Takeda UK Ltd v F. Hoffmann-La Roche AG

I am going now to give a decision on the first question raised in this CMC which is in dispute, which is the trial date for this action.

Experience shows that trial dates in major patent cases are often as hotly disputed as the substance of the cases themselves, and so it has proved on this application.

The listing window for this case is now between July and November 2019. Takeda seeks a trial date in June 2019, in the hope that a judgment might be handed down before the hearing of an infringement action bought by Roche in Germany in February 2018. The hearing of the German infringement action is set for 18th July 2019, with judgment likely to follow approximately four weeks later. Roche disagrees that a German court, in particular the Düsseldorf court, will be influenced by the decision of the UK court.

Takeda also say that it is important to have the trial in June rather than October, as contended for by Roche, on the basis that this is a very important product commercially and it is necessary to obtain commercial certainty as far as is possible. It is true that it is important to obtain commercial certainty, in so far as that is possible, as soon as possible, and that is reflected in the Practice Statement on the listing of cases for trial in the Patents Court, issued on 7th December 2015 (“The Practice Statement”). That provides that the Patents Court endeavours to bring cases on for trial, where possible, within 12 months of the claim being issued.

Mr. Waugh QC, in the light of the Practice Statement, said that a claimant had a legitimate expectation of having his trial heard within 12 months. I do not accept the words "legitimate expectation". The Practice Statement says what it says; that this is something that the Patents Court tries to do as far as possible. The Practice Statement also says that the parties will be expected to start to consider potential trial dates as soon as is reasonably practicable after the service of the proceedings. Mr. Tappin QC says that in the present case Takeda did not do that.

The main objection taken by Roche to the hearing of the trial in June rather than in October is that it will lose the services of counsel that it has chosen to instruct. I recognise that it is important for parties to be able to have the counsel of their choice, particularly where they have already been instructed and had some involvement in the case, and therefore this is a factor that I will certainly take into account. On the other hand, certain counsel are very busy, and it is not of itself a reason to delay a trial which will otherwise be ready for hearing. Also, in the present case, there is no evidence about the extent of involvement of Roche's current counsel. The proceedings are at a relatively early stage, having been commenced in April 2018.

Roche says that Takeda have been guilty of delay in seeking to obtain the trial date. The basic chronology is that the case was issued by Takeda in April 2018. Roche filed a defence on 11th May 2018, the proceedings having been started as an action for revocation. On 25th May 2018, the parties attended the listing office to set a date for the CMC. Takeda requested a CMC to be heard in early July, but apparently gave no reason. Roche sought the CMC later in July, to allow it to consider infringement materials and consider whether to bring a counterclaim. At that stage Takeda did not indicate that it sought a trial in June 2019. The first time the trial date was raised was on 13th June 2018 by counsel's clerk when Takeda sought a listing in the week beginning 17th June 2019.

I do not consider that it is fair to say that Takeda have delayed or been deleterious in seeking to obtain a trial date. I do think, however, that Takeda could have applied for a trial date shortly after issuing proceedings. It did not have to wait for the CMC to do so. On the other hand, had Takeda applied for a trial date earlier, the arguments before the judge would be very similar to the ones that I have heard today.

By contrast, Takeda accuse Roche of deliberately seeking to delay proceedings. I think there is very little evidence of that beyond the fact that they wish to have the trial heard in October rather than in June, for the understandable reason that otherwise they may lose their counsel.

I now consider whether it is in fact practical that the UK court will be able to deliver judgment in order for the German court to take account of it. The evidence is that the UK Patents Court is fairly busy during June and July, but none the less, a listing is available on 4th June, probably in a five-day window.

In my view, it is important to give Takeda at least the opportunity of obtaining a judgment from the UK court, which may have some influence on the Düsseldorf court hearing the infringement action. By a decision of the Bundesgerichtshof, dated 15th April 2010, Xa ZB 10/09, Roll-Forming Machine, the Federal Supreme Court held that:

The UK courts are always very interested to see decisions of our German colleagues and judges of other EPC Contracting States pertaining in particular to equivalent patents. If I were hearing an infringement case in the UK, I would be very interested to see what decision the German courts had reached.

