Merck Sharp Dohme Corp & Anor v Teva Pharma BV & Anor

In this action Dr Justin Turner QC appears for the claimants instructed by Hogan Lovells and Mr Simon Thorley QC leading Mr Mark Chacksfield instructed by Winston & Strawn appears for the defendants.

This is a claim for infringement of European Patent (UK) No. 0,582,455 which is in force until 3^rd August 2013 and SPC/GB00/035 which is in force until 19 November 2013. The first claimant is the patentee and proprietor of the SPC and the second claimant is the exclusive licensee. I will refer to the claimants as BMS (singular). The patent and SPC relate to efavirenz, a non-nucleoside reverse transcriptase inhibitor. Efavirenz inhibits viral replication and is used in the control of HIV infections. It is used both as a sole agent (sold under the trade mark SUSTIVA) and as part of a co-formulation with emtricitabine and tenofovir (sold under the mark ATRIPLA). UK sales of SUSTIVA in the twelve months to December 2012 were $22 million. Efavirenz is an important product for BMS.

The defendants can be referred to simply as Teva (singular). On 20^th October 2011 the Committee for Medicinal Products for Human Use (CHMP) issued a positive opinion on the grant of a marketing authorisation for Teva’s generic efavirenz product to be sold under the name Efavirenz Teva. On 9^th January 2012 the Commission granted that marketing authorisation to Teva.

The issuance of the CHMP opinion is a matter of public record. BMS believed that there was a risk Teva was going to launch its efavirenz product in the UK before the SPC expired and its solicitors wrote to Teva on 1^st December 2011. Teva refused to say what its intentions were for efavirenz, explaining that its policy was that its plans were confidential and that it was not obliged to tell BMS (or the court) what its intentions were. BMS interpreted Teva’s stance, in the overall context, as evidence of a threat and intention by Teva to launch before the SPC expired. BMS started these proceedings on 9^th February 2012. BMS applied for an interim injunction. Teva applied to strike the action out on the basis that it disclosed no arguable cause of action. I heard both applications together on 8^th March 2012. I dismissed the application to strike out and granted an interim injunction in a judgment handed down on 15^th March 2012.

The action proceeded and this is the trial. Teva accepts, for the purpose of this action in this jurisdiction alone, that the patent and SPC covers Efavirenz Teva. There are no technical issues arising in relation to the patent. Indeed the patent was not opened at trial. Teva had the opportunity in these proceedings to challenge the validity of the patent and/or the SPC but did not do so. One might therefore ask what the action is actually about. The answer is as follows.

In summary BMS contends that at the point the action began Teva was threatening and intending to infringe the patent (i.e. launch before 19 November 2013). Teva contends it was not. Teva contends that it made the position clear on 23^rd April 2012, in its response to a Part 18 Request from BMS, which states:

“The defendants have not at any material time had any plans to supply pharmaceutical products containing efavirenz for disposal in the United Kingdom before 20November 2013.”

Teva says that at the relevant date (9th February 2012) as a matter of fact it did not have any intention to launch efavirenz before 20 November 2013. Thus the action must fail. Mr Thorley submits that while there may (or may not) be costs consequences arising from the fact that Teva did not make this position clear until after the claim began and after the Defence was served, the action must be dismissed because Teva simply did not intend to infringe. He put the matter this way at the start of his closing submissions (Day2 p53 ln14):

“The issue -- and it is a very small one between us, but it is quite an important one because one needs to know for the future where one is going -- is if we were to write in response to a letter before action what we have now put in our Part 18 response, at no material time have we an intention to launch before the expiry date, whether thereafter a party can start proceedings for quia timet relief and obtain the sanction of the court behind that to say, "You have got to give either a contractual undertaking or be subject to an order of the court that you will not do that" -- obviously the position we adopt on this side of the court is once we have said we are not going to do something, thereafter of course somebody can start proceedings saying, "We do not believe you." If necessary, it can come to trial, as this one has, and the court will then have to assess as to whether you were right or wrong in saying, "I do not intend."

Obviously what has happened in this case is we did not come up with that letter at the outset, for reasons that you have heard of, and we accept that if we are correct that they have failed to prove that there is an actionable threat, none the less the court has options with regard to costs for the fact that we did not put our cards on the table earlier on. The fact that we did not does not mean that we lose the action.

What one has here is really a very simple question of fact that your Lordship has to decide as to whether at the date this action started -- or possibly at the date that the interim injunction was granted, and in some cases at the date of trial, but obviously not here, because we have been injuncted -- in fact there was a threat. The fact that they perceived that there was a threat is neither here nor there.”

BMS agrees that the point is important but does not agree that there was no proper basis to bring these proceedings. It submits the action should succeed and an injunction to be granted against Teva until the expiry of the SPC.

There was no argument that if a threat at the date of issue was established, which Teva denies, then in this case a final injunction until the expiry of the SPC would follow. Thus the only question I have to decide concerns the position at the date the action started.

BMS provided witness statements from Charles Caruso, Executive Director, Legal at the first claimant; Amadou Diarra, European Vice President and General Manager for UK and Ireland for the second claimant; and Stephen Bennett, a solicitor and partner at Hogan Lovells. The point of this evidence was to explain why by early 2012 BMS were worried that Teva were about to launch efavirenz. Mr Diarra summarised the reasons for this as follows:

Two points made by Mr Diarra require some explanation at this stage.

