UK: Wyeth's Meningitis B Vaccine Patent Valid And Infringed By GSK's Bexsero Vaccine; Henry Carr J Explains Patentability

Wyeth's Meningitis B (Men B) vaccine patent (the Patent) has been held valid by the English Patents Court, following a challenge by GlaxoSmithKline (GSK). GSK's Bexsero vaccine was found to infringe the Patent, but Wyeth has indicated that in light of the public health requirement for effective vaccines against Men B, it will not seek injunctive relief.

In a comprehensive judgment that is likely to be cited frequently in the future (GlaxoSmithKline UK Ltd v Wyeth Holdings LLC [2016] EWHC 1045 (Ch) (12 May 2016)), Henry Carr J took the opportunity to explain the legal tests on many aspects of the law regarding patentability.

Background

Meningitis B remains a substantial health problem in most countries. The quest for vaccines that provide effective immunisation against a broad spread of the Neisseria meningitidis bacterial strains responsible for Men B is longstanding.

GSK's Bexsero Men B vaccine was given regulatory approval at the European level in 2013. Wyeth/Pfizer's Trumenba vaccine awaits approval in Europe.

GSK sought to 'clear the way' for its Bexsero vaccine by challenging the validity of Wyeth's European patent, number EP (UK) 2,343,308 (the Patent), on multiple grounds. Wyeth counterclaimed for infringement of the Patent by Bexsero.

GSK's multiple grounds of challenge presented Mr Justice Henry Carr, the newest of the Patents Court judges, with an opportunity to peg out the legal principles underpinning many aspects of the law regarding patentability. In this article, we explain Henry Carr J's guidance, as well as the outcome for the parties to the dispute.

The skilled addressee

Henry Carr J noted that there was no dispute between the parties as to the legal principles governing the identification of the skilled addressee of a patent, which he summarised as follows:

For the Patent in issue, the skilled team would comprise scientists with an interest in Neisseria vaccines, including those with experience in microbiology and vaccinology, and would be led by a person having more than 10 years' post-doctoral experience. The earliest claimed priority date being in 2001, the team would have had access to basic bioinformatics expertise, but not to techniques considered "cutting edge" at the time.

The common general knowledge

Henry Carr J cited (without quoting) the summary of the legal principles in respect of 'common general knowledge' set out by Arnold J in KCI Licensing v Smith & Nephew [2010] EWCHC 1487 (Pat), and approved by the Court of Appeal in that case ([2010] EWCA Civ 1260). He also noted that he agreed with the following analysis of Sales J in Teva v Astrazeneca (asthma) [2014] EWHC 2873 (Pat):

Henry Carr J noted that the passage did not mean that all material available online constitutes common general knowledge. Rather, it "indicates that material which the skilled addressee knows to be available online and which is generally accepted as a good basis for further action (such as material which might be found offline in a text book or key journal article) may constitute common general knowledge" (emphasis added).

Turning to the present dispute, the judge noted that in 2001 it was no longer thought that a polysaccharide-based vaccine would be developed for Men B. There were also challenges with developing a vaccine using "OMV" technology: it was considered that in order to confer protection against a sufficient variety of strains it would be necessary to use a large number of different PorA proteins, which was undesirable.

However, while the limitations of PorAs were well known, the notional skilled team would nonetheless have considered that they could have utility in multicomponent vaccines, in combination with other proteins. In this respect the judge preferred the evidence of GSK's expert, Professor John Heckels, over that of Wyeth's expert, Professor Dlawer Ala'Aldeen.

The "reverse vaccinology" approach - the use of the genome sequence to attempt to express and characterise proteins which might be potential vaccine candidates - was an important development which was also part of the common general knowledge. The Men B genome was sequenced in the late 1990s.

The patent and its construction

The invention underlying the Patent was described in the specification as the providing of Neisseria "2086 proteins", including in recombinant forms, isolated forms, and in both lipidated and non-lipidated forms. The proteins were described as eliciting bactericidal antibodies to multiple neisserial strains. Antisera to a single 2086 protein was described as able to kill a range of strains, responsible for the majority of the Men B disease isolates in significant geographical areas. It therefore had the potential to reduce the number of components in the vaccine, as compared with the use of PorA proteins, which required 6-9 variants.

