Convention and the Patent Co-operation Treaty have in the territories to which those Conventions apply.”
11.02. I am grateful to Mr Gratwick for his able exposition of the history both of convention and statute which he traced from the Statute of Monopolies to the enactment of the 1977 Act. This is to be found in the transcript Day 1 pages 6 et seq of the argument in the Court of Appeal. It forms an important factual matrix against which to interpret the relevant provisions of the 1977 Act. It confirms, in my judgment, Mr Gratwick’s submission that for the purpose of interpreting the 1977 Act it must be viewed in the context of a departure from much of the authority and usage of previous patent law. The 1977 Act must be primarily construed in the context of the Community Patent Convention and the corresponding provisions of the European Patent Convention, the Community Patent Convention and the patent Co-operation Treaty (“the European Convention”). Indeed in relation to all those sections of the 1977 Act with which this appeal is concerned, section 130(7) specifically provides that the sections have been framed so as to have the same effect as far as practical in the United Kingdom as the corresponding provisions of the European Patent Convention. If observance to this provision involved what might appear to be a heretical departure form established practice then this must be laid at the door of the legislator, and taken as disclosing the intention of Parliament to observe the European Patent Convention. One example in point will suffice.
11.03. Mr Gratwick referred us to paragraph 4.31 of the 13th edition of Terrell on The Law of Patents where reference is made to the classic dictum by Lord Russell in Electrical & Musical Industries Ltd v Lissen Ltd, (1939) 56 RPC 23, 39:-
“The function of the claims is to define clearly and with precision the monopoly claimed, so that others may know the exact boundaries of the area within which they will be trespassers. Their primary object is to limit and not to extend the monopoly. What is not claimed is disclaimed. The claims must undoubtedly be read as part of the entire document, and not as a separate codument; but the forbidden field must be found in the language of the claims and not elsewhere,”
Article 69 of the Convention, paragraph 1 provides:-
“The extent of the protection conferred by a European patent or a European patent application shall be determined by the terms of the claims. Nevertheless, the description and drawings shall be used to interpret the claims.”
The Protocol on the Interpretation of Article 69 of the Convention provides:-
“Article 69 should not be interpreted in the sense that the extent of the protection conferred by a European patent is to be understood as that defined by the strict, literal meaning of the wording used in the claims, the description and drawings being employed only for the purpose of resolving an ambiguity found in the claims. Neither should it be interpreted in the sense that the claims serve only as a guideline, and that the actual protection conferred may extend to what, from a consideration of the description and drawings by a person skilled in the art, the patentee has contemplated. On the contrary, it is to be interpreted as defining a position between these extremes which combines a fair protection for the patentee with a reasonable degree of certainty for third parties.”
It may well be that an interpretation of the claim or claims which achieves a position between the two extremes referred to in the Protocol combining a fair protection for the patentee with a reasonable degree of certainty for third parties will not always be entirely consistent with an interpretation of the claim within the classic dictum of Lord Russell. In the presence of such conflict the Protocol should prevail.
11.04. Mr Gratwick submitted, in my judgment rightly, that when looked at as a whole in the foregoing context the 1977 Act provided a complete code dealing with the application for and grant of a patent and thus displaced any residual common law element which previously had been preserved by succeeding statues. In particular the Act provided complete codes dealing with the application and grant of a patent (section 14), the protection of the monopoly against infringement (section 60) and suits for the revocation of a patent (section 72).
11.05. Mr Gratwick submitted that Whitford J was not entitled to import the requirements of section 14(5) which specify the nature of the application into the provisions of section 72(1) so as to entitle him to revoke the patent for a failure to comply with that subsection. In support of this submission Mr Gratwick emphasised that as a pure exercise of construction, where a provision in section 14 is intended to have the dual purpose of permitting the Comptroller to refuse the grant of a patent under section 18 and also to permit third parties to sue under section 72 for revocation of the patent after grant, the particular requirement is mentioned specifically in both sections (see section 14(3) which is found repeated with minor adjustment in section 72(1)(c)). Similarly the restrictions upon the grant of an application for a patent amounting to an amendment of a previous application provided in section 76 are preserved for the benefit of a suitor under section 72(1)(e) if the amendment should not have been made.
11.06. Two other provisions should be noticed in this regard. Before the grant of the patent the rights of third parties affected are extremely limited. Under section 21 they may make observations which the Comptroller is obliged to consider in accordance with the rules. They may not, however, make oral representations with or without the assistance of counsel. This is the limit of opportunity given to interested third parties under the 1977 Act to oppose the grant of a patent. When compared with the provisions of section 14 and the appeal procedure provided in section 14(4) and section 85(1) of the Patents Act 1949 the provisions of the 1977 Act represent a significant extension of the unsupervised jurisdiction of the Comptroller. Section 14 of the 1949 Act provided as a ground for revocation of a patent that “the complete specification does not sufficiently and fairly describe the invention or the method by which it is to be performed.”
11.07. The extension of the unlimited powers of the Comptroller is reflected in the European Patent Convention. The requirements of an application for a European patent are contained in Part III of the Convention. Article 82 provides for unity of invention. Article 83 provides:-
“The European patent application must disclose the invention in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.”
Article 84 provides:-
“The claims shall define the matter for which protection is sought. They shall be clear and concise and be supported by the description.”
Part V of the Convention deals with “Opposition Procedure”. In this part Article 100 provides the ground for opposition in the following terms:-
“Opposition may only be filed on the grounds that:
(a) the subject-matter of the European patent is not patentable within the terms of Articles 52 to 57;
(b) the European patent does not disclose the invention in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art;
(c) the subject-matter of the European patent extends beyond the content of the application as filed, or, if the patent is granted on a divisional application or on a new application filed in accordance with Article 61, beyond the content of the earlier application as filed.”
There are no provisions for opposition by interested third parties at the stage of the application before grant by the examining divisions of the European Patent Office. The significant matter is, however, that whilst Article 83 is a ground upon which opposition can be based under Article 100, Article 84 is not. This proposition was confirmed by the Techinical Board of Appeal in their decision in Naimer’s Patent (Decision T23/86 dated 25 August 1986)  Official Journal EPO 316. Finally, Article 138 which provides for the revocation of a European patent under the law of a contracting state does not include a failure to comply with Article 84 although it extends to Articles 52 and 57 and a ground that the European patent does not disclose the invention sufficiently clearly.
11.08. With great respect to Whitford J I am forced to the conclusion that on this aspect of the matter Mr Gratwick’s submissions are made out to the extent, if any, that the learned judge purported to import into section 72(1) of the 1977 Act the provisions of section 14(5) as a further ground for revocation of the patent in suit. In this respect the judgment cannot be upheld.
11.09. I have come to this conclusion, after not a little hesitation, in spite of the powerful arguments proposed by Mr Jacob. Summarising, I hope not unduly eclectically, these were:-
1. That claims unsupported by the invention, ie lacking “fair basis”, or too widely drawn have as a matter of long legal tradition and authority been vulnerable to attack in either infringement or revocation suits; that it would be highly undesirable if claims deficient in this respect, having escaped detection during the examination process in the Patent Office, should thereafter acquire immunity to challenge and that, therefore, Parliament could not have intended the 1977 Act to have had this result.
2. That the 1977 Act did not have this result. The definition of “invention” in section 125 is “that specified in a claim”, which is different from the sense in which the word is used in the Articles of the Convention. “The invention or claim as specified” must satisfy section 1(1) and if it is drawn too widely it will fail for lack of invention or obviousness, ie is “not new” and is, therefore, not patentable under section 72(1)(a).
3. That there was nothing inconsistent in a decision that a claim which should have been refused under section 14(5) of the Act should also fail under section 72(1)(a).
11.10. Mr Jacob relied mainly upon two authorities, British United Shoe Machinery Co Ltd v Simon Collier Ltd, (1909) 26 RPC 21 and Mullard Radio Valve Co Ltd v Philco Radio & Television Corp of Great Britain Ltd, (1936) 53 RPC 323. The first comment which must be made is that in each case a primary ratio decidendi was that the claims impugned lacked subject matter, ie were too widely drawn. This ground which at the date of the decision was available as an avenue of attack in revocation proceedings, or as a defence in an infringement action, is, if I am correct in my view of the meaning of section 14(5) and section 72(1), not available to Wellcome in these proceedings. It is, however, necessary to look a little further into each of these cases.
11.11. In the Simon Collier case the invention sought to be patented was an automatic cutting machine which by means of an eccentric cam was able to vary the relative distances between the cutting knife and the guide used for producing the “scotch” edge for shoes. It was held that a claim for all automatic means of varying the relative positions of the knife and the guide was because of its width not proper subject matter and that the patent was, therefore, invalid. This would not be a defence available to an infringer under the 1977 Act. However, it appears from the report that under the present statute the patent would still have been held invalid for lack of novelty under section 72(1) and section 1(1). Although the idea of varying the relative positions of the knife and guide for the purpose of producing an ordinary scotch edge was not new, there was no evidence in any of the prior specifications quoted in argument of an automatic means for doing this. However, to use a cam for this purpose was well recognised. At page 50, line 41, Parker J said:-
“Even had I been able to construe the 15th claiming clause as seeking protection only for the use of a cam, or its mechanical equivalent, to vary automatically the relative positions of the knife and guide for the purpose of producing a bulged ‘Scotch’ edge , I do not think I could have upheld the patent. The validity of such a claim would, I think, depend upon whether the principle of using a cam to vary the relative positions of parts of a machine were a new principle or an old principle. If new, a claim would undoubtedly be good; if old, the claim might be bad on the principle of the cases already quoted. At the date of the plaintiffs’ patent, the employment of a cam to vary the relative position of parts of the machine was well known.”
11.12. In the Mullard Radio Valve case the invention related to the discovery of a solution to a problem which had been experienced in the design and operation of thermionic vacuum tubes. Basically, these valves consisted of a highly exhausted tube containing a number of electrodes which, when heated, emitted electricity by the process of thermionic emission. The cathode when heated sends out negative electrons in transit to a positive anode. In between the cathode and the anode a third device, namely a control grid, was inserted. The function of the control grid was to prevent an accumulation of negatively charged electrons gathering in a cloud around the cathode rather than joining the stream towards the anode. This resulted from negative repulsion amongst the electrons themselves overcoming the attraction of the positive charge on the anode. The control grid drew this cloud of electrons towards the anode. This triode arrangement resulted in an uncovenanted fall in the anode potential. This could be counteracted by the insertion of yet another electrode called a screening grid between the control grid and the anode upon which there was a positive potential. There was still a risk of the anode current being reduced at the expense of an increase of the screening grid current when the anode potential fell below the potential of the screening grid. The invention was the placing of a third electrode in the region of the screening grid and the anode which would avoid this loss of current in the anode. Claim 1, which was not in suit, was in these terms:-
“A circuit arrangement for amplifying electric oscillation by means of one or more thermionic discharge tubes connected in series or cascade characterized in that the discharge tube of the last stage of amplification comprises a screening grid kept at a constant high potential between the control grid and the anode and that such discharge tube is so arranged by the introduction of an auxiliary grid kept at a constant and relatively low potential that when the anode potential falls below the potential of the screening grid the increase of the screening grid current at the expense of the anode current will be substantially avoided.”
Claim 2, which was in dispute, as amended read:-
“A discharge tube having at least three auxiliary electrodes between the cathode and the anode characterized in that the auxiliary electrode nearest to the anode is connected to the cathode so as to be maintained continuously at the cathode potential.”
11.13. On the basis that the insertion of the third electrode at a low potential, ie connected to the cathode, solved the problem created when the screening electrode was inserted between the control electrode and the anode electrode and that this was a new “invention” Romer and Maugham LJJ in the Court of Appeal and Lord MacMillan in the House of Lords agreed that claim 1 had subject matter and was valid. This was in disagreement with the Master of Rolls who was against the claim on the ground of want of novelty. Claim 2 failed both in the Court of Appeal and in the House of Lords on the width of the claim and lack of substance. It will be sufficient to cite a short extract from the speech of Lord MacMillan at page 345:-
“Now discharge tubes were well known articles before this Patent was taken out, and the insertion of grids or auxiliary electrodes between the cathode and the anode was no novelty. The virtue of the Claim must reside, if anywhere, in the characteristic feature that in the discharge tube claimed the auxiliary electrode nearest to the anode is directly connected to the cathode. If the Claim is good, no one can without infringement make or sell a discharge tube containing three or more electrodes in which the electrode nearest the anode is directly connected with the cathode. Is the Patentee entitled to such monopoly? He is entitled to say that a discharge tube so constructed is new, for discharge tubes with such a connection were not previously known or made; and he is entitled to say that such a discharge tube has patentable utility in the sense that it can be used and will work as a discharge tube. But is the Claim justified by the inventive idea which the Patentee has disclosed? A patentee may make a most meritorious discovery and may give an entirely adequate description of his inventive idea and of the manner of putting it into practice, but when he comes to formulate the claim to his invention he may claim a monopoly wider in extent than is warranted by what he has invented. The Patentee had told us quite definitely that his invention deals with the case of a final amplifier which comprises a screening grid between the control grid and the anode and that he has invented means by which, in such a case, the screening grid current is prevented entirely or partially from increasing at the expense of the anode current when the anode potential falls. The problem which he set out to solve and the disadvantages which he professes to overcome relate solely to discharge tubes with a screening grid between the control grid and the anode. His discovery was that, if in a discharge tube with a screening grid between the control grid and the anode he inserted between the screening grid and the anode an additional ‘suppressor’ grid, he achieved the advantageous result which he describes. That is the ambit of his invention and for that he is entitled to protection. But Claim 2 makes no reference to screening grids or control grids at all. It simply speaks of three or more electrodes irrespective of their functions as screening grids or control grids or suppressor grids or of their arrangement relatively to each other.”
. . .
“The article claimed is a discharge tube in which by virtue of its construction the grid nearest the anode will always be kept at the same potential as the cathode, and its construction will no doubt always ensure this. But the inventor’s purpose was not to invent a discharge tube which should have this feature. If his problem had been to invent a tube in which the grid nearest the anode should always be at cathode potential, his problem could have been easily and obviously solved. Anyone could have told him that all he had to do was to establish a physical connection between the grid nearest to the anode and the cathode. His problem was to obviate disadvantages which arose when a screening grid was interposed between the control grid and the anode, a quite special and definitive problem. I do not think that he is entitled to prevent anyone in future from making a valve in which the grid nearest to the anode is connected with the cathode merely because he had discovered that in a valve in which the other grids are utilised in a particular way this connection will give the special advantage which he discovered. It is because he has sought to do so by Claim 2 that this Claim is in my opinion too wide. The fact that an article of obvious construction is discovered to give a valuable and new benefit if employed in a particular way does not entitle the discoverer to prevent everyone else from making that article.” (Emphasis provided)
11.14. The speech of Lord MacMillan and the rejection of claim 2 in Mullard’s case supports Mr Jacob’s submission that Genentech’s claims, save only for claim 9, are too wide but not his general proposition that a claim which is not supported by the inventive step described in the specification can be attacked under section 72(1) as being not patentable as such. Indeed, during the course of argument, Mr Jacob was constrained to adjust his submission to the effect that if the claim extends beyond that which has been invented then it is to that extent not for an invention and does not fall within section 1 of the Act (see Day 11 page 27D, Court of Appeal transcripts):-
“Mr Jacob: May I rephrase it in this way? If your claim — and that is the claim as such — extends beyond what you have invented, then a claim is to that extent not for an invention, and does not fall within section 1.”
The ambit of the invention was in the idea that the insertion of a suppressor grid with the appropriate potential between the screening grid and the anode would prevent the drop in the anode current. Claim 1, which was accepted as patentable, claimed this as incorporated in a discharge tube. The patentee was entitled to say that a discharge tube so constructed was new. Claim 2, which did not so define the claim, was held to be too wide.
12.01. Section 1(1)(a) of the 1977 Act requires that the invention must be “new”. For these purposes an “invention” is defined as “that specified in a claim of the specification . . . as interpreted by the description and any drawnings contained in that specification . . .” The word “invention” is used in this sense in the ensuing paragraphs of this judgment unless the context demands otherwise. The difficulties in this case arise from the fact that in one sense “nothing is new”, to state a questionably dangerous aphorism. Recombinant DNA technology was recognised as possible and had been the subject of published reports before the priorty date. The existence of a substance in nature described as t-PA obtainable from Bowes melanoma cells with its special therapeutic qualities was known and others besides Genentech had appreciated that recombinant DNA technology was a likely solution to the difficulty of obtaining commercially viable supplies of this important protein. In another sense it can equally be said that Genentech by the exercise of considerable skill, application and ingenuity discovered information about t-PA which had previously eluded the efforts of some of the finest scientific teams in the world and that this discovery enabled the development of a pharmaceutical process of immense value to the community at large and to Genentech in particular.
12.02. The question is: “Can the co-ordination of a nubmer of well recognised techniques which result in the discovery of a formula hitherto unascertained defining the molecular structure of a natural substance which itself is already known to exist, amount to ‘an inventive step’?” In answering this question the court must apply the provisions of the 1977 Act in the light of Article 69 of the Convention and the Protocol. A further unusual but critical feature of the case is the evidence that other teams of research workers were using the same of equivalent techniques to those described in the specification — if not always before the priority date, then certainly before any material information was imparted by Genentech. The impact of this evidence may well have an effect on whether the invention claimed is either new (section 1(1)(a)) or whether it involves an inventive step (section 1(1)(b)). It was for this reason I felt it necessary to burden the earlier part of this judgment with some account of these activities. Section 2 provides a codified formula against which the question whether an invention is new is to be judged. It is to be assumed to be new as long as it does not form part of “the state of the art”. “The state of the art” also receives a codified definition and is to be taken “to comprise” (note NOT “to include”) “all matter . . . which has at any time before the priority date been made available to the public . . . in any way” (section 2(2)). The definition except in special circumstances not relevant to this appeal is unlimited in all dimensions bar two:-
“(i) It must exist before the priority date;
(ii) It must have been made available to “the public”.
12.03. This definition of “the state of the art” is also imported into the definition of “inventive step” in section 3. For the purposes of section 3, the state of the art provides the background to the question of “obvious to a person skilled in the art”. The answer to the question: “What is the public?” would appear in a case such as this to be that community of research workers skilled in the art in general; but not, I would think, merely known to one or two individual research workers pursuing their own experiments in private.
12.04. Thus in deciding whether an invention is new within section 1(1)(a) one must look solely at any matter which has been “made available” to the public. It is an objective question of fact and has nothing to do with the subjective qualities or knowledge of anyone. Nearly all the processes mentioned in the specification of the patent in suit had been published and the references are given. This is a feature to which I have already drawn attention. One matter which had not been made available to the public simply because it was not known at the time were the details of the amino acid sequences of the protein and the DNA sequences for the t-PA gene molecule (hereinafter referred to as “the figure 5 data”). The figure 5 data was not made available until the publication of the report in January 1983. This would appear to be an answer to section 1(1)(a); but can it, however, be asserted that even if the invention is patentable under section 1(1)(a) it fails under section 1(1)(b)? First it must be said that the wording of section 3, in my judgment, places the burden of proving “obviousness” firmly on the shoulder of the person attacking the patent. On this aspect I agree with the submissions of Mr Gratwick. If Mr Jacob is to succeed he must, therefore, discharge the burden of negating “inventive step” which in the context of this appeal means establishing obviousness or lack of invention. Before coming to consider this issue I must divert for a moment to consider an issue which arises under section 1(2)(a).
12.05. There is no dispute that the determination of the full amino acid and DNA sequences coding for the t-PA protein constitutes “a discovery” in the sense that by their efforts Genentech “discovered” the full details of the structure of t-PA and the DNA which coded for it disclosed in figure 5 of the specification. For this reason it was not argued to the contrary that without more section 1(2)(a) would apply to prevent it being an invention. The words that have given rise to difficulty, however, are “but the foregoing provision shall prevent anything from being treated as an invention . . . only to the extent that a patent . . . relates to that thing as such.” There are two conflicting views at first instance as to the significance of these words. The judgment of Whitford J in the present case was on the basis that the effect was to restrict the ban imposed by section 1(2) in relation to the matter set out in the sub-paragraphs to inventions which related to these matters and did not include a manner of using the matter concerned either in a process or in the creation of an artefact. If this were not so, Whitford J would not have held in favour of the patent in relation to claim 9. However, on 7 April 1987 Falconer J had delivered a judgment to which Whitford J was apparently not referred in which he, Falconer J, came to a contrarty conclusion. He held that “an invention was unpatentable if the inventiveness was contributed only by matters excluded under section 1(2). In assessing whether or not an application related to an excluded thing, therefore, it was necessary to take into account “whether the non-excluded features were already known or obvious”: Merrill Lynch Inc’s Application  RPC 1. In this case the application in suit related to a data processing system for making a trading market in securities. A known computer system was used, controlled by a program coded in any known programming language. The examiner rejected the application on the ground that the invention as claimed was unpatentable under section 1(2). On appeal the hearing officer upheld the examiner on the claim as presented but allowed an opportunity to file amended claims before publication. He held, inter alia, “The qualificaton at the end of section 1(2) meant that anything in an excluded category was not to be treated as an invention and could, therefore, not contribute to the required novelty and inventive step.” On appeal to the Patents Court Falconer J dismissed the appeal without prejudice to the filing of amended claims on the ground already indicated and held further that Article 52(2) and (3) of the European Patent Convention had the same meaning.
12.06. In order to understand this position it is necessary to touch upon the invention alleged to have been made in the claims in the Merrill Lynch case. The following extracts from the specification are taken from the report of the judgment:-
“Automated Securities Trading System Background and Summary of the Invention:
This invention relates to business systems and, more specifically, to an improved data processing based system for implementing an automated trading market for one or more securities. The system retrieves and stores the best current bid and asked prices; qualifies customer buy/sell orders for execution; executes the orders; and reports the trade particulars to customers and to national stock price reporting systems. The system apparatus also determines and monitors stock inventory and profit for the market maker.'”
. . .
“‘The system may be implemented by any digital data processing equipment per se well known to those skilled in the art eg, any common bus system interconnecting a digital processor, manual data entry terminal apparatus, one or more memories (one of which contains the controlling program), and output signalling apparatus such as a cathode ray tube and printer. The system may be coded in any program language per se well known to those skilled in the art. The process variables may be of any form which conform to the constraints of the particular language being used and the below listed variables are for purposes of illustration only”‘
The report of Mr MF Pilgrim, the principal examiner acting for the Comptroller, described the alleged invention in these terms:-
“I consider that the ‘means’ specified in claim 1 relate to features which either would be present in a conventional business computer system or define essential functions required for the performance of the business method. Consequently this claim contains nothing which could be considered to constitute a new technical structure or to produce a technical effect in the sense in which this term is used in the Guidelines. I conclude therefore that this claim does not define a patentable invention. As indicated above I do not consider that the first amended version differs from the original claim 1 in any significant way and conclude that this version also does not define a patentable invention.”
This description is taken up by Falconer J at page 8, line 39:-
“. . . and as the principal examiner pointed out, what is described is a computerised system arranged to analyse customers’ buy and sell orders against certain criteria and to execute those orders which are deemed to qualify.”
There seems on this limtied study of the facts that there was little evidence of an inventive step at all. The matter was being considered as a “mathematical formula” rather than a discovery within the terms of section 1(2) and, maybe, there lies the distinction between Merrill Lynch’s Application and the present case and also the decision of the Technical Board of Appeal in Vicom’s Application (Decision T208/84),  Official Journal EPO 119. This decision is a decision of a relevant convention court of which judicial notice must be taken under section 91 of the 1977 Act.
12.07 The opinions of the Technical Board of Appeal as reported in the vicom decision are summarised in the following propositions:-
“1. Even if the idea underlying an invention may be considered to reside in a mathematical method (cf Article 52(2)(a) and (3) of the European Patent Convention (the ‘EPC’)), a claim directed to a technical process in which the method is used does not seek protection for the mathematical method as such.
2. A computer of known type set up to operate according to a new program cannot be considered to form part of the state of the art as defined by Article 54(2) EPC>
3. A claim directed to a technical process which is carried out under the control of a computer program (whether by means of hardware or software), cannot be regarded as relating to a comptuer program as such (cf Article 52(2)(c) and (3) EPC).
4. A claim which can be considered as being directed to a computer set up to operate in accordance with a specified program (whether by means of hardware or software) for controlling or carrying out a technical process cannot be regarded as relating to a computer programme as such (cf Article 52(2)(c) and (3) EPC).”
In paragraph 6 of their decision the board dealt with an issue central to the present discussion:-
“5. There can be little doubt that any processing operation on an electric signal can be described in mathematical terms. The characteristic of a filter, for example, can be epxressed in terms of a mathematical formula. A basis difference between a mathematical method and a technical process can be seen, however, in the fact that a mathematical method or a mathematical algorithm is carried out on numbers (whatever these numbers may represent) and provides a result also in numerical form, the mathematical method or algorithm being only an abstract concept describing how to operate on the numbers. No direct technical result is produced by the method as such. In contrast thereto, if a mathematical method is used in a technical process, the process is carried out on a physical entity (which may be a material object but equally an image stored as an electric signal) by some technical means implementing the method and provides as its result a certain change in that entity. The technical means might include a computer comprising suitable hardware or an appropriately programmed general purpose computer.
“6. The Board of Appeal, therefore, is of the opinion that, even if the idea underlying an invention may be considered to reside in a mathematical method, a claim directed to a technical process in which the method is used does not seek protection for the mathematical method as such.” (Emphasis provided)
At page 13 of the Merrill Lynch report, Falconer J, after referring to paragraphs 5 and 6 of the Technical Board of Appeal’s decision, sought to distinguish them in the following terms:-
“It is plain that in referring to a technical process in paragraph 6, the Board is referring to a process as considered in paragraph 5 providing a resulting change in the physical entity on which the process is carried out. In paragraph 12 the Board stated:
’12. The Board is of the opinion that a claim directed to a technical process which process is carried out under the control of program (be this implemented in hardware or in software) cannot be regarded as relating to a computer program as such within the meaning of Article 52(3) EPC, as it is the application of the program for determining the sequence of steps in the process for which in effect protection is sought. Consequently, such a claim is allowable under Article 52(2)(c) and (3) EPC.’
Again I understand the Board to be using ‘technical process’ as in paragrah 5, ie one which produces a practical or technical effect on the physical entity.”
12.08. With great respect to Falconer J I have come to the conclusion that, whilst the decision in Merrill Lynch’s Application was probably correct for the reason already indicated earlier, the broad expressions of policy were neither necessary for the decision nor can they be reconciled with the judgments of Cozens-Hardy MR and Fletcher Moulton LJ in Hickton’s Patent Syndicate v Patents & Machine Improvements Co Ltd (1909) 26 RPC 339 (“Hickton’s case”) (see para 21.11). In rejecting the submission of Mr Pumfrey as to the meaning of the words “as such” which were in line with the opinions expressed by the Technical Board of Appeal in Vicom and preferring the submission of Mr Thorley, for the Comptroller, as to the meaning of the words “only to the extent that” as they are found in section 1(2), Falconer J placed an undue emphasis upon those words which effectively distorted the general meaning of the section itself.
12.09. In my judgment the plain and ordinary interpretation to be given to the words “only to the extent that” in conjunction with “relates to that thing as such” is derived from taking the two phrases together as meaning that any of the matters listed in sub-paragraphs (a) to (d) shall not be an invention for the purposes of the Act. Semble, otherwise they would have constituted inventions and shall only to the extent that the application or patent relates to that step as such be disqualified. Applying this approach to the facts of the present case, a claim for the figure 5 data would be a claim to a discovery as such and only to that extent would be disqualified by section 1(2). If, on the other hand, the discovery is one that was not known in the state of the art at the material date and was not obvious to a person skilled in the art, then it is capable of forming the substratum of invention so that if it is applied in a technique or process or incorporated in a product then it would be patentable. This is in line with the judgments in Hickton’s case and the opinions expressed by the Technical Board of Appeal in Vicom’s case. It is also what I believe to have been the view of Whitford J see page 566, line 12:-
“It is trite law that you cannot patent a discovery, but if one the basis of that discovery you can tell people how it can be usefully employed then a patent or invention may result. This in my view would be the case even though once you have made the discovery the way in which it can be usefully employed is obvious enough.”
12.10. Mr Jacob based his arguments on section 1(2) on the following propositions:
1. Whatever discovery has been made, irrespective of section 1(2), a patentee may not claim as “inventive” artefacts or processes owing nothing to what he has done.
