Infringement of second medical use claims: Foreseeing intentions

Helen Christie | December 2015

Last month saw Mr. Justice Arnold, sitting in the High Court, rule on the latest stage of an ongoing battle between two pharmaceutical heavyweights, Pfizer and Actavis. A long line of decisions (including two hearings before the Court of Appeal and four at first instance) has culminated in a ruling that has at last helped to clarify some points of law relating to the infringement of second medical use patents. These are directed to purpose-limited products for treatment of specific medical conditions (“substance X for treating Y”).


The case at issue concerned a patent owned by Warner-Lambert, a subsidiary of Pfizer, for the drug pregabalin. Pregabalin is dispensed for the treatment of epilepsy, general anxiety disorder (GAD) and for neuropathic pain. Pfizer owned the initial patent which covered the drug itself. However, when further research indicated that it could also be effective in treating neuropathic pain they obtained a second medical use patent to the use of pregabalin in the manufacture of medicaments for the treatment of pain (claim 1) and specifically neuropathic pain (claim 3).

After the SPC (supplementary protection certificate) for the original patent was allowed to lapse, Actavis began seeking marketing authorisation to sell generic pregabalin for the treatment of GAD and epilepsy under the trade name Lacaent. Pfizer sought to prosecute Actavis for infringing their “pain patent” and Actavis, in turn, sought to invalidate the pain patent.

As one might expect, the enforcement of rights in this type of patent is not straightforward. The traditional way to deal with the issue is for generics companies to market their drugs using what is known as a “skinny label”, which lists non-patented uses only on the packaging. The consensus has been, however, that this method is not effective – prescriptions often don’t include information as to the indication for which a drug is prescribed, so that pharmacies dispensing the drugs have no way of judging whether a branded or generic product should be handed out.

In one of the many earlier court rulings, Pfizer sought an interim order requiring Actavis to undertake a number of actions on top of the usual skinny label packaging. These included writing to developers who provide software to pharmacists requesting that they change their software to facilitate the prescribing of Lyrica by brand name, as well as packaging drugs in an outer cellophane wrapper with a warning. These were all, probably rightly, dismissed as excessive by Justice Arnold who made the point that Actavis should not be expected to procure others to act when this was not entirely within their power.

The Swiss Claim Form

Claim 1 of the pain patent is worded as follows: “use of [pregabalin] or a pharmaceutically accepted salt thereof for the preparation of a pharmaceutical composition for treating pain”. This is the Swiss claim form (use of substance X in the manufacture of drug Y for treating Z). It has now been superseded by purpose limited product claims (substance X for treating Y), however many patents with claims in the Swiss claim form have not yet lapsed.

The claim structure is intended to protect the interests of the patentee without putting medical practitioners at risk of infringing, however it throws up several questions. Does the wording “for treating” require intent on the part of the manufacturer that the drug be used to treat the specific indication claimed? What level of intent is required in a case where it is likely that a drug will be dispensed somewhere along the line (by an unknowing pharmacist, for example) for the patented use?

In an earlier judgement, Mr. Justice Arnold ruled that “an intention on the part of the manufacturer that the product would be used for treating the patented indication” was required in order to infringe a claim in the Swiss form.  Intention, he says, is objective, not subjective, and can be judged on the facts. This intention or mental element was likely contemplated when the Swiss claim form was conceived because if “for” could simply be read as meaning “suitable for” then the patent would cover all medical uses and not just the second medical use as intended.

Lord Justice Floyd in the Court of Appeal came to the conclusion that a manufacturer infringes such a claim if he can “reasonably foresee the ultimate intentional use for pain” and that he does not need a “specific intention or desire”. This line of reasoning led Floyd to the conclusion that Pfizer had an arguable case for infringement.

The Decision

Arnold considered in detail the Court of Appeal’s conclusion, and attempted to set out how the test they had laid down should be applied.

He pointed out that a test based purely on a manufacturer being able to foresee use of their drug for the patented indication falls down in that infringing goods and non-infringing goods cannot be distinguished (and all goods cannot be seen to be infringing if a large proportion of them are not in the end being used for the patented indication).

He concluded that in order to infringe, Actavis must have been able to foresee that a large enough proportion (enough not to be de minimis) of people further down the chain of supply would intentionally prescribe pregabalin for the patented indication. A pharmacist handing out generic pregabalin against a generic prescription without knowledge of the indication for which the drug had been prescribed did not amount to intentional use. In this case, the requirement of foreseeability of intentional use was not fulfilled and Actavis was found not to infringe.

While the decision has gone some way towards clarifying the law in this important area, the saga is not over yet. There may well be questions asked about the use of the de minimis principle in Arnold’s test. Is it possible to put a number on how many people an infringer can predict will intentionally use the product for the patented use? How will this be determined if so? Pfizer is likely appeal the decision (their patent was also found invalid), so it looks like those interested will have to wait a little longer for a final conclusion.