Teva UK Ltd & Anor v ICOS Corporation & Anor

This is a claim concerning a drug called tadalafil. Tadalafil has a marketing authorisation for the treatment of male erectile dysfunction and benign prostatic hyperplasia and for those indications is sold under the brand name Cialis. It also has a marketing authorisation for pulmonary arterial hypertension and for that indication is sold under the brand name Adcirca.

There is an SPC on the compound based on a patent whose validity is not being challenged in these proceedings and which expires in November 2017.

The trial is due to start in June, in about two or three weeks’ time, and this is the pre-trial review. Three patents are in issue at that trial and it consists of parallel but distinct actions between the defendant (ICOS) and third party (Lilly) on one side and the claimants and various generic pharmaceutical companies on the other. ICOS and Lilly will be referred to in this judgment as Lilly. The generic companies are Teva, Actavis, Generics UK (Mylan) and Actelion. The precise positions of the generics differ in detail.

This application relates to the dispute with Teva and the issue arises in relation to EP (UK) 1,200,090 entitled “Pharmaceutical formulation comprising a beta-carboline and its use for treating sexual dysfunction”. The validity of that patent is not in issue. The only issue is infringement.

Lilly seeks an order requiring Teva to provide samples of its tadalafil product so that the samples can be tested. Samples of […] are also sought but the issue really focuses on […]. If I make such an order it will be inevitable that the 090 patent cannot be dealt with at the trial. If I require samples and therefore necessarily adjourn part of the trial, Mr. Abrahams submits that I should direct that the point of construction of the 090 patent should be dealt with at the trial in any event. I will come back to that.

Each side blames the other for this predicament. Lilly says that the Product and Process Description (PPD) served by Teva was effectively conclusory in the sense that it was based on Teva’s approach to claim construction. Lilly submits that Teva did not provide the information necessary to decide infringement on Lilly’s approach to claim construction.

[Teva contends a relevant characteristic to determine infringement is feature A ] On that basis its PPD explains what [feature A is.] […] Teva submits that as a result it does not infringe. Lilly says that the relevant [characteristic is feature B], the PPD does not state what that is and so, until samples are analysed, one cannot tell.

Teva says that Lilly has known about Teva’s position all along and has known about the nature of the information which was actually available and is in the PPD for months, and Lilly only has itself to blame. Teva also submits strongly that the correct construction of the claim is [that the relevant characteristic is feature A ].

Moreover, Teva submits that even on Lilly’s construction (which it says is wrong) [Teva’s product does not] infringe this claim anyway. Really this is all a delaying tactic to try to put off the trial and interfere with Teva’s proper attempts to clear the way in good time for the expiry of the SPC in November 2017. The timing of the trial date has taken into account not only a trial at first instance but also the possibility of appeals. So I should refuse this application.

First, I will deal with the patent. The patent is concerned with small particles for the drug. […]

The relevant parts of claim 1 are as follows:

Teva’s point is that […] is a clear indication that, although literally one may read the claim as if it is talking about [feature B], in reality what it is talking about is [feature A]. […] Therefore on the facts of that case there was no infringement. […]

In my judgment, Teva has a good arguable case on construction in general and that there are factors of the kind which applied in […] which could well lead to the same conclusion.

However, in my judgment, Lilly also has a good arguable case that the claim could be construed to refer to [feature B]. […] This may gain some support from the evidence of the two experts available for this application. They are Prof. Douromis for Teva and Prof. Frijlink for Lilly. […]

I will turn next to the evidence about the product in issue and infringement. Prof. Douromis expressed the clear view that the claim needs to be read as referring to [feature A] and on that basis there is no possible claim for infringement. I can see the force in the Professor’s opinion in that clearly, if Teva’s case on construction is right, then the conclusion follows. However the debate is about the premise.

As regards Lilly’s construction Prof. Douromis says that it is very difficult to [determine infringement if the claim refers to feature B] and he also says that given […], it is unlikely that [feature B would be satisfied anyway]. In this respect also there is a measure of agreement between him and Prof. Frijlink. The latter also agrees that it is difficult to [determine infringement if the claim refers to feature B] but expressed a view that it can be done and refers to a number of techniques which he would like to develop in order to achieve that result. These issues are ones for debate at trial and Teva will seek to rely on them to support its construction. However, as I have already found, both sides’ constructions are at least properly arguable.

As for […], Prof. Frijlink says that while he agrees that it is not likely, he maintains that it is possible that [there will be infringement if the claim refers to feature B]. Prof. Frijlink’s opinion is that one cannot form a concluded view about [infringement if the claim refers to feature B] based on the information in the PPD and he wishes to test samples.