2. Merely knowing something new about something old is not enough — see Chamberlain & Hookham Ltd v Mayor, etc, of Bradford (1903) 20 RPC 673.
3. A patentee may not claim a new property of some old substance or a new chemical reaction or process in the form: “I claim this newly discovered phenomenon as harnessed to make useful artefacts.”
4. When a man has made a discovery he must do more: he must disclose a manner or manners of harnessing that discovery, and his claim must be limited to that manner or those manners.
12.11. I have no difficulty in accepting the first three of Mr Jacob’s propositions so long as the third proposition is directed to the discovery and not to the artefact or process altered by the incorporation of the discovery. In my judgment there may be a critical distinction to be drawn between a claim to new knowledge or to a discovery “as such” which is not patentable under section 1(2) of the 1977 Act and a claim to a method embracing a discovery which may well be an invention which is patentable. This proposition appears from the judgments in Hickton’s case and is in accordance with the reasoning of the Technical Board of Appeal in their decision in Vicom (see paragraph 12.07 (ante)). In Hickton’s case the Court of Appeal, consisting of Cozens-Hardy MR, and Fletcher Moulton and Buckley LJJ were considering a patent relating to the manufacture of lace on twist lace machines:-
“The method of traversing described in the specification was by endwise movement of the comb-bars known as ‘shogging’. Traversing the carriages and bobbins in a lace machine by ‘shogging’ the comb-bar was old, and equalising the bobbin threads by interchanging the bobbins by hand was old. The plaintiffs alleged that it was new to traverse in a levers and go-through machine; that the alleged anticipations all related to net-making machines, and that in them traversing was for the purpose of making pattern and not of equalising. The defendants contended that, if this were so, yet that the use of the old device to effect equalising was not subject-matter for a patent, no ingenuity being required when once the idea was given.”
Per Cozens-Hardy MR at page 346:-
“The point is this:- It is said that the machines, lever machines, or lace machines, are perfectly old and well known; but it was also well known that their use involved the loss of a very considerable percentage of thread, because some of the bobbins had to do much more work than the others. The evil thus ascertained was attempted to be remedied by the operator putting his hand in and changing a few of the most exhausted bobbins and putting them in the place where some less exhausted bobbins were to be found. This operation was costly in time, and it was really not highly efficient, because it did not get that equality and evenness which is so desirable and essential in manufacturing things of this kind. The Patentee had got the idea that the equalisation of the use of the thread might be obtained by ‘shogging’ the comb-bar and thus securing that each bobbin does first an easy task, and then a hard task, with the result that if you ‘shog’ through the complete width of the strip of lace you produce a practical equality in the working of all the bobbins and thus greatly minimise, in fact almost destroy, the loss of thread which existed before. Was that a new idea, was it a meritorious idea, was it a useful invention? I think it was upon the evidence plainly a new idea. Certain it is that, although need for some invention to obviate this waste had been before the eyes of all persons engaged in the trade, it had never been discovered before. It was new in that sense, and it was none the less new because the operation of ‘shogging’ was as old as the hills. It was a new application of ‘shogging’, an application never applied to a machine of this class, nor for any analogous purpose. It is true that ‘shogging’ beind old, ‘shogging’ was applied in other lace machines, but it either was so applied as not to produce the equalisation which was aimed at here, or if by accident it did produce the equalistion, it was not a use so analogous to that which is found here as to render the thing so obvious as to compel us to state that there is no subject-matter in the patent. Utility is not denied and infringement is not denied, and I think the conclusion to which we are bound to come is, that there was subject-matter in this patent which entitled the plaintiffs to the relief which they sought by their action.
The learned judge in his judgment states a proposition, which, with the greatest possible respect, seems to me to be a great deal too wide. ‘An idea may be new and original and very meritorious, but unless there is some invention necessary for putting the idea into practice it is not patentable.’ That, I venture to say, is not in accordance with the principles which have hitherto been applied in Patent cases, and I do not think it ought to be recognised as the law.”
Per Fletcher Moulton LJ at page 347, referring to the patent granted to Bolton and Watt in respect of the steam engine:-
“The particular invention there was the condensation of the steam, not in the cylinder itself, but in a separate vessel. That conception occurred to Watt and it was for that that his patent was granted, and out of that grew the steam engine. Now can it be suggested that it required any invention whatever to carry out that idea when once you had got it? It could be done in a thousand ways and by any competent engineer, but the invention was in the idea, and when he had once got that idea, the carrying out of it was perfectly easy. To say that the conception may be meritorious and may involve invention and may be new and original, and simply because when you have once got the idea of it is easy to carry it out, that that deprives it of the title of being a new invention according to our patent law, is, I think, an extremely dangerous principle and justified neither by reason, nor authority.
I have taken the case of Bolton and Watt with the condenser, but I can give another. Take the case of the safety valve for boilers. The man who first discovered the idea of a properly weighted valve in the boiler solely for the purpose of relief, if the pressure rose too high, would have been making a most valuable and meritorious invention. So soon as he conceived that idea of guarding against the danger of explosion the carrying out of the idea required no invention at all. In my opinion, invention may lie in the idea, and it may lie in the way in which it is carried out, and it may lie in the combination of the two; but if there is invention in the idea plus the way of carrying it out, then it is good subject-matter for letters patent. As a matter of fact in the present case I have not the slightest doubt that there is good subject-matter.”
and per Buckley LJ at page 348:-
“No doubt you cannot patent an idea, which you have simply conceived, and have suggested no way of carrying out, but the invention consists in thinking of or conceiving something and suggesting a way of doing it. In my humble judgment the learned judge has gone wrong in that he forgot that that was so. I think you can have a patent for an idea, which is new and original and very meritorious, if you suggest a way of carrying it out. If you do not do so suggest you cannot no doubt have a patent; but the learned judge goes on to say: ‘Unless there is some invention necessary for putting the idea into practice it is not patentable.’ If you have got a new idea, a new conception of manufacturing something, or a new way of manufacturing something and you suggest a way of carrying it out, that is patentable. Here the inventor makes the lace — an old thing; he makes it by cheaper means, because he saves the wasted thread in the bobbin, and he does that by employing, for the first time in a new manner and for a new purpose, the old operation called ‘shogging’. That, I think, is a summary of the situation. To my mind that is plainly patentable.” (Emphasis provided)
Mr Gratwick submitted that using the figure 5 data by incorporating the cDNA in a well recognised expression plasmid is directly analogous to the inventiveness of using “shogging” in Hickton’s case. In my judgment there is considerable force in this submission. There was nothing new about “shogging”; what was new was the idea of using “shogging” to achieve the solution of a well known problem, namely the waste of thread resulting from the differential demands for thread made upon the various bobbins in the manufacturing process. Before the idea was conceived this waste had been reduced by the manual adjustment of the bobbins, itself a time consuming and wasteful process. In Hickton’s case only the idea was new, everything else was well tried and used. In the present case the only thing that was new was the figure 5 data. t-PA was well known and could be obtained by the laborious process of extraction from natural sources; or by processes such as expression hybridisation which were well known but extremely laborious in the case of a low abundant DNA. The discovery of the figure 5 data was new and its use in a well defined method was clearly patentable.
12.12. Proposition 4 may, in my judgment, be too restrictive if a narrow interpretation is put on the words “must be limited to that manner or manners”. Mr Jacob relied upon the judgment of Falconer J in the Merrill Lynch case; but, as I have already indicated, I do not agree that this judgment can be supported in some of its wider dicta. I do not agree that this judgment can be supported in some of its wider dicta. Leaving aside the question of obviousness and on the assumption that the discovery of the sequence in the figure 5 data was a discovery at all, then in my judgment the incorporation of that discovery in the form of a clone DNA insert in a plasmid would be a harnessing of the discovery so as to avoid section 1(2) of the 1977 Act notwithstanding that there was nothing new in using a plasmid or inserting a DNA clone into it. What was new was the product plasmid and also probably, the well worn process itself using the new discovery.
12.13. Mr Jacob went some way towards conceding this approach when dealing with the decision of the Technical Board of Appeal in Vicom’s application.
“Mr Jacob: Mr Lord, the difference, I suggest, is this: that when you come to look at the claim, it all depends on what is claimed. If, as a result of your new theory, or as a result of your new computer program, you end up with a new theory, or as a result of your new computer program, you end up with a new physical thing, in some way or other, that is to say features in your claim which are new, essential characteristics of the claim, then you have gone more than just mere theory, or more than just mere computer program, and you are into the area of patentability.”
12.14. We were referred by Mr Gratwick to number of authorities dealing with the question of discovery in relation to patentable claims. It is necessary to refer shortly to some of these. In May & Baker Ltd v Boots Pure Drug Co Ltd (1950) 67 RPC 23 the patent in suit claimed the manufacture of a class of products shortly termed sulpha-thiazoles. This class was very large and the specification contained statements that these new bodies “find application in therapeutics” and have “chemo-therapeutic activity in certain diseases”. On petition by the respondent company to revoke the patent on the ground that the statements as to therapeutic value of the new bodies were untrue the patentees admitted that the statements could not be substantiated for the class in general. They applied for leave to amend the specification so that in effect it only claimed two bodies sulpha-thiazole and suplpha-methylthiazole whose therapeutic properties had been described in detail in the original specification and which had proved to be of great value. In the House of Lords the patentees lost on procedural grounds (leave to amend the claims) but the speeches contained important views upon which reliance was placed by Mr Gratwick, in particular a passage in the speech of Lord MacDermott at page 50, lines 12 to 50:-
“The nature and extent of the invention claimed by the amended specification is readily ascertainable when this document is read in the light of a knowledge of the art. It is the manufacture of two new compounds of therapeutic value — sulphathiazole and sulphamethylthiazole. The problem was to find some substance at once less toxic and more effectual against certain infections than sulphanilamide. Each of the new compounds provided a solution. They were produced by known methods from known materials and the inventive step lay in the discovery of their virtue as drugs. This achievement was not the result of applying any known principle or law of nature. It was entirely empirical. No doubt the inventors used skill and experience of a high order in selecting a profitable line of research. But they had no means of knowing that success had crowned their efforts until the new products had been tested by experiment. There is conjecture but no certain knowledge as to how these drugs arrest infection. The chemistry of the human body still holds many mysteries, and all the organic chemist can do in this sphere is to explore in what he thinks a promising direction and then subject his results to rigorous trials in order to find if his hopes have been well founded. That such was the state of affairs at all material times is, I think, beyond question on the evidence. Tomorrow it may be different, for the bounds of knowledge are ever widening. But for the purposes of this appeal the invention claimed by the amended specification can only be regarded as empirical in character and dependent upon the proved worth of the specific substances described by the inventors.
Before proceeding to consider the original specification and the nature of the invention it claims it will be appropriate to mention two matters which, while this particular art remains in an empirical state, appear to me to be necessary consequences of that characteristic. In the first place an invention in this chemo-therapeutic field must be in respect of a substance which has actually been produced. There cannot be an empirical discovery in respect of a bare formula. And secondly, the discovery of each new compound having a therapeutic value is a separate invention. If the inventor is bound to say — ‘I have made a new substance which I find has therapeutic value, but I cannot be certain that any other substance, no matter how similar its molecular structure, will have such a value until I make and test it’ then as it seems to me, the inventive step he has taken must attach to the single substance he has made and to it alone. And if he has made and proved several such substances the position must, I think, remain the same for, while the art retains its empirical nature, the worth of each new substance is a new discovery. But when the inventor can say that his inventive step is such that each of the various new products which manifest it must have therapeutic value, and that although some of them have never been made, then, as I see the matter, the state of the art will have changed. It will have lost its empirical nature, at least to some extent, and the chemist will have found some law or principle by which he may predicate therapeutic effect in advance.”
12.15. In relation to this authority Mr Gratwick was at pains to emphasise to the court that his claim was not to a substance with therapeutic value as was the case in the May & Baker case but a claim to a discovery of something that nobody knew:-
“Mr Gratwick: . . . Everybody knows it. What they do not know is how to make t-PA synthetically. My clients made a discovery, an important and valuable discovery, in finding, as one can put it, either the insert, or finding the identity of the insert, its sequence, which would enable t-PA to be made synthetically.”
. . .
“My Lord, I am not claiming because it has therapeutic value. This is the point. It is the central point in this issue between myself and my learned friend. My position is this: we have discovered synthetically how to make that material of known therapeutic value. It is inherent in the discovery that we have made that we enable you to make a host of other substances as well; a whole host of them.”
. . .
“My learned friend puts his case on the basis that I have not discovered how to make it, he is saying no more than everybody knows how to discover a goldfield. You just go out into the wilderness with a hammer and chip at the rocks. My Lord, in truth, to say I have discovered how to make it does include the identity or sequence of that insert because you cannot say you have discovered how to make t-PA synthetically unless you have also discovered the identity or the physical embodiment of that insert. To say otherwise is merely to say you know something about how to make it; but you do not know how to make it.”
These submissions by Mr Gratwick were made in answer to questions from the court and, in my judgment, contains the central point of this appeal as he indicated, namely: What is the “discovery” and what is the “invention”, and how are these related to the inventions as claimed in accordance with the defintion in section 125 of the Act? Mr Gratwick correctly submitted that the “inventive step” in both aspects relates not to the therapeutic qualities of t-PA but how to solve the problem of producing t-PA in quantities viable for commercial exploitation. This is in the context that naturally obtainable sources were considered to be wholly inadequate. The “discovery”, if this is the correct word, consisted of determining the DNA sequence so that an expression vector could be constructed which incorporated the gene encoding for t-PA. Subject to questions of obviousness, both elements necessary to establish an inventive step are, in my judgment, present. There are the discovery of plasmid resulting in the latter’s alteration and adaptation by the incorporation of the new idea. This is in line with the judgments in Hickton’s case and the speech of Lord MacDermott in the May & Baker case where it refers to the empirical discovery having special properties leading to an inventive step.
12.16. In one sense the figure 5 data is not a discovery such as the therapeutic drugs in the May & Baker case, or the species of micro-organism in Chinoin’s Application  RPC 39, but rather the discovery of a formula leading to a commercially desirable technical process analogous to the processes in the Vicom and Hickton’s cases. Whether the figure 5 data is more appropriately considered under “discovery”, “mathematical method” or under the general rubric “among other things” may not matter. Once it is accepted as “new” in the sense “not available to the art”, it is probably best considered in its own particular light. As I have already pointed out, the figure 5 data was not part of the state of the art at the relevant date. In my judgment it does not lose its quality as a discovery merely because it was information about something which was known to exist and had in small quantities been isolated from other substances, including proteins with which it existed in nature. This leaves the question, which hitherto I have studiously begged, namely the impact of the words of section 3 of the 1977 Act.:-
“An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art . . .”
Although this is part of the issue of “Patentability” I propose to deal with it in a section of its own.
13.01. During argument much use was made of a “Dr Simpkin” — a hypothetical character who might well have made a thirty-first character in the catalogue of Theophrastus. As a shorthand mode of expression Dr Simpkin was clearly useful; but beyond this with respect to his promoters his usefulness must be qualified. The Act speaks in unqualified terms of “a person skilled in the art having regard to any matter amongst all the matters (whether a product, a process, information about either, or anything else) which has at any time . . . been made available to the public . . .” Although the epithet “ordinarty” has occasionally been added. I venture to think that the application of any qualification is dangerous. Even a degree of initiative, which is only another word for inventiveness, cannot be excluded and with great respect to Whitford J the path along which he ventured, namely the identification of actual scientists who have “Simpkin-like” or “supra-Simpkin” qualities, was one which should be trod with a degree of reservation. Expressions such as “a person ordinarily skilled in the art” or “a hypothetical workman of ordinary skill” had gained common usage in patent law but had never before 1977 achieved statutory status. Such expressions had significance in the world of inventing engineers producing blueprints upon which the skilled mechanics worked. In the modern world of science such as the sphere un which teams of highly qualified molecular biochemists, protein sequencers, protein chemists and others work, the artisan has receded into the role of laboratory assistant and the others have become segregated into groups of highly qualified specialists in their own spheres of all whom must possess a high degree of inventiveness. If this were not so the specalists would not have been there at all. Clearly some “prima donnas” in the form of Nobel Prize laureates can still be identified, but I do not think that these comparisons can be more than marginally helpful.
13.02. All are skilled in the art and the question posed by the section is whether the “invention” would be obvious to such a person with all the knowledge which had been made available to the general class of such scientists at the material date. If I may express a preliminary thought at this stage, the question is: What is it that must be “obvious”? It was certainly obvious that t-PA would consist of a collection of amino acids associated in some kind of defined molecular structure: but it was not known at the material date what that structure was. Does it make any difference that the information which was not available could have become available (a) without any substantial effort on the part of a team of persons skilled in the art; (b) only after considerable effort on the part of such persons, or should some other criterion be applied to define the word “obvious”? There is considerable attraction in the propositon that (a) is obvious, whilst (b) is not obvious within section 3; but such a test may be too simplistic.
13.03. One case which is of assistance in this consideratio involved “high technology” although in a different field is Valensi v British Radio Corp  RPC 337. The judgment of the Court of Appeal, against which leave to appeal was refused by the House of Lords, contains some helpful passages to which I would like to refer at page 377, line 7:-
“We think that the effect of these cases as a whole is to show that the hypthetical addressee is not a person of exceptional skill and knowledge, that he is not to be expected to exercise any invention nor any prolonged research, inquiry or experiment. He must, however, be prepared to display a reasonable degree of skill and common knowledge of the art in making trials and to correct obvious errors in the specification if a means of correcting them can readily be found. (Emphasis provided).
We agree with the judge that where the art to which the invention relates is so complicated and difficult as that of colour television it would be inappropriate to suppose the addressee to be a workman on the shop floor. The jduge took the view that in 1939 anyone considering manufacturing apparatus within ‘Valensi’ patent would refer the specification to one of the four or so existing design research teams in this field, and that such a team should be taken as the notional addressee. Such a team would include men such as Dr White, the very highly qualified gentleman called as an expert witness for the plaintiffs, or Mr Bedford, the very highly qualified gentleman called for the defendants. In our view this is too wide a departure from the principles laid down in the case referred to. Dr White and Dr Bedford were already, in 1939, men with high inventive capabilities, and it is somewhat artificial to suppose them studying the invention and trying to make it work while studiously avoiding any inventive step of their own. In 1939, when research into colour television had been going on for a number of years, it is obvious that between the leading experts and any manual workers who were construcing experimental sets there must have been skilled technicians from whom a team could be constituted which would possess the common knowledge of those in the relevant arts but would not have the capacity to make elaborate additions to or modifications of what was disclosed and illustrated in a specification. It is to a team of this category of people that we think the specificaton should be taken to be addressed.
Further, we are of the opinion that it is not only inventive steps that cannot be requied of the addressee. While the addressee must be taken as a person with a will to make the instructions work, he is not to be called upon to make a prolonged study of matters which present some initial difficulty: and, in particular, if there are actual errors in the specification — if the apparatus would really not work without departing from what is described — then, unless both the existence of the error and the way to correct it can quickly be discovered by an addressee of the degree of skill and knowledge which we envisage, the description is insufficient.”
13.04. It is true to say that in the passage cited, the court was considering not obviousness but “sufficiency”: but by analogy the approach adopted by the court in defining a person skilled in the art in high technology cases is valuable. Mr Prescott submitted that unless Mr Gratwick could establish the submission which he made in the Valensi case, namely that there was a different standard of skill to be expected of the person skilled in the art in secton 3 from the addressee of the patent for the purposes of section 72(1)(c) he would be in difficulties since the skills required to product t-PA applying the specification were no less than the skills required to determine the figure 5 data. Viewed purely from the necessity to exercise the skills generally required in what might loosely be called “genetic engineering” this submission clearly has some force; but it ignores the extra knowledge available to the addressee of the patent, namely the figure 5 data itself. In passing I would say that I see no mandate for applying a different meaning to the words “skilled in the art” in section 72(1)(c) from that in section 3. The more apposite question is whether, and if so to what areas of the science does the word “obvious” extend beyond the “state of the art” as defined. The authority upon which Mr Gratwick relied in the Valensi case was King, Brown & Co v Anglo-American Brush Corp (1892) 9 RPC 313; but this does not directly assist in the case, nor did I understand Mr Gratwick to be relying in this case on any such distinction in the meaning to be attributed to the words “a person skilled in the art” in the two sections.
13.05. Mr Gratwick placed considerable reliance upon the complexity of the operations, and the skill and persistance shown by the Genentech researchers which enabled them to discover and produce the full cDNA clone for t-PA, as constituting an inventive step. He referred to part of the dissenting speech of Lord Diplock in American Cyanamid Co (Dann’s) Patent  RPC 425 at 451, where Lord Diplock referred to the:-
“. . . searching for and finding hitherto unidentified strains of micro-organisms existing in the natural state from which useful new antibodies can be prepared by what is now a well-known standard process. The task of finding such a strain of micro-organism calls for the exercise of technical proficiency and is laborious and very costly, for the odds against success are large. It is not easy to see what inventive step, as distinct from the mere exercise of proficiency and practice, is involved in this kind of research, but the result of success in it is a new product useful to humanity which does not exist in nature. If such research is to be encouraged in a competitive society, the monetary rewards of success must be assured to those who undertake the expense; and the means of doing so in this and in most other countries with comparable social systems is by according to the successful discoverer of the new product the controlled and limited monopoly granted for inventions under the national patent laws.”
. . .
“It is in my view consistent with this basic policy to treat the kind of research involved in the discovery of a strain of micro-organism from which a new and useful antibiotic can be prepared as an activity which entitles the person who undertakes it to a temporary monopoly under the Patents Act 1949, of the product of his success. I accept, therefore, the extension of the concept of ‘invention’ to include antibiotics which are discovered through this kind of research.”
However, as Mr Jacob pointed out, there is a clear distinction between the final product in the American Cyanamid case, which was hitherto unknown, and t-PA which was known to those skilled in the art.
13.06. Mr Gratwick also relied upon what are known as the selection patent cases. In these cases there exists some general field of knowledge, eg a class of chemicals known to be dyestuffs or to have certain recognised properties, within which there may exist particular members or groups of members which are found to have special properties not possessed by the other members in the field. He cited from the locus classicus — the judgment of Maugham J in IG Farbenindustrie AG’s Patents (1930) 47 RPC 289 at 321:-
“It may be observed that chemical patents in recent years have consisted of two sharply divided classes. The first class is that of patents based on what may be described as an originating invention, that is, the discovery of a new reaction or a new compound. Such patents may be called for brevity ‘originating patents’. The second class comprises patents (the so-called selection patents) based on a selection of homologues and substitution derivatives of the original compounds which presumably have been described in general terms and claimed in the originating patent.”
“It is evident that the inventive step involved in the originating patent, for example, such a step as the Griess reaction to which I have above referred, differs in kind from the systematic investigation or research required to ascertain that some of the combinations possible under the originating patent, for example, selected arylides and diazo compounds made according to the prior specifications, possess distinctive and it may be unexpected properties. The question has been raised in the present case whether it is possible to show subject-matter in respect of a selection patent in the sense in which I use the word. I have come to the conclusion that such a patent may well be valid and that properly considered there is no more difficulty in such a case in establishing subject-matter than there is, say, in a mechnical or a combination patent. It must be remembered, of course, that the selected compounds have not been made before, or the patent would fail for want of novelty. If the selected compounds, being novel, possess a special property of an unexpected character, for example if a monoazo dye were to be made by selecting components not hitherto employed which resulted for the first time in a green dye, I cannot see that the inventive step essentially differs from the step involved in producing a new result by a new combination of well-known parts or indeed from using the common and well-known factors (cranks, rods, toothed wheels and so forth) employed in mechanics in the construction of a new machine.”
13.07. Again, it must be observed that in the IG Farben case Maugham J was dealing with a known substance which turned out to have unexpected properties, whereas the known properties of t-PA were themselves the “raison d’etre” of the research in the first place. The observations of Maugham J were approved by the House of Lords in Beecham Group Ltd v Bristol Laboratories Ltd  RPC 153.
13.08. Mr Gratwick further submitted that the further step of determining the figure 5 data was impossible to the man skilled in the art, however long and persistently he worked within the existing state of the art but without taking an inventive step. In support of this proposition Mr Gratwick relied upon a number of propositions amongst others to which I have already made passing reference in paragraphs 5 to 7 (supra). These were the following:-
(1) If Dr Simpkin, when he came to screening his cDNA clone library, followed the Suggs paper he would have failed. He would have washed away the target clone and been left with “background”. Reference was made to the Suggs paper in the textbook by Professor Maniatis. Although this was published shortly after the priority date, it was in draft at the material time and, moreover, was based on the teaching at CSH. If Dr Simpkin washed the target colony away, he would never realise his mistake and would assume that the librry did not include the target clone.
(2) The work carried out at Umea by Drs Ny and Edlund was beyond the capacity of Dr Simpkin in at least two respects. He would not have had the initiative or skill to switch to using the XeNOPUS laevis oocyte system to transcribe mRNA for t-PA when the rabit reticulocyte lysate system failed, and he would not have had access to the amino acid sequence data which was made available by Dr Wallen which enabled the Umea team to have oligonucleotide probes synthesised by Kabigen AB in Stockholm. Additionally he would not have devised “negative controls”.
The first proposition promoted a good deal of controversy. Mr Jacob produced calculations to show that in the case of searching for the t-PA clones the chance of losing the target clone and retaining a false positive or positives was in any event insignificant. More significant in my judgment, however, was the evidence that normally skilled workers did not wash out all but one or two colonies but as a matter of prudence always stopped well short of this position. Again, I am not convinced that there is any relevance in proposition 2. The work from which Dr Wallen derived the amino acid sequence data was routine laboratory work. In relation to the use of oocyte cells, Mr Gratwick was on stronger ground, and it could be argued that this decision might have been outside the capacity of Dr Simpkin. Furthermore, the introduction of negative controls in the screening process was submitted to have been beyond the ability of Dr Simpkin. I would not, however, be prepared to decide the issue of obviousness on these points alone.
13.09. I have, therefore, come to the conclusion that I cannot accept ths submission without some qualification and I am not sure that it goes to the real question, namely, whether or not the necessary step from the idea of making t-PA by recombinant technology to its eventual attainment could fairly be described as an “obvious one” to the teams of highly qualified persons skilled in the art. Mr Jacob identified this distinction at Court of Appeal transcript Day 11, page 37B:
He (Mr Gratwick) slips between obviousness of the idea and obviousness of how to implement the idea.”
Having the idea is certainly not enough — the legends and philosophy of transmutation in Greek and Arab cultures could not be said to have been discoveries; nor, I would venture to think, would be the efforts of the alchemists in the thirteenth century to convert base metals into gold be described as obvious, even though they had objectives which obviously had desirable ends and were attended with hopes of success, now known to be unfounded.
13.10. Mr Jacob based his assault on obviousness on a number of propositons which follow. He submitted that there was a strong commercial incentive to discover the amino acid and DNA sequences of t-PA and that there was a high prospect of success. He distinguished on these facts the expression of opinion of Lord Diplock in his dissenting speech in American Cyanamid Co case and other cases upn which Mr Gratwick relied:-
“Mr Jacob: Persistence alone is not quite enough and an important part of it is the expectation as you do this, because what we find in this case is a reasonably high expectation of success, and a series of stages through which people have to go to get there, but they know what they are looking for, whereas in the Lord Diplock case they did not know what they were looking for and just hoping that something would turn up, and they did not know what that something was.”
13.11. Mr Jacob also relied upon Johns-Manville Corp’s Patent  RPC 479, in which the Court of Appeal held that an alleged invention was “obvious and clearly did not involve any inventive step” if it could be proved, as it was, that a person versed in the art would assess the likelihood of success sufficient to warrant trial. This case involved the use of a flocculating agent in a process of manufacturing asbestos cement pipes. Notwithstanding the evidence of an expert called by the patentee the Court of Appeal upheld conclusions reached by the Patent Office and the Oatents Appeal Tribunal that the idea of using such an agent in the manufacture of cement asbestos was obvious having regard to information in prior publications. Referring to the evidence of this expert Diplock LJ said, at page 495:-
“In so far as this witness obtained literature about flocculating agents used in other industries, and realised, as soon as he heard of them, that polyacrylamides were well worth trying out as flocculating agents in his own industry of manufacturing asbestos cement pipes, his evidence confirms the opinion of the superintending examiner and the Patents Appeal Tribunal that the idea of trying out these newly-introduced flocculating agents in the filtration process in that industry would be obvious to persons ‘versed in the art’.
