Nothing to see here, Lilly wins in kicking out Genentech's Talz progeny patent (for now)

When it comes to the validity of European patents, the national courts of the EPC member states apply the same law as the EPO. None-the-less, UK judges sometimes come to different conclusions to that of the EPO. As a consequence, a patent granted and maintained by the EPO may very well be revoked by the UK (England and Wales) court. Potential infringers may seek legal certainty that any patent granted by the EPO covering their product would not be valid by seeking an Arrow declaration. However, what if you have already knocked out a patent in the UK courts and the EPO then grants a virtually identical divisional patent? Does a patentee have the right to take two bites at the cherry? This was the question considered in Lilly v Genentech [2020] EWHC 261 (Pat).

The case relates to Taltz, Eli Lilly's biologic drug for psoriasis. Taltz is Eli Lilly's marketed anti-IL-17A/F monoclonal antibody ixekizumab. Genentech owns a family of European patents claiming IL-17A/F antibodies and their use for the treatment of psoriasis. Last year, in a dispute between Lilly and Genentech, Mr Justice Arnold (as then was) in the UK High Court found one of these patents (EP1641822) invalid. In particular, the patent claim directed to the IL-17A/F antibody was found to be obvious. The use of the antibody to treat psoriasis was found to lack plausibility at the priority date, applying the principles of Warner-Lamber v Actavis ([2018] UKSC 56). The UK High Court decision has been appealed. However, the patent in question has also subsequently been revoked by the EPO Boards of Appeal, with the effect that the EP(UK) patent is considered never to have existed. 

A second bite of the cherry?

Genentech owns another granted patent in the family. This divisional patent (EP2784084) has the same priority date and effective filing date as the parent patent disputed in the UK High Court. Genentech brought a second action against Eli Lilly, based on this divisional patent. Lilly counterclaimed that the divisional patent was invalid (Lilly v Genentech [2020] EWHC 261 (Pat)). The parent and divisional patents are considered “virtually identical” (para. 44) [Why were multiple near-identical patents granted? See Appendix below]. Lilly therefore argued that the case contravened the legal principle of res judicata i.e. that the same case should not be heard twice. In other words, Lilly argued, that Genentech should not be given a second bite of the cherry and that “this is the evil that issue estoppel is there to prevent”. Lilly also cited the cost, length and complexity of the proceedings for the parent patent (the costs ran to over £11m). 

Second bite at the cherry?
Would estoppel amount to an injustice to Genentech?

Genentech argued that it would be an injustice to prevent them from pursuing the case that the divisional patent was valid and infringed. The appeal for the parent case lost all of its value following the decision of the EPO Board of Appeal to revoke the European patent. However, Genentech argued that the fact that they had been given leave to appeal the UK High Court decision demonstrated that “they must have had a good prospect of success” (para. 63). Notably, the ground on which the EPO had revoked the patent, added matter, was dismissed by Mr Justice Arnold in the UK High Court - a reminder that the EPO approach to added matter may not be fully aligned with that of the UK.

The judge had some sympathy with Genentech’s argument that it would be injustice to prevent them pursuing the case. However, the judge also found that any potential for injustice was outweighed by the length and expensive of a re-run of the trial: “[t]his cannot be in the public interest and would be an injustice to Lilly” (para. 64). The judge also felt that it was in the hands of the Court of Appeal whether the appeal should continue [Merpel: This is unlikely to be of any comfort to Genentech, given that the parent patent “no longer exists” following revocation by the EPO]. The judge therefore considered whether the criteria for estoppel (i.e. preventing the arguments from being heard again) were satisfied for each of the grounds under which validity was challenged.

When is an issue fundamental?

A party is prevented from raising an issue again in a case if the issue was “fundamental” to a previous final decision between the same parties. Genentech argued that for an issue to be “fundamental”, the issue must have been “legally indispensable to the conclusion” (para. 55) of the previous decision. Thus, if the previous decision could have been reached by alternative grounds, there is no estoppel. The judge noted that, if accepted, such a submission would have far reaching consequences in the field of patents. Patents are usually challenged on multiple grounds. A patent may, for example, be found invalid for obviousness based on two different prior art documents. Genentech argued that if a patent was found invalid on more than one ground, neither ground could be considered “fundamental” to the decision. Thus, estoppel should not apply to either ground.