First the advantage to the first generic mover. It is not in dispute that, as a generalisation, if a generic drug company is able to start selling the generic version of a successful branded product before any other generic is available, large profits can be made in a short time. Before any generic is on the market for a given drug, the market is a monopoly and the level of the price for the branded drug being sold by the originator company is relatively high. When the first generic starts selling, this first generic does not have to offer a large discount over the existing branded price in order to take market share. While the market is a duopoly between the branded product and the generic product, the generic company can expect to be able to sell at a price which is in effect a modest discount off the branded price. This is much higher than the generic’s cost of sales. When more generic companies enter the market, in order to compete they need to offer discounts against each other and the price falls rapidly to a level governed by the generic companies’ costs of sales (plus a profit margin). That is much lower than the branded price. Any given real market may not operate in this way for any number of reasons but as a general proposition, both sides accept this “first generic mover” advantage is a well understood idea.

Second, atorvastatin. In 2010 Teva engaged in correspondence with the patent holder (Warner-Lambert) stating that they intended to launch generic atorvastatin in November 2011, which was after the expiry of the relevant SPC, but then surreptitiously launched the product on a very large scale in June 2011, before expiry, and without any notice to the patentee. The matter came before Floyd J on 20^th June 2011, the day Warner-Lambert discovered that Teva’s atorvastatin was on sale. Teva’s counsel appeared on the application at very short notice. Teva’s position was that they believed the patent was invalid. In paragraph 6 of his judgment (Warner Lambert v Teva [2011] EWHC 1691 (Pat)) Floyd J noted the evidence that one million packets had been sold in that morning alone. In paragraph 12 Floyd J held that the case was an appropriate one for granting relief without notice and said:

“The position until today appears to have been that both the claimants and Teva were proceeding on the basis that there would be no launch of this product until November 2011. Whatever the position may be about the obligation on a defendant to clear the way, there is nothing to stop the defendants from at least informing the claimants that they intended to launch at an earlier date. Instead of taking that course, they have, as Mr Thorley characterises it, “launched surreptitiously” and without any form of notice. What is more, I am unable to see at the moment why it is that as long ago as 2010 the Defendants did not take steps at least to let their contentions about the validity of the patent be known to the Claimants. They had nothing to hide, so far as their intention to market a product was concerned, for their interest in the market for atorvastatin has been shared with the Claimants. I fail to see why it is that they, therefore, did not start those proceedings in at least 2010 if it was really their contention that this patent is invalid.”

So BMS fear that the same thing may happen again in this case. Teva will launch at risk of infringing the patent or SPC and without prior notice to BMS.

Teva did not cross-examine any of BMS’s witnesses.

BMS also served a Civil Evidence Act notice in relation to various internal Teva emails. BMS submits that these emails indicate that in the period between October 2011 and these proceedings being commenced, once Teva knew it had obtained a positive CHMP opinion and would therefore have a marketing authorisation which made a launch of efavirenz “first and alone” prior to November 2013 feasible from a regulatory standpoint, senior staff in Teva started a review to give active consideration to such a launch. The term “first and alone” is an expression used in internal Teva materials to refer to launching as the first and only generic in a market. In other words to be the first generic mover. BMS contends that the evidence from Teva does not establish what the outcome of that active consideration was. It may simply have been overtaken by events. BMS argues that there is no evidence Teva ever rejected a strategy of launching efavirenz at risk nor is there evidence it would not adopt one now if it were to be available.

Teva called evidence from Emma McDonagh, presently the EU Product Portfolio Manager with the Project Initiatives Group of Teva; Natasha Bloch, a solicitor working in Teva’s European IP Litigation team; Jonathan Rousell, Head of European Regulatory Affairs, Generic Registrations for Teva Pharmaceuticals Europe; Charles Robinson, Pipeline Product Manager for Teva UK Ltd All the Teva witnesses were cross-examined.

Each of Ms McDonagh, Ms Bloch, Mr Rousell and Mr Robinson were good witnesses who gave their evidence fairly and to the best of their ability trying to help the court. A point was taken about Ms McDonagh. It related to her readiness to accept a point about a reference to first and alone in some internal emails. In my judgment it did not undermine her evidence as a whole.

The point of the evidence of Teva’s witnesses was to establish Teva’s case that it never had an intention to launch efavirenz in the United Kingdom before 20November 2013. The witnesses explained how Teva’s internal processes worked in relation to efavirenz.

Although not comparable to the work the originator has to do to find and develop a safe and effective product in the first place, in order to be in a position to launch a generic drug, a considerable amount of work has to be done over a lengthy period of time. The company has to obtain or make the active pharmaceutical ingredient (API) and develop a suitable formulation for it, possibly by reverse engineering the originator’s formulation. Once a potentially viable formulation has been produced, bioequivalence studies must be carried out to compare the product with the reference product and stability studies must be done. Once the product has passed the bioequivalence and stability tests, the company can be comfortable that it has enough data to obtain regulatory approval. [……….] For generic approval, the authorisation depends on the existence of the originator company’s marketing authorisation but the rules on data exclusivity mean that a generic authorisation cannot be applied for until a certain date has passed. In order to file for generic approval, the formal application has to be prepared and filed and then the regulator will consider the application. [……….] She also estimated that the relevant Centralised Procedure for obtaining marketing authorisation can take around 12 months (Mr Rousell estimated 15 months on average). Ms McDonagh and Mr Rousell explained that there may be delays in that process. Once the marketing authorisation is obtained there may be further delays because the marketing authorisation needs to be amended. Also in some countries in Europe further time is needed for pricing and reimbursement approval. Although pricing and reimbursement approval is not needed in the UK, since the marketing authorisation is obtained by the centralised European procedure, the timing takes this into account. Mr Rousell explained in cross-examination that the process of developing a generic product and obtaining approval is a significant investment, of the order of hundreds of thousands to millions of pounds.