Now, as I think I may have already said, each side blames their opponent for the predicament that we have reached, which is that on the basis of Lilly’s case on construction, it is plain that the matter cannot be decided at the trial in three weeks’ time.

Before I get into the correspondence and how we have arrived at where we are, there are two matters which I wish to address.

First, ultimately, the overriding objective (CPR r1.1) is what counts. That is to deal with cases justly and at proportionate cost. Two particular points which are important in that regard:

This is clearly a case of some commercial significance. From what I have learned from reading the materials and having heard the rest of the pre-trial review before this application (which involved more parties than the two parties to this application) this is a dispute about a pharmaceutical market which obviously is of significant value.

Second the compound SPC does not expire until November 2017. There is still time, even if this action were to be adjourned for a period, to ensure that a first instance trial could be heard in good time to allow Teva to make its decisions, if it wins the case, well in advance of November 2017. Mr. Abrahams referred to the possibility of a trial in autumn term of this year and it seems to me that if this matter is to be adjourned, it should certainly not be adjourned any later than that. I have particularly in mind the current pressures being felt in the Court of Appeal in relation to hear-by dates. That is one reason why this trial has already been scheduled to be heard in June 2016.

Second, I need to refer to the principles that relate to product and process descriptions.

Mr. Hinchliffe, who appeared for Lilly, referred to the judgment of Arnold J in Starsight v Virgin Media, [2014] EWHC 828 (Pat) emphasising that the requirement of the rules are that full particulars are required to enable all issues of infringement to be resolved, with an emphasis on the word “all”. I agree with Arnold J, indeed I think I said as much in Vringo v ZTE[2015] EWHC 818 (Pat). It is an important point that the Product and Process Description is not there simply to resolve the issues on one side’s construction. The PPD needs to have enough information in it to resolve the issues on both sides’ constructions of a patent claim. A similar point, although it was the other way round, can be seen in RIM v Visto (Floyd J) in [2008] EWHC 335.

I also can refer to the current Patents Court Guide and, in particular, to paragraphs 10.4 and 10.5. The point in 10.4 is simply to emphasise that material omissions or inaccuracies in a PPD can lead to costly adjournments with consequential adverse orders, including as to costs. Paragraph 10.5 explains that in so far as a party is not able to verify that the PPD is a true and complete description of all relevant aspects of its product or process then the correct course is for the party to verify the parts that he is able and serve a disclosure list which may or may not contain documents in relation to the remainder. This paragraph of the Guide therefore illustrates the point that at its heart the exercise of producing a PPD is one of disclosure, even though the PPD is produced in lieu of the actual disclosure of documents.

Just on that score, as I will explain below, part of what has happened is that Teva has explained recently that it does not know [what feature B is]. Lilly submitted that that meant that Teva was already in breach of the order for directions for failing to carry out a disclosure exercise of the kind envisaged by paragraph 10.5 since it now knows that it is not able to verify a description which includes that information. However Mr. Abrahams submits his clients cannot be said to be in breach of the order that was made in these proceedings. I agree. The order was in paragraph 14 of the directions made on 14^th December 2015. It is in a fairly conventional form and consists of an order that in lieu of disclosure Teva will produce a PPD and a signed statement verifying the matters which are set out in paragraph 6.2 of Practice Direction PD 63. While I agree with Lilly that paragraph 10.5 of the Chancery Guide is relevant, I do not believe it could be fairly said that Teva is already in breach of that order for failing to spontaneously carry out a disclosure exercise after finding out that certain relevant information was not available. In a case governed by directions of the kind given here, when the party who produced a PPD then becomes aware that there are matters which cannot be described or verified, their obligation is to make that clear and inform their opponent promptly so that appropriate steps can be taken. An order for disclosure or samples may well follow but that will depend on the circumstances.

In this case Teva initially produced a PPD based on what it understands the proper construction of the claim to be. There was no reason why at that stage Teva should have had any inkling that Lilly might advance a different construction and so, given the way patent cases are pleaded, producing the PPD that way was an appropriate thing to do. However now we know Lilly’s case on construction, it is clear that the information in the PPD is not complete in that it does not describe [feature B]. What has emerged very recently is that Teva does not know what [feature B is] because […].

Does that mean that Lilly is blameless? I do not believe it does because it could have made its case on construction clearer at an earlier stage than it did. However the real question is: when did Teva first know what Lilly’s case was about [feature B]? If Lilly did not explain that to Teva, then Teva can hardly be criticised for not appreciating the information needed to be produced. However, once that point was apparent to Teva and on the footing that the information was not available, as Teva have now explained that it is not, the obvious thing to do in this case would be for Teva to offer samples.

I will turn to the correspondence. The first PPD was provided on or about 11^th January 2016. In response Simmons & Simmons, on 19^th January, wrote referring to a likelihood of requiring samples. However, it can fairly be said looking at that letter, that this was not focused on the idea that [feature B] was relevant. It was focused on a different question, which was to verify that what was said in the PPD was true.