The judge was not convinced by Genentech’s arguments. Particularly, the judge found that even if there were alternative grounds for invalidation, estoppel could apply to both grounds. In the present case, Lilly could be said to be relying on two grounds of attack that the claims were obvious. The judge concluded: “I see no injustice in there being issue estoppel in respect of both issues and the issues fundamental to each of them” (para. 61). The judge concluded that Genentech was estopped from arguing that the antibody claims of the parent case were valid.

Fundamental but different?

Genentech accepted that issue of plausibility was fundamental to the decision by the High Court that the treatment claims were invalid (para. 67). In the case relating to the parent patent, Lilly had successfully argued that these claims were invalid for lacking plausibility at the priority date. Genentech had attempted to argue in response that plausibility should be assessed on the filing date, not the priority date. Genentech had particularly tried to argue that the plausibility of the same subject matter may change because of, for example, advances in the common general knowledge. Genentech argued that if plausibility could be demonstrated for the filing date but not the priority date, this should result in a loss of priority, but not lack of sufficiency. However, the judge did not consider this argument because it was submitted too late in proceedings.

Given that Genentech’s argument on plausibility at the filing date had not been admitted to the proceedings relating to the parent patent, Genentech argued that they were not therefore estopped from bringing this issue with respect to the divisional. Lilly argued in response that the fact that the argument had not being submitted in time, was Genentech’s own fault. Raising the issue in proceedings for the divisional patent, Lilly argued, was thus an abuse of process.

The judge was sympathetic to Genentech: “I do not believe that Genentech should be shut out from arguing plausibility at the filing data simply because they …left it too late to…include that argument”. The judge thus concluded that Genentech was estopped from argued that the claims for the treatment of psoriasis were plausible on the priority date. However, the judge found that Genentech had “a real prospect of defending the claims […] were plausible at the filing date […] on the basis that the common general knowledge had developed sufficiently between the priority date of the said patent and [the filing date]”. The judge therefore refused to provide a judgement on these claims without a full trial (para. 94 and 95). 

The dispute can therefore be expected to continue. Lilly has succeeded in preventing Genentech from bringing the issue of infringement and validity of the the antibody claims to court again. However, the fight continues over the validity of the claims directed to the antibody for the treatment of psoriasis, and the issue of whether this claim was plausible at the filing date of the patent. 

Estoppel beyond divisionals – Double-patenting

The issue of estoppel in the UK may not be limited to divisional patents. Estoppel may also arise in cases of double-patenting, i.e. where the claims of two patents or patent applications are identical. There is a referral pending before the Enlarged Board of Appeal (EBA) of the EPO on the issue of double-patenting (G 1/19). The EBA has been asked whether double-patenting should be prevented in cases of priority, e.g. where the priority claim is to a previous European patent application (See T 0015/01, r. 26). If double-patenting is permitted in these circumstances, the principle of estoppel may apply to any attempted enforcement in the UK.

Final thoughts

An important feature of the European patent system is that there is no estoppel between the EPO and the national courts. Arguments raised during opposition at the EPO may be raised again in the national courts. The European patent system thus contrasts with that of the US. In the US, issues that were or could have been brought in inter-partes review or post-grant review may not be brought again in court. 

Whilst the EPO and national courts may consider the same arguments, they may also come to different conclusions. This can lead to the grant and maintenance of European patents containing claims identical to claims that have already been invalidated by the UK. In such a case, the judge in Eli Lilly v Genentech confirms the principle that patentees should not be permitted to argue the same issues again. However, the judge adds an interesting nuance. In particular, if there were lines of argument that a patentee did not pursue in a first invalidity action, a patentee may be permitted to bring another action based on a second virtually identical patent.  