At the outset the company has to decide to develop a given product. [……….] Another key date in the timeline is the PLMD (Preliminary Limiting Marketing Date). Ms McDonagh explained that this is the date before which the product cannot be launched for legal (i.e. patent) reasons, i.e. normally the date the originator’s relevant compound patent or SPC expires.

For Efavirenz Ms McDonagh’s evidence was that the PLFD was on 25 May 2009, which was the date on which BMS’ data exclusivity expired. The PLMD as far as the UK was concerned was set for 19^th November 2013, i.e. the date the SPC expired. Teva noted a separate PLMD of August 2013 for some European countries, presumably ones with no SPC. A key part of Teva’s case is that the PLMD for the UK was never altered from 19^th November 2013. I accept that the PLMD did indeed never change but the significance of that is a matter addressed below.

[……….] She exhibited a document from 19 November 2009 (exhibit 3). This showed the PLMD of 19^th November 2013, an anticipated date for regulatory submission of August 2010 and an anticipated date for obtaining marketing approval of 21 October 2011. As Ms McDonagh points out this shows the earliest launch date is just over two years after the anticipated approval date.

The responsibilities of the various witnesses for efavirenz over time was as follows. Ms McDonagh’s responsibility as EU Product Portfolio manager for efavirenz lasted from February 2009 until passed to Dirk Schmidt in September 2010. Mr Schmidt did not give evidence but his responsibility only lasted for two months until December 2010 when the application for marketing approval was submitted. There is no reason to suspect any ulterior motive on Teva’s part for not calling Mr Schmidt. Once the application was filed responsibility passed to the UK representative, in this case Mr Robinson the Pipeline Portfolio manager. [……….] It is a little surprising that David Lawrence did not give evidence but Mr Robinson explained, and I accept, that he attended the monthly launch team meetings and occasionally chaired them. That is a fair explanation for not calling Mr Lawrence. Mr Robinson explained that the launch date shown in the minutes of these meetings from September 2011 to February 2013 remained at 20 November 2013. He exhibited all the minutes. The one for the April 2012 meeting includes a note “Bristol Myers Squibb have successfully PI us, therefore cannot launch before 20 November 2013”. This reflects the grant of the interim injunction in this case. Finally, Ms Bloch had responsibility for the efavirenz in the European IP Litigation team from February 2011 to January 2012 when the project was handed over to Alia Knight.

The final witness I should mention is Mr Roussel. His evidence dealt in detail with the process of obtaining regulatory approval and explained the overall timing. In his opinion the regulatory approval timeline for efavirenz was relatively short but not extraordinary. He said it was not unusual to have a marketing authorisation in the UK in the 3-4 years before the intended launch date and he saw nothing unusual about the timing of the Efavirenz Teva marketing authorisation. I accept that Mr Roussel’s evidence shows that the fact Teva obtained a marketing authorisation for efavirenz nearly two years in advance of patent expiry does not of itself establish that Teva intends to launch before patent expiry. However his cross-examination bears on this issue. He accepted that he was familiar with the idea that part of Teva’s strategy is to launch at risk of patent infringement in some cases (e.g. atorvastatin). He also accepted that obtaining an early marketing authorisation could facilitate that strategy although in his opinion it was not the primary driver about having an early marketing authorisation. There was then a passage of evidence as follows:

7

Q. Now, Teva's strategy means, and so let me just remind you

8

about the strategy. The strategy I am putting to you is the

9

willingness and the opportunistic strategy of Teva launching

10

at risk. That is the strategy -- the at risk strategy, I will

11

call it. Now, Teva's approach, we will be saying to my Lord,

12

is given that strategy, that whenever they get a marketing

13

authorisation, there is a risk that they will launch before

14

rights expiry. Do you understand the proposition I am putting

15

to you?

16

A. I do, yes.

17

Q. Because they have this launch risk strategy, if they get an

18

early marketing authorisation, then there is a risk that they

19

will launch before the rights expire. Would you disagree with

20

that?

21

A. No, I would not disagree with that; there is a possibility.

Based on Mr Roussel’s evidence, I have no reason to infer that Teva’s motivation in commencing and scheduling its efavirenz project was driven by an intention to launch efavirenz before expiry of the BMS patent. Nevertheless it is also plain that if Teva does manage to obtain a marketing authorisation for a product sufficiently in advance of expiry to make a launch at risk feasible, that creates an opportunity which Teva well understands for it to launch at risk before expiry.

As I have mentioned already, the CHMP issued a positive opinion in relation to Teva’s application for marketing approval on 20^th October 2011.