I give no guarantee that the UK court will be able to render judgment before the German court reaches its decision. Indeed, it may well not be possible to do so in what appears to be quite a complicated case. It may be, given the state of the court's diary, that the parties in this case will be given a deputy judge. That is a matter for listing. Furthermore, it may be that a judge will consider this is an appropriate case in which to give a result, plus relatively brief reasons for the result, and to give a full judgment later.

I have taken account of the fact that this is not a case where Takeda needs to seek expedition and thereby push other litigants further down the queue. If the case is heard in June 2019 that will be in accordance with the endeavour of the Patents Court in accordance with the Practice Statement.

Therefore, for all these reasons, I have decided, on balance, and with some hesitation, to order that this case should be fixed, as is suggested by Takeda in this draft order, on a date after 1st June 2019.

I am now going to deal with an application for specific disclosure in relation to the plea of insufficiency and the alleged squeeze between insufficiency and non-infringement and/or obviousness as pleaded in the grounds of invalidity. This requires me to recite a little of the background to the alleged invention.

Antibodies typically contain a series of carbohydrate structures, also known as "sugar chains" or "glycans" that are attached to the antibody's constant region. The profile of glycans for a particular antibody is known as the antibody's "glycan structure" or "glycosylation pattern".

The patent in suit is concerned with glycans attached to a nitrogen atom of the amino acid asparagine ("Asn") at position 297 of the constant CH2 region. Because of the binding to a nitrogen atom, this form of glycosylation is also called "N-glycosylation".

Claim 1 is to human monoclonal antibodies of the IgG1 or IgG3 type, wherein the carbohydrate (glycan) structure attached to the antibody’s Fc region is highly fucosylated (by "at least 99%"), whilst it is almost free of other particular molecules which are present in the amount of 1% or less.

At paragraph 4 of the grounds of invalidity, Takeda have pleaded a number of insufficiency objections which are far too lengthy for me to recite. However, I shall merely refer to paragraphs 4(a) and (b), which states:

"4(a) If, which is denied, the citations pleaded in the particulars to paragraph 1 above do not anticipate the claims, then, in the alternative, the Claimant will say that the claims are so unclear that the skilled person is unable to tell what falls within and without the scope of the claims such that the specification of the Patent is insufficient.

"(b)

The claims of the Patent (other than claim 5) require the amounts of fucose, NGNA and/or N-terminal alpha 1,3 galactose within the sugar chain of the relevant antibody to be “analyzed by Liquid Chromatography/Mass Spectrometry (LCMS) peptide map analysis”. Such peptide map analysis using LCMS requires selection of (1) appropriate apparatus, (2) appropriate experimental technique(s), and (3) appropriate experimental parameters. None of these variables are specified in the specification or the claims of the Patent, yet the choice of each will affect the LCMS analysis inter alia in that different apparatus, different techniques, and different experimental parameters will result in different amounts of fucose/NGNA/N-terminal alpha 1,3 galactose being reported in any given sample. In these regards, the Claimant will rely, inter alia, on an article by Reusch et. al entitled “Comparison of methods for the analysis of therapeutic immunoglobulin G Fc-glycosylation profiles-Part 2: Mass spectrometric methods” published in MAbs. 2015;7(4):732-42."

The case for disclosure is summarised in Mr. Jenkins's first witness statement at paragraphs 20-45.

Turning to the order sought, it is agreed there should be disclosure, the dispute being about the correct interpretation of the order or, to put it another way, what order I should make which will include, if appropriate, that which Takeda is seeking. The draft order provides:

"On or before 19 September 2018, each party shall make and serve on the other a list of documents relating to:

(a)

any experiments, and data derived therefrom, as to the glycosylation profile of the deposited cell line referred to in paragraphs [0014] and [0071] of the Patent and any experiments and data (including any operating procedures or other descriptions (including calibration) of the LCMS peptide map analyses) underlying Tables 2, 3a and 3b of the Patent."