‘His failure to perserve with his experiments, when he found that the skin of asbestos cement upon the felt filter was too thick, would be cogent evidence for the appellants if the invention claimed in their specification included an adjustment to the speed of the filter belt.’
That is, if the key to success was to be found not just in the idea, but also in this further thing you must do, and that was a feature of their claim, then that would be relevant, but there is not a word about this in their specification.
‘If, appreciating the necessity for such an adjustment involved any inventive step, the specification could be attacked upon the alternative ground set out in section 14(1)(g), namely that it ‘does not sufficiently and fairly describe the invention or the method by which it is to be performed.’ But it is (and so far as the appellants are concerned, it has to be) common ground that, once the idea of adding polyacrylamides to the asbestos cement slurry used in the manufacture of asbestos cement pipes has been tried out, and the thicker skin of asbestos resulting from the improved filtration observed, the necessary adjustment to the speed of the filter belt to obviate any deleterious effect upon the qualify of the final product would be obvious, notwithstanding that the respondents’ own research manager did not find it so.'”
13.12. Mr Jacob also emphasised that such was the way in which the specification had been drawn and in light of the failure to deposit plasmids the skills required to implement the specification were no less than the skills required to perform the full operation of preparing t-PA by recombinant DNA technology. In adopting the last paragraph of Diplock LJ’s speech just cited, he submitted that either Mr Gratwick failed on obviousness or, if he did not fail on this aspect, he must fail because the patent did not sufficently describe the invention. This was a submission echoed by Mr Prescott. I deal further with this submission in paragraph 14.12.
13.13. On this aspect of the case Mr Jacob had, in fact, to contend that Whitford J was wrong to have ignored his submissions on obviousness. These would have been inconsistent with Whitford J’s finding in relation to claim 9. Mr Gratwick, on the other hand, submitted that the facts of the Johns-Manville case were quite distinguishable. Mr Gratwick referred the court to the consideration of this case and the case of Olin Mathieson Chemical Corp v Biorex Laboratories Ltd  RPC 157, in the judgment of Graham J in American Cyanamid Co v Ethicon Ltd  RPC 215. This case concerned a patent relating to the manufacture of surgical sutures. It is perhaps helpful to cite three short paragrahps from the judgment of Graham J at pages 266 and 267:-
This objection– see paragraph 2 of the Particulars of Objections — though based as is usual on the effect of the prior documents cited, was in fact argued on paragraph 2(b) of the Particulars. In effect what the defendants contend is that there was nothing inventive in their plaintiffs’ original basic conception and in their invention as claimed in claims 1 and 2. This alleged invention, it is said, was obvious because on the authority of Johns-Manville Corporation’s Patent  RPC 479, it was at the material date ‘obvious to try’ to make a suture out of polyhydroxyacetic ester, but if one then carried out this obvious step one obtained something which had no utility for practical purposes. The argument continues that admittedly commercially useful and valuable sutures such as Dexon and XLG were obtained and put on the market from 1970 onwards, but this was entirely due, as a matter of substance, to processes and methods discovered subsequently by the plaintiffs in the further development of their sutures. This argument was referred to in the case as the Tetra-Molectric argument from the case Tetra-Molectric Ltd v Japan Imports Ltd  RPC 547, where the same argument, but applied to essentially different facts, was successful. Both the arguments that an invention was ‘obvious to try’ and that the invention was entirely due to later research are sound arguments if the facts justify them, but one must be certain that they do before invalidating a patent on either ground. In, my judgment, here the facts do not justify the application of either argument for the following reasons.
“In the Johns-Manville case the claim was for a process of producing shaped asbestos cement articles by a filtration process which was characterisecd by the addition of slurry of a small specified proportion of a water-soluble high molectular weight polyacrylamide. Now although this particular flocculating agent had only recently come on to the market, it was known and sold as a flocculating agent and was recommended for use as such by its manufacturers at the date of the patent, in fact, in the proportions actually claimed by the patentees. Diplock LJ, after reciting these essential facts, went on to make it clear that obviousness had to be related to the particular facts and the particular claim of the case in question. There the facts were such that the person versed in the art would assess the likelihood of success as sufficient to warrant actual trial of that particular substance. There is, to my mind, all the difference between the assessment of the likelihood of success in that case, where the actual substance was already on the market and suggested by its makers as valuable for flocculating purposes, and the case here where no such facts exist.”
“Though in any given case the decision on obviousness in such circumstances must be a question of degree (as indeed Mr Gratwick accepted, Speech 13 page 8F), it seems to me that to regard as obvious the choice of a starting material not then available on the market, which had only been suggested as one of thousands of other candidates for a research programme to find an end product with particular qualities, is a very far cry from suggesting the trial of a starting material available on the market, whose characteristics and capabilities are known, for use in a process where those capabilities are required. Here I have no hestitation in saying that this is not a Johns-Manville case and that, in the circumstances here as shown by the evidence, it was ‘not obvious to try’ PHAE within the meaning of Lord Diplock in that case.”
13.14. With respect to Mr Jacob’s attractive argument, I have come to the conclusion that in the circumstances of the present case little is to be gained from assessing either the prospects of success or the commercial motive for further research, and that Mr Gratwick is correct in submitting that the real question is whether the Dr Simpkin holding his PhD but having no extraordinary inventive capacity could within a reasonable time have discovered the figure 5 data, because if he did then it is common ground that he would have been able to construct the appropriate expression plasmids required to express t-PA. It is, therefore, necessary to consider whether Mr Jacob had discharged the onus which rests upon him to show that Dr Simpkin’s team would have achieved the objective without exercising inventive capacity.
13.15. I have to confess that I have found this to be a difficult aspect of this appeal. Returning to the sheet anchor of the 1977 act itself, and considering the real discovery upon which the invention is based, it cannot be said that the figure 5 data was obvious to a person skilled in the art at the priority date. The existence of the data was obvious but what it was was unknown. It was, therefore, something which was not obvious to anyone and which was the target of a considerable number of different research teams exercising, I have no doubt, very great skills if not inventiveness. The present position is clearly distinguishable from the state of affairs (Johns-Manville case) where persons skilled in the art, although ignorant of the necessary fact, can, by ready reference to an available source, discover that fact.
At the same time, it must be said that the man-months, if not years, spent by the four teams working on the project, by their quantity alone, would tend to disprove that the figure 5 data could ever be described as obvious in the sense suggested by Mr Gratwick as being the proper test, ie to be found in the way or in the path of the worker. This is in contrast to the discovery of something which was not in the way although readily recognised as a desirable objective.
13.16. Mr Gratwick, in reply, based his case on the failure, on the part of the respondents, to discharge the onus upon them:-
“(Mr Gratwick) I make it clear that it is not my case, and nor has it ever been, that what other teams did on individual matters was itself inventive. We simply did not go into that enquiry in the evidence because it is no part of my case. I simply say such and such a thing was not generally available and was of help to the team, and then I throw down the challenge to my learned friend, ‘How do you say Dr Simpkin could have managed without it?’, and if he cannot give an answer which satisfies your Lordships then in my submission his case fails.”
13.17. The numerous and technical steps involved in this process, in my judgment, carry the matter beyond the obvious activities of those skilled in the art, as opposed to the activities of those bent upon research and discovery. In a sense, the whole existence and activity of persons skilled in the art is one of invention and discovery and this serves only to highlight the artificiality of attempting to import the test of obviousness demanded by section 3 of the 1977 Act into this world of high technology. Giving proper weight to the canons of construction which are binding upon us with particular reference to Article 69 and its Protocol, I do not think that fairness will be achieved as between the patentee and the rest of those involved in the practice of the art concerned if, through the application of the principles of obviousness, it was to be held that there was no inventive step involved in the discovery of the figure 5 data. In coming to this conclusion I have applied the test which I tentatively proposed in paragraph 13.02, which I think is in line with the approach adopted by Lord Diplock in American Cyanamid (Dan’s) Application and the judgment of the court in the Valensi case.
13.18. A few aspects of Genentech’s work between October 1980 and December 1981 will suffice to exemplify the point made by Mr Gratwick that in considering this effort one must look at the work and the discovery not as individual achievements to assess their inventiveness but as the total work and the result. As indicated in the Points of Agreement No 5 the probability of obtaining a particular cDNA clone was uncertain. The history shows that never before had a clone of the length of cDNA for t-PA having an mRNA with such low abundance been successfully isolated.
In determining whether the overall effort was within the capacity of the hypothetical team of Drs Simpkins one must bear in mind the inventiveness and initiative involved in the orchestration of the individual efforts of each specialist within their own sphere such as that achieved by Dr Goeddel assisted by Dr Pennica. In particular I have in mind the decision to shift the area of investigation of the amino acid sequencing from the N-terminus towards the C-terminus of the A chain on receiving the report from Dr Shiveley in April 1981 (see paragrah 4.07 supra). This led to the identification by William Kohr at the end of August 1981 of the mino acide sequence WEYCD.
There were other exceptional features, including: the decision by Dr Goeddel to obtain a human genomic clone library; to reprobe the libraries produced by Dr Holmes with the WEYCD probe and a polysomal probe concentrating on selected potential positives which led to the identification of Clone 25-E-10 (see paragraph 4.13 supra); and the recognition, notwithstanding the disappointing results from the reprobing exercises by Dr Pennica, of the amino acid sequence EEEQKF in the information obtained from William Kohr during preceding weeks; and finally, the work of Dr Goeddel’s instructions by Dr Pennica in January 1982 upon the human genome library which led to the identifiction of PA17 and the projection of the enquiry to the N-terminal region of the protein.
Although each of these individual steps were technical operations recognised by the state of the art, their orchestration into the successful discovery was, in my judgment, well outside the capacity, with its limited inventiveness, of the hypothetical team of Drs Simpkin without a lead given by a person of supra-Simpkin qualities.
13.19. I am fortified, in coming to this conclusion, by three other factors. Firstly, this was clearly the view of Whitford J (see page 591, lines 15-25). Secondly, regardless of the various degrees of success achieved by the competing teams, to which reference has already been made, all of them were either under the control of, or had access to, scientists of greater merit and inventiveness than the ordinary person skilled in the art and, in most cases, consisted of individuals themselves possessed of that degree of inventiveness exceeding the normal amount of initiative with which the hypothetical person skilled in the art is supposed to be endowed. I believe that this emphasises the artificiality of the criteria involved in this particular exercise and exposes the weakness of relying upon the various points of progress made along the road towards the synthesis of t-PA reached by each of the teams concerned in their own individual contexts.
Finally, my conclusion is in accord with the view of the Technical Board of Appeal, as expressed in Decision T292/85 dated 27 January 1988, in the appeal by Genentech Inc against the decision of Examining Division 023 of the European Patent Office, dated 15 March 1985, refusing patent application No 78300596.0 pursuant to Article 97(1) EPC. This decision, although concerned to a great degree with the “Polisky et al paper” published in the proceedings of the National Academy of Science USA, 1976, contains general opinions of value.
The question whether there was a discovery which was new was not in doubt (see paragraph 5 of the Reasons for the Decision). On the question of inventive step there is a very full and detailed consideration of circumstances which are in parallel to those in the present appeal. It would not be helpful to burden this judgment with a full recitation; but one quotation, hopefully not unduly eclectic, may be helpful:-
“6.8 In spite of the fact tht Polisky points away from the present invention and towards a different strategy to achieve proper expression of the whole DNA gene, the skilled person might have been keen and anxious to solve his problem already recognsied in the literature as the problem of biochemistry. If by some inspiration he had hoped to improve the less promising second suggestion in the paper and to try the use of bacterium’s own regulon and even to do something about proper reading frame, he would have been likely to choose a relatively simply DNA to produce a small polypeptide for insertion in order to test the idea. This is because the synthesis of small genes was relatively simple and so was the identification of corresponding polypeptides in the result. On the other hand, the synthetic preparation of large genes was cumbersome or impossible and it was difficult to obtain, identify and reverse transcribe the information from natural mRNA to cDNA for insertion, and then test the result. This would have been an enormous task and unless there was full knowledge as to the structure of the insert and the resulting polypeptide, the experiment would be useless to demonstrate whether or not the problem has been solved. Thus, a modest start with a small polypeptide, possibly encouraged by the report that some atostatin was successfully translated, would have been his aim, which should, as we now know, have led him into a failure in view of the destruction of the result of proteolytic enzymes.”
14. Invention and claims
14.01. I have arrived, after what I fear may have been a great deal of circumlocution, at the following conclusions:-
1. That the work carried out by Genentech was inventive in the sense that it led to a “discovery”, namely the complete DNA sequence coding for t-PA, together with the identification of the start and stop codoms to orientate it within the correct reading frame for recombinant DNA expression.
2. That Genentech also discovered the sequence of amino acids making up the full length mature molecule of the protein t-PA and a great deal about the features of its molecular structure that endowed it with “fibrin-binding” properties. These were crucial to the commercial target which was in everyone’s sights.
3. That the information imparted by the discovery, namely, the figure 5 data, was not “obvious” within the meaning of the word in section 3 of the 1977 Act.
4. That the figure 5 data was capable of being “the idea underlying (an) “invention” within the sense adopted in the Vicom decision cited earlier in this judgment and the judgments already cited in Hickton’s case. Apart from Mr Jacob’s submissions based on Falconer J’s judgment in the Merrill Lynch case there was little argument about the patentability of the two plasmids referred to in claim 9 which Whitford J indicated he would have held to be valid. However, Mr Gratwick indicated that a patent limited to claim 9 would be virtually useless and Mr Jacob told the court that whilst he did not accept that such a patent would be valid, his clients were not using these vectors and were not, therefore, infringing. Mr Jacob submitted, however, that these claims would fail, in any event, for lack of subject matter.
14.02. It must be recognised that the distinction between Mr Jacob’s submission based on Chamberlain & Hookman Ltd v Mayor of Bradford, and Mr Gratwick’s submission, based on the decision in Vicom and the judgments in Hickton’s case, is, in the present circumstances, a fine if not subtle one. The question to be asked is whether the “invention”, as claimed within the criteria of section 125 of the 1977 Act, is a method embracing a new discovery, in which case it escapes disqualification under section 1(2)(a); or whether it is merely a claim in the form: “I claim this discovery harnessed to make useful artefacts”, in which case it amounts to a claim to the discovery “as such” and, therefore, is not an “invention for the purposes of this Act” (see section 1(2)).
Thus, a claim describing a method embracing the discovery can be an invention for which a claim has been made or a patent granted within section 125(1); whereas a claim which does not describe a method embracing the discovery is not an invention since it is a claim for the discovery as such and cannot be an invention for the purpose of section 125(1). The latter claim fails not because it does not comply with the provisions of section 14(5), although the grant might well be withheld under this section; but because it cannot be an invention for the purposes of section 125(1). Section 125(1) cannot convert something which is not an invention into an invention merely because it purports to be claimed as such in a claim of the specification.
14.03. Section 125 provides that the extent of monopoly shall be specified in a claim for the specification subject to interpretation by reference to the description and any drawings contained in the specifications and the Protocol to Article 69 of the European Patent Convention. In my judgment one must look, in the first instance, to the wording of the claim or claims themselves in order to see whether the essential characteristics of the invention in the claim are in line with the criteria in the decision in Vicom and the judgments in Hickton’s case. If they are not, then, in the circumstances of this case, the claim cannot be describing a process or product which could be an invention for the purposes of the Act. It would either be a claim for the discovery “as such”, or it would not refer to anything done by the patentee which could be an invention regardless of section 14(5).
14.04. Although the speeches in Mullard’s case (supra) were dealing with width of claims and ambiguity, Mr Jacob was, I think, entitled to rely upon them for guidance as to how to approach the requirements in section 125 that the invention must be specified in the claim. At page 351 per Lord Roche:-
“On grounds of law two main contentions were raised in objection to the judgments in the Court of Appeal. The first was that they seemed to require the inventive step to be stated in the claim, and that was not necessary. The second was that they seemed to require the purpose or use of the article claimed for to be stated in the claim, and that was not necessary. In my view the Court of Appeal did not make such requirements but something different and in the circumstances something that was essential to the validity of the claim. It is true that an inventor need not state in a claim the reasons that have led him to his inventon or the stage or stages by which he has arrived at it. But the essential characteristics of his actual invention he must state. As to use or purpose it seemed to be contended for the appellants that to include it in the claim was always unnecessary and even inadmissible. Cases were cited in support of this contention, but I do not think they support it” (Emphasis provided)
(at page 352)
“The true principle I conceive to be this: That some articles when seen or described speak for themselves and further description as to function or purpose is needless. On the other hand, in some cases function is part of the invention and in such cases it not only can but must be included in the description to secure the validity of a claim.
“If reference to function is therefore necessary, mere reference to amplifying derived from the title is clearly, as I think, an inadequate description of the invention. At the risk of repetition I state my view to be that, wheres the invention was of a valve of a particular kind distinguished from other valves by the position and potentials of its various electrodes in a circuit, in claim 2 the major part of the characteristics of the invention find no place. Moreover, if function be thus an essential element in the invention and the claim for an article unrelated to function, then it is I think illegitimate to resort to the body of the Specificaiton to find the invention and thence derive subject-matter for the claim.”
Subsequently, Lord Roche referred to an extract of the speech of Lord Loreburn in Natural Colour Kinematograph Co Ltd v Bioschemes Ltd (1915) 32 RPC 256 at 266:-
“Some of those who draft specifications and claims are apt to treat this industry as a trial of skill in which the object is to make the claim very wide upon one interpretation of it in order to prevent as many people as possible from competing with the patentees’ business and then to rely upon carefully prepared sentences which it is hoped will be just enough to limit the claim within safe dimensions if attacked in court . . . This is all wrong. It is an abuse which a Court can prevent whether a charge of ambiguity is or is not raised in the Pleadings.”
At page 353 Lord Roche continued:-
“Ambiguity which was in question in that case is not one form of species of the genus wrong description. Excessive width of claim which is here in question is another, and in many cases it may be a more mischievious offence against the Statute than ambiguity. It is to be observed that I am not dealing with insufficiency of description in the sense of insufficiency of working directions in the specification.”
14.05. Mr Jacob also supported his attack on the claims in the present case by reference to Crabtree & Sons Ltd v Hoe & Co Ltd (1936) 53 RPC 443. Although this case also was dealing with the breadth of claim and, therefore, would not be available under section 72(1) of the 1977 Act, Mr Jacob was, in my judgment, entitled to rely on the general propositon that a patentee should escape the dilemma referred to in this extract of the judgment of Green LJ at page 450:-
“When this paraphrase is examined, it will be seen that it is worded in such a way as to escape a dilemma, each horn of which would destroy the appellants’ case. That dilemma is as follows: If the claim so far as it relates to interlocking is to be limited to the interlocking mechanism described in the body of the specification, there is no infringement; if, on the other hand, the claim is so broad as to include all means of preventing premature contact, it amounts to no more than the statement of a problem, since the inventor would merely be saying: The way to prevent premature contact is to provide means for preventing it. A patentee is not entitled in this way to obtain a monopoly of all possible means of remedying a known defect, nor did Mr Shelley argue that he could.”
By analogy with this judgment Mr Jacob submitted (Court of Appeal transcript Day 11, page 38H):-
“My friend’s claim is that the way to make t-PA genetic engineering is by making it by genetic engineering. That is what his claim is, and he is claiming (sic ‘not’) a monopoly of all possible ways of remedying a known defect but achieving a known objective. It is the same thing for practical purposes. That is that case. We see it is put on lack of subject-matter, that was the pleading.”
This submission appears to me to assert that nothing claimed by Genentech, including the figure 5 data, amounts to an invention, in which case there cannot be patentable invention upon which section 1 of the 1977 Act could operate. For reasons already appearing in this judgment I am unable to accept this argument in its full extent. On the other hand, I can accept Mr Jacob’s alternative argument, namely, that a claim which extends beyond the invention achieved is, to that extent, not an invention at all for the purposes of the Act. It is, I believe, another way of saying that for a claim to be an invention within section 125(1) and section 1 in the type of invention with which this appeal is concerned, it must describe the method which embraces the idea. If it does not, it is not a claim to an invention at all. (See the speech of Lord Macmillan cited in paragraph 11.13 and the speech of Lord Roche cited in paragrpah 14.04 in Mullard’s case).
14.06. It is necessary to consider in a little more detail the two stages at which Genentech can fairly claim that their “discovery”, ie the figure 5 data, has been the underlying discovery of an “invention”. There are two “products” which this discovery enable Genentech to make which were not available in the prior state of the art, namely, a specific nucleotide probe with which to extract substantial lengths of cDNA coding for t-PA, and expression vectors, including the DNA gene coding for t-PA. No claim has been made in respect of the former, which in any event would be of little or no commercial value now that the full molecular structures are known.
It remains to consider whether any, and if so which, of the other claims can be brought under the umbrella of a claim to an inventon which incorporates the figure 5 data as its underlying discovery, in contrast to a claim to the figure 5 data as such.
14.07. Mr Gratwick submitted that since Genentech made the discovery anyone subsequently using that information in any way should pay tribute. He included, in this submission, a person who used some new vector or method hitherto undiscovered which might, in the future, be itself patentable, but which used the figure 5 data. I regret that I am unable to follow this line of argument, which seems to me to be tantamount to claiming “an invention” consisting of the discovery as such.
14.08. Mr Jacob, on the other hand, submitted that if there was a patentable invention at all it should be confined to the plasmids actually used and specified in the specification (ie those described in claim 9). I wish to consider this submission a little further. The construction of vectors incorporating the t-PA gene forms no part of the state of the art agreed in the Points of Agreement, but the matter is dealt with in some detail in the description in the specification.
It is sufficent, for this judgment, merely to mention, by way of example, two processes of expression and amplification of t-PA in two plasmids: plasmid pET-PAER 400 transfected with dihydrofolate reductase treated with methotrexate (see paragraph E2B) and plasmid pETPFR treated in a similar way (see paragraph E3). These processes are set out in detail with references to the previous publications upon which they were based; but each involves the incorporation of the full t-PA encoding gene and are, therefore, new constructs. Each of these processes, in my judgment, could be claimed as methods embracing the figure 5 data and would be patentable. These parts of the specification must, however, be considered in the context of the paragraph at page 52, lines 13 to 16:-
“Notwithstanding that reference has been made to particular preferred embodiments, it will be understood that the present invention is not to be construed as limited to such, rather to the lawful scope of the appended claims.”
14.09. If, as I think may be the case, a claim to an invention under section 125 limited to plasmids in which the t-PA encoding gene is incorporated would be an invention which would be patentable and would not be a claim to the discovery as such, the above paragraph effectively negates this approach and exposes any wider claims to “invention” to the disqualification in section 1(2) of the Act.
14.10. The authorities seem to support the proposition that where the discovery is of a new substance, a patent can be claimed for that substance “however made” and will be valid as long as the specification discloses one method of manufacture, even if that method be not the most favoured one. This proposition does not apply, however, where, as in the case of t-PA, the discovery is merely part of the process by which a product already known to exist with properties already described can be manufactured. In the latter case the discovery can only form the basis of an invention limited to the method of producing the known artefact, ie a process patent.
The “vehicle” by which the “discovery” can become an invention need not itself be “new”, in the sense of section 2(1) in order to escape the disqualifcation in section 1(2)(a) if the principles in Hickton’s case apply. For this reason I cannot accept, without some qualification, Mr Jacob’s alternative submission that if his main submission that without the deposit of a cell line claims 9 and 19 failed to comply with section 72(1)(c), then Genentech should be limited to what they had actually claimed under claim 9. I am able to accept part of Mr Gratwick’s submission that Genentech should be entitled to protection not only in respect of their plasmids detailed in their specification but also in respect of plasmids or vectors readily available within the state of the art, or immediate derivations or variations thereof. I would extend this to include plasmids or vectors incorporating genes resulting from minor adjustments to the molecular structure of the gene disclosed in figures 4, 5 and 12 of the patent. I think that this accords with the sense of the Protocol to Article 69.
However, as I have already indicated, Mr Gratwick’s contention that Genentech should be protected against any use of this information, howsoever this may be achieved in the future, is, in my view, not a claim to a method embracing the discovery but rather a claim for protection of the discovery as such. There are two reasons for this: first, the claim is speculative as to the method by which the discoverey is to be embraced; secondly, the method is not capable of description and it must fail under section 72(1)(c). Again, I believe that this limitation does justice between the patentee and the public and is in accordance with the Protocol to Article 69.
14.11. I must revert to Mr Jacob’s further submission that the specification did not describe the process of expression by mammalian cell line sufficiently to enable the addressee of the patent to proceed to produce t-PA by expression and replication and that without the deposit of a cell line, the claims would fail under section 72(1)(c). In judging this question, as with the other issues raised in this appeal, I approach the solution by assessing the evidence afresh and forming my own opinion, see Benmax V Austin Motor Co Ltd (1955) 72 RPC 39. As I have already mentioned in paragraph 14.08 of this judgment, full details of the use of certain expression vectors are given in the specification leading to the expression and replication of t-PA in both prokaryotic and eukaryotic cells, and I can see no necessity for Genentech to have deposited a cell line.
14.12. In the appendix to his skeleton argument Mr Gratwick set out an expanded interpretation of claims 3, 16 and 17, relying upon the definitions and content of the specification. As an aid to understanding the claim this was clearly a useful excercise; but it must be remembered that under section 125(1) reference to the specification is for the purposes of interpretation only. As was made clear in Mullard’s case reference to the specification never was a means of restricting a claim which was otherwise too wide or lacked substance. In the present context one is not dealing with the width of claim in the old sense but the principles are the same when considering the invention claimed within the terms of section 125(1) in order to determine whether in law it is an invention which is patentable under section 72(1).
14.13. I now turn to consider the claims in the context of what I believe to be the only patentable invention which was achieved by Genentech, ie a method embracing their discovery of the full molecular structure of DNA coding for t-PA and the full amino acid sequences of the latter. Since in my view the method is an integral part of, and essential to, the invention it must be a method that is clearly identified and defined (see per Lord Roche in Mullard’s case). This must exclude any speculative element. Claims 1 to 6 all refer to the final protein product in one form or another and prepared by one method or another. In my judgment these are not inventions which fall within the class of invention recognised in the judgments already cited in Hickton’s case, ie a method of exploitation in which the “discovery” is used to achieve a desirable product, albeit one already known to science. They fail for the reasons claim 2 failed in the Mullard’s case, whilst claim 1 succeeded (see paragraphs 11.13 and 14.04 (ante)). Futhermore, they do not describe a method which embraces the discovery. I cannot, therefore, differ from the view taken by Whitford J that these claims cannot be supported.
14.14. I now turn to claims 7 and 8 which do relate to cloning vectors and expression vectors respectively. At Day 10 of the transcripts in the Court of Appeal, page 54B-55B, in answer to questions put by the court during the submissions being made by Mr Jacob, Mr Gratwick confirmed the way in which he puts the case in relation to these two claims and claims 16 and 17 in these terms:- “Mr Gratwick; My Lord, may I make no bones about it: as I put it in my happy way about a baking tin and the dough, once he has got the dough, any baking tin will put it in the oven for him, and we say that this is where claims 16 and 17 and claim 18 come in.
Dillon L J: And claim 7?
Mr Gratwick: Well, claim 7 is a cloning vector and that will only produce more DNA, whereas claim 8 will produce t-PA, and that is why I drew the distinction.
Dillon L J: But 7 and 8 are still classed generally; it is any old cloning vector and any old expression vector.
Mr Gratwick: Provided that the insert is that appropriate to numbers 1 to 527, figure 5.
Dillon LJ: That is the function effect of it really, is it not?
Mr Gratwick: Yes, my Lord.
Dillon LJ: Any old cloning vector that has the effect?
Mr Gratwick: No, my Lord, it is the vital difference between the generalised tool, which is the cloning vector (pBR320 to pBR345X1) and anything else; that plus a sequence (an insert: so it is clear I am talking about a molecule), an insert which codes for acids 1 to 527. That, we say, is what claim 7 is about. Claim 8 is about the same thing, save that the vector also has the mechanism for expression of t-PA as opposed to merely duplication of DNA inserts — that is because the claim relates to an expression vector. Then claims 16 and 17 are the complete process. If your Lordships would look at claim 17, the words are in the bundle, never mind what we put in the appendix to our skeleton. ‘A replicable expression vector capable of expressing the DNA sequence encoding human t-PA2’. I hope that makes my position clear to your Lordships, and I hope, helpful.”