When the positive opinion was obtained senior staff wanted to consider whether there was any chance of an earlier launch date anywhere in Europe. Massimo Berzigotti the Head of European Portfolio management sent an email to that effect on 21^st October. The email notes that efavirenz is a $144 million product (presumably a figure for Europe as a whole). The email shows that the barrier to early launch is regarded as the legal/patent position. I find that in the period after 21^st October 2011, Teva gave active consideration to the possibility of launching efavirenz at risk of patent infringement and it is clear that one of the five countries for which Teva gave this question active consideration was the UK. One email referred only to Spain but I reject the suggestion (if made) that Teva’s review did not include consideration of the UK. It is also clear, and I find, that none of the witnesses before me were told at any time that Teva had decided to launch efavirenz in the UK before 19^th November 2013. Moreover I accept that if such a decision had been made for efavirenz Ms Bloch, the solicitor in the European IP department who was at the relevant time responsible for efavirenz, would have been told shortly afterwards. She was not informed of any such decision. Subject to the point that a decision might not necessarily have been communicated to Ms Bloch immediately (which I accept) the fact that Ms Bloch had not been informed of such a decision is clear evidence that no such conclusion was reached within Teva in the period Ms Bloch was responsible for efavirenz which ended in January 2012.

The most likely explanation for the materials I have been shown and the evidence as a whole is that in October 2011 Teva started to give active consideration to the possibility of launching efavirenz in the UK before 19^th November 2013 but had not concluded that consideration by the time the relevant people inside Teva received the first letter from BMS’s solicitors. After that no final decision was taken, probably because the issue became bound up with the legal proceedings.

BMS submitted that another relevant factor was Teva’s responses to the letters from BMS’s solicitors and to the claim itself. A few points arose but the only important one was the assertion in correspondence that Teva regarded its launch plans as confidential and had a policy of not divulging them in these circumstances. BMS’s unchallenged evidence gave examples of Teva being prepared to tell solicitors acting for originator companies in similar circumstances what its plans were – in the sense of saying things like “We do not intend to launch before expiry of the patent and will give you notice if we change our mind.” That means that whatever Teva’s policy is or was, it has not been universal. The policy seems to have been formulated sometime between autumn 2009 and the end of 2011. I do not think it was created simply for this case or anything of that kind.

However I reject the idea that this information is of a character that a sophisticated patent litigant like Teva could credibly claim that divulging to the relevant originator (under suitable undertakings to prevent its wider dissemination) that Teva had an unequivocal plan to launch a product only after the SPC expired would cause Teva any damage at all. On the other hand a plan to launch efavirenz at risk before expiry of the SPC would be highly commercially sensitive and the last person Teva would wish to tell would be BMS, precisely because BMS would then be able to take steps to start an action and resolve the patent issue in advance of the launch. The true motivation for Teva’s policy was shown in the cross-examination of Ms Bloch. She accepted that the policy maximised Teva’s chance of being able to surreptitiously launch the product without being injuncted and that Teva stood to make money from such an at risk launch.

Finally BMS submitted that Teva were clearly prepared to launch at risk if they chose to do so, which I accept, and that Teva were able to launch at short notice if they wished, which I also accept.

There was a dispute in the evidence about whether the particular approach employed by Teva in the atorvastatin example could be applied to efavirenz because while atorvastatin was sold to retail pharmacies, efavirenz is sold into hospitals and is the subject of tenders. Mr Robinson gave some oral evidence in chief on this which BMS objected to. It did not seem to me that the objection was well founded given the questions Dr Turner had asked an earlier witness. However I am not satisfied that the particular nature of the market for efavirenz is such as to rule out Teva launching efavirenz at risk before the SPC expired. For one thing if it was so clear cut that efavirenz could not, as a practical matter, be the subject of an at risk launch, one might have expected that to be mentioned in the emails before the court. The staff would not have wasted their time actively considering the patent/legal side of an early launch if it was simply not feasible for efavirenz. Mr Robinson’s testimony was that it would be risky for Teva to enter into tender contracts. I can see that, although I am not convinced it is a risk Teva might not be prepared to run. Moreover I do not accept tendering is the only way to sell into this market. Product can be sold directly without a tender agreement. It would be difficult for Teva today to sell directly to the non-retail purchasers since Teva has no sales force directed at those purchasers. However I remind myself that the issue I have to decide concerns the position at February 2012. If Teva had decided in February 2012 to launch first and alone sometime before November 2013, it had nearly two years to make the necessary arrangements. In my judgment Teva could do that if it wanted to.

A point arose about the PLMD. I can deal with it as follows. In paragraph 35 of her statement Ms McDonagh states that “Once it became clear that the CHMP would be issuing a positive opinion on Teva’s MA application and once the positive opinion was issued, there were discussions within Teva about when the earliest launch date for this product was and whether there was any possibility to launch earlier than had previously been suggested.” She referred to emails on that topic and, in her second statement, addressed the internal Teva emails relied on by BMS in their Civil Evidence Act Notice which also bear on the same point. This is relevant because although, as Mr Thorley submitted, Ms McDonagh’s responsibility in relation to efavirenz related to the period from February 2009 to September 2010 when she was EU Product Portfolio manager for new launches, in her evidence she addressed a much wider period, commenting upon and giving evidence about many matters outside that timeframe. In particular her witness statement addressed in terms the period after the marketing authorisation was granted. In closing Mr Thorley submitted that [……….] Although this submission might seem to be at odds with the fact that Teva had put her forward as a witness giving evidence directed to this period, the forensic purpose of the argument was to support a submission that I should not place weight on certain evidence Ms McDonagh gave about the PLMD in cross-examination which related to the time after the marketing authorisation was granted. Ms McDonagh said in cross-examination that if someone decided to launch in the face of patent protection, they would not ask for the PLMD to be changed. If correct this would undermine Teva’s reliance on the invariant PLMD.