Appendix - Why were there two patents? Genentech leaps over the inescapable trap

The parent patent fell foul of a so-called "added matter trap". A granted European patent may be revoked on the grounds that the claims includes subject-matter not contained in the filed patent application (
Article 100(c) EPCArticle 123(2) EPC). A patent may be amended during opposition proceedings to overcome a ground of opposition (Rule 80 EPC). However, amendments that broaden the scope of the claims post-grant are not permissible (Article 123(3) EPC). A patentee may thereby find themselves caught in the inescapable trap of an added matter objection that cannot be overcome by deletion the offending feature. Deleting the feature would broaden the claim scope and there is not permissible post grant (Case Law of the Board of Appeal, II.E.3.1). 

Genentech found themselves in such a trap during opposition of the parent IL-17A/F antibody patent. The patent claimed an antibody, defined by its IL-17A/F antigen and a functional feature. The Opposition revoked the patent on the ground that the functional feature was added matter. Genentech could not remove this feature. Such an amendment would have broadened the scope of the claim. The Opposition Division's decision was upheld on appeal. 

The potential for inadvertently falling into an added-matter trap means that it is wise to keep a pending divisional application, at least during the opposition period. Broadening amendments are possible during prosecution, and so offending added matter features can be deleted before grant. This was the strategy employed by Genentech. The granted claims of the IL-17A/F divisional patent did not include the functional feature.
Nothing to see here, Lilly wins in kicking out Genentech's Talz progeny patent (for now) Nothing to see here, Lilly wins in kicking out Genentech's Talz progeny patent (for now) Reviewed by Rose Hughes on Wednesday, February 26, 2020 Rating: 5

1 comment:

  1. Once again, an extremely useful write up from Rose Hughes, which sets me thinking.

    I don't do chem/bio but for me, claiming medical devices, the chance of inadvertently and despite my best efforts falling into the Art 123(2)/(3) inescapable trap during EPO prosecution are higher than they should be, and getting ever higher. Hence, the advice to a client, to have a divisional pending as one goes through the opposition period is well-taken.

    Spoken about here are divisionals with "virtually identical" claims. Not unusual, if we are talking about avoiding the EPO's inescapable trap, with its ever more bizarre and unworldly assessments under the (no joke) "Gold Standard". A change of one word can escape the trap, without changing the substance of the subject matter claimed.

    I think that the root of the problem lies in the word "derivable". It means one thing to me but something else entirely to many Examiners and Technical Members at the EPO. Could it be that the EPO has a special difficulty with Art 123(2) EPC, not found in other Patent Offices and courts, because examination at the EPO is carried out by Examiners with a first language other than the language of the proceedings: it results in them concentrating excessively on the letter of the wording of one phrase or sentence of the specification, and not enough on the overall teaching, delivered to the notional skilled technical reader by the specification as a whole. After all, the reader is deemed to be technically skilled and to be reading the document with a mind willing to understand, and hungry to learn all they can from the written content of the document.

    Compare, for example, two recent TBA Decisions, T1442/16 and T524/17, on what the skilled reader makes of a paragraph of disclosure that opens with the words "In an embodiment....." What is "derivable" from such a statement? The way I see it, the Rapporteur of the one Board is (bravo) present in the real world and is alive to the issue while the other hasn't even begun to think the issue through, instead merely trotting out the unreal and unrealistic EPO "party line".


All comments must be moderated by a member of the IPKat team before they appear on the blog. Comments will not be allowed if the contravene the IPKat policy that readers' comments should not be obscene or defamatory; they should not consist of ad hominem attacks on members of the blog team or other comment-posters and they should make a constructive contribution to the discussion of the post on which they purport to comment.

It is also the IPKat policy that comments should not be made completely anonymously, and users should use a consistent name or pseudonym (which should not itself be defamatory or obscene, or that of another real person), either in the "identity" field, or at the beginning of the comment. Current practice is to, however, allow a limited number of comments that contravene this policy, provided that the comment has a high degree of relevance and the comment chain does not become too difficult to follow.

Learn more here:

Powered by Blogger.