14.15. Mr Gratwick returned to these matters during his reply (see Court of Appeal transcripts Day 16, page 17H et seq). The points he made can, I hope, be summarised as follows. He referred to the list of plasmids available at the material time shown in a book called “Gene Cloning” to which the court has access. This table listed a number of plasmids under their serial numbers, including pBR345 and pBR322 and others.
Having made the figure 5 data available a competitior could avoid the use of pBR322 and, if the claim was limited to products of that plasmid, then, by using the readily available alternative pBR345 and the information from the discovery, construct an equally effective cloning plasmid. This Mr Gratwick submitted, would deny the whole purpose of the protection sought.
Mr Gratwick extended the same argument to claim 8, which related not to the cloning vector but to the expression vector. Otherwise the points arising on each were the same. So long as the claim is limited to plasmids readily available within the state of the art which are used to be converted to a new product in which the cDNA clone is incorporated, I accept Mr Gratwick’s argument.
I would also, subject to detailed argument, be prepared to extend the protection to plasmids unknown at the priority date but proven to be immediate derivatives of known plasmids. But claims 7 and 8 refer to unrestricted classes of cloning and expression vectors any one of which would be protected provided it comprised a DNA sequence encoding t-PA.
I must confess that I have more difficulty in deciding whether or not these can be fairly described as “methods” embracing the figure 5 data, or something very close to it. In the final analysis I have been influenced by the very general nature of the category comprised by cloning and expression vectors, which might include molecules entirely unknown at the moment and quite different from the plasmids and vectors mentioned in the specification.
For this reason I have come to the conclusion that the “method”, ie vector, is really undefined and speculative. The claims are, therfore, claims to the discovery as such and do not refer to inventions for the purpose of the Act, and for this reason must fail.
14.16 Claims 10 to 15 must all also fail if my views about claims 1 to 8 are correct. It is, therefore, not necessary to make any further comment about them.
Claims 16 and 17 are process claims which were expanded by reference to the definitions in the specification and set out in the appendix to the appellant’s skeleton argument. This appendix formed the basis of Mr Gratwick’s submissions. Claims 16 and 17, as expanded, read as follows:-
“16. A process which comprises expressing in a recombinant cell a DNA insert encoding a polypeptide which either
(a) has the sequence of amino acids identified as numbers 1 to 527 of figure 5, or
(b) is a derivative resulting from allelic variations and modifications as herein defined of the polypeptide having the said sequence and which possess the properties of
(i) catalyzing the conversion of plasminogen to plasmin
(ii) binding to fibrin – and
(iii) immunological response
substantially the same as those possessed by human t-PA having the said sequence
which insert comprises
(I) the DNA sequence set forth in figure 5 encoding amino acids numbers 1 to 527
(II) a DNA sequence differing therefrom which encodes the said amino acids
(III) a DNA sequence encoding a derivative polypeptide as defined in (b) above
17. A process for producing a polypeptide which either
(a) has the sequence of amino acids identified as numbers 1 to 527 of figure 5
(b) is a derivative resulting from allelic variations and modifications as herein defined of the polypeptide having the said sequence and which possesses the properties of
(i) catalyzing the conversion of plasminogen to plasmin
(ii) binding to fibrin
(iii) immunological response
substantially the same as those possessed by human t-PA having the said sequence
which process comprises:-
(A) preparing a replicable expression vector containing a DNA insert comprising
(I) the DNA sequence set forth in figure 5 encoding amino acids numbers 1 to 527
(II) a DNA sequence differing therefrom which encodes the said amino acids
(III) a DNA sequence encoding a derivative polypeptide as defined in (b) above
(B) transforming therewith a host cell culture to obtain a recombinant host cell
(C) culturing the said recombinant host cells under conditons permitting expression of said DNA sequence to produce said polypeptide
(D) recovering said polypeptide”.
Mr Gratwick, particulary in reply, dealt with claims 16 and 17 by reference to claim 17 (see Court of Appeal transcripts Day 16, page 19H):
“Mr Gratwick: . . . If your Lordships have claim 17 in front of you, at page 79, in the middle of the page: ‘A: preparing a replicable expression vector containing a DNA insert comprising (i) the DNA insert set forth in fig 5 encoding amino acids 1 to 527’. That is all I have addresed my argument to so far. What I am saying at this point is, so far as that is concerned, it is not right that the pantentee should be limited to the replicable expression vector of pBR322. It is right that he should be entitled to cover the use of whatever replicable expression vector is appropriate to the host cell — I pick the words in B below, ‘transforming therewith a host cell’. That is my first point focussing solely on (i) the precise fig 5 sequence for acids 1 to 527. I say I should not be confined just to the replicable expression vector in fig 9; and the reason simply is that the choice of particular expression vector is not at the core of my invention. My invention is combining that which is a replicable expression vector with that which is a DNA insert according to figure 5. That is the new and inventive thing which I have given the world and which the world could not get without me.”
Again, I see no objection in this submission as it stands, and feel that it is within the spirit of the Protocol to Article 69, provided that it relates, as I think it does, to a defined method by which the figure 5 data is harnessed. However, the subsequent subparagraphs of 17A (as expanded in the appendix) refer, in far more general terms to different DNA sequences arising from the amino acids numbers 1 to 527 set out in figure 5. 17A (II) provides:-
“A DNA sequence differing from (figure 5) which encodes the same amino acids.”
This is a reference to the degeneracy feature of the genetic code as previously described in this judgment, which would enable a scientist, using the amino acid sequence information, to construct a DNA sequence containing the alternative nucleotides which would encode for the same amino acids. Here again, I would not object to a claim which covered the use of the “discovery” in a vehicle plasmid with the alternative gene construction. However, one must not lose sight of the narrow basis of the invention, ie a method embracing the discovery.
In his later submissions Mr Gratwick went much further than this. In his reply he referred to Decision T292/85, already mentioned in paragraph 13.19, of the Technical Board of Appeal in Genentech’s Application (see page 34 of Day 17 of Court of Appeal Transcripts):-
“So far as concerns the derivatives in our claim, I would invite your Lordships’ attention again, if I may, to the EP decision in my client’s case. That is T292/85, paragraph 3.1.5 on page 12.
‘The above examples show that the need for a fair protection governs both the considerations of the scope of claims and of the requirements for sufficient disclosure. Unless variants of components are also embraced in the claims, which are, now or later on, equally suitable to achieve the same effect in a manner which could not have been envisaged without the invention, the protection provided by the patent would be ineffectual. Thus, it is the view of the Board that an invention is sufficiently disclosed if at least one way is clearly indicated enabling the skilled person to carry out the invention. Consequently, any non-availability of some particular variants of a functionally defined component feature of the invention is immaterial to sufficiency so long as there are suitable variants known to the skilled person through the disclosure or common general knowledge, which provide the same effect for the invention.’
That directs particularly at the vectors.
‘3.1.6. The Examining Division’s tentative suggestion that such terms should be restricted to those available in the art has no basis in existing law. Unless broad, yet proper terminology is allowable, subsequent investigations by third parties might be encouraged to concentrate on finding alternatives outside the claims instead of trying to pursue progress through dependent inventions. The lack of recognition of the full significance and the interdependency of technical contributions could adversely affect progress in the area of microbiology and biochemistry.’
My Lords, what we say is that is absolutely sound, not only in EPO law, but always has been in English law, and that our new construct disclosed in detail on the basis of pBR322 property justifies a claim, having as basis any vector, now or hereafter invented, which enables the DNA insert to be expressed.
Such later vectors are indeed likely to be discovered. They are likely to be inventive. They are likely to be of universal application, and the inventors of them will have a patent for them. If Genentech want to use that new vector to make t-PA, quicker, or better, or whatever it is, they will have to pay a royalty, and in the same way, if the inventor of that new vector wants to make t-PA, he will have to pay a royalty to Genentech. We say rightly so. Each has an invention. Each is entitled to his tribute, and, as I believe I told your Lordships in opening — and certainly it was mentioned by Graham J in Olin Mathieson — if the two parties in that situation cannot come to reasonable terms, then the law provides that a compulsory licence may be given.”
I am not satisfied that the opinion of the Technical Board of Appeal went so far as Mr Gratwick’s submission just cited. The reference to “broad yet proper terminology” is, to say the least, imprecise. It is purports to refer to developments in the art which are neither direct nor immediate, I would respectfully have to differ. However, the reference to “dependent inventions” encourages me to think that there is not too wide a gulf between this judgment and the decision of the Board. In any event, in my judgment, Mr Gratwick’s submission claims far too wide a protection.
14.17. I can consider claims 16 and 17 together in the context of the question: Does the claim amount to a claim to an invention comprising a method embracing the figure 5 data within Hickton’s case concept or is it a claim to the figure 5 data harnessed in some useful way not defined and, therefore, merely a claim to the figure 5 data as such?
Each claim is restricted to a process involving either the figure 5 data or an allelic variation or modification thereof having the critical properties of t-PA in exactly equivalent terms. Although the method in claim 16 is restricted to expression in a recombinant cell, there is no further definition of the method or type of vector to be used other than (whatever it may be) plasmid, bacteriophage or some vector presently unknown. Claim 17 fails to define the method in the sense that the nature of the vector is wholly at large.
14.18. As in the case of claims 7 and 8 I have found this part of the case an extremely difficult one in which to draw the lines between:-
(a) A claim to an invention which satisfies section 125(1), section 1(2) and section 14(5) and which is, therefore, patentable;
(b) A claim to an invention which satisfies section 125(1) and section 1(2) but which does not comply with section 14(5) and in respect of which a patent should not, therefore, have been granted but which has managed to satisfy the examiner and cannot be challenged under section 72(1); and
(c) A claim to discovery as such which cannot, for the purposes of the Act, be an invention which includes section 125(1).
With not a little hesitation I have come to the conclusion, whilst recognising the near-esoteric distinctions involved, that claims 16 and 17, as presently drafted (even in their expanded form), cannot be said to be claims to be a method embracing a discovery rather than claims to the discovery as such. They cannot, by virtue of section 1(2), be claims to inventions for the purpose of the Act. The fact that, in any event, they would also fail under section 14(5) is not relevant for the purposes of the present enquiry. I cannot part with this aspect of my judgment without commenting that I would have probably taken a different view if claims 16 and 17 had been more specifically drafted.
14.19. Claim 18 depends upon claims 16 and 17 and needs no further comment, whilst claim 19 would stand if the claims were restricted on claim 9, even if expanded in some acceptable way, or with claims 16 and 17 if these were capable of suitably restrictive amendment. Claim 20 requires no separate treatment.
In the event, in my judgment, the “invention” as presented in the specification is not a patentable invention within the terms of the Act. During the course of argument Mr Gratwick extended invitations to the court to consider the question of applications to amend the claims in appropriate circusmtances. Mr Jacob, however indicated, at the same time, that he would, in the circumstances of this case, make the strongest objection possible to such applications.
In those circumstances, I do not think it proper for me to make any comment whatever at this stage upon this particular topic. For my part I would have been willing to hear further argument from counsel if Mr Gratwick was minded to make such an application in respect of claims 7, 8, 16 and 17. I have, however, had the opportunity of reading in draft the judgment to be delivered by my Lords and, in view of their judgments, I do not think it appropriate to extend any invitation to Mr Gratwick to make any application for leave to amend the Patent.
In the circumstances, and for the reasons I have given, I would dismiss the appeal.
Had it been necessary to consider the respondent’s notice for the reasons already appearing I would have been against Mr Jacob on what might be termed the section 14(5) — section 72 point, which he did not strenuously argue, and on the Merrill Lynch point in agreement with the views which I believe will be expressed by my Lords: and also against him on the obviousness/inventive step point upon which I have the misfortune to differ from my brethren.
Finally, I would add, in view of what I said at Day 1, page 2H of the transcripts concerning the role to be played by Dr Brenner, that Dr Brenner has played no part in the judicial decision-making process, nor has his advice had any influence upon that process, other than to provide a scientific factual background. Having said that, I would wish to repeat my acknowledgement of the estimable assistance he has given to the court.
JUDGMENTBY-2: DILLON LJ
DILLON LJ: I adopt gratefully the detailed account of the facts of this case given by my Lord in his judgment. I would also wish at the very outset of this judgment to express my thanks to Dr Brenner, for his explanations of the scientific terms and procedures involved. Without those explanations, I would not, I apprehend, have come even remotely near to achieving some glimmering of what the scientific issues in the case are about. I would emphatically endorse the comment of Whitford J at page 558 of his judgment that “to those unfamiliar with the developments in the field of genetic engineering, the language which is necessarily used by workers in the field is by no means easy of comprehension.”
The judge decided this case against the patentees, Genentech, and therefore revoked the patent, on the grounds that the claims in the patent were not supported by the description, were neither clear nor concise, and did not define the matter for which the applicant sought protection. In other words the claims failed to satisfy the requirements of subheads (a) (b) and (c) in section 14(5) of the Patents Act 1977.
Such failure could have provided grounds for revocation of the patent under section 32 of the Patents Act 1949. But the permissible grounds for revocation of a patent under the 1977 Act are comprehensively set out in section 72 of the 1977 Act, and they are much more limited than the grounds for revocation which were available under the 1949 Act; in particular the permissible grounds for revocation under the 1977 Act do not include the ground (i) under section 32 of the 1949 Act “that the scope of any claim of the complete specification is not sufficiently and clearly defined or that any claim of the complete specification is not fairly based on the matter disclosed in the specification.” The permissible grounds for revocation under section 72 of the 1977 Act do however include as ground (a) “that the invention is not a patentable invention”, and the learned judge in the present case felt able to hold at page 592 lines 28-31 that “a specification which contains claims whicg are not supported by the description or which are neither clear nor concise and which fail to define the matter for which the applicant seeks protection is not a specification directed to a patentable invention”, but “is one in respect of which a patent should never have been granted.”
In my judgment, the learned judge was wrong in holding that he could for those reasons revoke the patent.
The 1977 Act was enacted to give effect to this country’s adherence to the European Patent Convention, and that in turn was entered into by the contracting countries in order to establish a unified patent law throughout all those countries. Although there had been previous international patent conventions, there is no reason whatever to suppose that that unified patent law, as enacted in the 1977 Act, would be in all respects identical with the previous United Kingdom patent law. Section 72 of the 1977 Act is categorical in providing that a patent may be revoked on (but only on) any of the grounds specified in the section. Furthermore section 74 specifies in subsection (1) the only proceedings in which the validity of a patent may be put in issue, including proceedings for revocation under section 72, and then categorically provides in subsection (2) that the validity of a patent may not be put in issue in any other proceedings, and in subsection (3) that the only grounds on which the validity of a patent may be put in issue (whether in proceedings for revocation under section 72 or otherwise) are the grounds on which the patent may be revoked under that section. There is no general or inherent power in the court to revoke a patent if the court feels that it should not have been granted, or should not have been granted in such wide terms.
One of the other permissible grounds for revocation in section 72 is ground (c):- “the specification of the patent does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art.” That clearly reflects section 14(3) of the 1977 Act, but there is nothing in section 72 to reflect section 14(5), and the fact that section 14(3) is deliberately picked up in section 72, while section 14(5) is not, makes it impossible to imply that failure to comply with section 14(5) in relation to the claims in a patent is to provide grounds for revocation under section 72. Ground (a) of the grounds for revocation in section 72 “that the invention is not a patentable invention” refers back to section 1, which defines what is a patentable invention, and to sections 2, 3, and 4 which explain that definition, but it does not, in my judgment, refer back to section 14(5).
If, therefore, there is a patentable invention and a patent is granted, that patent cannot be revoked — nor can its validity be disputed — on the grounds that the claims in the patent as granted do not comply with section 14(5). Section 14(5) reflects the provisions of Article 84 of the European Patents Convention. I see no escape, on the scheme Convention and of the 1977 Act, from the conclusion of the appeals board of the European Patent Office in the cases of Naimer’s patent (Decision T23/86)  Official Journal, EPO 316, and Mobil Oil Corp’s Patent (Decision Gr01/84)  Official Journal EPO 299, decided on 24 July 1985 that in the absence of any application by the patentee to introduce amended claims on objection under Article 84 is inadmissible in opposition or revocation proceedings, and that opposition, or revocation, proceedings are not a continuation, involving third parties, of the examination proceedings in the Patent Office.
It follows, incidentally, that in considering in the course of the examination of an applicaton in the Patent Office, whether the claims satisfy the requirements of section 14(5) of the 1977 Act, the Patent Office ought to have very clearly in mind that it is undesirable to allow claims the object of which is to cover a wide and unexplored field or where there is no disclosure in the specification which is in any way coterminous with the monopoly indicated in the claims.
This appeal must therefore be decided on grounds other than those on which the judge based his decision in the court below.
On the facts, the only relevant ground for revocation under section 72 is ground (a) and the crucial question is whether or not the invention the subject of the patent was a patentable invention. The points somewhat faintly advanced by Mr Jacob under section 72(1)(c) in relation, in particular, to the extra ‘G’ in figure 5 of the patent are not, in my judgment, made out on the facts.
Section 1 of the 1977 Act provides by subsection (1) that a patent may be granted only for an invention in respect of which the conditions specified in the subsection are satisfied. This is a statutory definiton of what is a patentable invention, and the specified conditions relevant to the present case are (a) that the invention must be new and (b) that it must involve an inventive step. Subsection (2) however declares that (inter alia) a “discovery” or a scientific theory or mathematical method or a program for a computer is not an invention for the purpose of the Act, subject to the qualificaiton that the foregoing provision shall prevent anything fom being treated as an invention for the purposes of the act only to the extent that a patent or application relates to that thing as such (my emphasis).
Section 2 of the 1977 Act provides by subsection (1) that an invention shall be taken to be new if it does not form part of the state of the art. The state of the art is then defined by subsection (2) as being in effect all matter made available to the public anywhere before the priority date of the invention. Section 3 of the 1977 act then provides that an invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the state of the art.
What is unquestionably new in the present case is the discovery of the nucleotide and amino-acid sequences that make up the DNA in the protein t-PA, and these are disclosed in figure 5 of the patent. By itself, however, that discovery is merely a discovery and not an invention. The patentable invention, if there is one, must be found in the application of the discovery.
It is common ground that the patent could have been drafted with greater clarity. There is no consistory clause to set out what the inventor coneives the invention to consist of, but there is no need in law to have such a clause. What the patent is intended to be about is however indicated in the opening paragraphs on page 1, viz:-
“The present invention relates to human plasminogen activator, correspnding to that found in human serum and/or tissues, and to novel forms of compositions thereof and particularly to the means and methods for its production to homogeneity in therapeutically signficant quantities.
The present invention arises in part from the discovery of the DNA sequence and deduced amino acid sequence of human plasminogen activator. This discovery enabled the production of human plasminogen activator via the application of recombinant DNA technology, in turn enabling the production of sufficient quality and quantity of material to initiate and conduct animal and clinical testing as prerequisites to market approval, unimpeded by the restrictions necessarily inherent in the isolation methods hitherto employed involving production and extraction from existing cell culture. This invention is directed to these associated embodiments in all respects.”
his is amplified on page 9 in the following passage dealing with allelic variations and modifications, viz:-
“The potential exists, in the use of recombinant DNA technology, for the preparation of various human tissue plasminogen activator derivatives, variously modified by resultant single or multiple amino acid substitutions, deletions, additions or replacements, for example, by means of site directed mutagenesis of the underlying DNA. Included would be the preparation of derivatives retaining the essential kringle region and serine protease region characteristic generally of the human tissue plasminogen activator described specifically herein, but otherwise modified as described above. All such allelic variations and modifications resulting in derivatives of human tissue plasminogen activator are included within the scope of this invention, as well as other related human extrinsic (tissue-type) plasminogen activators, similar physically and biologically, so long as the esssential, characteristic human tissue plasminogen activator activity remains unaffected in kind.”
The point is made that such variations and modifications can be easily achieved, and may indeed occur naturally.
When this appeal was opened the claim put forward as being the principal claim in the patent was claim 3, which is for “human tissue plasminogen activator as produced by recombinant DNA technology.” Claim 1 is to be construed as covering the same ground as claim 3 and adding nothing, and so may be disregarded. Claims 2 and 4 appear to be much wider than claim 3, and Genentech has not sought to support them on the hearing of this appeal; for the present I put them to one side.
Other clais cover important steps in the process of producing human t-PA by recombinant DNA technology in reliance on the discovery of the sequences. Thus claim 7 is for a recombinant cloning vector comprising a DNA sequence encoding human t-PA claim 8 (which as the argument proceeded came to be relied on as the most important claim) is for a replicable expression vector capable, in a transformant micro-organism or cell culture, of expressing a DNA sequence according claim 7, and claim 9, a very limited claim which the judge would have been prepared to uphold, is for the actual plasmids created and used by Genentech in producing human t-PA or its equivalent by recombinant DNA technology. Claim 10 is then for a microorganism or cell culture transformed with the vehicle according to claims 8 or 9. Claim 16 is for a process which comprises expressing DNA encoding human t-PA in a recombinant host cell, and claim 17 elaborates the steps in such a process starting with the preparation of a replicable expression vector in accordance, in effect, with claim 8. Claim 19, which the judge would also have been prepared to uphold, is a limited claim for the process of producing human t-PA actually described in the patent. Claims 9 and 19 do not much matter since they are so narrow and can easily be circumvented; it is not suggested that Wellcome has infringed them.
As I see it, these claims all depend on use of the sequences discovered by Genentech, and without knowledge of those sequences human t-PA could not have been produced by recombinant DNA technology. But apart from the use of the sequences there is nothing new in the invention described in the patent. In particular it is agreed between the parties’ expert witnesses for the purposes of the proceedings that at the relevant time, in early May 1982, the ordinary skilled worker, once in possession of a complete cDNA sequence encoding t-PA, was capable of manipulating it so that the protein could be expressed in bacteria, yeast or mammalian cells, and it is also agreed that while most workers in the field would have tried to express t-PA in E coli, most workers would ahve also planned for expression in a mammalian cell line such as CHO.
Equally importantly, the patent specification in describing what Genentech did to discover the sequences and to arrive at the plasmids referred to in claim 9 which produced artificial human t-PA or its apparent equivalent contains copious references to the prior art and to known procedures and contains nothing to suggest that Genentech supposed that there was any new or original departure from, or improvement on, established procedures in what Genentech did, either in the procedure followed to discover the sequences or in applying the discovery.
It is submitted therefore, in argument for the respondents in this court, that the patent therefore falls foul of the qualification in subsection (2) of section 1 of the 1977 Act and reliance is placed on the words which I emphasised in referring to that qualification earlier in this judgment, viz:- “to the extent that a patent . . . relates to that [discovery] as such.” This submission must be valid if the reasoning of Falconer J in Merrill Lynch Inc’s Application  RPC 1, is correct. In that case Falconer J upheld the reasoning of a principal examiner in the Patent Office which was expressed as follows:-
“I consider that the proviso in section 1(2) . . . makes it clear that the inclusion of an excluded thing in the subject matter of a patent application does not inevitably mean that the subject matter is unpatentable. However this subsection does not say that anything is prevented from being treated as an invention ‘only if’ the application relates to that thing as such; instead the words used are only to the extent that; I am of the opinion that this wording . . . should be understood as meaning that anything which is in an excluded category is not to be treated as an invention and consequently its inclusion cannot be considered to contribute to the required novelty and inventive step.”
Merrill Lynch was concerned with a computer program, and Falconer J’s conclusion is summed up by his acceptance, at the top of page 12, of a submission that the wording “only to the extent that” means that there cannot be a patentable invention in so far as the invention resides in the computer program itself, but if some practical (ie technical) effect is achieved by the computer or machine operating according to the instructions contained in the program and such effect is novel and inventive (ie not obvious) a claim directed to that practical end will be patentable notwithstanding it is defined by that computer program.
Such a conclusion, when applied to a discovery, would seem to mean that the application of the discovery is only patentable if the application is itself novel and not obvious, altogether apart from the novelty of the discovery. That would have a very drastic effect on the patenting of new drugs and medicinal or microbiological processes. It would also represent a drastic change from English law as previously understood. Thus in Hickton’s Patent Syndicate v Patents & Machine Improvements Co Ltd (1909) 26 RPC 339, Cozens Hardy MR said at page 347:-
“When once the idea of applying some well-known thing for a special and new purpose is stated, it may be very obvious how to give effect to that idea, and yet none the less is that a good subject-matter for a patent.”
Fletcher Moulton and Buckley LJJ both rejected as fallacious the view taken in the court below that “an idea may be new and original and very meritorious, but unless there is some invention necessary for putting the idea into practice it is not patentable.” This established view is reflected in the judgment of Whitford J in the present case where he said at page 566:-
“It is trite law that you cannot patent a discovery, but if on the basis of that discovery you can tell people how it can be usefully employed, then a patentable invention may result. This in my view would be the case, even though once you have made the discovery, the way in which it can be usefully employed is obvious enough.”
On this point, I prefer the view of Whitford J to the reasoning of Falconer J. In so far as a patent does not, in my judgment, relate only to the discovery as such, even if the practical application may be obvious once the discovery has been made, even though unachievable in the absence of the discovery. When I look at the claims of the patent in the present case other than claims 2 and 4, they are not claims for the discovery of the sequences as such, but claims in relation to the practical application of the discovery in the production of human t-PA (defined on page 9 of the specification — as “corresponding to t-PA otherwise native to human tissue”) by recombinant DNA technology. In my judgment, therefore, this patent does not fall foul of subsection (2) of section 1 of the 1977 Act.
I should add that though I disagree with the reasoning of Falconer J in the Merrill Lynch case, it does not in the least follow that I disagree with the result of that case. It would be nonsense for the Act to forbid the patenting of a computer program, and yet permit the patenting of a floppy disc containing a computer program, or an ordinary computer when programmed with the program; it can well be said, as it seems to me, that a patent for a computer when programmed or for the disc containing the program is no more than a patent for the program as such.
In my judgment, therefore, the novelty of the discovery of the sequences is sufficient to make the inventions as claimed in the patent “new” for the purpose of section 1(1)(a) of the 1977 Act, even though the practical use of the discovery was a matter of the application of known technology.
The next question, which is fundamental to this appeal, is whether the invention involved an inventive step or was obvious to persons skilled in the art.
This is always a difficult question where highly specialised science or technology is involved. The judge was of the view that there was an inventive step — see his judgment at page 596 lines 14 to 21. Otherwise he could not have been prepared to uphold even claims 9 and 19. But it is clear that in this court we have to evaluate the evidence afresh and form our own independent opinion, as is explained in Benmax v Austin Motor Co Ltd (1955) 72 RPC 39.
It is common ground that the persons skilled in the art in the present case, for the purposes of section 3 of the 1977 Act and the test of obviousness, would be a team of persons, each of whom would be the holder of a PhD in a relevant field of science. They are thus necessarily persons of very considerable intellectual capacity, but by force of the terms of the section whatever is obvious to them has to be obvious without any inventive step of their part.
It is agreed between the experts that at the relevant time, in early May 1982, the production of quantities of t-PA adequate to treat human patients suffering from blood clots was known to be a desirable objective and it is further agreed that the Bowes melanoma cell-line was then available and was known to be a source of t-PA and of the corresponding mRNA. It was of course then known that the nucleotide and amino-acid sequences of t-PA existed, though their composition was not known. All the steps taken by Genentech in finding out the composition of the sequences and applying that knowledge to produce human t-PA, as defined in the specification, by recombinant DNA technology were applications of known technology, and no step was by itself inventive.
It is not in doubt that empirical research industriously pursued may lead to a patentable invention. This was stated by Lord Simonds in May & Baker Ltd v Boots Pure Drug Co Ltd (1950) 67 RPC 23 at 34, a case concerned with the discovery of two new sulphathiazole drugs, and the application of those drugs for medicinal purposes. What Lord Simonds said was amplified by Lord MacDermott at page 50 as follows:-
“The problem was to find some substance at once less toxic and more effectual against certain infections than sulphanilamide. Each of the new compounds provided a solution. They were produced by known methods from known materials and the inventive step lay in the discovery of their virtue as drugs. This achievement was not the result of applying and known principle or law of nature. It was entirely empirical. No doubt the inventors used skill and experience of a high order in selecting a profitable line of research. But they had not means of knowing that success had crowned their efforts until the new products had been tested by experiment.”
The patent merited by such empirical research ought to have been limited to the actual new drugs discovered. No question of obviousness was however raised in that case, but that, as it seems to me, was because the drugs discovered were indeed new and the profitable line of research selected by the inventors was not obvious.