Lilly submit that it was not as clear as it might have been that the […] data in the PPD was referring to [feature A]. I accept that up to a point and the issue did need to be nailed down however, as Mr. Abrahams has pointed out, actually, given the details which were in the PPD, it was pretty obvious that [the data in the PPD was referring tofeature A] rather than [feature B].

More information was sought and a further version of the PPD was provided in February. As Mr. Abrahams has pointed out, that made it clear at paragraph 7 that [the data in the PPD was referring tofeature A]. At that stage, therefore, it must have been clear to Lilly that Teva’s position was as I have explained it to be. The critical question, in my judgment, is what happened after that.

The important matter is a letter of 26^th February 2016, which I was shown by counsel. What matters is the second page — page 32 of the bundle — in which Lily’s solicitors respond to paragraph 7 of the PPD, and state as follows:

At that stage, Teva knew, or should have known, that there was an issue. It is clear from this passage that the possible need for samples was linked to the absence of [information about feature B]. We know now that, in fact, Teva did not know […], but that information did not emerge until 6^th May.

It was suggested in the evidence from Teva’s solicitor that it must have been implicit and, therefore, obvious to Lilly that Teva did not have the information about [feature B]. I do not agree. There is nothing in the materials that I have been shown which date from before 6^th May that would have indicated or should have indicated to Lilly that Teva in fact did not have that information.

I should also mention a significant point addressed in Ms Weekes statement. She is a solicitor acting for Teva. She explained that the lateness of the explanation that Teva did not have the information was erroneous and that it was something that was intended to be explained in earlier correspondence. I accept that it was a slip but ultimately the stance taken by Teva in correspondence after 26^th February, refusing to agree to samples but not explaining that it did not have the relevant information either, made the problem that now arises inevitable.

Lilly’s position after 26^th February made sense. It was that Lilly wished to be provided with information in the PPD about [feature B]. If that information had been provided, then the matter would have rested there and there may well have been no need for a sample. It was only once it became clear that Teva was not in a position to provide that information that it became necessary to produce a sample. Providing a sample for testing is the only way, fairly, to resolve this question of fact.

Mr Abrahams submits that one can see from Lilly’s evidence that it will be necessary for Lilly and its expert to develop a new method [in order to address feature B] and, therefore, that has a bearing on the issues that I have to decide. He refers to paragraph 58 of Ms Noor’s witness statement, which refers to the evidence of Prof Frijlink that an effective methodology […] has to be developed and will take three months. Mr Abrahams is right about the evidence but I do not agree that it has a significant bearing in the issues. A […] method does have to be developed and the fact that this is so may be relevant to the issue of claim construction. However that is all a matter that will have to be resolved at trial. As long as that work can be accommodated in a fair timetable, it has no other significance at this stage.

Ultimately, standing back and thinking about all of this, I do not strongly criticise Teva for the predicament we are now in since the matter was ultimately made clear in correspondence, as it should have been. However time was short and it should have happened earlier. I am not persuaded that Lilly should be deprived of the ability to analyse a sample simply because of the impending trial date. What should have happened is that shortly after 26^th February, Teva should have explained to Lilly that they did not know the answer to the question that was being asked. Teva should not simply have refused to answer the question. Then the problem would have emerged earlier and if it had this matter might still have been tried in June. But that did not happen. I will therefore direct that samples to be provided, and to that extent this matter has to be adjourned.

The outstanding question I have to decide is whether to go ahead anyway and have a hearing in June as part of the trial in relation to the 090 patent and to decide the point of construction. Mr Abrahams urges me to do that. The difficulty I have is that the only reason that the 090 patent would have to be involved in the June trial at all is this infringement issue. As Mr Hinchliffe submits, when the testing is done there may be no infringement case at all. In that case the 090 patent would not have to trouble the court at trial.

Equally well, I bear in mind Teva’s understandable desire to have this matter resolved at first instance and on appeal if it possibly can be before the expiry of the SPC. If decision on construction of the 090 patent did go in Teva’s favour then to decide the 090 point on construction first and then do the testing afterwards may lead the whole thing to get in a mess. There could be two trials, one on construction and another on the facts following some months afterwards, leading to more uncertainty and cost.

In my judgment, the fair and appropriate thing to do is to direct that the samples be provided, to ensure that the trial, if there is to be a trial, of the infringement issues takes place in the autumn term — that is to say, between 1^st October and Christmas of this year. I am quite satisfied that that will be possible. That will allow all the matters to be resolved in an expeditious way and will still give Teva plenty of time to have a proper degree of commercial certainty. That is what I will do.