UK: Neurim v Mylan: On And On It Goes

In the preliminary issues trial of the Neurim v Mylan battle, take two (concerning a divisional patent), the Patents Court (Meade J) has decided that no issue estoppel arises from Neurim's previous win on the parent patent. Mylan (now Viatris) failed on its allegations of abuse but will now proceed in its attempt to convince Marcus Smith J that the divisional patent is invalid in a heavily case managed procedure.

We have previously reported on the Neurim v Mylan battle "take two" concerning a divisional patent (EP 3 103 443) granted on 30 June 2021 (expires 12 August 2022), when Neurim successfully applied for an expedited preliminary issues trial (Neurim v Mylan: Take Two). That article also covers (some of) the twists and turns of the original Neurim v Mylan battle on the parent patent.

In December Meade J heard a three-day trial of the preliminary issues. Following the decision by Mellor J to permit a trial of preliminary issues, matters developed and Meade J describes the list of issues as being "14 paragraphs with many sub-paragraphs" but notes it was not much used. As is usual at trial the proceedings boiled down to a few key questions, being:

1. "Whether Mylan is issue-estopped from challenging the validity of the Divisional in the light of the 2020 action.
2. Whether Neurim's conduct in amending the Divisional is an abuse.
3. Whether Neurim's conduct amounts to an abuse of a dominant position (on the assumption of dominance and market definition)." ([15] of the judgment)

In brief the answers were (i) no, Mylan could not appeal at the relevant time so issue estoppel does not arise; (ii) no, Neurim's conduct was not abusive (but there is a pleading issue that limited the scope of this decision); and (iii) the competition arguments were contingent on issue estoppel arising and furthermore raised complicated novel issues of law and therefore were considered unsuitable for decision in the circumstances.

It is not intended in this article to delve into the details of the ways that issue estoppel arise and affect parties litigating a divisional patent. That subject, covering the events of Lilly v Genentech as well as Neurim v Mylan, is worthy of its own article. Instead, we shall address the issue estoppel and abuse points which were decided and then explain the novel approach to deal with the resulting situation that has been adopted by Meade J.

Issue Estoppel

The key issue on the law was the proper approach where it is the "winner" that is said to be estopped, especially if they could not appeal. Considering the relevant case law Meade J noted that appealability is a useful test but not the ultimate test. There was also a significant discussion of the nature of an appeal and, having considered various authorities, Meade J approved paragraph 52.0.6 of the White Book and said "I find the terms "result" and "outcome", to be contrasted with "findings" or "reasons", to be helpful, and indeed more so than "judgment". I appreciate of course that "judgment" is one of the words in the relevant statutory provision, but if used loosely it may be rather ambiguous as to whether reference is being made to the reasons in a judgment, or to the outcome of it."

That being said, Meade J analysed the real issue as the proper characterisation of the result of the previous proceedings, the reasons for the result, and who was the "winner", and when. The problem was, and as explained here, that Neurim was initially the winner. However, even though during part of the time Mylan could have filed an appeal (as a loser), it was always recognised by the judge and the parties that the December 2020 final order was provisional because a Technical Board of Appeal ("TBA") hearing was imminent.

Meade J decided that the when, therefore, was after the TBA hearing. At that time the facts changed and Mylan was the winner. Neurim's claim was denied and a (further) final order was produced reflecting this result. The reason for the result was the events at the TBA and thus the reasons in the December 2020 judgment were not fundamental to the eventual result. No issue estoppel therefore arises.

The practical effect of this decision is complex and is dealt with below under the heading Case Management.

Abuse of Process

Mylan's defence alleged Neurim's application to amend the divisional patent was an abuse of process. The amendment seeks to put the divisional into the same form as the parent patent that was the subject of the December 2020 judgment.

Perhaps unfortunately for patentees that wish to have certainty on what they may and may not do with divisional patents, a pleading point was taken that prevented Mylan from running the argument they developed at the trial. The focus by that time was on the overall strategy of using a divisional as a back-up (a common occurrence) and the abuse of suing twice on the same invention in the UK. This, Meade J said, would be an important and complex argument but as it had not been pleaded it would be unfair to allow it to proceed. He also noted that the UK court has already developed the Arrow declaration as a way to deal with the problem and can use its powers of case management to deal appropriately with the practice of using divisionals as back-up to a key patent.