There are also observations in relation to scientific discoveries, in particular in the field of antibiotics, by Lord Diplock in his dissenting speech in American Cyanamid Co (Dann’s) Patent  RPC 425, where he said at page 451:-
“Since the original discovery of the therapeutic uses of antibiotics and of the methods of aerobic fermentation by which they can be produced from micro-organisms to be found in nature, further advances in this field of medicines have been achieved by searching for and finding hitherto unidentified strains of mico-organisms existing in the natural state from which useful new antibiotics can be prepared by what is now a well-known standard process. The task of finding such a strain of mico-organism, calls for the exercise of technical proficiency and is laborious and very costly, for the odds against success are large. It is not easy to see what inventive step, as distinct from the mere exercise of proficiency and practice, is involved in this kind of research, but the result of success in it is a new product useful to humanity which does not exist in nature. If such research is to be encouraged in a competitive society, the monetary rewards of success must be assured to those who undertake the expense; and the means of doing so in this and in most other countries with comparable social systems is by according to the successful discoverer of the new product the controlled and limited monopoly granted for inventions under the nationa patent laws.”
And at page 452:-
“It is in my view consistent with this basic policy to treat the kind of research involved in the discovery of a strain of micro-oranism from which a new and useful antibiotic can be prepared as an activity which entitles the person who undertakes it to a temporary monopoly under the Patents Act 1949, of the product of his success. I accept, therefore, the extension of the concept of ‘invention’ to include antibiotics which are discovered through this kind of research. But if this is to be done without defeating the basic policy of the Act one must also accept the necessary corollaries in the interpretation of those parts of it which deal with the information to be provided as the counterpart of the temporary monopoly granted to the inventor.”
These observations were adopted by Whitford J in the present case, and, as general statements of the law they are not challenged by Mr Jacob for the respondents. They were however obiter in that there was no issue in the American Cyanamid case as to the patentability, as such, of the new antibiotic which had been discovered; the question was whether the disclosure in the patent was inadequate because strains of the newly discovered micro-organism had not been deposited and so were not available to the reader of the patent. Lord Diplock was therefore not concerned with any question of obviousness. Whitford J seems however to have deduced in the present case from Lord Diplock’s observations that as the work of Genentech involved, as he put it at page 596, laborious and costly effort amounting to rather more than the exercise of proficiency, it therefore involved an inventive step and cannot have been obvious to persons skilled in the art, within the meaning of section 3 of the 1977 Act. I have great difficulty in seeing the logic of this, since (as I have already said) Lord Diplock was not concerned with obviousness, and in the American Cyanamid case a new product had been found by following a line of empirical research which may well not have been obvious to anyone.
Lord Diplock was indeed concerned with obviousness, as Diplock LJ, in the case of Johns-Manville Corp’s Patent  RPC 479. At pages 493-4 he expressed the view that the case that an allegedly inventive idea was at the priority date “obvious and clearly did not involve any inventive step” would have been made out if before the priority date the man skilled in the art would have thought the idea well worth trying out in order to see whether it would have beneficial results. He took the view that it would be enough that the person skilled in the art would assess the likelihood of success as sufficient to warrant actual trial, without postulating prior certainty of success. In Olin Mathieson Chemical Corp v Biorex Laboratories Ltd  RPC 157, Graham J formulated the question at pages 187-8 as being whether a notional research group at the relevant date would have been directly led to try a certain idea, in the expectation that it might well produce a useful result. Again certainty of success was not postulated. In Phillips (Bosgra’s) Application  RPC 241 at 251, Whitford J held that to render an invention obvious it was not necessary that the materials in question should have been the first choice of the notional research worker; it was enough that the materials were “lying in the road” and there for the research worker to use.
By the various tests set out in the immediately foregoing paragraph it was indeed obvious, in my judgment, to the person skilled in the art to set out to produce human t-PA by recombinant DNA technology. At least four teams did just that at about the same time. The evidence of those at Leuven and Umea as to their choice of projects is particularly relevant. The end was a known desiderandum, and to produce by oligonucleotide probing was, if not the first choice of each team, an early choice which lay in the way, ready to hand. It was indeed at one stage submitted for Genentech, as I understood Mr Gratwick, that it was not made out that the person skilled in the art would have known of the process of screening a library by the use of oligonucleotide probes, but in my judgment the evidence is ample that the process was well-known.
If indeed no one else had set to clone t-PA and produce human t-PA by recombinant DNA technology, the position on obviousness would, I apprehend, have been very different, and it could then have been said that a novel idea for a project, carried through with painstaking skill and proficiency to a successful end, involved an inventive step and merited patent protection. Even so, there would have been scope for much argument on the proper width of the claim.
As it is, however, the main argument for Genentech on the question of obviousness to persons skilled in the art has been that the hypothetical team of persons skilled in the art, even if they had thought of trying to do so, would not have succeeded in discovering the sequences. This is put, not on the basis that there was any thing novel or original in the technology used by Genentech in discovering the sequences (since admittedly there was not) but on the basis that the person “skilled in the art” has come by judicial interpretation to mean the person only moderately skilled in the art. Indeed the running thread that showed through Mr Gratwick’s skilled int eh art. Indeed the running thread that showed through Mr Gratwick’s very detailed analysis of the evidence was that in his submission whenever anything was required to be done, in the discovery of the sequences, that required skill in the art, the “person skilled in the art” would, almost ex hypothesi, have been able to do it.
The foundaiton for this submission appeared to be the decision of this court in Valensi v British Radio Corp  RPC 337. That case was however concerned with a different question, namely, the question, which now arises under sections 14(3) and 72(1)(c) of the 1977 Act, whether the specification of a patent, or application, discloses the invention clearly enough and completely enough for it to be performed by a person skilled in the art. It was held, and the good sense of the conclusion is obvious, that the hypothetical addressees were not — or were not merely — persons of exceptional skill and knowledge, like the expert witnesses in that case, but skilled technicians from whom a team could be constituted which would possess the common knowledge of those in the relevant arts but would not have the capacity to make elaborate additions to or modifications of what was disclosed in a specification. The application of that to the question of obviousness and inventive step under section 3 of the 1977 Act might in an appropriate case be that if a new line of research or some other original step would only have occurred to a particularly brilliant or inventive person in the art, and not to persons skilled in the art generally, then that line of research or original step was inventive and not obvious. That is not, however, this case since the production of human t-PA by recombinant DNA technology was an obvious choice as a line of research, and the discovery of the sequences was achieved by known technology without any original step. In such a case as the present, the ability of the person skilled in the art, or moderately skilled in the art, to perform known techniques has, in my judgment, nothing to do with the question of obviousness or inventive step under section 3. I cannot see that there is an inventive step in the use by Genentech among the processes by which they discovered the sequences of a known process not devised by Genentech which only persons of high skill in the art — of whom there are many — have the ability to perform.
I am reinforced in this conclusion when I come to consider the evidence.
Mr Gratwick relied on three points in particular as demonstrating, in his submisson, that to discover the sequences would have been beyond the capacity of the hypothetical persons skilled in the art, viz:-
(i) the difficulty of protein sequencing so as to cleave amino acids from the t-PA protein and so achieve a sequence of low redundancy of five consecutive amino acids to form a probe for screening a library of cDNA clones;
(ii) the misleading suggestion in the Suggs paper of 1981 in the prior art, which was also cited in Professor Maniatis’ manual published shortly after the priority date of the patent, that in the washing down of the clones from the library with increasing intensity to find the true positive to which the probe was hybridised the last one left will be the one wanted; and
(iii) the difficulty of the actual process of screening the library where the protein in question is as long and complex as t-PA.
As to (i) the evidence clearly shows, to my mind, that the protein sequencing was technically the most difficult operation among the processes followed by Genentech to discover the sequences. It was a vital operation, since oligonucleotide probing was not possible without the prior isolation of the short sequence of five or so amino acids to form the probe. Protein sequencing required special equipment and tended therefore to be done in laboratories where there were people used to handling that equipment.
That said, however, it is important to keep in the forefront of the mind the Points of Agreement between the Experts in Document A-18, which my Lord has already set out in full, and in particular paragraphs 13 to 17. Known techniques of protein chemistry make it possible at the material date for such a five amino-acid sequence to be isolated and identified. Although the work required was painstaking and difficult, the techniques were known and the spinning cup sequenater generally used was commmercially available. The more sensitive gas-phased sequenater, of which Dr Hewick was one of the inventors, was being introduced. With the exception of CSH, all the teams which set out to produce human t-PA by recombinant DNA technology succeeded in obtaining the requisite 5 amino-acid sequence to form a probe.
Genentech initially sought to have the sequencing done for them by Dr Shively of The City of Hope Research Institute in California, but it seems that he was not successful in obtaining a sequence at the N-terminal, probably because he did not have sufficient material, although Genentech themselves had ample material available to them. Genentech therefore had the sequencing done for them in house by Mr Kohr on a spinning cup sequenater. He isolated and identified the sequence WEYCD, from near the centre of the protein, by, as I understand it, cleavage by trypsin (as to which see paragraph 14(c) of the Points of Agreement). At the University of Leuven, Mr Opdenakker and his supervisor Dr Volckaert had the sequencing done in the biochemistry laboratory of Professor Rombauts in the same University, and the same sequence WEYCD was achieved; on the dates this must have been by the use of a spinning cup sequenater.
Drs Ny and Edlund at the University of Umea had the sequencing done for them at the Karolinska Institute in Stockholm, where there was a spinning cup sequenater in the laboratory of Dr Jornvall. A different short sequence was isolated, but it seems to have been equally effective for the purpose.
At GI, Dr Hewick carried out the protein sequencing and obtained by June 1982 sequence data covering the first 24 amino acid positions starting from the serine N-teminus of the protein. He used the more sensitive gas-phase micro-sequenater of which he was one of the inventors. He said in his affidavit that obtaining the sequence data by using a spinning cup sequenater of the commercially available type would not, in his belief, have presented him with any particular difficulty, although it would have taken longer than using a gas-phase micro-sequenater; I see no reason to doubt this assertion, in view of the experience of others, even though he admitted in cross-examination that he himself would have needed 14 days training on a spinning cup sequenater before he used one.
The only exception was CSH, but their trouble was that they did not have a protein chemist available for the work. Their spinning cup sequenater operator Dr Smart was known to be about to leave CSH and in fact left in November 1981 to join a rival firm, and another experienced protein chemist they had, Dr Drickamer, was engaged on other work and not interested in assisting with this project. They therefore made arrangements with GI.
Against this background, I am unable to accept Mr Gratwick’s submission on the facts (even assuming relevance in law) that in early May 1982 persons skilled in the art would not have been able to carry out the protein sequencing necessary to isolate and identify the necessary short 5 amino-acid sequence — WEYCD or its equivalent — to form a probe. The team of persons skilled in the art would have included a protein chemist or protein sequencer.
There is another way of looking at this point. I find nothing in the specification to suggest that it is of any significance that the protein sequencing for Genentech was in the event done in house by Mr Kohr rather than by the outside expert, Dr Shively, whom they originally approached. Had Dr Shively done the sequencing successfully, Genentech would, it would seem, have made the same patent application. Could it then have been said that, though the process was known and the machinery was known, all protein sequencing was necessarily inventive, and not obvious to the persons skilled in the art because all persons skilled in the art habitually sent their protein sequencing out to specialist laboratories to do for them? Surely not.
As to point (2) the short answer is that the error in the Suggs paper does not appear ever to have caused anyone any practical problem. This is no doubt because persons engaged in these highly complicated fields of science are not lacking in common sense and practical experience which teaches them sensible procedures in carrying out experiments. The point is dealt with by Professor Maniatis in his evidence when he was cross-examined on the Suggs Paper and said at T6/13 E-F:-
“I think that the interpretation of that remark is carried far beyond what any investigator with common sense would do. I do not think you can extrapolate from that statement and conclude you have washed down to the highest stringency. Anyone with common sense and experience in cloning would wash to a point where the signal noise ratio looked reasonable, would pick the clones and test them. The fact that it has not been a practical problem and many other laboratories have isolated clones of this nature would argue that that is true. It is simply common sense.”
Dr Ny in his oral evidence said that he and Dr Edlund increased their temperature in hybridisation and in washing until they got fewer and fewer clones that gave signals. But when asked whether they went on until they had only one clone left he said “no, we did not go that far, because we were afraid of losing the positive clone so we stopped when we had a reasonable number of clones . . .” The general tenor of the evidence of Dr Harris of Celltech is, as I read it, to the same effect; the judge said at page 574 of Dr Harris that he had no reason to suppose that Dr Harris would have been any less successful than Genentech were.
In my judgment the person skilled in the art would have used his common sense and would not have slavishly followed Suggs into error.
As to Mr Gratwick’s point (3), that screening a library with oligonucleotide probes for the cDNA of t-PA is difficult, Dr Ny said in his affidavit in Bundle B2 at paragraph 26 that:
“compared to obtaining the first fragment, it would have been a relatively straightforward matter to prepare a further cDNA library or libraries and use the cDNA fragment that we already had as a probe to find further fragments, so that a full length cDNA could be prepared. However the work would probably have been quite time consuming.”
Dr Stark, Genentech’s expert, gave evidence in T8 at 47G which I understand as meaning that once a cloned probe is obtained, there is nothing difficult to do any more. Dr Sambrook, who was not himself a protein chemist, said that the problem was to get the sequence, and he did not foresee a problem with the later probing if they got to that stage. He seems to have been referring to getting the 5 amino-acid sequence, which in Genentech’s case was WEYCD, since he goes on to refer to the Beckmann spinning cup sequenater at CSH, and to explain that that was being used for other purposes most of the time.
If therefore the protein sequencing to produce the short sequence WEYCD or its equivalent, to form a probe, was not beyond the capacity of persons skilled in the art, the screening of the library by probing the full cDNA sequence was also not beyond their capacity.
Mr Gratwick has certainly further submissions in relation to timing.
He says first that the person skilled in the art — or hypothetical team of such persons — would not have allowed himself the time to carry out the exhaustive, albeit routine, work which Genentech carried out to discover the sequences. Genentech seem, however, to have themselves taken some sixteen months for their work from the first arrival of Bowes cells until the first full length clone or full length construct was available; although this was a rapidly developing field of science, every stage involved very painstaking experimentation. Why should it be assumed that the person skilled in the art would be allowed less time? Beyond that, however, the question whether an invention would or would not have been obvious to persons skilled in the art at the relevant data has, in my judgment, to be considered broadly, as a touchstone of the originality of the supposed invention. The question is a hypothetical question posed by the statute and it is, in my judment, beside the point to consider such matters as how long the hypothetical team of persons skilled in the art would have been allowed for their work, or, for that matter, how much money from what source would have been available to finance them.
Mr Gratwick makes the further submission that as Genentech’s work took them 16 months, the question of obviousness should be considered not at the date in May 1982 which is the priority date of the patent (viz the date of the patent application, with convention priority, in USA), but as at a date some 16 months earlier, with the consequence that the true question ought to be whether the persons skilled in the art could have completed, before the priority date of the patent, the discovery of the sequences disclosed in figure 5 of the patent, and the construciton of a replicable expression vector or construct in accordance with claim 8. The priority date of the patent is, however, the conventional date for answering the question of obviousness or inventive step, and the question itself involves consideration of a hypothetical person, or team of persons, skilled in the art. In answering that question regard has to be had under section 3 to any matter which forms part of the state of the art, and under section 2 the state of the art includes any matter made available to the public at any time before the priority date. The state of the art has therefore necessarily to include the Opdenakker paper, Bundle O, Tab 3, which is of a very great importance in the art but was only published in February 1982 and was not available 16 months before the priority date. It necessarily follows, in my judgment, that the question of obviousness and inventive step has to be considered wholly as at the one date, the priority date of the patent, without regard to how far various hypothetical teams would by then have got if they had started at various earlier dates.
For the foregoing reasons it follows, in my judgment, that claim 3 of the patent, and therefore claim 1, claim 5, which elaborates claim 3, claim 6 in so far as dependent on claims 1, 3, or 5, and claims 7 to 20 all fail for obviousness and want of inventive step.
It remains to consider claim 2 and 4 which Genentech do not seek to uphold. These read as follows:-
“2. Human tissue plasminogen activator unaccompanied by associated native glycosylation.
“4. Biologically active human tissue plasminogen activator in essentially pure form, unaccompanied by protein with which it is ordinarily associated.
These are wider than claim 3, in that they purport to extend to human t-PA (by which I understand the draftsman to mean artificial human t-PA and not human t-PA occurring naturally in human cells) however produced, and are not limited to human t-PA produced by recombinant DNA technology.
With a patent like this, which has a succession of ever wider claims, Mr Gratwick must, in my judgment, be right in submitting that the court is only concerned with the widest claims, such as claims 2 and 4, in so far as they exceed the next widest, viz claim 3 Mr Gratwick is therefore right, in my judgment, in submittng that had the point been pleaded — and it has not been — these claims could clearly have been revoked under section 72(1)(c) of the 1977 Act on the ground that there is no disclosure or teaching in the patent to enable the person skilled in the art to perform the invention the subject of these claims.
Mr Jacob submits, by reference to Mullard Radio Valve Co Ltd v Philco Radio & Television Corp of Great Britain Ltd (1935) 52 RPC 261 in this court and (1936) 53 RPC 323 in the House of Lords, that, if claims are framed more widely than is justified by the inventive step relied on, the claims can be revoked for want of subject-matter and therefore want of inventive step and want of patentable invention under section 72(1)(a).
Since under the scheme of the 1977 Act matters relating to the width of the claims in a patent are to be dealt with in the office at the examination stage under section 14(5) and section 14(5) does not provide grounds for the revocation of a patent once granted, I would be reluctant to hold that a patent can be revoked on the ground that there is no subject-matter and no patentable invention merely because the claims have been framed too widely. But in the present case, on my conclusions in relation to claims 3 and 8 etc, there is nothing whatever in the patent by way of inventive step to support any claim in the patent. Accordingly I would revoke claims 2 and 4 as well.
The point made in the preceding paragraph that claim 2 fails because there is nothing whatever in the patent by way of inventive step to support it is, I think, a very much narrower point than the points discussed by Mustill LJ in his examination in depth of the nature of an invention and in subsequent passages in his judgment. As at present advised, I would not accept many of Mustill LJ’s conclusions, but I do not find it necessary to accept those conclusions or decide on their validity in order to decide this case. I am particularly concerned that his conclusions would lead to the result that a newly discovered drug or medicine is not patentable. I believe that that result would be contrary to the intentions of the states which were parties to the European Patent Convention; indeed, I have always understood that the Republic of Italy altered its patent laws in order to give effect to the harmonisation envisaged by the Convention, by making drugs and medicines patentable in Italy, where they had not been patentable before.
There is, as it seems to me, a further difficulty in Genentech’s way over claim 2 and also over claim 4. Claim 2 purports to cover the whole of what was intended to be covered by claim 3, plus more. But if claim 3 fails for obviousness and absence of a patentable invention, claim 2 would have to be amended to exclude from its scope what has been held to be obvious. On the wording of judgments of the Appeals Board of the European Patent Office in the cases of Naimer and Mobil Oil Corp referred to above, however, an application by the patentee to amend claim 2 would let in the objections under Article 84 of the Convention, and section 14(5) of the 1977 Act that the claim was neither clear nor concise and did not define the matter for which the applicant sought protection and was not supported by the description.
In the result, therefore, I agree wiht the result reached by the judge, albeit by a different route, that the patent shuld be revoked, and I would dismiss this appeal.
Many other points were discussed in argument, but I do not find it necessary to advert to them.
JUDGMENTBY-3: MUSTILL LJ
MUSTILL LJ: I agree with Purchas LJ that the claims as formulated cannot be sustained, with the exception of claims 9 and 19, and I agree with Dillon LJ that, quite apart from this, the inventions claimed, however formulated or re-formulated, fail for want of an inventive step. On both these grounds, therefore, I conclude that the appeal should be dismissed.
Since, however, my route to this conclusion is not precisely the same as either of those taken by my Lords, I will add some reasons of my own. In doing so I am grateful indeed to adopt in full the exposition of the technological and scientific background, and narratives of the work undertaken by the various teams, contained in the judgment of Purchas LJ. Moreover, since each of the judgments just delivered contains an account of the evidence and the principal authorities, I need not cover that ground again in detail. Many more points were raised and cases cited in the skilful arguments of counsel than it is possible to explore in a reasonable compass. I have done my best to take account of them all, but will concentrate on those which, to my mind, are closest to the heart of this appeal.
The issues fall into two distinct groups, relating the intrinsic patentability of the various inventions claimed, and to the existence of an inventive step or steps. I will consider them separately.
The first group of issues: patentability
During the argument we have traversed some very difficult terrain. The difficulties are not, however, all of the same kind. On the issue of obviousness the facts are technical and their implications are hotly in dispute. By contrast, the difficult questions of law and construction arising in relation to the first group of issues, taxing as they are, seem to me to turn on facts which were essentially common ground. In order to set the scene I will endeavour to state these facts in a way which involves no controversy, all the assertions being directed to matters as they stood on 5 May 1982.
1. Tissue-type plasminogen activator of human origin was a substance of proved therapeutic value. The idea that it would be desirable, if possible, to synthesise a protein having the same amino acid sequence and spatial configuration as t-PA of human origin, but lacking the impurities inescapable if the protein was extracted from human tissue, was known and indeed was obviuos.
2. The following passage in the specification accurately stated the level of achievement of the relevant art in May 1982:-
“Recombinant DNA technology has reached the age of some sophistication. Molecular biologists are able to recombine various DNA sequences with some facilitiy, creataing new DNA entities capble of producing copious amounts of exogenous protein product in transforming microbes and cell cultures. The general means and methods are in hand for the in vitro ligation of various blunt ended or ‘sticky’ ended fragments of DNA, producing potent expression vehicles useful in transforming particular organisms, thus directing their efficient synthesis of desired exogenous product. However, on an individual product basis, the pathway remains somewhat tortuous, and the science has not advanced to a stage where regular predictions of success can be made.”
3. In the words of the specification (page 4):-
“It was perceived that the application of recombinant DNA and associated technologies would be a most effective way of providing the requisite large quantities of high quality human tissue-type plasminogen activator (earlier referred to as human plasminogen activator), essentially free of other human protein. Such materials would probably exhibit bioactivity admitting of their use clinically in the treatment of various cardiovascular conditions or diseases.”
4. Whatever method or recombinant technology might be adopted would involve the making of a construct by ligating cDNA (“the insert”) into a vector. The insert would be a DNA molecule consisting of the appropriate number of nucleotides arranged in such an order as would enable, in an appropriate host environment, both the replication of the insert and its expression as the desired protein. The vector would probably be (in its natural form) either a bacterium (such as E coli) or a bacteriophage or a mammalian cell. It was likely that at least some of the vectors employed would themselves be recombinant constructs, synthesised by ligating various lengths of synthesised or alien DNA into the natural molecule of the vector. Suitable vectors were either readily available or could be synthesised for the specific purpose without exceptional difficulty.
5. Whichever recombinant method was used would involve the creation of a library of clones embodying cDNA molecules of which either — (a) one at least would be the desired insert in its full length, or (b) there would be fragments of such length and sequence as could be ligated together to form the desired insert in its full length. It would then be necessary to separate from the library either the full length of the insert or the fragments from which it was to be created.
6. Any researcher would have to find a way of retrieving a full length of the desired insert from the library, if it contained one, or otherwise retrieve such fragments as would when pieced together amount to a full length.
7. If the strategy adopted by the research team involved the prior ascertainment of partial amino acid sequences for the natural t-PA the team would be also likely to ascertain by known methods and nucleotide sequence data for cDNA fragments embodied in clones which showed promise. The purpose would be to verify that these sequences were consistent with the amino acid sequences already obtained. The research team would also need to carry out at least partial sequencing of the fragments in order to devise a strategy for the subsequent extraction either of a full-length clone or of clones embodying a fragment or fragments sufficient to complete the whole length of the desired insert.
8. Once the research team had extracted what appeared to be a clone containing a full length of the desired insert a sequence for the nucleotides would be obtained (again by known methods) in order to verify, by comparison with the ascertained lengths of amino acid sequences, that the clone was what it was believed to be. Knowledge of the complete nucleotide sequence would lead, as a matter of course, to knowledge of the amino acid sequence of the desired protein.
9. Once the full nucleotide sequence had been ascertained, it would no longer be necessary to use these methods for extracting the insert or its fragments from the library. Any team in possession of the sequence information could proceed directly to use a full-length probe to extract an insert for use in an expression vector. It would, however, still be necessary to construct a library.
10. Armed with the sequence information, any subsequent team could explore the benefits of amino acid substitutions in the desired protein by directed mutagenesis of the DNA sequence.
Given that this was the known state of affairs on the relevant date, what was it that Genentech achieved? The answer seems to me as follows:
First, they won the race. The goal was known, and others were trying to reach it. Genentech got there first.
Second, the goal was to find a means of making, and having found it actually to make, the desired protein — a substance identical to that which already existed in nature.
Third, they reached the goal by a route the general nature of which was already practised.
Fourth, the success was due to the fact that they were the first to create by recombinant means a full-length insert.
Fifth, on the way to the goal they constructed a number of organisms, of which the two expression vectors referred to in claim 9 were examples, which had never existed before. Some of them contained the full-length insert. These constructs were of no value except as a means to an end.
Sixth, on the way to the goal they discovered the nucleotide and amino acid sequences of “natural t-PA”. This discovery was not in itself a goal, and it is a fair inference that Genentech would not have set out to achieve it simply as a matter of pure research.
Seventh, the publication of Genentech’s work was of value to subsequent workers in two respects. It demonstrated that the desired protein could be made by recombinant methods within the existing technologies; and (by communicating the sequences and the restriction map) enabled the subsequent workers to reach the goal by a more direct route, at less expense and in a shorter time, and with a lesser risk of failure.
Eighth, the publication of Genentech’s work also enabled other workers to know what route Genentech had taken. But nobody would ever wish to take the first part of the route again, or to traverse any of the later parts in precisely the way described.
What does Genentech claim?
Before the validity of the claims can be assessed it is necessary to understand what is being claimed. This should be easy, for if the addressee of the patent cannot understand the limits of the monopoly granted to the patentee he is in peril of an unintentional infringement. In the present case a comprehension of the claims is far from easy; indeed their opacity lies at the heart of many of our present problesm.
The claims themselves are laconic, but the writings which precede them are long and diffuse, and cast in a most unusual form. This may well be due to a recognition that the patent applied for was of a novel type, difficult to assimilate into ordinary English and European patent law. Both the claims and the description seem designed to claim a protection of concentric zones of generality, no doubt in the hope that if one or more claims failed, others would still survive in a form sufficient to give Genentech the commercial advantage which the application was intended to secure. However this may be, problems have ensued for all concerned.
Mr Gratwick made no secret of the embarrassment created for parts of his argument by the definitions to which I shall later refer, and found it necessary to proffer written expansions of some of the more crucial claims. Mr Jacob, in his turn, had to frame his arguments on the facts by reference to the meaning of claims which, for the purposes of other parts of his submissions, he invited the court to treat as meaningless. For the court itself, the task of approaching difficult questions of law and the facts in the light of twenty widely different claims, complex enough in itself, was made more difficult still to handle by the uncertainty as to what the claims really mean.
This being so, I see no alternative but to begin by tackling the interpretation of the claims, an exercise which must be approached in the spirit demanded by section 125(1) of the Act, read in the light of the Protocol on the Interpretation of Article 69 of the Convention. Since several of the claims are ambiguous or meaningless when read in isolation, it is necessary to describe and quote from the specification in some detail. But first I must set out the title which reads as follows:
HUMAN TISSUE PLASMINOGEN ACTIVATOR, PHARMACEUTICAL COMPOSITIONS CONTAINING IT, PROCESSES FOR MAKING IT, AND DNA AND TRANSFORMED CELL INTERMEDIATES THEREFOR
The specification itself begins with two important general statements about the nature of the invention claimed.
“The present invention relates to human plasminogen activator, corresponding to that found in human serum and/or tissues, and to novel forms compositions thereof and particularly to the means and methods for its production to homogeneity in therapeutically significant quantities.
“The present invention arises in part from the discovery of the DNA sequence and deduced amino acid sequence of human plasminogen activator. This discovery enabled the production of human plasminogen activator. This discovery enable the production of human plasminogen activator via the application of recombinant DNA technology, in turn, enabling the production of sufficient quality and quantity of material to initiate and conduct animal and clinical testing as prerequisites to market approval, unimpeded by the restrictions necessarily inherent in the isolation methods hitherto employed involving production and extraction from existing cell culture. This invention is directed to these associated embodiments in all respects.”