Those paragraphs and that table expressly refer to what has been described as sub-clones 1-8. It is agreed that disclosure of the experiments underlying the glycosylation profile, etc., of sub-clones 1-8 should be disclosed.

The issue, however, arises from the fact that in the European Patent Office, Roche relied upon an expert declaration which revealed the fact that, in addition to the eight sub-clones specifically mentioned in the patent, work was done on another six sub-clones which I shall call sub-clones 9-14, which the patentee classified as being outside the scope of the patent. Mr. Tappin contends that given that they are classified as outside the scope of the patent, they are not relevant to the issues to be decided in those proceedings.

I should make it clear that I am always cautious about overloading patent cases with expensive disclosure, which may be of peripheral, if any, relevance. However, in the present case, there is no evidence that the disclosure sought would be onerous for the patentee to obtain or particularly costly. Indeed, Mr. Waugh submitted, and Mr. Tappin did not dispute, that given that experiments and data concerning sub-clones 1-8 is to be disclosed, the exercise of extracting sub-clones 1-8 and omitting the work on sub-clones 9-14 might be more costly than simply giving all 14 together.

I think there is force in that, and I also bear in mind, given the size of this case, disclosure would be proportionate.

As to relevance, it seems to me that at least a part of Takeda's case is that classification as to what falls inside and what falls outside the claim is extremely difficult. That may or may not be correct, but it does at least prima facie mean that the way in which the patentee classified sub-clones as falling inside and outside the claim may be of some relevance.

Furthermore, if, for example, the patentee classified a sub-clone as outside the claim because the relevant percentage was 98.5% and Takeda were able to prove that approximately the same percentage exists in the alleged infringement, that may also assist in relation to a case of non-infringement.

Of course, I have no idea whether this will in fact turn out to be the case, but I do consider that in all the circumstances, disclosure is appropriate, not just of sub-clones 1-8 but of sub-clones 9-14 as well, and the order should expressly so provide.

In relation to the second category of documents of which Takeda seeks disclosure, the category defined as documents relating to the public availability and glycosylation profile of Simulect before the priority date, Simulect is a product that was sold by Novartis before the priority date. Takeda's case is that Simulect fell within the scope of the claims of EP 809, the patent in suit, and therefore its public prior sale amounts to anticipation.

That case of anticipation was advanced by Novartis during what was apparently the latter stages of an appeal in the European Patent Office. It appears that after Novartis had filed their evidence and submissions, some kind of agreement was reached between Novartis and Roche, because Novartis did not appear at the Board of Appeal hearing and withdrew its opposition.

Takeda's case is that it might be the case that Roche themselves have investigated the glycosylation profile of Simulect and/or in the course of any dealings with Novartis they would have been provided with documents relating to the analysis of the glycosylation pattern of Simulect perhaps carried out by Novartis. In both cases it is said that such documents would be relevant to the issue of whether Simulect anticipates EP 809.

As to the second proposition, that Novartis may have provided documents relating to the analysis of the glycosylation pattern of Simulect to Takeda, which they chose not to put into the EPO, I find this to be speculative in the extreme. One might imagine that in the course of the opposition, Novartis put their best case forward in the EPO. Therefore, I think it extremely unlikely that any such documents exist.

It is, of course, possible that Roche themselves investigated the glycosylation pattern of Simulect, again in evidence that they chose not to put into the EPO, which took the case further than the evidence advanced by Novartis, but I have say I am doubtful. In any event, whatever doubts I may have about that appear to me to be settled by paragraphs 52-54 of the witness statement of Mr. Gilbert, who says:

"52.

As to paragraph (b) of the proposed order, in its disclosure report dated 12 July 2018, Roche confirmed that 'all documents [relating to the Simulect Request] in Roche's control have been disclosed in the EPO proceedings, which Takeda are therefore already in possession of'. In preparation of Roche's disclosure report MCS made various enquiries in accordance with Roche's disclosure obligations under the CPR. The result of those enquiries is as set out in the disclosure report (as quoted above), which is signed with a statement of truth by Graham Burnett-Hall, one of my partners who is also working on this matter.

"53.

