UK: Empirical Research Inventions Allowable Generic Claim Scope

In science there is only physics; all the rest is stamp collecting¹.

This discussion is with particular, but not exclusive, reference to practice in the pharmaceutical industry for claiming new chemical entities, and in particular to the feasibility of defining a genus of compounds that validly covers compounds that have not yet been made and tested.

In the early days of the pharmaceutical industry claims were to single compounds. The well-known Felix Hoffmann US patent No 644077 granted in 1900 for aspirin provides an example. Another researcher had previously described acetylsalicylic acid, but Hoffmann demonstrated that the compound that the earlier researcher had made did not possess the correct properties. Hoffmann had developed a new process that gave "the real acetyl salicylic acid" which "exhibits therapeutical properties" and claimed:

Compounds isolated from nature and produced synthetically could also be claimed. An example is provided by US patent 3880888 (Firmenich S.A.) which reports the isolation of the compound 3,4,7-trimethyl-2-oxo-1,6-dioxa-spiro[4.5]dec-3-ene from an essential oil derived by distillation of Burley tobacco. The compound is said to provide a mild herbal or woody note to the tobacco, and is claimed as a composition of matter consisting essentially of a compound of the formula

As is well known, the pharmaceutical industry was an offshoot from the dyestuff industry. From the beginning of the 20^th century onwards, the practice developed within that industry of claiming classes of dyestuffs generically by means of structural formulae. US-A-1744172 provides an example of so-called Markush practice as it had developed by the mid 1920’s. The patentees had claimed a novel class of azo dyes which had the benefit of producing durable colours on cotton and other vegetable fibres, and which are defined in claim 2 of the patent (with slight editing) as follows:

The patentability of the compounds of this genus flows from the link between the chemical structure and the newly discovered and valuable properties possessed by the compounds of that structure. That formulation is consistent with the question put forward by Sir Stafford Cripps K.C. in UK proceedings between Sharp & Doehme v Boots⁴ which concerned the manufacture of certain alkyl-substituted resorcinols which were known as antiseptics (at that time chemical product claims were not allowed in the UK). Published papers had already disclosed that the C₁-C₄ members of the class had been made and that antiseptic activity increased as the number of carbon atoms in the alkyl chain increased. The patentees averred that because drug research was empirical, the skilled person knew in advance neither that the higher alkyl resorcinols could be produced by the claimed method nor what their antiseptic properties would be. The compound that proved to be of therapeutic value was n-hexyl resorcinol which is of formula:

The defendants replied that because therapeutic activity increased in the known compounds from C₁ to C₄, it was obvious to make the higher members of the series. Sir Stafford Cripps argued that the question concerning inventive step that the court had to decide was based on whether there was an obvious link between structure and value, and was as follows:

• that he could make valuable therapeutic agents by making the higher alkyl resorcinols?" (Paragraphing and emphasis added).

The same link between structure and value is implicit in the requirement for utility in US law, as is apparent, for example, from the following passage from Brenner v Mason which calls for the product to be useful, i.e. to have a "specific benefit":

In the azo dyestuff case referred to above, it is clearly credible that the whole class of compounds would dye cotton with the indicated combination of colour and resistance to boiling. However, the credibility of similar Markush claims for pharmaceuticals or agrochemicals is open to question since human, animal and living plant cells are more complex than cotton fibres.

Finding pharmaceutically active compounds has been at least until recently a task of empirical research of the kind pioneered by Thomas Alva Edison, and the rules applicable to patents of this category have been known at least since the decision of the US Supreme Court in the Incandescent Lamp case which concerned a third party patent which was alleged to be infringed by the electric lamps being made and sold by Edison licensees. The third party patent purported to monopolise all carbonised fibrous or textile materials for use as filaments for electric lamps, as is apparent from the following main claim:

The evidence adduced in the case showed that the researches carried out by the third party could not support such a broad monopoly and that the field was unpredictable. The factual background and the reasoning of the court will be apparent from the following passages from the Opinion:

The views of the US Supreme Court coincide with those of the English judges. In the case of claims to newly discovered compounds, the basic principle is as set out by the House of Lords in May & Baker v Boots⁷ which explains that generic claims are permissible where there is a rational basis for them, and sets out the following principles for research in an empirical art:

(i) in a field where progress is by empirical discovery, an invention must be in respect of a substance which has actually been produced, since there cannot be an empirical discovery in respect of a bare formula;

(ii) each new compound is a separate invention since its worth is a new discovery; but

(iii) where the chemist has found some law or principle by which he can predict therapeutic effect in advance so that each of a group of new products will be of value, the art will have lost its empirical nature, at least to some extent (i.e. a Markush claim may be allowable).

The factual situation in May & Baker is of interest. The invention claimed in the patent in issue GB-A-533495 covered sulfathiazole (M & B 760) and sulfamethylthiazole, the first of which was the most important of the sulfa antibacterials and is of the formula indicated below:

The specification did not have specific claims directed to the two exemplified active compounds. It only had generic claims to benzene sulfonamido thiazole derivatives, the genus of compounds claimed being described in words rather than by a structural formula. The description explained that the NH₂ group could be substituted amino (e.g. alkylamino, N-acyl-N-alkylamino or aralkylamno) and that the benzene ring could have alkyl substituents. The thiazole ring could also be substituted. The resulting genus covered tens of millions of compounds, and although it covered compounds additional to the two compounds exemplified that had been found to be active, it admittedly also covered compounds that were inactive. The patentees therefore proposed to replace their generic claim by a claim to the two exemplified compounds, arguing that it was apparent on the face of the specification that they were drawing an inference from their specific data and should not be penalised if that inference proved unfounded. In a split (3:2) judgment the House of Lords rejected this approach and said⁸:

The generic claim was therefore invalid, and in the absence of specific claims to the two exemplified compounds, the patent was held to be unenforceable.

The same principles were followed by the UK House of Lords in Biogen v Medeva⁹:

In the EPO, although technical analysis is conducted within the formal framework of problem/solution analysis, the outcome principally determined by whether the claimed subject matter leads to a new effect, some new function or advantage that cannot be predicted from the teachings of the prior art references considered collectively¹⁰. Szabo made that clear, for the chemical field at least,¹¹ when he said that:

This observation is supported by the decision in the well-known case of Agrevo/Triazoles¹², which concerned a class of herbicidal triazole sulphonamides. The dispute was whether the whole class of compounds had to herbicidal in order to establish patentability, or whether it was enough for the applicants to show that:

(a) some of the claimed compounds were herbicides; and

(b) the cited art did not teach the skilled person to make any compound within the class.

The origin of the objection was that the genus of compounds claimed included open-ended definitions that covered very large substituents and the Examining Division considered that it was not plausible that the claimed compounds had utility for all possible substituents. The applicants argued that the Board need only consider whether the skilled person would make the claimed compounds starting from the prior art teachings – i.e. the relevant question was what the public would do, not what the inventor had achieved. The Board rejected the applicant’s argument in uncompromising terms, and based its opinion on the legal principle that the extent of the patent monopoly should correspond to and be justified by the technical contribution to the art that is contained in the specification. It went on to say that mere structural ingenuity was not sufficient. If the result that the skilled person was seeking to achieve was simply obtaining further chemical compounds, then all known chemical compounds were equally suitable as starting points and all known methods of transformation might be used, so that the selection of particular compounds to be made was a mere arbitrary choice. For that reason

The UK decision in Olin Mathieson v Biorex¹³ provides an example of a generic definition in the pharmaceutical field that was not successfully challenged. The tranquiliser chlorpromazine (Largactil) had been patented in 1950 and provided a breakthrough in the treatment of mental illness. It was of formula:

The discovery of chlorpromazine stimulated structure-activity investigations aimed at the production of compounds with improved properties. The patentees had discovered inter alia the drugs trifluopromazine and fluphenazine, which were based on a ring CF₃ substitution and were of formulae:

The patent in issue GB-A-857547 (which had been divided from GB-A-857546 claiming a wider genus) covered a genus of compounds of the formula:

wherein A is a lower alkylene radical and B is a basic nitrogen-containing radical with fewer than twelve carbon atoms¹⁴. The compounds were said to be antihistaminic, anti-emetic and tranquilising agents and to be useful for the treatment of psychotic states, the compound trifluopromazine being disclosed as more potent than chlorpromazine. The defendants proposed to manufacture a compound called trifluoperazine (Stelazine) that was covered by the generic claims of the patent in issue but had not been exemplified or specifically disclosed. It was of formula:

An initial allegation of non-enablement was not maintained, and the argument centred on whether the disclosure supported the broad generic scope of the independent claims. The Court’s decision to uphold the generic claims was based primarily on the facts that:

(a) the side chains covered by the patent in issue had been disclosed in earlier patents concerning phenothiazines and had been shown to be active, (b) there was no reason to suppose that change of the ring substituent from Cl to CF₃ would affect the range of side chains that was compatible with activity,

(c) for any given side chain the CF₃ substituent in the 2-position gave the highest activity, and

(d) all the CF₃-substituted phenothiazl compounds that the plaintiffs had made and tested were therapeutically active.

A further instance where the same principles were adopted arises in Chiron v Murex¹⁵ which concerned the sale of test kits for the hepatitis C retrovirus. The patentees had isolated form an infected chimpanzee called Rodney a polynucleotide that contained the instructions for making the antigenic determinant of the virus i.e. the site on the viral particle to which antibodies of the immune system attached themselves. They had then determined the sequence of that sample. These were the critical steps in identifying the cause of HCV infections and in harnessing the HCV antibody:antigen reaction to produce diagnostic test kits. The patent claimed inter alia:

The defendants alleged that the claim did not relate to a new class of chemicals because the members would be different chemically or biologically from one another, and because there was not prescribed a formula that linked them. The patentees replied that the link was their discovery of the virus and the presence of the antigenic determinants of the virus to which HCV antibodies would bind. The Court held that the facts differed from those in Biogen and stated:

"We prefer the submissions for Chiron. We will take claim 1 as the paradigm. There is nothing in that claim as a matter of language or in the specification as a whole to differentiate between one polypeptide falling within the claim and any of the millions of others. At the other end of the scale, there is no doubt that it was the finding and sequencing of the Rodney virus genome which was of the greatest importance for this discovery and analysis enabled the antigenic determinants to be found in the protein expressed by the sequence. In view of the fact that an antigenic determinant can only be defined by reference to the antibody which binds to it, and the further fact that the immune response of an individual produces a whole range of antibodies to any given virus it is not possible to define antigenic determinants by reference to any common chemical formula. Thus the invention is the chemical comprising at least 10 amino acids in which there is an antigenic determinant to which an antibody to HCV will bind. In our view, both in substance and in form, this is a single invention properly defined by the common denominator of the existence of an antigenic determinant of HCV, notwithstanding that the resultant polypeptides will have divergent characteristics in other respects. The invention of one is the invention of all of them because that which is common to all is of the essence of the discovery and that which distinguishes one from the other is irrelevant to that discovery."

Monsanto et al v Merck¹⁶ concerned infringement of a patent covering non-steroidal anti-inflammatory drugs. Pumfrey J. at first instance handed down a 78-page opinion on 4th February 2000.

By way of background, US-A-3743656 (Brown) had previously disclosed a class of diaryl heterocyclic compounds said to be anti-inflammatory agents and of the formula:

in which:

the heteroatom X could be O, S or –NH-;

Z could be hydrogen;

Ar and Ar₁ could be phenyl or substituted phenyl in which the substituents could be F or CH₃SO₂; and

R represented an alkyl group.

Amongst the specific disclosures in the Brown specification was a sub-class of compounds of the formula:

in which X and R have the meanings given above.