(I have added emphasis, here and elsewhere.)
The bibliography referred to in the last-quoted passage contained a list of 17 patents or patent applications and no less than 79 publications by various authors, dated between 1970 and 1982. (These were not cited as prior art in the present litigation, presumably because they are acknowledged by the patent itself.)
There follow two passages containing general descriptions. The first is of human t-PA, and need not be described. The second describes the technology, and is of some importance in relation to the arguments on the state of the art:
“Recombinant DNA technology has reached the age of some sophistication. Molecular biologists are able to recombine various DNA sequences with some facility, creating new DNA entities capable of producing copious amounts of exogenous protein product in transformed microbes and cell cultures. The general means and methods are in hand for the in vitro ligation of various blunt ended or ‘stick’ ended fragments of DNA, producing potent expression vehicles useful in transforming particular organisms, thus directing their efficient synthesis of desired exogenous product. However, on an individual product basis, the pathway remains somewhat tortuous and the science has not advanced to a stage where regular predictions of success can be made.” (pages 4-5)
There then follows:
“DNA recombination of the essential elements, ie, an origin of replication, one or more phenotypic selection characteristics, an expression promoter, heterologous gene insert and remaining vector, generally is performed outside the host cell. The resulting recombinant replicable expression vehicle, or plasmid, is introduced into cells by transformation and large quantities of the recombinant vehicle obtained by growing the transformant.” (page 5)
Later in the same secton we find:
“The present invention is based upon the discovery that recombinant DNA technology can be used successfully to produce human tissue plasminogen activator (t-PA), preferably in direct form, and in amounts sufficient to initiate and conduct animal and clinical testing as prerequisits to market approval. The product human t-PA is suitable for use, in all its forms, in the prophylactic or therapeutic treatment of human beings for various cardiovascular conditions or diseases.” (page 6)
“The product produced by genetically engineered microorganism or cell culture systems provides an opportunity to produce human tissue plasminogen activator in a much more efficient manner than has been possible, enabling hitherto elusive commercial exploitation. In addition, depending upon the host cell, the human tissue plasminogen activator hereof may contain associated glycosylation to a greater or lesser extent compared with native material. In any event, the t-PA will be free of the contaminants normally associated with it in its non-recombinant cellular environment.
“The present invention is also directed to replicable DNA expression vehicles harbouring gene sequences encoding human tissue plasminogen activator in expressible form, to microorganism strains or cell cultures transformed with them and to microbial or cell cultures of such transformed strains or cultures, capable of producing human tissue plasminogen activator. In still further aspects, the present invention is directed to various processes useful for preparing said gene sequences, DNA expression vehicles, microorganism strains and cell cultures, and specific embodiments thereof. Still further, this invention is directed to the preparation of fermentation cultures of said microorganisms and cell cultures.” (page 7)
Next, there was a list of what were essentially captions for the twelve figures accompanying the specification. Figures 1 to 3 and 8 illustrated some of the steps which had led to the ascertainment of the nucleotide sequence. Other steps, namely the construction of the cloning vector pPA25E10 for the long partial sequence and the expression plasmid pARIPA containing a full length insert, were set out in figure 6 and the results of Genentech’s research, in the form respectively of a restriction map and a depiction of the full-length nucleotide and derived amino acid sequences, were contained in figures 4 and 5. Figure 7 was the reproduction of an assay for functional activity of pARIPA. Figure 9 was a schematic diagram for the construction of the expression vector pt-PAtrpl2 coding for direct expression in E coli, and figure 10 showed an assay for such expression. Figure 11, which has played no part in this case, depicted the construction of two other expression vectors, for producing t-PA in conjunction with DHFR protein, and Figure 12 represented the “current understanding” of the protein structure of human t-PA.
Of these illustrations it would appear that only figures 5, 6, 9 and 11 (and possibly also 4) had any materiality to the definition and the working of the invention.
Resuming the text of the specification, there is next an important series of definitions. These are too long to quote in full, but the following passages are particularly to be noted:
“As used herein, ‘human tissue plasminogen activator’ or ‘human t-PA’ or ‘t-PA’ denotes human extrinsic (tissue-type) plasminogen activator, produced by microbial or cell culture systems, in bioactive forms comprising a protease portion and correspondng to those tissue plasminogen activators otherwise native to human tissue. The human tissue plasminogen activator protein produced herein has been defined by means of determined DNA gene and deductive amino acide sequencing . . . (page 9)
“The potential exists, in the use of recombinant DNA technology, for the preparation of various human tissue plasminogen activator derivatives, variously modified by resultant single or multiple amino acid substitutions, deletions, additions or replacements, for example, by means of site directed mutagenesis of the underlying SNA.
. . .
“All such allelic variations and modificiatons resulting in derivatives of human tissue plasminogen activator are included within the scope fo this invention, as well as other related human extrinsic (tissue type) plasminogen activators, similar physically and biologically, so long as the essential, characteristic human tissue plasminogen activator activity remains unaffected in kind. (page 9)
. . .
“To summarize in the present invention, human t-PA thus has a functional definition, it is capable of catalysing the conversion of plasminogen to plasmin’ binds to fibrin, and is classified as a t-PA based on immunological properties as set forth hereinabove. page (11)
. . .
“. . . In the present specification, ‘plasmid’ and ‘vector’ are used interchangeably as the plasmid is the most commonly used form of vector. However, the invention is intended to include such other forms of expression vectors which serve equivalent functions and which become known in the art subsequently hereto. (pages 12-13)
“‘Recombinant host cells’ refers to cells which have been transformed with vectors constructed using recombinant DNA techniques. As defined herein, t-PA is produced in the amounts achieved by virtue of this transformation, rather than in such lesser amounts, or, more commonly, in such less than detectable amounts, as might be produced by the untransformed host. t-PA produced by such cells can be referred to as ‘recombinant t-PA’. (page 13)
After these definitions there follows a section entiled “Host Cell Cultures and Vectors”, the contents of which are summarised in the introductory paragraph:
“The vectors and methods disclosed herein are suitable for use in host cells over a wide range of prokaryotic and eukaryotic organisms.”
Section C of the specificaiton, entitled “Methods employed”, contains some brief and (as it would seem to me) randomly chosen descriptions of some of the techniques which could be used at various stages. The purpose of including this section is not clear, and neither party has placed any reliance upon it. It may, however, be noted that the construction of suitable vectors is said to employ “standard ligating techniques.”
Section D, headed “General Description of Preferred Embodiments”, contains a historical statement in very general terms of the “protocol” which was actually used by Genentech to ascertain the sequence and proceed to expression of t-PA. Purchas LJ has already set this out in his judgment and I need not repeat it.
Section E of the specification sets out in great detail the operations outlined in section D. Reference was again made to the figures, and the legends already given were repeated in a much expanded form. This section is long, and since no weight was placed on it in argument I will not attempt to summarise it, and need do no more than observe that (a) none of the steps is explicitly described as novel, (b) the narrative is supported by 34 references to the publication listed in the biography, and (c) the section has the following introduction:
“The following examples are intended to illustrate but not to limit the invention. In the examples here an E coli host culture and a CHO cell line suitable for the type of DHFR protein coding sequence to be introduced were employed as host cell cultures. However, other eukaryotic and prokaryotic cells are suitable for the method of the invention as well.” (page 22)
Section F describes how the compounds previously described can be formulated “according to known methods” to prepare pharmaceutically useful compositions, and section G contains various speculations about the structure of human t-PA. At the end of section G there is the following passage, intended to govern the specification as a whole and not just that section:
“Notwithstanding that reference has been made to particuar preferred embodiments, it will be understood that the present invention is not to be construed as limited to such, rather to the lawful scope of the appended claims.”
I confess that I have found the specification a very difficult document, containing as it does a mixture of background information, partial statements of prior art and knowledge, definitions of an objective, definitions of some of the terms used in the claims, and an elaborate description of the research processes which led up to the discovery or discoveries said to have been made, and of the processes which stemmed from those discoveries and led to the production of t-PA from a mammalian cell line.
It is, however, possible to discern that the patentees were claiming to have made two discoveries — namely, that the protein and its coding DNA had the sequences respectively set out in the top and bottom lines of figure 5, and that a recombinant method of producing t-PA could be made to work. The description also indicated that the inventions in respect of which Genentech would claim a monopoly extended to “human” t-PA, could be made to work. The description also indicated that the inventions in respect of which Genentech would claim a monopoly extended to “human” t-PA, whatever its attributes (so long as they retained the essential functional characteristics of “human” t-PA); the use of any recombinant methods for producing these substances whether or not those methods were at all the same as those which they had themselves adopted; and all types of expression vehicles for producing t-PA by recombinant means.
In my judgment this is, in the event, very much what one does find when at last the claims are reached. There are two conspicuous absentees from the list: a claim for the information contained in figure 5, and a claim for the insert itself. Present are a series of claims which may be arranged into eight groups, as follows.
Claims 2 and 4
I understand these to assert a monopoly in respect of all pure protein having the amino acid sequence of figure 5, and so configured spatially as to have the three functional characteristics set out in the passage from page 10 of the specification, quoted above.
Claims 1 and 3
These are product claims. In the context of the specification, I think it clear that the product claimed as the invention must conform with two requirements, namely that it has the attributes just mentioned in respect of claims 2 and 4, and that, as a matter of history, it must have been produced by recombinant technology: viz by the technique of ligating molecules (whether of natural or synthetic origin, and whether or not severed by restriction enzymes or other techniques) so as to make different molecules.
the monopoly extends to all t-PA conforming with these requirements, irrespective of whether (a) the recombinant DNA techniques used in the production of the t-PA bear any relation (except in the obvious respect that they are recombinatory) to the techniques used by Genentech and shown in the specification, or (b) the othe techniques used in the process bear any relation to thsoe used by Genentech, or (c) a knowledge of either of the sequences shown in figure 5 was necessary or even useful to the successful performance of the process.
Claims 7 and 8
These are claims for classes of products, the products being tools for use in a process aimed at the production of t-PA. Although the two claims are for some reason rather differently worded, the context makes it clear enough that they relate to constructs consisting of an insert ligated into a vehicle, the construct having the capability of promoting replication (Claim 7) and expression (Claim 8) in the appropriate host environment.
The monopoly extends to all such constructs, irrespective of whether (a) the vehicles to which the inserts are added are (as used by Genentech) plasmids, or bacteriophages, or other organisms, whether or not known or used at the priority date, and whether or not “equipped” with special features resembling those of the plasmids used by Genentech, and (b) whether the host organism is any of the organisms used by Genentech, or is an organism known or used at the priority date.
This is a product claim for two of the tools made and used by Genentech in their process.
Claims 16, 17 and 18
These are process claims. In the context of the passage quoted above from page 13 of the specification, claim 16 can be seen to assert a monopoly embracing any process which has as one step the use of a vector of any kind containing the insert to induce expression in a host cell of any kind. Clause 17 seems narrower at first sight, but does not, in my view (when read with the passage quoted from page 12 of the specification) do more than repeat the same idea in different words. So also for claim 18, except that the host cell is limited to a mammalian cell.
I take this to be a process claim limited, subject to minor divergences, to the same path to the goal as was traversed by Genentech. It, therefore, corresponds to claim 9. Neither party has suggested that it is of the least commercial value.
This claims the product of the processes covered by claims 16 and 17, and differs little in principle from claim 3.
Claims 5, 6 and 10 to 15
These are variants of other claims and it has not been suggested that any of them could survive if the others are struck down.
Before discussing these claims individually, it is opportune to consider in general terms how the court should interpret the 1977 Act, by reference to which their validity must be assessed. Here, I would offer two suggestions.
First, as to the reported cases. Because the industrial revolution happened when and where it did, and because some of the most penetrating legal minds, both on and off the bench, have directed themselves to the evolution of patent law, the United Kingdom is fortunate to possess a wealth of reported authority on this topic which is probably unrivalled elsewhere. It is not, therefore surprising that in a case so difficult as the present the parties have combed the books and brought before us dozens of citations in an effort to throw light on the various problems. Of course, this is not a matter for criticism, and I warmly acknowledge the diligence and skill with which the authorities have been assembled and deployed. Yet I believe that we should approach them with caution.
We are concerned here with a new statute which differs in important respects from the former law, not only as regards procedure, but also in the balance which it strikes between the interests of the researcher and of the public. We should not assume that the new Act is just the old English law re-written, or that statements of principle or passing observations on individual questions can now be culled from the reported cases and applied without reserve, however eminent the sources from which they are drawn. This is all the more so given that the source of the act is a treaty, and moreover a treaty written in three languages of equal status.
Whilst it would be senseless to ignore the obviously English parentage of some of the expressions used in the English text, it would also be a mistake to assume that the Act, which must be interpreted by other national courts and by the tribunals of the European Patent Office, can best be illuminated by researching the arcana of the English patent law, distilled from decades of reported cases by the very special analytical processes of the common law. I think it safer to concentrate on what the new statute says.
Moreover, it must be remembered that we are here dealing with a new problem. New not only because the technology is very recent, but also because the problems arising from the manufacture of an existing substance with known properties, by the application of a known technology, through the medium of a new discovery and novel intermediate artifacts (the vectors) have never (so far as I can see) been considered by the courts. Nor have I been able to discover a consistent thread of logical reasoning, running through the cases where the courts have had to consider all kinds of individual situations, which can be applied with any confidence to this novel situation. Thus, although the authorities where relevant must be given the greatest weight, we should not take it for granted that they are decisive in the context of the new Act.
My second general comment relates to the issues of policy which have been much pressed upon us in relation to more than one of the issues in contention. Naturally, the fate of the individual patent is of the greatest commercial importance to this individual patentee. But the importance of the present action far outstrips the ultimate decision itself, for it is impossible to arrive at a decision without expressing conclusions which apply to ths fledgling industry as a whole, and which (if reflected in courts elsewhere in the European Community and outside) may have an important bearing on its future. On the more general plane, the argument for the patentee is easily understood and widely voiced.
The kind of research which leads to a marketable form of t-PA is very expensive. The success rate is low. The benefits to suffering humanity are great. If a sufficient reward is not given in those instances where the research bears fruit, the industry will not attract the venture capital which it needs for survival, the research will cease, and humanity will continue to suffer.
The appeal of this proposition is undeniable, and indeed it reflects the reasons why we have a patent system at all, yet the arguments are not all one way. If the criteria for patentability are pitched too low there is a risk that mere hard work or superiority of resources, or simple good luck, will entitle a researcher to a monopoly, the commercial and social justification for which is by no means clear, given the risk of stultifying the development of the industry by open competition.
This tension has always been a feature of patent law, and the judges have often given effect to their views on the way in whch the competing interests of the applicant and the competitors should be resolved.. For a recent example, one need look no further than the speech of Lord Diplock in American Cyanamid Co (Dann’s) Patent  RPC 425. Even after 1977 there is still ample scope for the courts to adapt the existing principles to new situations.
I do not, however, believe that this is the occasion for the court to embark on major judicial legislation. I believe that this would be out of turn with the concept of a homogeneous patent system, regulated by a convention which has not long been in force. Moreover, I am not confident of knowing where the justice of the case may lie. It may be that the explosively new technology with which we are concerned has exposed some deep flaws even in the current regime; but if this is so, any necessary repairs must be effected by the legislation, not by the courts. I approach the present appeal on the footing that our task is to understand the Act, and apply it.
In the light of these observations, I return to the twenty claims now before the court. When their implications are spelt out as I have suggested, it seems to me plain that the monopoly granted to Genentech far outstrips any legitimate reward for their success in winning the race to the expression of the protein in a workable cell line. Mr Gratwick was indeed disposed to concede as much, as regards claims 2 and 4, and in effect invited the court to disregard them.
Tempting as the idea may be, I do not think that this is a course we can adopt. The learned judge had held these claims (amongst others) to be bad on a ground which, as all members of the court agree, was beyond his jurisdiction. There is no application to amend, and the claims must be reinstated unless their invalidity is affirmed on some basis: and if the latter, we must say what it is. Equally, although the two claims comprised in claim 9 are conceded by Mr Jacob to be valid, subject to the question of obviousness, we must at least look to see why this should be so if we are to strike down some or all of the others. Such an enquiry will enable us to test the law at the two ends of the spectrum whether the answers are regarded as plain, before transferring the reasoning to the more difficult cases in the middle.
Before addressing any specific claim, it is convenient to list the grounds upon which (leaving aside the question of an inventive step) the patent has been subjected to attack. These are as follows.
1. The claims do not relate to “inventions”.
2. They are not supported by the description in the specification, and therefore do not comply with section 14(5).
(3) Those parts of the events leading up to the making of the invention claimed which had any element of originality consisted of discoveries and are, therefore, excluded from patentability by section 1(2).
4. The claims assert a monopoly for what is no more than a trivial idea.
5. The claims relate to “products” which are not new.
6. The specification does not sufficiently inform the skilled person how the patent should be worked.
I now turn to the claims themselves, beginning with those numbered 2 and 4. In company with all the other claims except claim 9, these were struck down by the learned judge on the second of the grounds just stated, namely that they were not supported by the description, and hence failed to comply with section 14(5).
If I had thought that the learned judge had jurisdiction to take this course I would have agreed with his conclusion, and with the whole of his reasons for it, which I need not repeat. I now consider it clear that this is a patent which should not have been granted, at least in its present form. In saying this I intend no disrespect to the Patent Office, for we have had the advantage, denied to the examiner, of sustained and skillful adversarial argument, together with the invaluable assistance of a renowned authority, to help us in the penetration of an outstandingly difficult technology. Nevertheless, the application did not comply with section 14(5)(c), and should have been rejected.
Common sense would suggest that the matter now can and should be put right. Unfortunately this is not so. My Lords have already given the reasons, in terms with which I respectfully agree. The opening words of section 72(1) are simply too strong to enable the court, as guardian of the public interest, to assert an inherent power to revoke a patent on grounds not expressly conferred by the statute. This is a conclusion which I much regret, for it erects an obstacle not only to the trial judge’s route, but also to other grounds of objection which, as I shall suggest at a later stage, I also believe to be intellectually sound.
The result has been that the true shape of the issues has been distorted, and the conceptual difficulties, of what is undoubtedly a difficult area of patent law, have been much enlarged. It is, moreover, a ground for general concern, for the public will suffer if research in such a dynamic field is inhibited by the grant of monopoly which is much too wide and will suffer again if the Patent Office is driven into an excessively cautious attitude for fear of making, at a stage when the proceedings are not fully contested, a mistake which cannot afterwards be put right. Nevertheless, I see no room for doubt, and conclude that the learned judge’s grounds for decision cannot be upheld.
In expressing this conclusion I offer no opinion on whether Swift & Co’s Application  RPC 37, is still good law in the new statutory context, for that decision was concerned with the intrinsic patentability of the invention, and not with the contents of the application for grant.
his is not, of course, the end of the appeal, for it is rightly acknowledged, on behalf of Genentech, that if there is some other ground, enumerated in section 72, which applies to these claims, the court has power to give effect to it, however closely it may resemble the ground which we have been compelled to reject. (Mr Gratwick’s concession did not go so far as to accept that all the arguments now put forward by the respondents are open to them, for he maintained that some, at least, of them had not been pleaded. In part, this complaint is justified, but I am satisfied that no injustice has been caused by the opening up of these new matters, and in the particular circumstances of this appeal I think that it would be wrong to exclude them.)
It must, therefore, be considered whether any of the other grounds are fatal to claims 2 and 4. When doing so, and when reviewing the other claims, the court must be scrupulous not to revive, simply by giving it a different label, the judge’s correct but inadmissible conclusion that Genentech have asserted a monopoly wider than is justified by any invention described in the specification. At the same time, it is permissible to bear in mind that the grounds of objection have always been prone to overlap, and that the very same factors may lead to an irredeemable flaw in the patentability of the supposed invention, and to an impossibility of framing an application which complies with section 14(5).
In this light, I turn to the first ground of objection, which is that claims 2 and 4 do not disclose an invention. This raises an issue which does not seem to have been touched upon in any reported case. We are here concerned with a process for synthesising a substance identical to that which occurs in nature. The t-PA produced by the process is not “artificial” t-PA or “synthetic” t-PA, in the sense of artificial silk or synthetic rubber: ie in the sense of something which resembles the natural substance, or can perform a similar function, or act as a substitute. It is not ersatz. The t-PA which Genentech made is neither more nor less than t-PA.
It is true that the word “recombinant tissue plasminogen activator” may be a useful turn of phrase, but this should not be allowed to disguise the fact that “recombinant” describes, not the product itself, but its history. It is I believe, a failure to acknowledge this which has compounded the already substantial difficulties of relating these unusual claims to the framework of patent law, and which have diverted attention away from the fact that the success of Genentech lay, not in the invention of a new substance — for protein molecules with the amino acid sequences shown in figure 5 and the functional characteristics set out in the specification have existed since far into the distant past — but in the accomplishment for the first time of a method of creating that substance. Wrapped up as it may be in the product claims, I believe that in truth what Genentech have invented (if they have invented anything at all) is a process.
In the ordinary way, this point would be of no more than academic interest, for the effect of section 60(1)(c) is that the patentee can inhibit any unauthorised dealings with the products of his patented process, so that the validity of any associated product claim is neither here nor there. In the present case, however, I believe that the distinction is important. If I am right in the opinion, which the trial judge evidently shared, that there is something fundamentally wrong with at least some aspects of this patent, it is essential to analyse the claims and the invention accurately, in order to find out what is wrong, and to see whether the structure of the legislation permits a remedy.
This suggestion of a need to identify the invention leads me to a part of the case which I found most perplexing. Most of the arguments have been concentrated on the three conditions precedent to the grant of a patent set out in paragraphs (a) to (c) of section 1(1) – and understandably so, given the shape of the old law. But this approach tends to mask a more fundamental requirement which must be satisfied before a patent can properly be granted, namely that the apllicant has made an “invention”.
In my judgment, this requirement emerges clearly from the opening words of section 1(1):
“A patent may be granted only for an invention in respect of which the following conditions are satisfied . . .”.
Compliance with these four conditions turns an invention into a “patentable invention” (see the concluding words of section 1(1).
Section 1(2) then goes on to exclude certain matters from the scope of “invention” — not “patentable invention”. To my mind this shows that question whether the claim discloses anything which can be described as an invention must be answered in the affirmative before compliance with paragraphs (a) to (d) becomes relevant: and the wording of Article 52 in all three languages is even more plainly to the same effect.
I am fortified in this view by the Guidelines for Examination in the EPO which, in paragraph 1.1 of Chapter IV, say:
“There are four basic requirements for patentability:
1. There must be an ‘invention’.
2. The invention must be ‘susceptible of industrial application’.
3. The invention must be ‘new’.
4. The invention must involve an ‘inventive step’.”
In paragraph 2.2 the Guidelines go on to point out that:
“It must also be borne in mind that the basic test of whether there is an invention within the meaning of Article 52(1) is separate and distinct from the questions whether the subject-matter is susceptible of industrial application, is new and involves an inventive step.”
It might, at first sight, seem that this adds a wholly unnecessary complication, where paragraphs (a) to (c) do all that is necessary to define the permissible subject matter of the monopoly, and that it is absurd to speak of an invention which does not involve an inventive step — as one must be ready to do, if the interpretation just suggested is sound. But this is not so, as the example of an amateur experimenter working alone will demonstrate. It must quite often happen that such a person will, through the exercise of imagination and ingenuity devise a useful product or process. By doing so, he will, in a real sense, have made an invention. But it will also quite often happen that, although he does not know it the same device may already have been invented and published by soneone else: so that the experimenter’s invention, though real enough, is not patentable, for want of novelty.
Also, although it may seem paradoxical, his invention may also fail as a patent, on the ground that it involves no inventive step: for the latter concept is artificial, in that it relates the “step” to the objectively prior art, rather than to the actual knowledge of the inventor before he takes that step. The inventor may have made what for him is a giant leap, and yet find that through prior art of which he was unaware, his invention must be treated as obvious.
Thus, although the objection to a patent on the ground that it monopolises something which is not an invention will very often overlap another potential objection — and such an element of overlapping is nothing new in patent law — it is none the less a separate element which, in the appropriate case, ought to be separately investigated.
Turning to the specific question of a naturally occurring substance such as t-PA, there are several quite distinct types of new steps forward which may be taken by an ingenious researcher. Thus —
1. He may for the first time find out that the substance exists.
2. He may find out that it exists in a hitherto unsuspected place or form.
3. He may find out some of its properties, such as its molecular composition or physical structure.
4. He may find a new use for it.
5. He may devise a new way of extracting the natural substance from other natural substances, or purifying it from harmful admixtures.
6. He may devise a new way of giving the natural substance a new form.
7. He may devise a new method by which the substance is synthesised from other substances.
Now it seems to me clear that quite apart from section 1(2)(a), and quite apart also from section 60(1), which inferentially confines “inventions” to those relating to either a product or a process, the first three activities, unlike the last three, do not in ordinary speech amount to an invention. You cannot invent water, although you certainly can invent ways in which it may be distilled or synthesised. I prefer to say nothing in respect of the fourth activity, since it does not arise for consideration here, and may prove to be a source of difficulty in the future, except within the limited compass of section 2(6), for at present I do not see how the ascription of a hitherto unknown property to a known substance can be either the invention of that “product” or the invention of a “process” — although some authorities seem to assume that it can.
So also with t-PA. A properly framed claim for a process for synthesising, refining or transforming the substance could be capable of founding a patent, but I am unable to see how claims 2 and 4 could by any stretch be described as claims to an invention. In my judgment, they should fall at the very first hurdle.
It might perhaps be suggested that section 125 provides an answer to this point. I disagree. The purpose of the section is to tell the reader where to find the invention, to which he will apply the various tests of section 1: namely, in the claim. But the claim is no more than a claim, and if what the applciant asserts is not an invention he cannot make it such simply by including it in a claim.
This then is the conclusion at which I arrive by reading the plain language of the Act and the Convention. Nevertheless, in a field so dfficult it must be acknowledged that what appears at first to be plain may, to others, seem not plain at all, and that the authorities may indeed show that the first impression was plainly wrong. I have, therefore, examined the reported cases to see whether they compel a different view of the legislation. it seems to me that only three call for discussion.
Perhaps the closest is Chinoin’s Application  RPC 39. The applicant had by examination of soil samples identified a species of micro-organism (denoted MNSR) which had not previously been known. Evidently the applicant had initially discovered one strain (“the parent strain”) of this species. All strains of MNSR could be cultivated so as to produce in a greater or lesser degree the known antibiotic sysomycin. One such strain (SI/109), derived by breeding or mutation from the parent strain, was regarded by the applicant as particularly suitable for this purpose. The application includes two relevant claims:
1. A very broad generic claim to the MNSR species.
2. Strains of MNSR derived from SI/109.
An example of strain SI/109 was deposited in support of the application. The patent strain was not deposited.
Objection was taken by the examiner to various aspects of the application. On an appeal to Falconer J, from a decision of the hearing officer rejecting the application, it was conceded, on behalf of the Comptroller, that claim 3 was valid. Falconer J stated that this concession was rightly made (page 47, line 24). He went on to hold, however, that claim 1 was invalid, because in the absence of a deposit of the parent strain there was nothing in the patent to show the man skilled in the art how to work the invention. Evidently the learned judge would have reached a different conclusion if there had been such a deposit.
Two points may be made on this case straight away. First, it was not directly concerned with the present problem; for it had plainly not occurred to anyone that the known end-product sisomycin could be patented, simply because the applicant had found a new source for it.
Secondly, the deposited strain SI/109 was seemingly a new product, obtained by breeding or other manipulation from the parent strain. What does seem to me rather puzzling was the apparent willingness to accept that the whole species could have been patented if there had been either a deposit of the naturally occurring patent strain or a sufficient description of it. The answer may be that correctly construed the claims related to a process utilizing the species, not the species itself (the claims are not quoted verbatim in the report). This would fit in with section 14(4), which provides as follows:
“(4) Without prejudice to sub-section (3) above, rules may prescribe the circumstances in which the specification of an application which requires for performance the use of a micro-organism is to be treated for the purposes of this act as complying with that subsection.”
