UPC Weekly - ‘Off label’ and off the hook: UPC finds no infringement of second medical use patent

2025 Week 20

For many years, the EPO has recognised that the novelty and inventiveness of a ‘medical use’ claim can be derived from a mechanism or effect that is recited in the claim. In Sanofi / Regeneron v. Amgen, the UPC Düsseldorf Local Division (LD) followed the EPO opposition division in upholding such claims in Sanofi’s EP 3 536 712 B1 (EP’712). The court found that the lipoprotein(a) [Lp(a)]-lowering effect of anti-PCSK9 antibodies was not previously disclosed and was not obvious from the prior art. 

However, the Düsseldorf LD also found that the sale of Amgen’s anti-PCSK9 antibody Repatha® does not infringe EP’712, despite a reduction in Lp(a) being disclosed among the ‘pharmacodynamic properties’ listed in its “Summary of Product Characteristics” (SmPC). The SmPC does not disclose Lp(a) reduction among the ‘therapeutic indications’ for which the drug is approved. Moreover, the court was not convinced that Amgen was marketing Repatha® in a way that would reasonably lead to it being prescribed to achieve this particular effect.

There is a certain symmetry in, on the one hand, recognising patentability of uncovering a previously unknown biological effect within a known medical treatment and, on the other hand, only finding infringement if the treatment is performed specifically to achieve that effect.

Background

This is the latest instalment in the long-running dispute between Sanofi / Regeneron and Amgen, in relation to their respective cholesterol-lowering drugs Praluent® and Repatha®.

The case marks the UPC's first substantive ruling on infringement of a second medical use claim and so it offers interesting insights into the enforcement of such patents in the new court system. Indeed, any European case law on infringement of medical use claims in thin on the ground. Most of the leading national decisions deal with Swiss type claims, which the EPO phased out in 2011 (which is when Sanofi’s patent application was filed). 

Sanofi’s EP’712 (exclusively licensed to Regeneron) claims the use of anti-PCSK9 antibodies for reducing lipoprotein(a) levels in patients with elevated serum concentrations. 

Amgen’s product, Repatha® (evolocumab), an anti-PCSK9 antibody, is primarily indicated for lowering low-density lipoprotein cholesterol (LDL-C) and is approved in Europe for treating hypercholesteraemia and mixed hyperlipidaemia. 

However, it does also have the effect of lowering Lp(a) levels, a fact that is confirmed in the SmPC accompanying its EU marketing approval.

Sanofi and Regeneron initiated infringement proceedings against Amgen, alleging that Repatha® infringed EP’712 by virtue of its Lp(a)-lowering effects. Amgen countered with a revocation action.

Second medical use claims

Second medical use claims protect the use of known substances for new therapeutic applications, because therapeutic methods per se are excluded from protection in Europe. Under the European Patent Convention (EPC), such claims are typically drafted in the form "Substance X for use in treating disease Y." 

As the LD explains carefully in the present decision, such claims are product claims but - exceptionally - their scope is limited by the purpose of the product. ¹They can hence be directed to a known substance or composition, with “notional novelty” being derived from a specified new medical use of that product. As the European Patent Office has long since accepted, and as the UPC in the present decision appears to endorse, the concept of a ‘new medical use’ is not limited to a new disease indication but also covers other novel aspects such as a new group of patients to be treated.²

The legal principles applicable to medical use claims have been the subject of considerable discussion, with national courts around Europe yet to reach consensus on a number of key issues. These include the correct test to be applied to establish infringement. The UPC’s comments on this matter will therefore be viewed with great interest.

Validity

One of the prior art documents stated that it was ‘conceivable’ that PCSK-9 inhibitors may, among other things, lower Lp(a). However, the court only found this to motivate the skilled person to consider using PCSK-9 inhibitors to achieve this effect. For a finding of obviousness, “[a] motivation only to consider is not enough, the skilled person needs to be motivated to implement” the teaching³.

The existence of more obvious ways of lowering Lp(a) alongside some teachings against a link between PCSK-9 inhibition and Lp(a) reduction persuaded the court that the skilled person would not be motivated to implement the idea of using PCSK-9 inhibitors to lower Lp(a).

Infringement

The test adopted by the UPC to assess infringement of second medical use claims is essentially two-pronged, as it has both objective and subjective elements.⁴

According to the court, infringement occurs if:

1. The alleged infringer markets the product in a manner that leads or may lead to the claimed therapeutic use (the “objective” element), and
2. The infringer knows or should reasonably have known that the product would be used for the claimed purpose (the “subjective” element).

Holders of second medical use patents may take some comfort from the fact that the subjective element of this test (number 2, above) appears to be based on reasonable foreseeability, rather than being a ‘subjective intention’ test (as has sometimes been suggested)⁵ which would be more difficult to prove. 

The LD set out some factors which should be taken into account when considering the behaviour of the alleged infringer and assessed whether it meets the above test. These include:

• The extent of the allegedly infringing use.
• Market practices and the significance of the patented use.
• Actions taken by the alleged infringer to either encourage or prevent the product from being used for the patented medical use.

In setting out these general principles, however, the LD emphasized that a comprehensive analysis of all relevant facts and circumstances will always be necessary to determine infringement. They also noted that, although information associated with a pharmaceutical product, such as its packaging insert and SmPC, was likely to be very important to consider, it is not the only decisive factor. 

In the case in suit, the court examined whether Repatha® was prescribed specifically for lowering Lp(a) levels in the specified patient population and found that Sanofi and Regeneron had failed to provide sufficient evidence to determine that Repatha® was prescribed, or was likely to be prescribed, for this ‘off-label’ use. They also found no compelling evidence that Amgen had promoted Repatha® for the patented indication or that physicians prescribed it for lowering Lp(a) levels. 

The mere mention of Lp(a) reduction as a “pharmacodynamic effect” in the SmPC (notably, not in the context of the "therapeutic indication”) was deemed insufficient to establish infringement.⁶

Skinny labelling – necessary but maybe not sufficient to avoid infringement

Under the logic of this decision, it seems that the practice of ‘skinny labelling’ (where patented indications are deliberately excluded from the published therapeutic indication of a generic medicine) would not in itself be enough to avoid a finding of infringement. Other evidence could be used to persuade a court that a medicine is being marketed in a manner that leads (or may lead) to the claimed therapeutic use, and that the infringer knows (or reasonably should know) this. In this case, the UPC implied that, had evidence of actual prescriptions for the patented use been adduced, the finding on infringement might have been different.⁷ According to this judgment, the analysis of the relevant factors is always likely to be highly fact- and case-dependent.

Conclusions / future outlook

Given the high value of the case, these decisions are likely to be scrutinised by the UPC Court of Appeal in due course. The EPO decision to maintain the patent has already been appealed. 

There are other proceedings pending at both the EPO and the UPC (including an imminent UPC Court of Appeal decision on Amgen’s own second medical use patent EP 3 666 797, which was revoked by the UPC Central Division last year) as well as in the German national courts (in relation to EP 2 756 004, the parent of the Sanofi patent in suit in the present case). 

This battle therefore looks set to continue for some time.

1. Paragraphs 87 and 92 of the decision
2. Paragraph 93 of the decision
3. Paragraph 139 of the decision
4. Paragraph 182 of the decision
5. e.g. Lords Hodge and Briggs in Warner-Lambert v Mylan & Actavis, [2018] UKSC
6. Paragraph 191 of the decision
7. Paragraph 193 of the decision

This blog was co-authored by Julie Carlisle and Eliot Ward.