The specification of the Patent also identified specific 2086 proteins, by their amino acid sequences, including those referred to by the identification numbers "212", "214" and "216".

Claim 1 of the patent was to a composition:

1. containing at least one protein comprising an amino acid sequence having sequence identity greater than 95% to the amino acid sequence of any one of SEQ ID Nos: 212, 214 and 216, and
2. additionally comprising at least one PorA protein.

Other claims narrowed the claimed compositions to cover, for example, proteins with sequence homology greater than 95%, the addition of an adjuvant, and use as a vaccine.

Applying the principles of construction set out in Kirin-Amgen v Hoechst Marion Roussel [2005] RPC 9, and in Virgin Atlantic Airways v Premium Aircraft Interiors [2010] RPC 8, Henry Carr J made a number of rulings on the construction of claim language in issue:

• There was no reason for the skilled person to think that the patentee was using the term "PorA protein" as limited to any particular type of PorA protein, for example to "isolated" or "free" PorA. There was no technical reason to do so.
• Claim 3, which limited the language of claim 1 to proteins comprising "the amino acid sequence of any one of SEQ ID Nos: 212, 214, and 216" was limited to proteins having 100% sequence identity to the specified sequences. This was consistent with a) the specification referring to a sequence of levels of homology, and earlier claims being to broader levels of homology, b) the patentee choosing not to include a specified higher level of homology (e.g. 99.9%), and c) the evidence of the expert witnesses. Further, it was the judge's view that permitting a broader construction of claim 3 would not give reasonable certainty to the public, as required by Article 69 of the European Patent Convention.
• Claim language to "use as a vaccine", to "use in a method of inducing an immune response in a mammal" and to "use in a method of ameliorating or preventing N. meningitidis infection in a human", did not require more than a discernible effect in the treatment of Men B. The technical contribution of the patent was "a composition and vaccine which provides protection for a population of patients against a broad range of diverse strains, as a result of inclusion of the 2086 protein". It was not correct to characterise the technical effect of the patent in terms of individual patients. Nor did the 2086 protein have to be effective against every Men B strain. However, GSK's case that the claims covered non-functioning 2086 proteins, where the required efficacy came from PorA alone, was not accepted. This, said the judge, would defeat the object of the claimed combination and was not a fair interpretation of the claims, objectively assessed. The purpose of the inclusion of one or more PorAs was to extend coverage to strains against which the 2086 protein would be ineffective.
• The judge also noted that, contrary to the apparent understanding of GSK's expert witness, Professor Heckels, the reference to "complimentary" in the Patent did not mean that the combination of 2086 and PorA had to be synergistic.

Plausibility

Noting that lack of 'plausibility' underpinned many of the pleaded insufficiency allegations and the allegation of AgrEvo-type obviousness, Henry Carr J summarised the legal principles as follows:

The judge noted that once the misunderstanding of GSK's expert, of a need for a synergistic effect between the 2086 and PorA proteins, was removed, the expert evidence established that the disclosure of the patent was plausible.

The judge's reasoning included the following: the data in the patent were consistent with what the skilled person would expect i.e. that killing generally appeared to be related to the level of homology between the 2086 protein used to raise the sera and the test strain - that there were exceptions to this general trend would not surprise the skilled person; the threshold figure of 95% homology in claim 1 was a reasonable boundary on the basis of the data in the Patent; and although on the basis of the data contained in the Patent it would have been preferable to select a lipidated rather than a non-lipidated 2086 protein, it was clear that it was not necessary for the 2086 protein to be lipidated in order to raise bactericidal antibodies.

Hence GSK's challenges of insufficiency (general; squeeze arguments are addressed later) and AgrEvo obviousness (i.e. that the technical contribution was arbitrary) failed.

