Let’s talk antitrust - Servier case – The state of play in pay for delay

Video | May 2019 | 08:10

Video Details

Let’s talk antitrust - Servier case – The state of play in pay for delay
Marta Giner Asins Hello Richard. Thank you very much for this opportunity to discuss the Servier case, which is the latest episode in this never ending pay for delay saga. So, maybe for the benefit of those who have not read this page-turner decision – the Commission decision was already a page-turner, but the General Court has stayed in the same length – maybe you can walk us through the facts if you would.
Richard Whish QC Yes. Well, it’s a very interesting case. As you say, it’s one of several cases on pay for delay. We’ve had Lundbeck, and here we’ve got Servier and, of course, there is the United Kingdom case called Paroxetine, which is now on an Article 267 reference to the Court of Justice. So, very briefly, in Servier, Servier has a patent on a drug, Perindopril. There are generic companies that might be able to, would like to enter the market, but there is a patent. They litigate to challenge the validity of the patent and then we hear that a settlement has been achieved between Servier and the five generic firms. And to put it simply, if not simplistically, the patent litigation is abandoned, Servier pays a sum of money to the generic companies in return for which they agree not to enter the market. With one company, Krka, it’s not a simple pay for delay agreement. Krka abandons the litigation and then takes a licence to produce the drug in question at a royalty. Query, is the royalty a market royalty? The Commission says all of the agreements violate Article 101. The Commission also said that Servier had abused a dominant position by acquiring from the generic companies their technology in circumstances where Servier did not wish to use that technology for itself. But, of course, by owning it, it meant it wasn’t available to third parties. So, that was the Commission decision and very significant fines were imposed.
Marta Giner Asins My impression is that, probably that, the most interesting point of the case does not in fact concern pay for delay – it concerns market definition, and that is something that may concern other cases besides pay for delay.
Richard Whish QC Well, I think that’s right. So, to be clear, the General Court upholds the findings that the agreements violate Article 101 except in the case of Krka. But to me the more interesting part is the abuse of dominance because the Commission had said this drug, Perindopril, basically was the market, the drug produced by Servier and, of course, that it was dominant and, in doing so, it went through the anatomical therapeutic classification table down to ATC 5 and said that this was the market. Servier appealed and said that’s too narrow a formulation because actually this is a drug for which there are substitutes and price isn’t necessarily the issue. The question is whether doctors are willing to, and do in fact, prescribe other drugs. And Servier says we have got evidence that this does happen. And the General Court says yes, we agree with you Servier, we annul the Commission’s market definition, and then it’s like a house of cards – there’s no dominance and so there’s no abuse. I think that’s really interesting because it’s an interesting thought that in history the Court of Justice or the General Court has only ever twice disagreed with a market definition. One was in the Swiss Watchmakers case, which was actually to do with rejection of a complaint. There was a micro point in a merger case, Tetra Laval/Sidel. Other than that the Courts have always shown great deference to the margin of appreciation of the Commission and, here, the Court says no we don’t defer, we disagree, and I think that is an interesting moment in time.
Marta Giner Asins Well, from a legal counsel point of view, it’s key because it really opens a margin for action. As lawyers what we advise our clients is that they have to take into account the risk of an ATC 5 definition when they are assessing because for our clients they are assessing a priori more than a posteriori, so it’s very difficult to know the definition of the market.
Richard Whish QC What a cliché, but every case turns on its facts and you can’t possibly assume that ATC 5 is always correct. We have a case in the UK, Pfizer and Flynn which was ATC 5, but that was a very specific drug for the treatment of epilepsy where my understanding is that the doctor is very loathe to prescribe a different drug for this very specific and very, sort of, worrying neurological disorder. That’s a little bit different from high blood pressure, which is a little bit more prosaic. It’s serious, of course, but it’s just not the same drug so it comes down to the evidence and the facts of the case.
Marta Giner Asins And that’s a very good approach, I think, the General Court is taking. I was very interested about the reference to internal documents because that’s also a very sensitive subject for our clients. You always have, in the companies, documents sitting around saying our product is the best, it bears comparison to no other, and the General Court says here, well this has to be taken with a bit of perspective and with the promotional quality these documents have so ...
Richard Whish QC As you say, I am a great believer in common sense and, of course, companies puff their own products and so on and so forth. I still think the company needs common sense – be careful what you put in writing. It’s not particularly sensible to put up papers to the Board to say the reason for this merger is that we are going to become the dominant player in the world. I think that’s probably rather foolish. On the other hand, something which is puffery and exaggerated and whatever cannot possibly be dispositive in itself. Courts, competition authorities have to have a certain amount of common sense.
Marta Giner Asins Those are points that the pharmaceutical companies will be able to use in their day to day practice. Returning then to the pay for delay agreements, which was the main subject of the decision, I really understand that the General Court has maintained the state of play and has not essentially overturned what had been said before.
Richard Whish QC And how could it? Because Lundbeck is recent and so we have got questions such as, is the generic capable of being a potential competitor? Yes, actually, depending on the evidence again. The Court said that in Lundbeck. How could this Court possibly reach a different conclusion? Can these agreements be restrictive by object? Yes, they can in principle. Most of these were, but not the Krka agreement. It was different, and so I can only assume now that the pharma companies and their legal advisers will be looking to see to what extent a licensing arrangement, query, might one achieve the same effect as pay for delay without a pay for delay in the purest form of that term.
Marta Giner Asins And probably also being very careful about the way in which they determine the price for the licence.
Richard Whish QC Very careful, because basically the question is, am I inducing you to leave the market in return for a nice wodge of cash? And, of course, that can be achieved in different ways. I would assume that the answer is that if you do license you have to license under normal market conditions and if it can be shown that these conditions are not normal then you may have a problem.
Marta Giner Asins Well, thank you very much, Richard. That was very interesting and a good summary of the main elements of the Servier case.