The decision on abuse is limited to the act of avoiding of a written decision by the TBA, an allegation that the withdrawal of the TBA appeal was an acceptance that the parent patent was invalid and that this results in the amendment of the divisional to match the parent being an abuse. Meade J rejected that any abuse flowed from avoiding a written decision because the divisional might still have come to grant later with a written decision and the validity of the divisional can in any event be contested at the EPO. On the facts, Meade J found that Neurim had not accepted the parent was invalid. The withdrawal was a practical response to the situation at the EPO.

Meade J further held that Mylan was not being harassed looking at the matter more generally since at all times it knew there was a divisional held as a back-up which would lead to ongoing litigation uncertainty (inherent in the EPC system of divisionals). These views, in the absence of a fuller debate on the general proposition Mylan was prevented from pursuing, provide some comfort to patentees with divisionals held as back-ups. He also noted, at [166], further reasons specific to the case at hand.

Case Management

These decisions mean that Neurim can proceed with its application to amend but that, also, Mylan is permitted to run its validity case. However, Meade J has attempted to deal with "the real justice of the situation" by novel case management. He records Mylan's key concern as being the injustice of never having had a full chance to ask for permission to appeal on what is referred to as the lay-patient argument.

In effect, the lay-patient argument is the key and probably only ground of invalidity that Mylan will run as their invalidity defence. It is a particular plausibility insufficiency on the facts of the case and it succeeded at the TBA. It was also run in the 2020 action but Meade J noted that it was "not articulated very clearly". He also thought it was relevant to this argument that Marcus Smith J had assessed Mylan's expert as having been very poor. In the UK system, it is always difficult to succeed on a point, even a good one, if your expert does not come up to standard under cross-examination.

Of course, given this finding by Marcus Smith J, a party attacking the divisional patent would wish to be able to rely on fresh expert evidence. However, this is not possible for Mylan given events and the court's approach to case management (generally and in the specific case). Presumably this is why Mylan's complaint ultimately focussed on their lack of ability to appeal the point.

Meade J's first inclination was to "deal with the trial of this action (a) on the basis only of the materials and arguments before Marcus Smith J, (b) by adopting and giving effect to the Main Judgment, and (c) by refusing permission to appeal just as Marcus Smith J did on 16 December 2020."

This conclusion appeared in a draft of the judgment circulated to the parties. However, Mylan objected to (b) and (c) and a further hearing was held to deal with the appropriate way forward. In brief: Mylan considered it an inappropriate way forward, in the circumstances, for a judge to adopt the judgment of another and that it had never had a proper chance to seek permission to appeal taking into account the decision of the TBA.

In the unusual circumstances of this case, Meade J accepted that it would be appropriate for the case on the divisional to proceed to Marcus Smith J for a decision on the evidence that he had heard. It will be for the parties to make submissions about exactly how that should be handled and he may simply adopt his prior decision (as Meade J had been planning to do). However, Mylan can also attempt to convince Marcus Smith J to change his original judgment, but only relying on the original evidence. Once Marcus Smith J comes to a decision (which may be with or without a hearing) it appears that permission to appeal will proceed in essentially the normal way for whichever party loses.

As an outside observer, this novel case management decision appears to adopt an awkward but fair position in-between a pure appeal and getting to run the lay-patient argument afresh. Between these parties the latter was unrealistic but it cannot be forgotten that Teva also stand in the wings ready to invalidate the divisional patent (they have been sued by Neurim but no defence had been filed at the time of the trial of preliminary issues). We expected that Teva will fancy its chances on the lay-patient argument with fresh evidence. That is for another day.

Finally, this decision has imposed a new duty on parties to litigation before the Patents Court to keep it informed about scheduling issues and changes to hearing dates in parallel EPO proceedings. See our article on this point here.