Added matter

Henry Carr J summarised the legal principles relevant to the case as follows:

Regarding the case law of the European Patent Office (EPO) derived from T-12/82 Diastereoismoers, concerning selections from multiple lists, Henry Carr J concluded:

GSK alleged that claim 1 of the Patent comprised an impermissible selection from three separate lists in the application as filed. The judge disagreed. In his judgment, the importance of the 2086 protein from strain 8529 was clear from the examples of the application; the skilled person would know (and was expressly told) that sequences with 95% homology would also have utility; and it was expressly disclosed that 2086 proteins could be combined with one or more PorAs.

Priority

Henry Carr J summarised the legal principles regarding a claim to priority as follows:

The first priority date

The judge had already held the Patent to be sufficient (plausible) over the application as filed, but the first priority document did not contain all of the examples of the application as filed. GSK alleged that this difference meant that claim 1 of the Patent lacked support in / plausibility in light of the first priority document.

The judge disagreed. He said that the disclosure of the combination of a 2086 protein with one or more PorAs was acknowledged by GSK's expert, Prof Heckels, as being plausible, and its use as a vaccine against Men B credible, in view of the examples that were included in the first priority document. Each of the 212, 214 and 216 sequences were unambiguously disclosed and enabled and the disclosure focused upon the group and used them to generate data in examples. Nor were they required to be disclosed without any other sequences, or to be called by the same identifier, in order for priority to be claimed.

Also, the fact that the first priority document disclosed the claimed combination of proteins, and other combinations, "or a combination of any foregoing in a form suitable for mucosal delivery" (which was not yet an established approach to vaccine delivery for meningococcal vaccines), did not limit the disclosure of the claimed combination to mucosal delivery.

The second priority date

Although strictly unnecessary to do so, Henry Carr J then explained why the Patent was also entitled to its second claimed priority date. The second priority document lacked the sequence listings for the 212, 214 and 216 proteins.

The judge noted as common ground that once the skilled person had sequence 212, he would be able to deduce numbers 214 and 216 from other information in the relevant examples. The question was whether, from the disclosure that was made (Example 2), the skilled person could derive, directly and unambiguously, using common general knowledge, sequence 212. The question was not whether the skilled person would have chosen to disregard the disclosure of Example 2 and taken a different route. Rather, it was directed at what would be the result if he did follow that disclosure.

Upon the evidence, the judge was satisfied that the work of the skilled person, as demonstrated by experiments relied upon by Wyeth, would not have resulted in anything other than the 212 protein and that they would not have constituted an undue burden. This work included waiting a few weeks for a sample of the 8529 strain to arrive, and PCR and sequencing taking a further two weeks. No specifics of experimental adjustments were relied upon as materially affecting the outcome of the work. The inclusion of nine errors over the consensus in Wyeth's experiments did not prevent the 212 protein being the inevitable result: "a variety of examples were shown in the literature where a consensus sequence did not mean that 100% of residues at any given position had to be the same".

Consequently, the claims of the Patent were entitled to the second priority date also.

Novelty

In the terminology of the UK Patents Act 1977, a patent will lack novelty if the invention has, before the priority date, "been made available to the public...by written or oral description, by use or in any other way".

Anticipation by prior art

For prior art (i.e. a written description) to deprive a patent of novelty, Henry Carr J noted that two requirements must be met, as was made clear by Lord Hoffmann in Synthon v SKB [2006] RPC 10:

GSK challenged the novelty of the Patent over four separate prior art disclosures, but none was judged as satisfying the above test. The following noteworthy points of guidance emerge from the judge's reasoning in this context:

The need for an enabling disclosure

One of GSK's anticipation challenges was based upon an abstract ("Andersen") comparing the antibody response to natural OMVs versus detergent-extracted OMVs from a particular strain of mice (H44/76). GSK alleged that the OMVs were known to comprise PorA protein, that they did in fact contain a protein (fHbp) now known to have homology greater than 95% to the amino acid sequence of each of the 212, 214 and 216 proteins, and that the skilled person could have determined that as a matter of course before the first priority date, using standard techniques of protein sequencing.