DuPont had been working on diaryl heterocyclic compounds from 1979/80 onwards, and had published in 1989 details of a compound called DuP 697 that was said to combine good anti-inflammatory characteristics with low gastric irritancy.

The properties of that compound had been described in a slide shown at a prostaglandin conference as follows:

• Active as an anti-inflammatory, antipyretic and in models of inflammatory pain.

• Selective inhibition of cyclooxegenase (COX) in cellular and enzymatic assays.

• Lack of platelet activity and safety profile may reflect in vivo results of selective CO inhibition.

• May be a useful tool to understand role of COX 1 and COX 2.

Despite its promise, DuP 697 was never commercialized because it turned out to be eliminated only very slowly from the body.

Soon after the prostaglandin conference the claimants had started work on developing diaryl heterocyclic anti-inflammatories using DuP 697 as lead compound. The object of their work was to discover another compound that was not only effective but also novel and hence patentable. The claimants found that:

• For cyclooxegenase activity the aromatic rings had to be located in adjacent positions on the heterocyclic ring.

• A CH₃SO₂ group in the p-position had to be present to inhibit the COX 2.

• The F atom on the other ring was not essential.

Their EP-B-679157 claimed a group of compounds said to exhibit anti-inflammatory or analgesic activity without causing erosion of the lining of the stomach because they selectively inhibit COX 2 (which produces inflammation) without substantial inhibition of COX 1 (which controls gastric secretion). The claimed compounds were therefore alleged to open up the possibility of an effective treatment for arthritis with reduced harmful side effects. They were of the formula:

in which:

R¹ and R² represent aryl or heteroaryl which may be optionally substituted with a specified range of substituents, provided that one of them is substituted with methylsulfonyl (CH₃SO₂) or sulfamyl (NH₂SO₂);

X is hydrogen or is selected from a long list of possible substituents, including, amongst others, hydroxy (OH) and halogen (includes Br); and

Y is selected from S, O and –NR¹- in which R¹ is selected from hydrogen and alkyl.

The specification mentioned by name more than 140 compounds, of which 15 had been exemplified as being made (the remainder presumably being prophetic), all 15 had been tested in vitro for activity (IC₅₀ against COX I and COX II) and five had been tested in vivo for anti-oedema or analgesia. No structure/activity relationship was disclosed except what could be gleaned from the claims. Unfortunately for the claimants, the most promising compound that they had identified within the class claimed in their patent was dropped because of poor bio-availability.

Merck had started work later than the claimants, but had succeeded in finding a compound MK 966 that was commercially exploitable, and that was of formula set out below:

The claimants had not made MK-966 and had no specific claim in their patent directed to that product. They therefore had the more difficult task of establishing that one of the broad generic claims in their patent was valid and covered MK-966.

In defence to the infringement proceedings, Merck adduced experimental evidence that a number of the compounds covered by the patent did not have the claimed properties, including the materials of Examples 1, 2, 5, 11 and 15 of the patent in issue, three of which were held at the trial not to demonstrate any anti-inflammatory effect. One further compound that the defendants tested, referred to as Compound 68, had been listed in the patent as preferred. It had the formula:

Despite being "preferred", Compound 68 proved to have all the undesirable features that the claimants aimed to overcome: it was selective for COX 1 and not COX 2, and it was a gastric irritant. The evidence included photographs of the gastric lesions that it caused. The claimants replied that the compounds made and tested by the defendants were not representative but had been carefully selected.

The Patents Court found that the patent was not valid and not infringed. The major conclusions of Pumfrey J. are summarized below:

• The invention with which the patent in issue was concerned was a class of compounds predicted to have anti-inflammatory activity and fewer or less drastic side effects owing to COX 2 specificity.

