Biogen v Sandoz, Viatris and Mepha (BPatGer, S2024_005 to S2024_007, 2025)

Dhenne Avocats · 11 October 2026

Our analysis

Summary

The Swiss Federal Patent Court dismisses Biogen’s applications for preliminary measures against three dimethyl fumarate generics, based on the Swiss part of EP 2 653 873, which claims the oral treatment of multiple sclerosis at a dose of 480 mg per day. Having rejected the added matter, insufficiency and novelty attacks, it holds that inventive step is likely lacking: an earlier phase II study left a gap between 360 and 720 mg that the skilled person was prompted to explore, and 480 mg was the obvious first choice. For dosage regime patents, a dose-response trend visible in an exploratory study is thus enough to found a reasonable expectation of success.

Facts and procedure

Biogen MA, the proprietor, and Biogen International filed three applications on 30 August and 4 September 2024 against the 120 mg and 240 mg capsules of Sandoz, Viatris and Mepha (paras 1 to 3), joined on 9 October 2024 (para. 8). The patent, filed on 7 February 2008 claiming a US priority of 8 January 2007, is a divisional of EP 2 137 537, which an EPO Board of Appeal revoked on 20 January 2022 (T 1773/16); granted on 20 July 2022, it was maintained by the Opposition Division in a form limited to relapsing-remitting multiple sclerosis, a decision under appeal (paras 17 and 19). The court notes diverging outcomes in five other countries, including the annulment of the French part by the Paris Judicial Court on 18 June 2025 (para. 17).

The applicable law

Under Article 77 of the Patents Act (PatG) and Article 261(1) of the Swiss Civil Procedure Code, the applicant must show a credible infringement and a harm not easily remedied, the threshold rising with the severity of the measure (para. 18). In summary proceedings a patent may be asserted in one version only; absent a clear choice, only the granted version is examined (para. 16). For sufficiency of a medical use claim, the court transposes the G 2/21 test: the effect must be encompassed by the teaching of the application as filed and derivable from it, and the proprietor may answer with post-published evidence a challenge itself based on such evidence (para. 34).

Question

Is a daily dose of 480 mg inventive over a phase II study that tested 120, 360 and 720 mg per day?

Decision

The 480 mg value is the end point of the narrowest disclosed range (480 to 720 mg) and results from a single selection from a single list, which is permissible (para. 33). The in vitro examples and the mouse model make the effect derivable (para. 35), and a stabilising effect is credible even in the primary progressive form (para. 36). Priority is left open (para. 39) and WO 2006/037342 does not destroy novelty, since it discloses the indication and the dose only through a double selection (para. 41).

The starting point is the May 2006 presentation of the phase II results (Kappos I) (para. 44). Nothing makes any advantage of 480 mg credible beyond efficacy comparable to 720 mg, so the problem is to provide an alternative daily dose (para. 46). Correcting for the baseline imbalance in lesion count in the 360 mg group, the skilled person would have seen an increasing dose-response relationship and little additional benefit at 720 mg over 360 mg (para. 48). With fewer gastrointestinal adverse events at 360 mg, the skilled person was prompted to explore the gap between 360 and 720 mg; the lack of statistical significance at the lower doses does not teach away in an exploratory study with about sixty patients per arm, and the available 120 mg and 240 mg tablets led naturally to 480 mg (para. 48). The result is the same on Biogen’s formulation of the problem (para. 48). As infringement of a valid right is not credible, the applications are dismissed (para. 49). Biogen bears a CHF 75,000 court fee and CHF 140,000 in party compensation in total (paras 50 to 52).

Key points for practice

• In summary proceedings the proprietor must assert its patent in a single version (para. 16).
• A dose-response trend visible in an exploratory study is enough for a reasonable expectation of success; statistical significance is not required (para. 48).
• Patent attorney costs invoiced to another group company are reimbursed only if the party proves it bore them (para. 51).
• Practical point: against a dosage regime patent, a close reading of the earlier phase II data, including baseline imbalances, may be enough to bridge the gap between the doses tested and the dose claimed.

Provisions applied

European Patent Convention

art. 76, art. 83, art. 87, art. 123, art. 138

National law

Arts 1, 7, 17, 26, 50 and 77 of the Federal Patents Act (PatG); Arts 26 and 36 of the Patent Court Act (PatGG); Arts 109 and 110 of the Private International Law Act; Arts 106 and 261 of the Swiss Civil Procedure Code

Other provisions

Lugano Convention, Art. 2; Paris Convention, Art. 4

Case law cited

EPO, G 2/21, G 2/98, T 1773/16, T 606/89, T 967/97; BGE 146 III 177; BGE 138 III 111; Federal Supreme Court 4A_490/2020; BPatGer S2024_008 (Lisdexamphetamin), S2019_007 (Tadalafil 5 mg), S2017_001, S2021_005 (Deferasirox)

Related decisions

• G 2/21, 23 March 2023, test transposed to sufficiency of the medical effect (para. 34).
• G 2/98, 31 May 2001, cited for the concept of the same invention in priority matters (para. 38).

Prepared by Dhenne Avocats from the text of the decision (official text, bundespatentgericht.ch), consulted on 11 October 2026. Only the official text is authoritative.