However, GSK's expert, Professor Heckels, characterised the task as "extremely laborious" and conceded that while the task was theoretically possible, it was very unlikely to work. Numerous steps would be required, any of which might fail, and with the outcome depending upon the conditions employed. An OMV containing fHbp was not the inevitable result, so Andersen did not amount to an enabling disclosure.

Another of GSK's anticipation challenges was based upon a patent application referred to as "869", published after the first priority date of the Patent (so this part of the judge's reasoning was not, strictly, necessary). GSK contended that a composition within the scope of claim 1 of the Patent was the inevitable result of carrying out Example 6 of 869. Example 6 contained no directions on how to obtain or produce the proteins in question, and GSK's case relied upon the skilled person selecting the appropriate antigens from a long list of possible antigens described in text elsewhere in the specification. The difficulty, said the judge, was that there was no link in 869 between the disclosure of Example 6 and the specific text that the skilled person needed to suggest; 869 did not anticipate.

GSK also alleged anticipation based upon four documents (the "Bernfield and Farley posters/abstracts") presented by Wyeth at the Twelfth International Pathogenic Neisseria Conference, between the second priority date and the filing date of the Patent (so again the judge's reasoning in respect of this challenge was not, strictly, necessary). The documents described the isolation of recombinant 2086 proteins from a group A meningococcal strain, antigens capable of eliciting bactericidal antibodies against meningococcal strains expressing heterologous PorA and heterologous P2086 proteins, and the potential use of P2086 proteins in vaccines, alone or in combination with other antigens.

The issue was whether the disclosure would enable the skilled person to obtain the sequence of 2086. The evidence of GSK's expert, Prof Heckels, was that, after conducting a simple experiment, and with reference to the Men A genome, the skilled team could have worked out that sequence from the disclosure of the Bernfield paper. Wyeth contended that the skilled person would have referenced instead the Men B genome, not found the relevant protein, and not realised this error.

The judge held that the simple experiment proposed by Prof Heckels would not inevitably have been performed by the skilled person at the second priority date, so the anticipation challenge would have failed. However, the judge said that had he concluded that the Patent was not entitled to either of the claimed priority dates, he would have considered it to lack inventive step in light of this disclosure.

The need for an individualised disclosure

GSK also challenged the novelty of the Patent based upon an international patent application ("885"), which concerned a composition comprising a Men B outer membrane preparation and "an immunogenic component selected from one or more of" a very large number of proteins, listed by reference to 11 other documents. In order for 885 to anticipate the Patent, the skilled person would be required to "pluck out" the 212 protein (referred to by a different code number) from a list of 1,510 proteins contained in just one of those 11 reference documents. There was no direction to the protein in question, nor any direction that the outer membrane component should comprise PorA.

Henry Carr J agreed with Wyeth that anticipation requires a "sufficiently individualised" disclosure. He said that this was made clear by the judgment of Jacob LJ in Dr Reddy's Laboratories v Eli Lilly [2009] EWCA 1362. Rejecting the argument that disclosure of a large class of compounds constituted a disclosure of each member of the class, and therefore deprived each member of the class of novelty, the Court of Appeal held that this was wrong both as a matter of a priori reasoning ("Where does a wise man hide a leaf? In a forest"), and because it was inconsistent with EPO case law that "an individualised description" is needed (Hoechst Enantiomers T 0296/87).

Anticipation by prior use

The principles governing a challenge of anticipation by prior use were explained by Lord Hoffmann in Merrell Dow v Norton [1996] RPC 76, and summarised by Henry Carr J as follows:

The judge also cited the House of Lords' guidance (citing the Enlarged Board of Appeal's decision in G1/92), in Synthon:

GSK challenged the novelty of the Patent, relying upon a prior use, in a vaccine developed at the Finlay Institute in Havana, Cuba. The Cuban vaccine composition included OMVs, which comprised PorA protein, and it was known to do so before the priority date. But did it contain a 2086 protein as required by the Patent?