• A claim is invalid for insufficiency if covers subject matter that owes nothing to the invention disclosed. For a claim to a class of compounds to be valid, it must be possible to make a well-founded prediction that they have a common property that unites them as a class. "If compounds having the features of the claim may or may not possess the qualities which the patent says unify the class, it cannot be said that the claim represents a true class at all. It is just a generalized description of a large number of chemical compounds. Such a claim is not analogous to a claim to a new principle, since the patentee has given no information, such as a structure/activity relationship, which enables the reader of the specification to draw any conclusions as to the properties of any particular compound without further experiment. All he has done is to describe the scope of the claim with spurious precision."¹⁷

• The onus was on the claimants to show that the defendants’ test compounds were exceptional, and they had not done so, nor was there any substance in the defendants’ criticisms of their experimental technique. The experimental evidence showed that the properties of compounds within the claimed class were unpredictable, and that there were a substantial number of compounds that did not exhibit either COX 2 specificity or diminished side effects. The most distinctive feature of the claimed compounds was the 3,4-disubstitution rather than the 2,3-disubstitution of DuP 697, but this did not give success on its own, and it was not possible to identify features that did. Accordingly the patent was invalid for insufficient disclosure.

The above decision was affirmed on appeal to the Court of Appeal¹⁸ where Aldous L.J said:

In parallel proceedings before the EPO, EP-B-0679157 was revoked on the ground that all the requests (sets of amended claims) submitted by the patentees related to intermediate generalizations that were not supported by the application as filed and hence contravened article 123(2) EPC¹⁹.

American Home Products and Professor Sir Roy Calne v Novartis²⁰ – You may be in trouble if you don’t Markush!

Rapamycin is a macrolide of the formula shown below. It was known prior to Professor Calne’s invention as an antifungal antibiotic of formula:

Professor Calne’s contribution was to discover that rapamycin could also be used as an immunosuppressant. The patent in issue was EP-B-0401747²¹, and its main claim read:

There was, however, a broadening statement in the description, which read:

The defendants’ compound had been made by replacing the OH group on the ring of the side chain with a 2-hydroxyethyl group. Trial of the present infringement proceedings had (presumably by agreement) covered only issues concerning infringement and sufficiency of disclosure, leaving other issues such as novelty and inventive step to be considered later if so required. The questions before the court were whether:

• The claim covered the defendants’ compound, which was not rapamycin but instead was 40-O-(2-hydroxyethyl)-rapamycin.

• The patent was bad for insufficiency if the claim did so. Possible grounds of insufficiency were that the claim covered a class of compounds whose scope was impossible to determine, and that in order to discover rapamycin derivatives within the scope of the claim a skilled person would have to embark on a prolonged and unduly burdensome research program.

• There was a distinction between ordinary claims and "second medical use" claims.

Laddie J handed down an opinion of the Patents Court on 6th December 1999. He held that the claims covered the defendants’ derivative and rejected the arguments concerning insufficiency, i.e. because of the fundamental nature of Professor Calne’s discovery it was legitimate for his patent to "reach-through" to other structurally similar compounds discovered by subsequent empirical research. His opinion made the following points:

• The word "rapamycin" used in the claims meant a single known chemical and did not mean "rapamycin and derivatives of rapamycin which exhibit rapamycin-like immunosuppressant activity."

• There was a distinction between the meaning of the words used in a claim and the meaning of the patent. In the present case, as in Kastner v Rizla²², the inventor had not made it clear that his intention was only to seek limited protection, but instead there were factors e.g. the statement about derivatives and prodrugs that showed that the intention was to cover more than rapamycin. If the specification were limited to a single compound, it would be virtually valueless because: "It would have disclosed to the art the novel seam of interrelated molecules but have claimed only one of them." Any reasonable addressee would have been surprised at such a limitation. Instead, the patent covered those derivatives of rapamycin that produced the same type of inhibition of organ rejection as did rapamycin itself.

• Because there was no sure way of predicting which derivatives would be effective, the inventor should not be forced into a choice between accepting narrow protection and putting forward a wider claim with neatly defined boundaries but where the appearance of certainty was an illusion and the claim was in truth a calculated gamble. "Where the technology makes prediction impossible it cannot be right that the law requires it."