For the avoidance of doubt and in response to Jenkins 1, MCS has made further enquiries and it has been confirmed by the relevant individuals at Roche that all documents which are referred to in entry 4 of Roche's disclosure report i.e. those that fall within the Simulect Request, were obtained by Roche in the course of the EPO proceedings concerning the Patent and therefore are available from the EPO on the public file and (it appears from Jenkins 1) are already in the possession of Takeda UK.

"54.

It follows that Roche's view is that an order for it to disclose documents relating to the Simulect Request is pointless as the position has already been set out (in the disclosure report and above)."

It seems to me that in the light of this, Roche have already made investigations into whether any documents falling in this category, which were not available on the EPO file, are in their possession, and the answer is that they are not.

In those circumstances, I think it would be unnecessarily costly and not required in order fairly to decide this case for any further order for disclosure to be made, and I decline to order disclosure within this category.

In relation to the dispute about whether the notice of experiments should be served on 17th October 2018 or 4th December 2018, it seems to me that given the fact that disclosure which may affect the need for, or design of, experiments is to take place on 19th September 2018, both sides are likely to require their chosen experts to consider the experimental protocol and it will be required also to consider the PPD, I think 4th December is the appropriate date to choose.

However I should make it clear that I, and I suspect any of the other Patents judges, will not be very sympathetic if this is then used as any kind of excuse to try to postpone the trial date. None the less, I think it is important that each side should have a full and proper opportunity to design their experimental protocols and I think the time that I have given will allow that.

In relation to the final contested issue between the parties on this case management conference, Roche seeks an order that Takeda shall on or before 3rd August 2018 identify those parties receiving profit from the sale and/or marketing of the Entyvio product in the UK. I should say at the outset that that order is somewhat too wide. What is in fact sought is disclosure of other companies within the Takeda group, if any, that are receiving profits from the sale of Entyvio in the UK. This follows from a chain of correspondence between the parties' solicitors whereby Roche sought information from Takeda concerning those entities involved in the commission of acts of infringement in the UK and also those parties who are group companies of Takeda and are receiving profits arising out of infringing sales in the United Kingdom. I refer in particular to letters from Marks & Clerk solicitors dated 8th May 2018, 18th May 2018, 4th June 2018 and 29th June 2018.

Takeda, which is responsible for the sales and marketing in the UK, by which I mean the claimant, has indicated that it will undertake to be liable for all infringing acts committed in the UK by any Takeda group entity. However, it has declined to reveal the identity of the entities, if any, involved in receiving profit from the sale of Entyvio in the UK on the basis that it is not relevant to the issue of liability.

Mr. Tappin submits that such companies are arguably involved in common design with the entities committing the acts of infringement, and thus are also liable to account for those profits. He postulates the case where, for example, the parent company is receiving the bulk of the profits from UK sales that its UK subsidiary is carrying out and suggests that a party cannot simply by clever accounting practices attribute profits from the sale in the UK to companies that do not do acts in the UK to avoid liability for those profits. Takeda is not willing to admit common design with any companies receiving the profits or to undertake to account for profits made on sales in the UK as if those profits were still made by it.

Mr. Waugh claimed that an order of this nature would set an unfortunate and unprecedented example, and indeed that the jurisdiction for making such an order had not been identified. I disagree. It has long been the case, even before the Norwich Pharmacal decision, that the court has jurisdiction to order disclosure from a potential wrongdoer of the names of other potential wrongdoers. Nor do I think that any great precedent is being set by this decision. It seems to me that there is no question of me deciding at this stage that any other company in the Takeda group is arguably involved in a common design. For that, I would need to see a pleading, and if it is objected to, of course, Takeda are entitled to apply to strike out once it is served.

However, it is the case that, on occasion, parties who have been found liable for infringement raised by way of defence to an account of profits that the relevant company to whom profits have been transferred has not been joined in the proceedings. In the present case, the claimant, Takeda UK, has not been prepared to disclose whether any transfer pricing has taken place nor the identities of any Takeda companies to whom profits have been transferred.

In those circumstances, I consider it is appropriate to make the order without any decision as to whether or not any such companies are in fact in a common design or as to whether any other case against them can be pleaded.