Accord Healthcare Limited v Novartis AG

For a patent claim covering distinct technical fields, obviousness may be tested from any legitimate skilled team working in a field covered by the claim. Each team must be defined realistically, with appropriate common general knowledge. For a product claim to a combination, it is enough that making the composition for animal testing would have been obvious; an expectation of effective treatment in humans is unnecessary. Plausibility requires the application, read with common general knowledge, to disclose a reasoned basis for the claimed therapeutic effect. Qualitative animal results can suffice. A specific combination shown to have a beneficial interaction is not an arbitrary collocation. Under the Regulation (EC) No 469/2009, the SPC product was valsartan and sacubitril, not their co-complex. The patent and SPC remained valid.

Accord sought declarations that Novartis’s patent for pharmaceutical compositions combining valsartan and a neutral endopeptidase inhibitor was invalid. The patent had expired, but it supported an SPC protecting Entresto. Accord’s intended product would contain separate salts of valsartan and sacubitril, whereas Entresto contained a co-complex. Accord challenged the patent on plausibility, obviousness, collocation and lack of technical contribution grounds, and challenged the SPC under Articles 3(a) and 3(b) of the Regulation (EC) No 469/2009. Novartis counterclaimed for threatened infringement. The central issues were the proper skilled person, the sufficiency of the application’s qualitative animal-model disclosure, the effect of the prior art, and whether the SPC product was the active ingredients or the co-complex.

Held: All attacks on the patent and SPC failed. The patent was valid while in force, the SPC was valid, and Accord’s intended product would infringe the SPC on the basis accepted by Accord. A separate infringement finding was unnecessary.

1. Skilled addressee. A broad product claim covering distinct technical fields may be assessed from the perspective of separate legitimate skilled teams in those fields. The skilled person must be identified by reference to the problem, the established field, actual teams at the priority date and appropriate common general knowledge. The singular wording of section 3 of the Patents Act 1977 did not prevent this approach. The court applied the guidance in Illumina v Latvia [2021] EWHC 57 (Pat) and Philips v Nintendo [2014] EWHC 1959 (Pat).
2. Plausibility. Under Generics (UK) Ltd v Warner-Lambert Co LLC [2018] UKSC 56 and Generics (UK) Ltd v AstraZeneca AB [2025] EWCA Civ 903, the application, read with common general knowledge, had to disclose a reasoned basis giving a reasonable prospect of the claimed therapeutic effect. A bare assertion or mere possibility was insufficient, but numerical data were not required. Taking the application as a whole, its specific references to valsartan and sacubitril, representative studies and hypertension animal models disclosed qualitative positive results for the combination. The plausibility threshold was therefore met.
3. Obviousness. For a product claim to a combination, it was sufficient that making the composition for animal testing would have been obvious; an expectation of effective treatment in humans was unnecessary. Neither Ksander nor Trippodo made the claimed combination obvious. Ksander disclosed promising sacubitril pharmacology but no ARB/NEPi combination. Trippodo used an ARB and NEPi as mechanistic probes in a heart-failure model and focused on ACEi/NEPi therapy. Its mixed results and the absence of a class effect for NEPis meant that neither the therapeutic combination nor the selection of sacubitril was obvious.
4. Collocation and technical contribution. Applying Sabaf v MFI [2005] RPC 10 and Illumina v Latvia [2022] EWCA Civ 1924, the combination was a single invention. The patent disclosed that valsartan and sacubitril interacted positively and dispelled uncertainty about an adverse interaction. Quantitative proof of synergy was unnecessary. The specific experimentally supported combination made a technical contribution over the broader disclosures in Darrow and Trippodo.
5. SPC. Under the Regulation (EC) No 469/2009, the product was the active-ingredient combination of valsartan and sacubitril. The co-complex, counterions and water had no independent pharmacological effect. The two-part test in Royalty Pharma Case C-650/17 was satisfied. The marketing authorisation authorised the two active ingredients for Article 3(b), and claim 1 covered the co-complex because it comprised both salts and did not require a conventional admixture.

The court would hear counsel on the form of order. Time for seeking permission to appeal would not run until after that hearing.