Mr Thorley’s submission is a fair reflection of the nature of Ms McDonagh’s job but I do not accept it means she was not qualified to give the evidence she had herself given, commenting on and explained the emails arising from after October 2011. In cross-examination she accepted she was not involved in discussions about launching a product in the face of patent protection and it was clear that there were limits to her knowledge about who took decisions after marketing approval had been given as well as on other matters. However the PLMD was something which was an important matter to Ms McDonagh in her work generally and was something she was able to give evidence about, including by reference to the period after marketing approval. In my judgment Ms McDonagh was not speculating when she said that the PLMD would not change if someone decided to launch in the face of patent protection.

On the same topic, Mr Roussel was asked about the PLMD. His evidence was that it could be moved and I understood him to mean that this included moving it forward in a launch at risk situation (day 2, Private, p88 line 4-7) but when asked if the PLMD was moved in the atorvastatin example he said “I do not believe it was. I do not know for sure.”

I find that Teva has no consistent policy about the PLMD in relation to a launch at risk of patent infringement. In a given case, if Teva had decided to launch a product before the originator’s patent has expired and therefore before what would otherwise have been the PLMD, the PLMD itself will not necessarily be changed to reflect the new earlier launch date.

Accordingly I reject the submission that I can rely on the fact the PLMD for efavirenz did not change as evidence that Teva had not decided to launch efavirenz at risk.

A claimant may ask for an injunction to be granted before a wrong is committed in order to stop it from happening. It is clear that the court has a jurisdiction to do this. The Latin tag quia timet (“because he fears”) is used which might indicate that the injunction is granted because of a fear that the wrong will be committed, but the famous words of Lord Dunedin in AG for Canada v Ritchie [1919] AC 999 at 1005 explain that “no one can obtain a quia timet order merely by saying ‘Timeo’; he must aver and prove that what is going on is calculated to infringe his rights.”

A number of cases were cited in argument. I will deal with the important cases in chronological order.

Paul v Southern [1964] RPC 118 is only really relevant for the graphic illustration of a cry of “I fear” in words of Harman LJ (at p123):

“It is exactly as if Mr Aldous in a sworn affidavit has said: ‘I am fearful that unless restrained by the Court Sir Lionel Heald will hit me on the head with a volume of the Patent Reports’. He then goes on and says ‘If he is not going to do it, it will not hurt him, so you may as well grant an injunction’. I have heard that argument before.”

Dr Turner relied in particular on Hooper v Rogers [1974] 1 Ch 43. That case was about neighbouring farmers. The defendant cut a deep track on his land in such a way as to expose it to soil erosion and risk undermining the plaintiff’s farmhouse and causing it to collapse. The plaintiff brought a successful action for damages in lieu of a mandatory injunction to compel the defendant to reinstate the land. The defendant appealed and the Court of Appeal (Russell, Stamp and Harman LJJ) dismissed the appeal. The judge found that there was a real probability of actual damage to the plaintiff unless the land was reinstated. The defendant contended the injury to the farmhouse was neither certain nor “imminent”. Russell LJ held that the word “imminent” simply meant that an injunction should not be granted prematurely. At p50 C-D Russell LJ said this:

“In truth it seems to me that the degree of probability of future injury is not an absolute standard: what is to be aimed at is justice between the parties, having regard to all the relevant circumstances.”

I dealt with Hooper v Rogers in my judgment in this case last year. I said this passage from Russell LJ showed that “There is not an absolute standard. What is aimed at is justice between the parties in all the circumstances.” I maintain that view.

Mr Thorley relied in particular on the words of Chadwick LJ in Lloyd v Symonds [1998] EWCA Civ 511 (Court of Appeal, Millett, Waller and Chadwick LJJ). The defendant’s barking dogs had caused a nuisance to the claimant but the trial judge had found as a fact that the problem had ceased at the end of May 1996, before proceedings were commenced on 20^th June 1996. Nevertheless at trial he granted an injunction. Chadwick LJ observed that the injunction was quia timet, it was not to restrain something that the defendants were doing at the time of the trial (or at the commencement of proceedings), it was to restrain something which the claimant alleged the defendants were threatening or intending to do. The learned Lord Justice said:

“On the basis of the judge's finding that the previous nuisance had ceased at the end of May 1996 the injunction which he granted on 7th January 1997 was quia timet. It was an injunction granted, not to restrain anything that the defendants were doing (then or at the commencement of the proceedings on 20th June 1996), but to restrain something which (as the plaintiff alleged) they were threatening or intending to do. Such an injunction should not, ordinarily, be granted unless the plaintiff can show a strong probability that, unless restrained, the defendant will do something which will cause the plaintiff irreparable harm -- that is to say, harm which, if it occurs, cannot be reversed or restrained by an immediate interlocutory injunction and cannot be adequately compensated by an award for damages. There will be cases in which the court can be satisfied that, if the defendant does what he is threatening to do, there is so strong a probability of an actionable nuisance that it is proper to restrain the act in advance rather than leave the plaintiff to seek an immediate injunction once the nuisance has commenced. “Preventing justice excelleth punishing justice” -- see Graigola Merthyr Co Ltd v Swansea Corporation [1928] Ch 235 at page 242. But, short of that, the court ought not to interfere to restrain a threatened action in circumstances in which it is satisfied that it can do complete justice by appropriate orders made if and when the threat of nuisance materialises into actual nuisance (see Attorney-General v Nottingham Corporation [1904] 1 Ch 673 at page 677).

….

In the present case, therefore, I am persuaded that the judge approached the question whether or not to grant a permanent injunction on the wrong basis. He should have asked himself whether there was a strong probability that, unless restrained by injunction, the defendants would act in breach of the Abatement Notice served on 22nd April 1996. That notice itself prohibited the causing of a nuisance. Further he should have asked himself whether, if the defendants did act in contravention of that notice, the damage suffered by the plaintiff would be so grave and irreparable that, notwithstanding the grant of an immediate interlocutory injunction (at that stage) to restrain further occurrence of the acts complained of, a remedy in damages would be inadequate. Had the judge approached the question on that basis, I am satisfied that he could not have reached the conclusion that the grant of a permanent injunction quia timet was appropriate in the circumstances of this case.”

(Emphasis added)

Mr Thorley also emphasised two further statements by Chadwick LJ in a later passage which in fact concerned payment on the cross-undertaking in damages:

“The question for the judge after trial is whether or not, in the light of the facts which have been investigated and determined at trial, the defendant has been liable to an interim restraint which would not have been imposed upon him if those facts had been established at the date of the interim order.”

and

“In the present case it seems to me that the judge ought to have appreciated that if the facts as he found them to be – namely that there was no nuisance current on 27^th June 1996 – had been known to the court on that day no injunction (a fortiori no injunction in that form) would then have been granted.”

Mr Thorley submits, and I accept, that this shows one needs to be alert to the distinction between an interim and a final injunction. Mr Thorley also submitted that in Lloyd v Symonds the court required the claimant to show a strong probability that the defendants would act in breach of the abatement notice. That is indeed what Chadwick LJ said in the passages above. It was in the context of a case concerning the exercise of the court’s jurisdiction to restrain a Statutory Nuisance which is subject to an Abatement Notice. In another passage, which I have not set out, Chadwick LJ indicated that that jurisdiction was one to be exercised with great caution.

In Linpac v Eagleton [1994] FSR 545 (Court of Appeal, Dillon, Hirst and Millett LJJ) the defendants had imported infringing copies of the plaintiff’s cups but denied knowledge that they were infringing. Until the writ was issued the defendants were prevaricating and it was only on the day the writ was issued that the defendants said they had stopped shipping the cups. The judge refused a permanent injunction on the basis there was no real threat to infringe. The Court of Appeal allowed the appeal, finding that the injunction was justified on a quia timet basis. Dillon LJ held that the test applied by the judge had been too stringent and that the claimant does not have to wait while prevarication continues to see whether further shipments or sales of goods take place. Both Hirst LJ and Millett LJ agreed with Dillon LJ. Hirst LJ added that the defendants’ conduct after the writ was issued was “quite sufficient to signify an intention to continue infringement after they had reason to believe that the cups were infringing copies”.

Millett LJ (as he then was) expressed the matter of the later conduct as being a case in which the defendants were “claiming the right to continue the acts of infringement of which they were charged”. A “claim of right” used to be regarded as a requirement before the court’s inherent jurisdiction to grant negative declarations could be invoked (see Re Clay [1919] 1 Ch 66) although today the jurisdiction is broader. Although he did not say so I expect Millett LJ had this in mind. If a patentee claims to a trader that they have the right to stop the trader from selling a product then the law has always allowed the trader to bring the patentee before the court and obtain a declaration under the court’s inherent jurisdiction that the product does not infringe. This is in addition to the statutory declaration of non-infringement regime in the Patents Act 1977. Millet LJ’s words put the matter the other way around. If someone claims the right to carry out acts of infringement, then the rights holder is entitled to bring them before the court and seek an injunction to restrain those acts.

In Landor v Azure [2007] FSR 9 the Court of Appeal (May, Neuberger and Wilson LJJ) heard an appeal from the grant of a quia timet injunction in an unregistered design case. The problem in that case was that although the defendant had given contractual undertakings at an early stage which were in force when the claim was started, the undertakings were later withdrawn. Neuberger LJ (as he then was, with whom May and Wilson LJJ agreed) held that the judge was entitled to grant the injunction at trial in those circumstances and said that the debate, which was whether there was no threat of infringement at the date of issue of proceedings, was an arid one. He rejected the argument as without merit and held that the courts have tended to take a pragmatic case by case approach rather than a technical view on the issue of whether it was appropriate for a claimant to have sought an injunction to restrain infringement of an intellectual property right. On this the learned judge cited Hooper v Rogers, Linpac v Eagleton and also LA Gear v Hi-Tec Sports (which was also considered in Linpac). He observed that at worst Landor had taken a chance when it issued proceedings and held that it was reasonable for Landor to have sought an injunction in the circumstances of that case.