• (after reviewing Biogen v Medeva²³ and Genentech I/Polypeptide expression²⁴). There is nothing wrong in principle in the patent covering both rapamycin and derivatives that work in the same way. "Sufficiency requires the monopoly to match the contribution. The fact that here, as in most if not all other pharmaceutical cases, there is no way of predicting with accuracy which derivative molecules will possess new and improved properties does not mean that a monopoly which covers all rapamycin derivatives which work extends beyond Professor Calne’s contribution. On the contrary, although his discovery of the effectiveness of rapamycin was, no doubt, empirical, once that discovery had been made, it made available to the art the opening through which a new class of immunosuppressants could be found. The discovery of other molecules which achieve the same or better results is no longer empirical. The addressee of the patent can work in a logical and predictable way from the one molecule disclosed in the patent to the similar molecules, which are derived from it. Unlike the examples given in Biogen, this patent is not attempting to cover all new immunosuppressants. It is directed only to those molecules which can be reached as a result of rapamycin having shown the way."

• The research necessary to find further derivatives did not impose an undue burden because the required iterative work was standard and expected for finding good candidates in the pharmaceutical industry, and the time and money required was only an indication that the research in this field was slow and expensive. However many derivatives had been disclosed in the patent, it could still be argued that the same process of synthesis and testing would be required to find further derivatives. If the only safe course were to limit the patent to the molecules that had actually been tried and tested, that would make patents for pharmaceuticals more or less valueless.

• The second use form of claim should make no difference to the outcome. "If Professor Calne had happened to find his novel use for rapamycin before it had been found to have fungicidal properties, his claims would not have needed to be in second use form. However, the invention would have been exactly the same and the requirement for sufficiency would have been exactly the same."

The Court of Appeal agreed with the High Court that "rapamycin" when used in the claims referred to the single known molecule and did not also refer to derivatives that also inhibited organ rejection. Aldous L.J. based his conclusion on a linguistic analysis of the specification, which showed a distinction was made between rapamycin and its derivatives, and on the fact that the specification did not disclose a single derivative that had been shown to work.

He went on to find on the basis of the "Protocol questions"²⁶ that the defendant’s compound was not a variant falling within claim 1 of the patent in issue. The defendant’s compound did not satisfy the test imposed by the second Protocol question, which in the interests of third parties required for an affirmative answer that the variant should be obviously or clearly immaterial. In the present case, although the defendant’s compound might have been a good candidate to try, it was not obvious that it would work as an immunosuppressant and in order to find that out you would have had to make and test the compound, which required research. The inventor had only discovered and described the second medical use of rapamycin, and it was left to others to find out which derivatives, if any, worked. Furthermore, it was unfair to the patentee to construe his claim in a way that was not intended. To ignore a limitation could render a patent invalid contrary to the wishes of the patentee. In answer to the third Protocol question, the skilled person would have understood that strict compliance with the primary meaning was essential because:

• throughout the specification "rapamycin" was used to denote the molecule itself, and it would be surprising if a different nomenclature was used in the claims;

• the inventor’s discovery had been defined in relation to rapamycin itself;

• if the patentee had intended to cover derivatives, he could easily have done so; and

• a claim covering rapamycin and rapamycin-like derivatives would not have been allowed by the EPO as it would have lacked support and been speculative. The broadening passage relied on by the claimants contained an unqualified reference to all derivatives and prodrugs of rapamycin whereas the EPO would not have permitted all derivatives to be claimed²⁷. Furthermore a test or standard was needed by which a skilled person could decide whether or not a particular derivative had a rapamycin-like effect and the specification did not disclose any such test.

