Arcus Biosciences (NYSE: RCUS) presented new clinical data from its ARC-20 platform study evaluating casdatifan, an investigational HIF-2α inhibitor, as a treatment for clear cell renal cell carcinoma (ccRCC) across multiple lines of therapy.

In the first-line setting, casdatifan combined with the anti-PD-1 therapy zimberelimab showed progression-free survival (PFS) landmark values of 86% at six months and 69% at 12 months in a cohort of 30 patients, along with a 50% confirmed objective response rate. These figures were compared against third-party benchmarks for approved therapies, though Arcus noted the comparisons are not head-to-head and differences in study design exist.

In the second-line setting, casdatifan combined with cabozantinib produced a median PFS of 14.5 months in 43 patients, compared to a benchmark of approximately 10 months for cabozantinib alone. The 18-month landmark overall survival (OS) rate was 77%, versus approximately 60% for cabozantinib benchmarks. The combination showed similar efficacy in patients who had previously received TKI therapy as in those who had not.

In heavily pre-treated patients, pooled monotherapy data from 121 patients showed a 35-month median OS, a 13-month median PFS, and a 36% confirmed objective response rate. The company stated these figures exceed benchmarks for the only currently approved HIF-2α inhibitor by more than one year of median OS.

No unexpected safety signals were reported across any cohort at the September 4, 2026 data cut-off date.

Arcus said its Phase 3 PEAK-20 study evaluating casdatifan combined with ipilimumab and nivolumab in the first-line setting is on track to begin by the end of 2026. The company is also enrolling the Phase 3 PEAK-1 study comparing casdatifan plus cabozantinib against cabozantinib alone in immunotherapy-experienced patients. Arcus additionally plans to initiate a registrational trial evaluating casdatifan plus tivozanib. Casdatifan has not received regulatory approval, and its safety and efficacy have not been established.