Roche Claims Best-in-Class Potential for Sefaxersen Without Disclosing How Much It Cut Proteinuria
- Roche said on 23 September 2026 that sefaxersen met the primary endpoint at a prespecified interim analysis of the Phase 3 IMAgINATION study in primary IgA nephropathy, with a statistically significant and clinically meaningful fall in 24-hour urine protein-to-creatinine ratio against placebo at week 37. Roche says the result shows best-in-class potential. It did not give the size of the reduction.
- The trial randomised 459 adults 1:1 for 105 weeks and stays blinded to assess kidney function by eGFR at week 105. Safety was consistent with earlier data, with no new signals. Sefaxersen is a liver-directed antisense drug that blocks production of complement factor B, given as a once-monthly injection designed for self-administration.
- Roche licensed the drug from Ionis, paying $75m upfront for an option in 2018 and $55m upfront for the IgAN licence in 2022, according to Fierce Biotech. Its nephrology head, Jay Garg, said in March he hoped week 37 data would support accelerated approval. The interim data will go to health authorities and an upcoming medical meeting.
- Roche puts peak sales at €1bn to €2bn. Its head of investor relations said in March that most analyst models carried about €400m by 2030. Both forecasts predate the readout.
CI read
The target is the same as Novartis’s Fabhalta. The release withholds the one number that would show whether Roche’s drug does better.
- The comparison figure is missing. Fabhalta cut proteinuria by 38.3% against placebo at nine months in APPLAUSE-IgAN (95% CI 26.0% to 48.6%), which supported its 2024 accelerated approval. Sefaxersen was measured on the same endpoint at 37 weeks, close to the same timepoint. Until Roche publishes its number, best-in-class is a claim with no figure behind it.
- Same target, different trade-offs. Both drugs block factor B. Fabhalta is taken orally twice a day and can raise cholesterol; sefaxersen is a monthly injection that showed no cholesterol effect in Phase 2. Fabhalta’s final analysis recorded serious infections in 6.7% of patients against 2.1% on placebo, and whether a liver-directed approach avoids that signal is something the full sefaxersen data must show.
- Roche is running Novartis’s regulatory playbook two years behind. Fabhalta won accelerated approval on nine-month proteinuria in 2024 and full approval this year on two-year kidney function. Sefaxersen has the first half of that path. Novartis, meanwhile, lists atrasentan and zigakibart alongside Fabhalta in its renal pipeline, so the incumbent will defend the space with more than one drug.