In London Borough of Islington v Elliott [2012] EWCA Civ 56 (Court of Appeal, Longmore, Patten, Rafferty LJJ) the court heard an appeal in relation to the roots of ash trees. The claimant’s house was next door to a house owned by the Council. The claimant was concerned that the roots would undermine her foundations. She wrote to the Council and did not receive a satisfactory response. The Council were not prepared to agree to deal with the trees. Between July 2008 and March 2009 there was no further correspondence between the parties. In fact the Council decided to remove the trees in December 2008 but when the action began in March 2009 the Council had not informed the claimant of this decision. As Patten LJ said (paragraph 21) if the claimant’s solicitors had taken the precaution of writing a formal letter before action to the Council before instituting the claim it seems likely that they would have been told of what was planned but they did not do that.

The case was being tried only in order to decide what to do about costs. The judge found for the claimant. The Court of Appeal allowed the appeal. At paragraph 29 Patten LJ distinguished between an interim injunction and a permanent injunction and said “a permanent injunction can only be granted if the claimant has proved at trial that there will be an actual infringement of his rights unless the injunction is granted.” Patten LJ went on to refer to authorities including the passages I have quoted above from Lloyd v Symonds, (noting it concerned nuisance by noise) and said at paragraph 32 that in this case the determining issue was whether, absent an injunction, there was imminent danger of actual damage (referring to Hooper v Rogers). At paragraph 36 Patten LJ said that the question of whether it was an appropriate case for the grant of quia timet relief has to be considered in the light of all the relevant circumstances known at the trial and drew a distinction between that wider consideration of relevant factors and the narrow question of whether the tree roots were likely to cause damage within a particular time frame. In paragraph 37 he explained Hooper v Rogers on the basis that:

“… the inevitability of subsidence attributable to the new track was such that nothing short of its removal would cure the problem. It was therefore realistic for the judge in that case to have taken the view that an injunction should be granted as the only means of preventing that risk from materialising. Questions of timing were less significant because the defendant landowner was not prepared to restore the slope underpinning the plaintiff’s property unless compelled to do so by an order of the court.”

At paragraph 38 Patten LJ said:

“But cases involving damage caused by trees are not necessarily so stark. […] The claimant has to show that an injunction is necessary in order to prevent the occurrence of the nuisance. The defendant is entitled to rely on his own rights and obligations as an adjoining landowner to cure the problem and it ought therefore in principle to be only in cases where the risk of damage is so imminent and the intransigence of the defendant so obvious that the court should ordinarily be prepared to grant an injunction in order to prevent a nuisance which does not yet exist. Mandatory injunctions of this kind are not justified merely on the ground that if nothing is done a tree on adjoining land may at some point in the future begin to cause damage to the claimant’s property”

At paragraph 39 Patten LJ noted that the judge’s conclusion that the Council would not have done the work without an order was difficult to reconcile with his finding that a works order had been signed in December 2008 nor was there any challenge in the pleadings to the Council’s defence that it intended to carry out the works. Patten LJ concluded that:

In these circumstances, it was not open to the judge in my view to hold that the injunction was necessary in order to prevent the potential nuisance from becoming an actual one. Although the claimants had initially to face a combination of delay and misleading information from the Council, it had by December 2008 at the latest resolved to remedy the problem by removing the trees. There was therefore no necessity for the grant of quia timet relief at the trial and the plea that the Council intended to carry out the work was a complete answer to the claim. If the appropriate rule to apply was that costs should follow the event then the judge should have dismissed the claim with costs.

Having considered the cases, my conclusions on the law are as follows.

One question before me is whether the intention of the defendant is the only relevant factor. Is the issue solely about the defendant’s subjective state of mind or do other factors play a part?

The principle I derive from these authorities is that the question the court is asking in every case is whether, viewed in all the relevant circumstances, there was a sufficiently strong probability that an injunction would be required to prevent the harm to the claimant to justify bringing the proceedings. In adding the word sufficiently to the word strong I do not mean to put a gloss on the words of Chadwick LJ, rather I am seeking to encapsulate the idea that the degree of probability required will vary from case to case depending on all the circumstances but that mere possibilities are never enough. To justify coming to court requires there to be a concrete, strong and tangible risk that an injunction is required in order to do justice in all the circumstances.

If a defendant really does, at the date of the proceedings, have no intention to do the act then in the majority of cases that will be conclusive of the question whether there was a sufficiently strong probability to justify proceedings. (e.g. London Borough of Islington). However it seems to me that the question is not confined to the defendant’s subjective intentions. A defendant’s overt acts must be capable of being relevant. To take an extreme case, if a man began taking actual preparatory steps to commit some unlawful act seriously damaging to the claimant and in infringement of the claimant’s rights and did so in full view of the claimant and well aware that the claimant could see them, he could hardly complain if the claimant started proceedings and the court decided to grant a final injunction to prevent it. A statement at trial that he had never intended to go through with it would get short shrift.

I bear in mind that intentions are not necessarily simple. A state of mind need not merely be either one thing or another. Also in this case the defendants are corporate entities to whom an intention can only be imputed.