If claim 1 had the broader scope for which the patentees had argued, then Aldous L.J. concluded that the specification would be insufficient. The requirements for sufficiency were (a) that the claim should be enabled over its whole scope²⁸ and (b) the skilled person should not be required to carry out research to ascertain how the invention is to be performed²⁹ (though he could be required to use appropriate skill and tenacity). In this case, if the patent had covered derivatives of rapamycin, it would have needed to teach how to perform the invention with such derivatives. The specification did not do so because although it told the skilled man where to start, it left him to ascertain by research which derivatives work, the number of derivatives was vast and the task of ascertaining those which would satisfy the functional part of the claim was vast and correspondingly burdensome. A claim covering derivatives would cover all molecules that would work while leaving it uncertain which ones did and how many of them there were. Such a claim did not reflect a class with a unifying characteristic but was a claim to a number of compounds with the number and identity being left to the skilled person to find out.

It was not true that a patent covering only a second medical use for rapamycin would be virtually valueless. The patent protected that use and the long and expensive work needed to obtain regulatory approval. Anyone who wished to market a derivative would have to make the derivative and carry out the long and expensive work to get the derivative on the market. The patent system should not be used to enable a person to cover more than he had described in sufficient detail to amount to an enabling disclosure, and in this case it would stifle research aimed at finding a derivative of rapamycin that was a better immunosuppressant than rapamycin itself. In support of the latter proposition Aldous L.J. quoted as apt a passage from the Biogen case in which Lord Hoffmann discusses the work of Samuel Morse and the Wright Brothers and concludes that:

The defendants’ compound contained up to 0.8% of rapamycin and the patentees argued that this was sufficient to establish infringement. Aldous L.J. rejected this argument on the ground that the medicament had to be essentially rapamycin and there had been no discovery that medicaments containing only 0.8% rapamycin had any therapeutic effect. Accordingly the Court of Appeal found that the patent was valid but not infringed.

In an empirical research field, and in particular in the pharmaceutical field, it is difficult for a pioneering inventor to formulate a generic claim that is both wide and scientifically supportable because there are no established structure/activity relationships that can be used as a basis for rational prediction. The difficulties are illustrated by the Rapamycin case where Professor Calne had researched the prior art concerning rapamycin derivatives, but that prior art was relatively sparse and potential sites for modification and the range of permissible changes at those sites were only beginning to be investigated. It was only after Professor Calne’s invention that significant numbers of patents disclosing structure/activity investigations for compounds within the rapamycin family started to appear. Similarly in the Monsanto case, although pharmaceutically active disubstituted thiophenes were known, the COX-2 target was relatively new and there was no body of knowledge concerning previous selective inhibitors of COX-2 on the basis of which valid predictions could be made. A reader of the specification would be struck by the limited information as to activity supported by reports of actual tests, and might view the claim scope as driven more by insight as to what was chemically available than insight as to biological effectiveness. This situation may be contrasted with that in Olin Mathieson v Biorex where a specific change was being made at a particular position within a well-explored family of compounds, and an existing body of knowledge of structure/activity relationships in parts of the molecule that were not subject to change provided basis for a claim of useful generic scope that nevertheless survived the challenge of litigation.

Two drafting points emerge. Firstly, as shown by the Agrevo and Monsanto decisions discussed above, when dealing with inventions concerning biologically active compounds, including open-ended substituent definitions e.g. alkyl or aryl in a generic claim is risky and should be backed-up by other claims with more specific generalisations in which the maximum number of carbon atoms is specified or e.g. an overall range of molecular weights is specified. Secondly, if the invention involves a specific change at a location in a known family of molecules, then it may be to the applicant’s advantage to discuss in detail the results achieved in structure/activity relationships at other locations within the molecules described in the prior art because this helps to show that in these aspects the genus claimed is not an irrational prediction but builds on previous successful experimental work, as in Olin Mathieson above.

There is no fully satisfactory way of claiming generic pharmaceutical inventions. The Markush approach has at least 80 years of use, but the validity of Markush claims if challenged by a resolute and well-resourced opponent or defendant should not be taken for granted. No judge is likely to treat a generic claim sympathetically following demonstrated inactivity within the genus. Over-broad claims should be avoided, and participation not just from chemists but also from biologists and molecular modellers can be helpful. Furthermore, it is important to claim the disclosed compounds individually as well as within the genus because the claims to the individual compounds may be the only valid claims.

Footnotes

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.