The language of this sub-section is not very happy, since a reference to the invention seems to have dropped out: one cannot “perform” a specification or an application. Nevertheless, the presence of the words “the use of a micro-organism” do tend to suggest that the subsection is not dealing with a case where the micro-organism itself, as distinct from some employment of it, is the subject of the claim. Or perhaps section 14(4) is a signal that micro-organisms are to be given a special status as regards patentability, in a similar manner to the conferring of such status on products for therapeutic use by section 2(6): see John Wyeth & Brother Ltd’s Application  RPC 545, at 565. At all events, the present problem was clearly never raised at all in Chinoin, and I do not find that the expressions of opinion on claim 1 need inhibit us now from giving to the word “invention” whatever meaning sits most appropriately with the language and structure of the Act.
Secondly, there is John Wyeth, supra. Here, in the interests of uniformity, a Divisional Court adopted a ruling of the Enlarged Board of Apepal on the meaning of Articles 52 and 54, in preference to its own opinion on the corresponding provisions of the English Act. The arguments in this case were directed to novelty, rather than the intrinsic patentability of a new use for a known substance, and it was in any case concerned to new therapeutic uses, which undoubtedly are given special treatment by the Act. I would prefer to reserve my opinion on the two opposing views of the particular questions which there arose, and which may call for re-examination on a future occasion. What does, however, seem clear is that the case decides nothing about the patentable status of a known substance with a known use, created by a new process.
Parenthetically, I would add that section 91(1)(c) does not, in my view, have the effect suggested in argument. The requirement that the court shall take judicial notice of the decision of the relevant Convention court is directed (like the remainder of the section) to evidentiary matters; in this case the mode of proof of matters which might otherwise have to be proved as foreign law.
The subsection does not give to rulings of other courts any greater status than they would otherwise possess, although, of course, the desirability for a uniform course of decision in matters touching the Convention is manifest (see, for example, section 131(7)), and the Board of Appeal as the central decision making body of the European patent system must be hearkened to with particular attention.
The third authority which calls for remark is American Cyanamid Co (Dann’s) Patent  RPC 425. There, the invention for which the patent was claimed consisted of the cultivation of certain strains of a known micro-organism with a view to the production of a new antibiotic. The claims extended both to the method of production and to the antibiotic itself. The micro-organism existed in nature in very many strains, and even a skilled researcher might not happen upon those relied upon by the applicants as the starting point for their process.
This is a case of the greatest importance in a number of respects, not least for the expressions of opinion on points not directly in issue, although it is essential to bear in mind that it was concerned with the former legislation, which differs in important respects from the Convention-based Act of 1977. I do not, however, find that it bears upon the particular point now under discussion.
The existence and nature of an invention was not in issue (see per Lord Reid at page 434 line 27 and page 437 line 28), the dispute being concerned with questions of description. It is true that Lord Diplock did, at page 452 line 30 in his dissenting speech, express an opinion on the width of the concept of “invention”. But this was a claim for the new product profyrimycin and its method of production, and not for the naturally existing, although hitherto undiscovered, starting material for the process. The patentability of the latter was, therefore, not in debate, although it is significant that Lord Wilberforce expressed in passing the opinion that it was not capable of being patent (page 448 line 10).
Accordingly, I consider that the authorities do not demand any qualification of the reason already expressed for holding that claim 2 and 4 ought to be revoked.
Even if this reasoning were not well-founded, I would arrive at the same conclusion on the ground that any invention disclosed by claims 2 and 4 is not new. Although this is Mr Gratwick’s preferred explanation of the conceded invalidity of these two claims, I confess to feeling some reservations about it. There is a real difference between devising (to use the old word) something which is not patentable because it is part of the state of the art, and not devising anything at all. To say that when Genentech succeeded in working out how to make t-PA they devised an old product seems to me an abuse of language. But still, if the first objection is unsound, I would adopt this one in its place.
Having reached this conclusion, I think it unnecessary to describe the other possible objections to claims 2 and 4, and will pass directly to claim 9, which is at the opposite pole. Here it is undeniable that Genentech have devised and made something in the shape of the two vectors there described, and that what they have devised is new, since no such constructs as the two vectors have ever existed before. No objection can be taken on the ground that the method of reconstructing the artifacts is not shown in the specification, or that they are of such excessive generality as to be no more than a statement of an idea.
There is, however, another matter which, when it first arose during the argument of the appeal, seemed to pose an unanswerable ojection, not only to claim 9 but also the other claims as well. The entire thrust of the claim to a monopoly, so far as it can be discerned from the specification, was that Genentech had been first to the sequences. Equally, the fact that Genentech had “won the race” to the sequences was at the heart of the arguments developed before the trial judge, and opened to this court on the hearing of the appeal, it being explicitly accepted (as was inevitable in the light of the Points of Agreement, paragraphs 24 to 26) that nothing subsequent to the ascertainment of the sequences was in any way outside the competence of one ordinarily skilled in the art.
Now the problem with this approach to the case is that the inventive step (assuming it to be such) which Genentech have chosen to make the foundation of their case amounted to the ascertainment of an existing fact of nature: ie, to a discovery. We are, therefore, remitted to section 1(2) of the Act, which reads as follows:
“(2) It is hereby declared that the following (among other things) are not inventions for the purposes of this Act, that is to say anything which consists of (a) a discovery . . .
. . .
” but the foregoing shall prevent anything from being treated as an invention for the purposes of this Act only to the extent that a patent or application for a patent relates to that thing as such.” (Emphasis added.)
Given the difference of opinion upon it, it must be accepted that this provision admits of more than one meaning. Nevertheless, I confess that on first reading I gave to it precisely that interpretation which had been adopted by the principal examiner and by Falconer J respectively in Merrill Lynch Inc.’s Application  RPC 1, at 9 11. The following citations will serve to make the point: “I consider that the proviso in section 1(2) of the present Act makes it clear that the inclusion of an excluded thing in the subject matter of a patent application does not inevitably mean that the subject matter is unpatentable. However this subsection does not say that anything is prevented from being treated as an invention ‘only if’ the application relates to that thing as such, instead of the words used are ‘only to the extent that’. I am of the opinion that this wording . . . should be understood as meaning that anything which is in an excluded category is not to be treated as an invention and consequently its inclusion cannot be considered to contribute to the required novelty and inventive step. This does not mean that a claim should be treated as if anything falling in an excluded category was completely excised but rather than the mere presence of such a thing cannot constitute the invention. For example, in many cases a machine tool controlled by a program might not be excluded from patentablity but if the physical components were wholly conventional and the system as a whole functioned in a conventional manner any distinction from the prior art would probably reside in the program itself and thus exclude the whole system from patentability.”
“. . . I think Mr Thorley is right in his submissom for the Comptroller, that Mr Pumfrey’s construction of the proviso or qualification in section 1(2) does not give effect to the wording ‘only to the extent that’. In effect, it replaces that wording by words ‘only if’ as though the concluding words of the qualification read ‘only if a patent or patent application relates to that thing as such’. That seems to me to be a much narrower construction of the qualification than the actual wording requires. Plainly, if a patent or patent application relates only to one of the matters specified in (a), (b), (c) or (d) of the subsection, eg a computer program, that would be excluded from patentability by the subsection, but in my view the excluding effect of the subsection is wider than that. It seems to me that the words ‘to the extent that’ contemplate that the subsection is also to be applicable to cases where the invention involves one of the excluding matters (specified in paragraphs (a), (b), (c) or (d) but does not relate to it only. Using the exemplification of an invention involving a computer program, Mr Thorley submitted that the wording ‘only to the extent that’ means that there cannot be a patentable invention in so far as the invention resides in the computer program itself, but if some practical (ie technical) effect is achieved by the computer or machine operating according to the instructions contained in the program, and such effect is novel and inventive (ie not obvious) a claim directed to that practical effect will be patentable, notwithstanding it is defined by that computer program. In my judgment, Mr Thorley was right in that submission.”
It has been argued for Genentech with much force that the old law was different, and that the combination of a new discovery and an obvious application of it can lead to a patentable new article and (semble) to a patentable new process as well: see Lane-Fox v Kensington & Knightsbridge Electric Lighting Co Ltd (1928) 9 RPC 413, and Reynolds v Herbert Smith & Co Ltd (1903) 20 RPC 123. I do not, for my part, regard this as germane. The members of the court were discussing the combination of a new idea, conceived by the patentee, namely the shogging of the comb bar, with an obvious way of carrying it out.
By contrast, the present appeal is concerned not with a new idea, but with an existing fact of nature, newly discovered. Ideas and discoverires have this much in common, that neither is patentable in isolation. The position seems to be the same in the United States, at least as regards computer programs which (as under the 1977 Act and the Convention) are not patentable per se, and which, according to the majority opinion in Diamond v Diehr and Lutton (1981) 450 US 175, may nevertheless found a patent when combined with an obvious embodiment. (Genentech also relied upon Hickton’s Patent Syndicate v Patents & Machine Improvements Co Ltd (1909) 26 RPC 359, but this does not mean that they stand on all fours).
All this is very well, but we are here concerned with the new European law. Is the result reached by the examiner and Falconer J in relation to a combination between a computer program and an obvious application of it so plainly unjust that it must be rejected at all costs? It is said to be absurd that a discovery in (say) the medical field of the utmost inventiveness and utility should not be capable of yielding a patent in conjunction with some obvious embodiment: for example, where an exsisting substance with a newly-discovered healing property is combined with a commonplace vehicle so as to make it therapeutically acceptable. But is this so odd? Perhaps a discovery ought to be patentable, but the Act has declared that it is not. There seems no reason why an unpatentable invention should be turned into the subject of a monopoly merely by the addition of something trifling.
At the same time, one must acknowledge the weight of the arguments the other way. The words “anything which consists of” in section 1(2), and “the subject-matter or activities referred to” in Article 52(3), can, I accept, be linked with “as such” to suggest that a claim fails only where it relates to a discovery and nothing else. This interpretaion gains support from the decisions of the Technical Board of Appeal in Koch & Sterzel T26/86 and in Vicom T208/84. It is, however, to be noted that in the former the Board declined to follow a decision of the Bundesgerichtshof in the Federal Republic of Germany (not cited to us), and that the report of the latter is so compressed as to seem, to me at least, almost incomprehensible.
This controversy raises a puzzling question. Since the point in section 1(2) applies to all the claims, it cannot simply be brushed aside on the basis that neither party is interested in claim 9, an infringement of which is not in contemplation. On the other hand, I am uneasy about deciding the appeal in the respondent’s favour on a point which was neither pleaded nor raised before the trial judge, nor even relied upon before this court until it emerged during argument. This is the more so because I am still not entirely convinced that the discovery of the entire nucleotide sequence was a step towards, or even preceded, the creation of the expression vectors which, by the conclusion of the argument, had become the lynchpin of Genentech’s case (together with the associated process in claim 17).
Curiously enough, given the crucial significance attached to this discovery — it will be noted that Genentech’s skeleton argument states that “The foundation of the patent as noted on page 1 1.24 et seq of the specification is Genentech’s discovery of a novel DNA sequence which encodes t-PA” — its precise place in the order of events is not disclosed by any document before the court. Undoubtedly, the Genentech team needed to know the sequence in order to verify that the source of the expression which they were obtaining was indeed a full-length sequence, and undoubtedly also a knowledge of the sequence would be of the utmost importance to subsequent workers. Yet Mr Gratwick accepted during argument that a course of events consisting of (a) the discovery of the sequence, (b) the creation of the final vector and (c) the production of the desired protein, could not be clearly demonstrated on the evidence. This uncertainty puts in doubt not only the factual assumptions of the argument on section 1(2)(a) but also raises the possibility that the close attention focussed on the discovery may have been misplaced.
Be that as it may, it is only fair to recognise that this particular group of problems was a very late arrival on the scene. If it had been foreshadowed in the pleadings, or explored at the trial, there can be no doubt that the factual position would have been cleared up, so that a proper basis for argument on section 1(2)(a) would have been laid. But this did not happen. In these circumstances, and in the light of the fact that I understand my Lords to be each of the firm opinion that the objection is not well founded, I will say only that if the logic of the view expressed by the principal examiner and by Falconer J in Merrill Lynch is to be overturned, the only available means is to emphasise the words “which consists of” in the opening passage of section 1(2). For the remainder of this judgement I will, however, assume, in company with my Lords, that the identification by Genentech of a discovery as the foundation of their patent is not fatal to its validity.
I now turn to claim 3. This was the centrepiece of Genentech’s case when the appeal was first opened. As the argument progressed, it began to recede and attention was directed primarily to claims 7 and 8, and to the associated process claims of which claim 17 is an example. With respect, I believe that Mr Gratwick was prudent to redeploy his clients’ arguments in this way, forto my mind (as I have already said) there is no difference between recombinant t-PA and any other kind of t-PA. If so, claim 3 must, like claims 2 and 4, be unsound. Genentech did not invent t-PA. At most, they invented a new way of making it. The same objection is, in my view, fatal to claim 1.
I turn next to claims 7 and 8. The points arising on these two are the same, and I will concentrate on claim 8 which is of the greater practical importance.
Claim 8 is in a different category from the claims previously discussed. It differs from claims 1 to 4, since it relates not to the final product t-PA but to an intermediate artifact, of value only as a tool for making t-PA. It differs from claim 9 in that the latter is limited to tools which Genentech actually made, whereas claim 8 comprises a generalised class of such tools, of which the two cited in claim 9 are instances.
It is here that we strike what seems to me the most difficult part of the appeal. The problem may be illustrated by Mr Jacob’s notion of arranging a series of possible claims concentrically as in a target. At the periphery, one finds “cheap and pure t-PA”. Then, working towards the centre one arrives at “cheap and pure t-PA, using oligo probing and genomic cloning, with pA-trp12 as the vector”. It seems plain enough that, assuming the requirements of section 1(2) to be satisfied (including of course the need for an inventive step), the claims at or near the bull’s-eye are patentable, and that those at or near the “outer” are not. But why not, and how does one identify the ring at which the claims change their character? Mr Jacob’s propostion is that as they progress towards the edge of the target the rings arrive at a degree of generality where all that is being claimed is an obvious idea: or, to express it differently, a general desideratum which could be written down without originality by anyone possessed of a passing acquaintance with the field. Mr Jacob would, therefore, say, if I correctly understand the argument, that the claim fails under section 1(2)(a).
Whilst I see the force of this proposition, I do not accept it for two reasons. First, because the argument does not lead to a solution, since even the most specific version of the claims (viz those contained in claim 9) could have been designed on the drawing board, in the sense that the shape of a plasmid molecule including “the nucleotide sequence encoding for t-PA” and the various other features of p-RIP and PAtrp12 could have been laid out in advance in a researcher’s study: and yet everyone agrees that these are intrinsically capable of being patented.
Secondly, because I do not believe that when addressing the issue of obviousness in the context of a product patent one ever does so by asking whether the idea of the product is obvious. I return to this point in the next-section of this judgment. Finally, because Genentech have not simply written down a class of products as an abstract intellectual exercise (not patentable because a bare idea is never patentable), but have written down a class, two members of which they have actually constructed.
There is, however, another way of expressing this objection, which is as follows. Beyond a certain degree of generality the claim is so wide as to embrace products which Genentech have not invented, and which others may invent in the future. Since no-one can obtain a patent for something which he has not invented, claim 8 must be invalid in part, and if invalid in part and unseverable the whole must fall away. This argument is quite different from the one examined in relation to claims 1 to 4. There, the complaint is that the subject-matter of the claim is not capable of ranking as an “invention” at all. Here, anything falling within claim 8 could be an invention; but the class includes products which Genentech have not invented.
I believe this argument to be intellectually sound, but am driven to conclude that it is defeated by section 72(1). The only legitimate way to discover whether the claim outstrips anything which was invented is to compare the inventions set out in the description which is the relevant invention by virtue of section 125(1)) with the invention comprised in the claim. Yet this is exactly the exercise to be carried out when deciding for the purpose of section 14(5)(c) whether the claim is supported by the description; an exercise which all agree can be performed only at the stage of examination. It cannot, in my opinion, properly be introduced into revocation proceedings merely by giving it a different format.
The same obstacle must, I believe, stand in the way of somewhat different argument, which I would express as follows. Let it be assumed, contrary to the decision of Merrill Lynch, that a discovery coupled with an obvious embodiment can amount to a patentable invention. Nevertheless, there must still be real embodiment. Here, there was no embodiment of the discovery, except to the limited extent shown in claim 9. The supposed embodiment set out in claims 7 and 8 did not really exist, and once these are stripped away there is nothing left of the invention, except a non-patentable discovery. If correct, this argument would form an alternative objection to some of the earlier claims, if my own prefered view that one cannot “invent” t-PA is to be rejected.
Whilst I acknowledge the force of this argument, I find myself unable to accept it, for it requires the court to ascertain whether the embodiment has actually been practised by the patentee. Since the only permissible source of the information on which to base this enquiry is the description, the task seems precisely the same as when considering whether the claim is fairly based: and this, as all members of the Court agree, is something which we are not permitted to do.
There remains, however, one further possibility, namely that the specification does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art. That this differs from an allegation that the claim is not supported by the description, and in that (unlike such an allegation) it may be brought forward in revocation proceedings, is clear from sections 14(3), 14(5) and 72(1). Significantly, the Guidelines say this, at D-III, 5:
“The following allegations, for example, do not constitute grounds for opposition: that the proprietor of the patent is not entitled to the European patent, that the subject matter of the patent lacks unity, that the claims are not supported by the description (unless it is also argued that the claims are so broadly worded that the description in the specification does not sufficiently disclose the subject-matter within the meaning of Articile 100, subparagraph (b)). . .”
One must, therefore, ask this: Given the benefit of the specification to work from, could a person ordinarily skilled in the art create without a further inventive act all the products stipulated in claim 8? If I have correctly interpreted this claim, at an earlier stage of the present judgment, it seems to me that the answer must be negative. Indeed, I believe that even an inventive person could not be guaranteed to construct all these products in the present state of the technology. If this is so, claim 8 and also the associated claim 7 must be apt for revocation.
I have hesitated before expressing this conculsion, partly because I am conscious of the risk that to allow the objection would let back into the case the trial judge’s ground for decision, which all are argeed is excluded by section 72(1), and partly because this way of putting the matter was not foreshadowed in the respondents’ Particulars of Objections. As to the former, however, it is undeniable that a patent, and an application for a patent, may be invalid on grounds which give different statutory labels to what are essentially the same flaws: and any qualms are largely allayed by my strong impression that, by whatever route attained, the conclusion is fair. As to the question of pleading, I entirely endorse the view that in the interests of an orderly and just trial an objector should, as a rule, be held to the case which he has reduced to writing. But the present litigation is out of the ordinary. Against the background of a complex and unfamiliar technology the parties have had to address a series of legal concepts which, to me at least, have seemed remarkably elusive. In the event there have throughout the proceedings been substantial reformulations and shifts on both sides. This is not a matter for surprise or criticism, but it does suggest that insistence to what would otherwise be healthy formalities would be out of place. I am satisfied that counsel for Genentech have been able to master a full response to this line of argument, and that there will be no injustice if it is given its full weight.
The arguments relating to the process claimed in clause 17 are substantially the same as those already discussed, and I will not occupy space by setting them out again. All the process claims are, in my view, invalid, with the exception of claim 19 which is maintainable for substantially the same reasons as claim 9.
In the result I would say that apart from claims 9 and 19, the patent should be revoked, whatever the position as regards the existence of an inventive step. I must, however, go on to discuss this question, partly it affects the validity of claims 9 and 19, and partly because (as I understand it) my Lords are not in agreement upon it.
The second group of issues: inventive step
Because the technology which is the background to the issue of obviousness is so novel and complex it seems, at first sight, that the difficulty in arriving at a solution is simply a reflection of the difficulty of the subject-matter. This is indeed so, in part. Not because the science is so hard as to be impenetrable, given the great help we have received from counsel and Dr Brenner, but rather because of the problems which a layman must encounter, in a situation where all the ideas and techniques seem objects of wonder, in distinguishing between those which are obvious and those which are not. Nevertheless, I have come to believe that questions of at least equal difficulty arise from the need to decide what the act means by “inventive step” and “obvious”, and what characteristics are to be attributed to the “person skilled in the art”. A solution to these problems will, in my judgment, take one much of the way, if not necessarily all the way, to a conclusion on the issue as a whole. I will, therefore, consider them first, beginning with the meaning of “inventive step” and “obvious”.
Before looking at the words of the Act, it may be convenient to explain why what might seem to be an empty exercise in verbal analysis seems to me central to the case. Take, for example, the word “inventive”. The layman might perhaps think of James Watt or Thomas Edison as typical inventors, and he might have in mind that the former’s conception of the separate condenser for a steam engine was the typical invention. James Watt started with the wasteful Newcomen engine and recognised (perhaps intuitively) that the key to greater efficiency lay in condensing the steam outside the cylinder rather than inside, as under the existing system. No doubt there were other technical difficulties to be overcome, but there was, in a real sense, a step from the prior art to Watt’s new device, and a step which can easily be recognised as inventive. The present case is in a wholly different world. All are agreed that Genentech did an excellent piece of work. The excellence seems to me, however, to have lain primarily in the field of management. Under the leadership of Dr Goeddel, the Genentech team identified a target for research; estimated the risk of failure and decided to take it; decided that recombinant technology would be a feasible method, and ascertained the routes which might be taken; assembled a team of highly competent workers, and furnished them with the necessary apparatus and supplies; saw that money needed to be spent, and spent it; and when obstacles arose, either overcame them, or changed course to evade them. All this was most praiseworthy, but is it the kind of activity which in law amounts to, or may involve as a part of it, an inventive step?
Again, what is meant by “obvious”, and what is it that must not be obvious if there is to be an inventive step? These questions raise no difficulties in the simple, almost old-fashioned case, where the mechanical engineer contemplates the existing art, identifies a possibility of improvement, and devises a way of achieving it. One has to ask whether the improvement claimed was an obvious way of fulfilling the need. The answer may be difficult, but it is not hard to pose the question. But here we are concerned with what was essentially a research and development project which, by its nature, could not be precisely “devised” in advance, however inventive its participants.
There would inevitably be unforeseen pitfalls, and problems which could only be solved empirically. In such a situation the way in which the question about obviousness is put will often, if not always, dictate the answer. Thus, taking the artifact p-RIPA as an example, if one asks whether its precise configuration was obvious at the time when the project began, the answer must be negative, since nobody knew what was the sequence of the DNA which would fill one segment of the circular molecule. A similar answer would be given to the question whether in the existing state of technology the artifact could be made. So also with the question whether the precise route by which the sequence would be elucidated could be forecast in advance. But if one asks whether the general nature of the route, or choice of routes, towards the goal was obvious, the answer may be different, as it may also be if the correct question is whether the methods of trial and error, or adjustment of technique, which would or might be necessary to overcome or circumvent any unforeseen obstacles, could have been forecast in advance.
I believe that the prime reason why these questions are so hard to formulate is that the structure and philosophy of the Act are not appropriate to an enterprise of the kind undertaken by Genentech — nor, I must confess, to an enterprise of the kind considered in Dann. Apart from this, however, there are two sources of difficulty within the wording of section 3 itself. The first concerns the definition in section 3, which reads:
“An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art . . .”
I refer to these words as a defintion, since it has been assumed throughout this case (and the contary is not suggested in any decision or text laid before us) that section 3 contains an exclusive definition of “an inventive step”: ie, that the expressions “involve an inventive step” and “is not obvious to a person skilled in the art” are synonymous. Indeed, throughout the argument of the appeal the expressions were used interchangeably, largely to suit the shape of the contention currently being advanced, and it would be quite inappropriate to cast any doubt on this at the present stage, although it may be wondered why, if the second term is included to tell the reader exactly what the first means, the first was used at all in section 1.
I do, however, think it legitimate to say that many of the problems seem to me to arise because (to my mind at least) the terms are not wholly synonymous, in their ordinary use, and that if section 3 contains an exclusive definition of obviousness, then one or both must be used in a rather special sense. Imagine a dull but efficient plodder who sets out on a voyage of discovery. He equips himself fully with all the necessary equipment, supplies and personnel, and pursues the enterprise with great tenacity in an entirely conventional manner. Eventually he makes a great new discovery. Many things would not be obvious about his enterprise. Whether he would find anything at all; what it would be when he found it; by what route he would get there; and so on. Yet he would not have done a single thing which could be described as inventive. The same must, I believe, be so in very many instances of valuable scientific research and development.
The second source of difficulty may be illustrated by contrasting the requirements of section 1(1)(a) and 1(1)(b). In each instance a condition is prescribed which must be fulfilled. Under paragraph (a) the invention must be new. Under paragraph (b) “it” (that is, the invention) must involve an inventive step. The invention which has to satisfy this test will be either a product or a process: see section 60.
In our case the claims are both kinds: claims 1 to 9 claim inventions in the shape of products, whether substances or artifacts; the remainder of the claims relate to processes. With relation to both categories of claim, the requirement of novelty imposed by section 1(1)(b) presents no problem. The elaboration of this requirement in section 2(1) shows that the test for compliance is purely concrete and objective. One simply has to ascertain the state of the art, and consider whether the object or process falls within it.
The requirement of an inventive step, in the light of the definition in section 3, is much less straightforward. One may take the case of a product claim for an artifact, such as a bicycle gear. One may see that a bicycle gear may be new, but for my part I do not understand how a bicycle gear can be obvious, or how a bicycle gear can “involve” an inventive step. Plainly, therefore, the requirement and its definition are not to be understood literally. It must be something about the gear which is to be inventive and non-obvious. What must that “something” be?
Since the notion of inventiveness and obviousness to the skilled person call up a reference to human reasoning, it is natural to assume that the Act requires the idea of the invention to be other than obvious to the skilled man. But this will not do, for it is common ground that an artifact is capable of being patented even though it was a known desideratum. Many years ago, an inventor could not have patented a heavier-than-air flying machine simply by writing down the concept, but equally the fact that the concept was capable of being written down in advance could not, in itself, exclude the rights of a person who had actually made one fly.
As a variant of this interpretation one may say that the test must be applied to the idea of achieving the object in the particular way claimed. This works well enough in the ordinary mechanical engineering case, but it will not work here. Nobody could have written down the precise composition of p-RIPA, since it was unknown, and in this sense it was not obvious: but to say that this made it “non-obvious” is to confuse novelty with inventiveness, a point to which I return a little later.
Another possibility is to fasten on the word “step”, and look at the steps which were actually taken by the inventor on the road to the first realisation of the invention: the first making of the product, the first working of the process. Yet this too is unsound, for the cases show that the history of the route by which the applicant arrived at the invention is not material to the validity of the patent, and may and indeed probably should) be omitted from the specification. Furthermore, I believe that this interpretation places too much weight on the choice of the particular word “step”, which in the context of a case such as the present must carry the meaning of “activity”, as the other two texts of the Convention make plain.
This being so, I see no alternative to an interpretation which requires the Office and the court to imagine the hypothetical skilled man, up to date with the prior art as circumscribed by the closing words of section 3, and looking towards the goal, whether or not precisely identifiable in advance, which will become the claimed invention if and when realised. Then, in a case like the present, which does not involve a simple leap from the prior art to the invention (as in the James Watt type of case) but rather entails a journey with numerous steps taken in sequence, the court must ask itself by what routes it would have been possible to proceed to the goal from the starting point. Then, the court must see what obstacles the skilled man would have faced on these routes, and must enquire how he could have overcome them, either in the way that the inventor himself overcame the obstacles on his chosen route or by circumventing or overcoming them in some other way, or by choosing another route from the outset, or by abandoning one route and choosing another.
Having identified these various expedients, the court must finally ask whether they could have been overcome by pertinacity, second technique or trial and error, with no more, or whether there would have been required a spark of imagination beyond the imagination properly attributable to the man skilled in the art. Only if the question is answered in the latter sense are the requirements of section 1(1)(b) fulfilled.
Thus far, I have proceeded solely by examining the words of the Act. Very properly, however, we have been referred to a considerable body of authority, almost all of it under the former law, on the nature of the inventive step. To my mind three cases stand out from the rest: May & Baker Ltd v Boots Pure Drug Co Ltd (1950) 67 RCP 23, I G Farbenindustrie AG’s Patents (1930) 47 RCP 289, and the speech of Lord Diplock in American Cyanamid Co (Dann’s) Patent  RPC 425. THese have called for the most anxious consideration, for if they were clear authority for the proposition that a patentable invention could be reached by the mere process of pursuing obvious lines of enquiry by trial and error until a favourable result chanced to turn up it would be necessary to consider whether, notwithstanding their high authority, they could now be regarded as consistent with the new legislation.