The way the matter is put in the Particulars of Claim contains the allegation that the defendant “threatens and intends” to infringe. I think this is a useful expression in that it encompasses both the defendant’s intentions and also the idea that the court should look from the outside at what the defendant is threatening to do. Both are relevant.

The question I will ask is whether, viewed in all the circumstances at the date the proceedings were issued, there was a sufficiently strong probability that an injunction would be required to prevent Teva from infringing. I will look first at the position without regard to Teva’s case about its actual intentions, then I will consider Teva’s actual state of mind and then draw the matters together and look at the position overall.

Viewed objectively from the stand point of February 2012, the UK market for efavirenz was large and valuable. A generic company which was prepared to take the infringement risk would have a real commercial incentive to do so by launching first and alone, i.e. as the first generic on the market. I do not accept that the fact Efavirenz is sold to hospitals rather than retail pharmacies undermines that incentive to any material extent. Launching before SPC expiry would be a rational thing for a generic company to do if it was prepared to bear the risk of patent infringement.

Looking at Teva’s position objectively, it is a large, successful seller of generic pharmaceutical products. It is demonstrably prepared to launch products without notice and at risk of infringing an originator’s patent if it chooses to do so. Atorvastatin is a sufficient example but there are others. Teva is demonstrably prepared to launch surreptitiously if it chooses to do so. Again atorvastatin is an example.

By 20^th October 2011 Teva had acquired a favourable CHMP opinion for its efavirenz product and in January 2012 it obtained a marketing authorisation. From the point of view of an outside observer, Teva plainly had evinced an intention to launch efavirenz sometime but as I said at the interim stage, that does not mean the marketing authorisation shows an intention to launch efavirenz before expiry. On the other hand having a marketing authorisation in place two years before expiry does make an at-risk launch feasible from a regulatory stand point.

The objective position in late 2011/early 2012 was that there was a real risk that Teva could launch efavirenz before 20th November 2013. An at-risk launch was a credible possibility, but not more than that. This was the position from the point of view of BMS prior to contacting Teva. Raising the matter in correspondence was appropriate and reasonable.

I think Teva’s actual intentions as at 9^th February 2012 were tolerably clear. Teva’s long term plans had a PLMD of 20th November 2013 but the possibility of launching in the UK before that date was under active consideration as a result of the timing of the CHMP opinion. From Teva’s point of view it was capable of making sense both commercially and practically: commercially in that the reward could be worth the risk and practically in that it was feasible. As at 9^th February 2012 I infer from Teva’s internal documents and also from Teva’s position adopted in response to the letters from BMS, that Teva actively wanted to keep that option available. A later decision to actually launch efavirenz before expiry would not be a change of mind, it would be the crystallisation of what was, as at 9^th February 2012, an option Teva was taking steps to consciously keep open.

What of the statement in the Response to the Part 18 Request on 23^rd April 2012 when Teva did purport to tell BMS what its intentions were? I remind myself that this statement of case is not evidence in itself but in any case the statement is not far from what I infer was Teva’s position. Teva did not, as at 9^th February, have concrete plans in place to launch before expiry and never had had but such plans could readily have been formulated. There was plenty of time between February 2012 and November 2013 to put in place everything necessary to launch at risk and without notice before 20 November 2013. If the option which I find was being kept open had been exercised for example in May 2012, there was plenty of time. In any case I am not persuaded that a statement of what Teva’s intentions were in April 2012, after the proceedings had been started, an interim injunction had been ordered and Teva had refused to make a statement about their intentions in the Defence, carries the same weight as a statement made before proceedings are issued.

To characterise Teva’s intentions at the relevant date as no more than an absence of plans to launch efavirenz before 20th November 2013 would be materially incomplete. If Teva had said what its actual state of mind was in response to a request from BMS, it would have said it had no plans to launch efavirenz before expiry but was actively considering the option of doing so.

Considering all the circumstances together, I remind myself that this is the trial of a patent action. Before me the patent is in force, Teva had the opportunity to challenge validity but did not and it is agreed that the patent would be infringed by a sale by Teva of efavirenz before November 2013. The option Teva was actively considering was to launch at risk, i.e. to sell efavirenz without BMS’ permission. It follows that what Teva was actively considering the option of doing was committing an act of patent infringement; an act which would be unlawful, would interfere with the claimant’s legal rights and would inflict serious damage on the patentee.

In other words Teva believed it had the right to sell efavirenz before 20th November 2013 without reference to BMS if it chose to do so. But Teva has and had no right to sell efavirenz before expiry without the permission of BMS. It had not positively decided what to do but if its private considerations arrived at the decision to launch, it was going to do so without warning. The only things which would prevent Teva from infringing would be an injunction or Teva’s agreement to give BMS advance notice of a decision to launch in order to give BMS time to obtain an injunction. Teva had the opportunity to agree to give BMS warning in correspondence but refused to do so. BMS had no other option. In my judgment in 9th February 2012 the probability that an injunction would be required to prevent Teva from infringing was sufficiently strong to justify legal proceedings.

I find that this action was justified. I should grant an injunction to restrain Teva from infringing the patent and SPC. This injunction is not an interim one. It is permanent in the sense that it remains while the patent and the SPC are in force and will expire with the expiry of the SPC.