Applying the above principles, Henry Carr J explained that in order for the use of the Cuban vaccine to anticipate, it had to be established not just that the vaccine contained the protein in question, but also that it was possible (at the priority date) for the skilled person to identify the presence of that protein and to reproduce it without undue burden.

Upon the evidence before the court, the judge concluded that in none of these respects was the burden satisfied.

Inventive step

Henry Carr J noted that the legal principles of relevance to the present case were as follows:

GSK's main challenge under this head was of obviousness over an international patent application ("922") published before the first priority date of the Patent. The object of 922's 'invention' was described as being to provide alternative and improved approaches for heterologous expression of proteins from Neisseria meningitidis and Neisseria gonorrhoeae. Data were provided comparing the properties of the proteins expressed under different conditions.

GSK's case was that data indicating good bactericidal titres and breadth of protection meant that there was a fair prospect of success for the '741' protein as a candidate vaccine against Men B. The 741 protein had an identical amino acid sequence to the 212 protein of the Patent.

Wyeth pointed out limitations of the test data provided in 922; and the judge concluded that 922 did not render the Patent obvious when viewed without hindsight, alternatively that the skilled team would not have tried the 741 protein as a vaccine candidate with a fair expectation of success.

The judge's reasons for reaching this conclusion included that 922 was not concerned with vaccine candidate identification, that the use of different experimental conditions meant that one could not compare results for different proteins, and that the information in 922 was not sufficient to conclude that the proteins discussed were promising vaccine candidates. Also, the fact that GSK's expert, Professor Heckels, had first considered 922 already having knowledge of the protein of significance to the case, was important evidence of hindsight in his evidence.

However, if both priority dates had been lost, the challenge of obviousness over the Bernstein paper (discussed above) would have succeeded.

Squeeze arguments

The judge noted that GSK had relied upon a number of alleged squeezes between the prior art and insufficiency, the submission being that the Patent, the priority documents and the application as filed were no more enabling than the prior art and so one or other of its challenges must succeed.

The judge did not accept this, noting that the former documents disclosed that protein 2086 was a promising, credible and serious candidate for inclusion in a Men B vaccine, that there would be a fair prospect that this antigen would treat a broad range of Men B strains, and that coverage would be increased by inclusion of one or more PorAs. In contrast, the prior art neither provided this disclosure nor rendered it obvious.

Infringement

Aside from the dispute as to the validity of the Patent, infringement was largely conceded, the judge noting that the only debate at trial in relation to infringement concerned claim 3 - which was limited to the specific 212, 214 and 216 proteins. The relevant protein in GSK's Bexsero vaccine having sequence identity of 99.2%, 98.8% and 99.2% to 212, 214 and 216 respectively, claim 1 and several other claims (but not claim 3) of the Patent were infringed.

Comment

Involving well known pharmaceutical companies, and concerning important vaccine products, the GSK v Wyeth dispute will be of widespread interest.

For commentators on the state of the law in the UK regarding patentability, the decision is also of considerable interest for its attempt to peg out, clearly and succinctly, the relevant legal tests. Addressing a broad swathe of the principles underpinning patentability, Henry Carr J's judgment represents a comprehensive starting point for the drafting of a textbook on the subject, and a good refresher course for those in need of reminding as to the current state of the law. A few additional thoughts are noted here.

Plausibility

Following the Autumn 2015 decisions of the Patents Court in Generics v Warner-Lambert, Merck v Ono and Actavis v Eli Lilly, appeals are pending. Interlinking many of the issues in those cases is the question of what is meant by the concept of 'plausibility', and how this concept plays a role in the principles underpinning the assessment of patentability.

In GSK v Wyeth, the challenge to sufficiency was one of 'excessive claim breadth', raising again the concept of plausibility. Henry Carr J reported no dispute between the parties as to the principles that should be applied; his summary of them draws upon his ruling in Actavis v Eli Lilly and states that plausibility is a "threshold test" which is satisfied by a disclosure which is "credible" as opposed to speculative.

Henry Carr J's summary of the law in GSK v Wyeth presents the Court of Appeal with a ready-made definition which it could choose to comment upon in another case - or not.