One possible approach would be to point out that the cases were different on the facts, since they discussed the situation where the research had disclosed a new product, or a new property of an old product. I do not feel able to take this course, for it would mean treating the cases as authority for the proposition that an invention becomes patentable simply by being new. I do not believe that this was ever the law, or was treated as such by the distinguished judges who took part in the cited cases. At all events, it is not the law now, as section 1 of the Act makes plain.
One must, therefore, look at the cases to see what they concerned. May & Baker was a case on amendment, turning primarily on the meaning of the original specifications, as compared with the narrower amended specifications, to see what invention they respectively described. The patentability of the invention was not directly in issue, nor was the distinction drawn between the invention (understood in the light of the description) and the inventive step which it must “involve” so clearly drawn as the current legislation requires. Nevertheless, there are important observations of Lord MacDermott (at 50) to be taken into account. Do these mean that methodical diligence will suffice? I do not think so, at least in unqualified terms.
In the first place, it appears that in May & Baker the choice of a profitable line of research required skill and experience of a high order (see per Lord MacDermott at 50).
Secondly, it may possibly be that, in the context of an entirely empirical process, May & Baker is authority for the view that the act of stumbling upon a favourable result by the kind of luck which might befall a skilled though unimaginative worker is to be regarded as inventive. If so, I believe that this attribution of a very special meaning to the word should be strictly confined to its own sphere, for otherwise the distinction plainly drawn by the Act between novelty and inventiveness will disappear.
Be that as it may, I do not consider that the decision is in point here. As I shall suggest, the choice of oligonucleotide probing did not require skill and experience of a high order, although it certainly required skill and experience. Nor was the field entirely empirical in the same sense as in May & Baker. In that case, as Lord MacDermott said, the applicant’s achievement “was not the result of applying any known priciple or law of nature”. By contrast, Genentech and their rivals were operating in a field where although there are, of course, still very many fundemental principles to be discovered, nevertheless quite sufficient was known of the theory to enable the teams to know what they were doing, and quite enough was known of the practice to enable them to set about the search for their known goal.
As regards Dann, I see the force of the argument that the statements of Lord Diplock were obiter, and that this court should decline to follow them. This would be a bold step, which I find it unnecessary to take. Attention has naturally been focussed in argument on the first part of the speech, up to line 35 of 452. If, however, the speech is read as a whole, it can be seen that it forms a single whole, and that the part relied upon is setting the scene for what Lord Diplock described as the “necessary corollaries in the interpretation of those parts of it which deal with the information to be provided as the counterpart of the temporary monopoly granted to the inventor”. It was with this question of the interpretation of the claims that the decision of the House was concerned, and on which Lord Diplock found himself the sole dissentient. If, therefore, the opinions contained in the latter part of the speech must be rejected as inconsistent with the views of the majority, then the passages cited form the first part, which are the opposite side of the same coin, must fall away as well.
Finally, as to I G Farbenindustrie, if the observations of Maugham J at 322 are not to be explained in the same way as I have suggested for May & Baker, then I must respectfully say that they are inconsistent with the wording of the new Act and must yield to it.
I pause at this stage to consider a contention advanced on behalf of Genentech, separate from the main line of their case. It runs as follows. On the priority date, Genentech were the only workers who had actually discovered the sequence and made a complete t-PA expression vector. By doing so they had gained knowledge and made a construct which could not at that moment have been written down or made by anyone else. Thus, by the very act of planting the flag, as Mr Gratwick put it, Genentech had demonstrated that the invention was not obvious, and hence that it involved an inventive step.
This argument may have some initial attraction when applied to a case such as the present, where in a difficult field several groups of skilled and intelligent workers had all set out at much the same time in pursuit of the same prize. Yet if it is sound it would hold equally good where any person of modest attainments could have made the artifact, and the only reason why the applicant made it first was that he started first, or because nobody else started at all. This cannot be right. First to the post is the test of novelty, but novelty is not enough.
It is true that in many instances the fact that no competitor has produced the invention may be a strong, and indeed on occasion clinching, ground for concluding that it was not obvious — as in the example of the external condenser, but where there is a road to be travelled towards the goal, the winning of the race may tell one no more than that the winner was first in the field, or richer or more determined. To hold otherwise would be to elide the separate requirements of section 1(1)(a) and 1(1)(b).
I now turn to the second group of problems arising under section 3, which concern the meaning of “a person skilled in the art”. The trouble is, of course, that in a case like this no such person exists. The successful pursuit of Genentech’s research required the development of techniques in more than one field: for example, protein sequencing, handling mRNA, building a library, making a probe. I am satisfied on the evidence that there was nobody who united in himself (or herself) all the knowledge and practical skills in each field to a sufficient extent to carry out any kindred project, even if assumed to be non-inventive, on his own. This fact has two corollaries, neither of which I understood to be in dispute. First, that the hypothetical person is a team of persons. Second, that since the search embraced a series of arts, the obviousness of any particular contribution to the ultimate success must be adjudged by reference individually to the hypothetical members of the team, attributing to each the appropriate degree of skill.
The next question relates to the staffing and equipment of the team, a topic on which there appears to be no authority. As to staffing, this arises because some of the teams sent the work of sequencing or probe manufacturing to workers or laboratories outside. I see no problem here. Once it is accepted that the hypothetical team is practising a variety of arts through a variety of individuals, it seems to me that it makes no difference whether they are conceived to be working together as a single unit, or whether the notional individuals against whose notional skills the obviousness of the invention is to be tested are regarded as sub-contractors. The standard for skills is the same.
The question of equipment is more puzzling largely, I believe, because traditional patent law, and indeed the current legislation, is ill at ease with this type of complex and rapidly developing new technology. It seems to me, however, that since we are looking to distinguish the inventive spark from a triumph of method, we should credit the hypothetical team with the best available equipment to see whether, so equipped, they could have found their way to a solution without exceeding the permitted maximum of inventive thinking.
This last observation serves to introduce the most difficult of the present group of problems. What is the permitted maximium? It may be objected that to state the question in this way at once introduces an error, for the authorities show (so it may be said) that the hypothetical team is to be credited with no inventive capacity at all.
In order to explain why I believe this objection to be unsound I must draw a distinction between section 3 and section 14(3). Each of these refers to the person skilled in the art, and it has been assumed that since the words are the same the person and his attributes must also be the same, whichever section is in play. In the case of the classical mechanical engineering patent, this is true. Whether one is asking if the addressee can read the drawings and the description, so as to be able to work the invention, or if the skilled man can proceed from the drawings and descriptions of the prior art to the new product or process without inventiveness, there is no difficulty in using the same notional skilled artificer as the touchstone. But the position here is different. Once given that we are concerned with a series of different arts practised in this complex field, it cannot be assumed that the arts in which the hypothetical persons are skilled will be the same whether they are addressees who start with the patent and try to make it work, or persons who start with the prior art and try to get to the patent. This is indeed obvious in the present case, since the amino acid sequencer who is a vital member of the discovering team will be redundant when the addressees are seeking to fabricate say the claimed expression vectors, since ex hypothesi they will know, not just the five or six bases which were derived en route to the discovery but the full length of the protein sequences.
Equally, I am not persuaded that the standard of attainment contemplated by the sections will always be the same (although it will often be the same) for the individual members of the notional addressee team as for the notional discovery team. To work the patent, the addressees follow the instructions, filling in gaps and clarifying obscurities by common sense and trial and error. This may require skill and experience, but not inventive facility. But the discovery team has no instruction to follow. Unlike the addressees it does not know the answers in advance, and in practice the people who constitute it will be required to exercise different talents from those who take information disclosed in the patent and put it into practice.
In my judgment this distinction lies close to the heart of what has seemed to me one of the most confusing aspects of this difficult case, namely the undoubted presence in the authorities of statements which seem to deny to the skilled man any inventive capacity at all. This surely cannot be applied in a literal way to cases like the present, for if it were so an invention could scarcely ever be obvious, since all the routes to the desired end are likely to throw up problems, more or less knotty. This problem does not arise in the context of section 14(3). If one is asking whether the patent teaches the reader how to work the invention, the fact that it cannot be worked unless the addressee himself makes an invention must be fatal, since it demonstrates that the description is insufficient. But where one is looking at the research team, one cannot treat them as dull plodders, for such people would not be members of the team at all, except as laboratory assistants. We have to envisage people who are skilled, and skilled in the art. Here we have a difficult art, in which the skill consists in a substantial degree of an ability to solve problems. It must, I consider, follow from this that the hypothetical skilled man must be credited with that particular ability in the appropriate degree.
This serves in my opinion to distinguish the present case from the line of authority summarised by this court in Valensi v British Radio Corp  RPC 337 at 377 as follows:
“We think that the effect of these cases as a whole is to show that the hypothetical addressee is not a person of exceptional skill and knowledge, that he is not to be expected to exercise any invention nor prolonged research, inquiry or experiment. He must, however, be prepared to display a reasonable degree of skill and common knowledge of the art in making trials and to correct obvious errors in the specification if a means of correcting them can readily be found.”
I would not venture to doubt the correctness of this summary, which is in any event binding on us as a matter of decision. But the eight reported cases discussed in the passage of which the quoted words are the culmination were all concerned with insufficiency, and their language makes it plain that they are concerned with the notional addressee, who is to set about putting the invention into practice. It cannot, I believe, be transferred uncritically to a research team operating in a field where some substantial measure of ingenuity is an essential qualificiaton for being involved in the enterprise at all.
It must certainly be acknowledged that this is not a complete explanation of the authorities, for they include cases on obviousness where a view of the skilled man has been expressed in terms not dissimilar to the one just quoted. Of these, much the most important, to my mind, is Technograph Printed Circuits Ltd v Mills & Rockley (Electronics) Ltd  RPC 346, for here we find a statement by Lord Reid, with whose speech Lord Morris of Borth-y-Gest agreed, that the mosaic of documents constituting the prior art is one which can be put together “by an unimaginative man with no inventive capacity”. This seems to have been a matter of concession, albeit by very learned counsel. Much more importantly, however, it was a concession made in the context of a case quite different from the present. The particular aspect of radio technology involved in the patent does not seem to have been one where special ingenuity was part of the art. Moreover, it was a case where a solution to the problem had been sought for many years wihtout success. In such a case, obviousness is hard indeed to establish (see per Tomlin J in Samuel Parkes & Co Ltd v Cocker Bros Ltd (1929) 46 RPC 241 at 248) particularly if, as in the Technograph case, the party contending for obviousness had called no expert evidence in support. This is quite different on the facts from the present situation, where as it seems probable that the various groups who set out in search of a solution would actually have found one within roughly the same time-scale. It is true that this factual distinction does not serve to explain away Lord Reid’s statement of principle, and others which can be found in the cases, but the point on the nature of the art remains the same. Where the art by its nature involves intellectual gifts and ingenuity of approach, it would, I believe, be wrong to assume that the hypothetical worker is devoid of these gifts.
Two further points, much pressed in argument, must be considered at this stage. The first relates to the prospect of success, viewed as at the start of the entrprise. Cases were cited to suggest that if the prospect was very low, and yet success was achieved, this would in itself be sufficient to establish inventiveness, even in the presence of what might appear to be routine laboratory work. I do not read the cases in this sense, nor could they be so read consistently with the statute. Quite plainly, the longer the odds against mere repetition of established techniques yielding the derived answer, the more likely it is that success was achieved either by intellectual activity beyond the norm or by good luck (if good luck is enough to make a patent). But this does not itself show that what made for success is anything other than the proper reward for diligent and skilled labour. It may be that such labour and the resulting success deserve a prize, but the law, as I read it, calls for something more.
Secondly, there are cases which discuss the question of fact whether it was obvious to try a known technique, or whether the chances of success were sufficient to warrant a trial (see American Cyanamid Co v Ethicon Ltd  RPC 215 at 266; Johns-Manville Corp’s Patent  RPC 479; Olin Mathieson Chemical Corp v Biorex Laboaratories Ltd  RPC 157 at 187-188; Tetra Molectric Ltd v Japan Imports Ltd  RPC 541 at 583-584). Whilst not doubting the principles discussed in these cases, I find them only tangential to the problems raised by the appeal. I can accept without difficulty that the decision to set out at all or to follow a particular signpost might, if dense clouds of fog obscured the entire subject, be sufficient in itself to demonstrate the necessary element of inventiveness. But surely this is not so here.
The project was the most difficult to have been tackled at the time, but the possible routes and the destination were known, even if nobody could foresee just what obstacles might be found on the way. This does not, of course, prove directly that the invention was obvious, and the facts must be examined at a later stage. But equally, it cannot, in my judgment, be assumed that inventivness must have been involved somewhere, just because a wager on success could have been placed at long odds.
What then are we to take as the standard by which the skilled man is to be judged? (In the foreign language text the equivalent expressions are “homme de metier” and “Fachmann”, both conveying the notion of professional or expert.) I believe that this question is incapable of being answered in vacuo. Not only will the standard differ from case to case, but there may well also be differences between the individual members of the teams. Just as the authorities show that the characteristics of the notional skilled addressee must be assessed by reference to the nature of the invention which he is expected to work (see, for example, Badische Anilin und Soda Fabrik v La Societe Chimique des Usines du Rhone 1897) 14 RPC 875, per Willes J), so also must the gifts to be ascribed to each member of the notional team which sets out in search of the invention depend upon the contribution which he is to make to the work of the team by the exercise of his own particular art. Some such arts demand what are essentially bench skills, rather than a creative intelligence. Others call primarily for long experience. Others again may need a more probing intellect. Still others may require a combination of all these qualities. There is no single standard. Nor do I have any faith in the proposition that formal qualifications are the test, or even a useful test. Possession of a doctorate in one subject or another does not necessarily make a person skilled in a particular art for the purposes of section 3. Nor, even in this complex field, is it always an essential qualification: as witness the case of Mr Kohr, who did not have a further degree yet was obviously an expert. Again, I do not find it very helpful simply to count the number of patents in which a person was named as one of the inventors to see whether he had a degree of inventiveness above or below the hypothetical norm.
In the end, obviousness is a jury question (see General Tire & rubber Co v Firestone Tyre & Rubber Co Ltd  RPC 457 at 498 and Johns-Manville at page 487 and 492). This question must be in the light of a highly artificial test which, although compressed into a few words in the Act, is impossible to elaborate with any degree of precision. The judge must simply form his own mental picture of the art and of the skilled practitioner, and see how it measures up against the task which he is assumed to be tackling: a judicial process, it must be admitted, which is a great deal easier to describe than to perform.
Obviousness: The Facts
The real thrust of Genentech’s argument is that the road to success was long and stony. The very low abundance of the right mRNA and the uncommon length of the protein combined, so the argument runs, to face the researcher with a problem of a new dimension which a group of ordinarily skilled men would have been unable to solve. Even omitting the steps which had to be taken after a full length of DNA was obtained, thee was still the need to decide upon the way to tackle the problem, and then having purified t-PA from the Bowes melanoma cells, to isolate the relevant mRNA, construct a library, find the amino acid sequence, synthesise a pool of oligoprobes, screen the library, distinguish the true hybrids from the apparent positives, and probe for the complete length of DNA. All this, say Genentech, would have been beyond the grasp of the hypothetical team.
This is an attractive argument. The subject is so new, its terminology so forbidding, its techniques so unfamiliar, and its short history so punctuated by feats of intellect and flashes of inspiration, that the layman is tempted to fall back in amazement, and assume that the whole business would be too much for any but the most powerful of minds. Yet this is not what the evidence shows, and the temptation must be resisted. It is scarcely disputed, and indeed the Points of Agreement make clear, that a number of steps involved no more than the application of established technologies. Adopting the approach previously suggested, I believe that the right course is to identify the individual obstacles in the way of the hypothetical workers, and see whether they would have been overcome or circumvented. Of these, it seems to me that three were the most formidable, and I will deal with them in turn.
Choice of cloning strategy
The first question is whether the hypothetical skilled worker would have set about his task by choosing the selection method which ultimately led to success, namely oligonucleotide probing. When considering this question it is important to remember that oligo probing was not necessarily the “right” method, viewed from the starting-point, even though it was the one which ultimately bore fruit. Certainly, it was the appropriate choice for mRNA of low abundance, but the relative abundance of mRNA coding for t-PA was unknown at the start, and it seems to have been a matter of surprise when it provied to be so low. Moreover, oligo probing required amino acid data which would be troublesome to obtain. It would, therefore, not be surprising if the skilled worker began with hybrid selection. This is what Ny and Edlund did, and it seems that with a larger team they would have succeeded. Cold Spring Harbor tried this method. Significantly, it is also what Genentech themselves did, and it was not until May 1981 that hybrid screening was finally abandoned.
To my mind, therefore, the question is not whether the skilled worker would necessarily have gone to oligonucleotide probing first. If it were the right question, one would have to consider the disputed Point of Agreement No 12, and whether the worker should have attributed to him an item of knowledge kindred to the “method of choice” in Professor Maniatis’ as yet uncirculated compilation. But to my mind what we ought to ask is whether the skilled worker would have tried some other method, found that he was bogged down and given up, without even addressing himself to oligo probing. I see nothing in the evidence to justify an affirmative answer. Oligo probing was there in the prior art as a known technique. The skilled worker would have got there, as a first or second choice, without what I would regard as an inventive step.
Genentech contend that the hypothetical skilled man would have been misled by the prior art, in the shape of the advice in the articles by Suggs and Wallace, into choosing a stringency for the washing-off process sufficient to leave only one apparently positive result. This process was likely to lead, in a substantial proportion of instances, to a dead end, because it took no account of the favourable binding properties of the GC bond. A mismatch with a high GC content would, so the argument runs, bind stronger than a perfect match with a lower content, so that washing down to a single positive would risk the earlier loss of the true hybrid. The worker would then, on finding that his apparent positive did not yield expression, tend to abandon the entire library on the assumption that it contained no cDNA. This was a risk against which there was no warning in the Maniatis compilation.
For my part, I am more than doubtful about whether it is shown that hybridising with probes based on the WEYCD sequence would, in fact, have yielded false positives in preference to a true hybrid. The formula are empirical, and as Dr Stark said, “the theory of this sort of business is still not exact”. Moreover, the evidence of Dr Brammar indicates tht the GC bond would be a problem only when screening a highly complex library. Quite apart from this, however, the weight of the evidence established, in my view, that the skilled worker would not (as Professor Maniatis put the matter during his evidence) follow Suggs’ article as a protocol. The choice of hybridisation conditions would be a matter of trial and error (Dr Goeddel), with the hybridisation conditions constantly altered because this is an empirical art form (Dr Brammar). One would try to get down to a handleable number of putative positives and then examine them another way (Dr Stark). Even Dr Harris, who came nearest to endorsing Genenetech’s submission did not, as I read his evidence, say that he would necessarily have gone for a single positive, but would rather have proceeeded step-wise, raising the temperature until he lost all the positives of the preceding step — not just a solitary positive.
It is true that the hypothetical worker might not have worked out the Ny/Edlund strategy of negative controls, although Dr Ny called it rather an obvious thing to do, and he and Dr Edlund can without any disrespect be called learners, since it was precisely in order to develop an expertise that they engaged on this project at all. Very probably the worker would not have found Woods solution of using tetramethyl ammonium chloride, but he could have got by without these expedients, as others did. I am satisfied on the evidence that Professor Maniatis was right in saying that to carry Suggs through literally would not be what an investigator with common sense would do, and that such an investigator would either proceed cautiously until he arrived at a number of positives small enough to handle (as did Dr Ny, who it may be noted, had read the 1979 Wallace paper), or, if he lost all his positives, would do what Dr Pennica did and back off to a lower stringency and try again. The problem, if there was a problem, was one which the skilled worker would have been able to solve.
In reaching this conclusion, I have left out of account the possibility of expression screening. This was a latter development and is, therefore, not relevant to the issue of obviousness, although it does provide a useful illustration of the way in which the width of Genentech’s claims may inhibit legitimate development by others. Nor have I taken acocunt of differential screening. Dr Opdenakker tried this first, because he had no amino acid sequences. Dr Opdenakker was only just setting out on his career. He seems to have done very well (his paper was read and used by Dr Harris). In spite of this worker’s comparative inexperience, he may perhaps have done better than the norm, and I am not sure that it would have been within the compass of the ordinary skilled worker to devise the idea of induction or phorbol esters, or that if he could have done so he could have made a success of the process as a whole.
Amino Acid Sequencing.
The ascertainment of a partial amino acid sequence of low redundancy is an essential starting point for oligo probing. Progress of all the teams was held up whilst waiting for this information, and three of the five teams experienced trouble of one sort or another. It is, therefore, an important question — perhaps the most single important question on the facts — whether the hypothetical worker (or team of workers) would have found himself confronted at this stage with an obstacle which he or they could not surmount. I believe that the answer to this question is almost pre-determined by the view taken on the issues of law previously discussed, but we must nevertheless examine the available evidence, which unfortunately is incomplete. In summary it was as follows.
Not long after Genentech began work on this project, they acquired a second-hand but un-used sequenater, which they modified in accordance with the advice of their consultant on sequencing, Dr Shiveley of the City of Hope Research Institute. According to Mr Totty, there was nothing unusual in having sequenaters which were “custom made” like this. The machine was not, however, used at first for the t-PA project, because Genentech were busy with other matters, and the sample of t-PA obtained fom Dr Collen was sent to Dr Shiveley for sequencing. Dr Shiveley was unsuccessful. He himself, and others concerned at Genentech, thought that this might be because the N-teminal was blocked. With hindsight, Mr WJ Kohr, who conducted this aspect of the project for Genentech, said that he thought that failure was due to the small quantity available. Matters cannot have been helped by a mechanical failure which caused Dr Shiveley to lose 11 cycles on the machine (T/8/16). Four months later, Dr Collen brought a larger, more satisfactory sample. There was a considerable discussion as to the best tactics to be adopted in the choice of technique to be adopted with a supply which was still very small. The first attempt was unsuccessful, but the subsequent runs, accompanied by the use of a different type of reverse-phase high-pressure liquid chromatography (HPLC) column yielded traces which, after careful analysis showed a number of possible probe regions. Of these, the most suitable was the WEYCD sequence which Mr Kohr passed to Dr Pennica.
For the purposes of the litigation, photocopies of Mr Kohr’s HPLC traces were handed to Mr NF Totty for examination, so that he could see whether (as Genentech contended) the task of identifying the amino acids from the traces was exceptionally difficult. If I correctly understand his evidence he successfully identified the first nine amino acids in the sequence. His exercise was deliberately performed under pressure of time, using only photocopies and wihtout access to the machine. He said, and this was not contradicted, that under more favourable conditions he could have identified fifteen or sixteen amino acids. It seems that if he had been in the position of Mr Kohr, he too would have found the WEYCD sequence.
I pause briefly at this stage to note that if the hypothetical skilled person is to be identified wholly or partly by reference to formal qualifications, and in particular (as was suggested) by enquiring whether they possessed doctorates, neither Mr Kohr nor Mr Totty would qualify. Since I do not think this is the test, I say no more about it, but it is perhaps worth noting that neither was in a position of high authority, skilled as they unquestionably were. Mr Kohr worked under the direction of a Dr Hawkins, about whom I believe we know nothing. Mr Totty was a chief research officer (the second highest technical grade) in day to day charge of running the instrumentation laboratory at the Ludwig Institut of Cancer Research.
At Genetics Institute, Dr Hewick received from Dr Rifkin a first sample which turned out to be impure and heterogeneous, so that the sequencing failed. A second and much larger sample was more satisfactory. It took Dr Hewick only six days to produce a sequence of 24 amino acids from the N terminus, which included a group of six amino acids of low redundancy. Dr Hewick was, however, using a new gas-phase sequenater and had participated in its design. Although this was offered on the market during 1981, none were supplied until quite soon before the priority date. I do not think it safe to assume that this machine would have been in the armoury of the hypothetical worker(s).
On the other hand, there is no doubt about the availability of the spinning cup sequenater, and Dr Hewick’s evidence was that he could probably have got enough data from such machine with a sample of the size with which he achieved success, although it woud have taken longer. He did not have a great deal of experinece with the spinning cup, and would have probably have needed a couple of weeks of training beforehand.
I should mention that the fact that Cold Spring Harbor Laboratory did not tackle the sequencing, but transferred it to Genetic Institute for a substantial potential reward, does not, in my view, advance the matter. CSH was fundamentally a centre for individual research. The skills employed there were of a high order, but did not cover the full range and could not be deployed in the same way as in a commercial organisation. There was a spinning cup sequenater in Dr Smart’s laboratory, but he was already planning to leave well before December 1981. Dr Drickanner might have been able to perform the sequencing, but he was not interested. After Dr Smart left CSH had no facilities of amino acid sequencing. The laboratory, therefore, did not muster a full team, in any sense relevant to the identificiton of the hypothetical skilled worker.
As regards Dr Ny and Edlund at UMEA, protein sequencing was not a skill which they possessed, and it it was done for them by others. Exactly how, and by whom, is not clear. In evidence Dr Ny said that Dr Wallen had a very outdated machine. Dr Wallen’s affidavit speaks of getting long amino acid sequences with a Beckman spinning cup sequenater. At all events, according to Dr Ny most of the protein sequences came out of a machine possessed by Dr Jornvall at the Karolinska Institue, run by Gunnar Pohl, one of Dr Wallen’s research students. We do not know why one of the amino acids in the sequence given by Dr Wallen in March 1981 was wrong, or how the mistake was discovered and corrected three months later.
About the Leuven/Celltech sequencing we know nothing, except that, in about the middle of 1982, Profesor Billau’s group, in which Dr Opdenakker was a graduate student, received from Professor Rombauts some sequence information including the WEYCD sequence.
These events show, to my mind, that on the priority date — as distinct from the position some months later, when the team could have equipped itself with a gas phase sequenater — the task of sequencing would have been troublesome, and that set-backs were a distinct possibility; but that there is no reason to suppose that a person skilled in the art of sequencing would have found the obstacles to be insuperable. That person would have got to WEYCD or another low-redundancy sequence in due course, within a reasonable time-scale. Skill and persistence would have been enough.
For these reasons I do not consider that the features raised by Genentech as the prime obstacles to success would have been insuperable by the skilled man. In saying this I am very conscious of the risk involved in seizing on a few aspects of the case, and drawing too close to them to achieve a proper perspective. As Fletcher Moulton LJ observed in British Westinghouse Electric & Manufacturing Co Ltd v Braulik (1910) 27 RPC 209, an argument addressed ex post facto to the effect that a new combination can be shown to be arrived at by starting from something known, and taking a series of comparatively easy steps should be viewed with suspicion.
I have already acknowledged, and am glad to repeat, that to this layman at least Genentech appear to have performed an excellent piece of work. But the excellence resided in a combination of tenacity, skill and managerial efficiency. As to tenacity, it is impossible to read Dr Pennica’s affidavit without admiration for the way in which she worked literally day and night for months in the most adverse physical conditions. But others worked hard too. When Dr Ny told Dr Pennica about his work, he said that it had nearly killed him. Should one not also credit the skilled worker with diligence and determination? As to skill, one sees it in abundance. But this is not the same as inventiveness.
Dr Pennica’s laboratory techniques excite admiration, but they would have been needed by someone working the patent, who would also have to create a library, and who cannot be credited with more skill than the hypothetical skilled person in his or her particular field, even if (as I have suggested) the hypothetical teams for the purposes of section 3 and 14(3) are not the same.
Dr Goeddel seems to have been a first rate team leader. He assembled the right people, the right machinery and the right money, and kept the project moving to a conclusion. His group reached the winning post first for a variety of reasons. I do not accept, however, that when his recipe for success is analysed it can be found to contain ingredients which, if I have correctly understood the law, were not available to the hypothetical skilled team of workers. The whole enterprise was well run, and it is not surprising that Genentech finished first. Yet it is inventiveness which counts, and I cannot find it here in any degree which exceeds the amount of resource to be expected of a group mustering the skills, remarkable as they seem to a layman, ordinarily to be expected of persons skilled in this most difficult array of arts.
It may well be said that the work done by Genentech seems worthy of a reward greater than a few months’ start on the road to a marketable product. Yet I am driven to conclude both that the monopoly claim exceeds any legitimate award, however exactly it ought to be formulated; and also that for want of an inventive step Genentech are not entitled to any reward through the medium of a patent monopoly. I would hold that all the cliams should be revoked.
I cannot part from the case without adding my own acknowledgement of the indispensable help which we have received from Dr Brenner. I would also wish to recognise the written as well as the oral submissions of counsel, which provided an invaluable framework for consideration of the issue, as well as the efficient manner in which the appeal documents were indexed and prepared.
Bristows Cooke & Carpmael; Reynolds Porter Chamberlain