The reach of the "common general knowledge"

On the reach of the "common general knowledge", Henry Carr J said that he agreed with Sales J's analysis, in Teva v Astrazeneca, that if it were obvious for the unimaginative and uninventive skilled person to conduct a search, the common general knowledge would include material that would readily be identified by such a search.

The judge then proceeded to qualify Sales J's analysis by saying that it did not mean that all material available online constituted common general knowledge, just that it indicated that material which the skilled addressee knows to be available online and which is generally accepted as a good basis for further action (such as material which might be found offline in a text book or key journal article) may constitute common general knowledge. How much of Sales J's analysis remains intact following this qualification is less clear.

In this context, the recent decision of the Court of Appeal in Richter Gedeon v Generics [2016] EWCA 410 Civ should be noted. The Court of Appeal dismissed an attempt to characterise the sort of information which a person skilled in the art would know he could look up as a "species of or extension of ... common general knowledge". Instead, the Court of Appeal approved the following summary of Arnold J in KCI Licensing v Smith & Nephew [2010] EWHC 1487 (Pat):

In some instances, this distinction will make a difference. As Lord Hoffmann stated, sitting in the Court of Appeal in Lundbeck v Generics [2008] EWCA Civ 311, "In order to anticipate a patent, the prior art must disclose the claimed invention and (together with common general knowledge) enable the ordinary skilled person to perform it". Information that is considered common general knowledge will therefore be in the head of the notional skilled addressee when reading prior art and interpreting what it discloses; information that is not common general knowledge will not.

Similarly, in the context of a claim to priority, the question is whether the skilled addressee can derive the subject matter of the claim directly and unambiguously, using common general knowledge.

In contrast, in the context of an obviousness challenge, the consequences of information not being common general knowledge are less binary. As Lord Justice Floyd succinctly put in Richter Gedeon, "where it is established that the skilled person would have been prompted by the state of the art to ask for a specific piece of information from a source which would be obvious, and that the information requested would have been freely given, a finding of lack of inventive step is...properly open".

Defining the test for obviousness

Turning to obviousness, Henry Carr J noted that: where it is alleged that a step is obvious to try, the question is whether the skilled person would do so with a fair expectation of success; how much expectation depends on the particular facts of the case; and including something in a research project is not enough to establish lack of inventive step.

To the well-known judgments in Conor v Angiotech and Medimmune v Novartis, Henry Carr J added the Court of Appeal's July 2015 decision in Teva v Leo. Our commentary on that decision is available here: it is an important decision as to the standard which must be reached for an 'obvious to try' challenge to succeed.

The approach to the instruction of an expert witness

Finally it is noteworthy that, in the context of its challenge of obviousness over 922, Henry Carr J considered the evidence of GSK's expert witness to suffer from hindsight. This was because the issue was whether it would have been obvious to the skilled person, in light of 922, to have tried one of the proteins described by 922, but when the question was first put to the expert, he already had knowledge of the significance of the relevant protein to the case.

Previously, in Medimmune v Novartis ([2011] EWHC 1669 (Pat)), Arnold J noted that the correct approach for the lawyers instructing the experts (in that case) to take was to ask the expert to consider, first, the prior art, then the priority documents and then the patent. This enabled the expert to form and express his opinions on the prior art without knowledge of the invention, and on the priority documents without knowledge of the patent. More recently, in American Science & Engineering Inc v Rapiscan Systems Ltd ([2016] EWHC 756 (Pat)), each party's representatives complied with the Medimmune guidance, but nevertheless received criticism in respect of the approaches they had adopted.

In the context of a complex patent dispute, the order in which an expert is instructed to consider documents relevant to the case will always require careful consideration.

Concluding comment

To conclude, GSK v Wyeth is a comprehensive, must-read judgment to add to the growing pile of decisions explaining the law regarding patentability from Mr Justice Carr; the first to be tested in the Court of Appeal is expected to be in the Actavis v Eli Lilly case, which is scheduled to be heard by mid-November 2016.

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.