Two-Stage Estimation of Overall Survival in the Phase 3 CheckMate 9ER Trial, Adjusting for the Impact of Subsequent Therapy

This is the authors' abstract. We don't add key points for this paper.

The CheckMate 9ER (CM9ER) trial demonstrated the superiority of cabozantinib plus nivolumab (CaboNivo) over sunitinib (SUN) in patients with untreated advanced renal cell carcinoma (aRCC). We conducted a post hoc analysis using CM9ER data (median follow-up 67.6 months) and two-stage estimation (TSE) to assess the efficacy of CaboNivo versus SUN, adjusting for the impact of subsequent treatment. Patients in the intention-to-treat (ITT) CM9ER population who discontinued their randomized study therapy were included in step 1 of the TSE, which used accelerated failure time modeling to estimate the extent that subsequent therapy extended survival in each treatment arm. Step 2 adjusted the CM9ER ITT population overall survival (OS) estimates to account for this impact; an adjusted hazard ratio (HR) for OS was then calculated to evaluate the relative impact of CaboNivo versus SUN without the effects of subsequent treatment. In total, 540 patients from CM9ER (CaboNivo, n = 264; SUN, n = 276) discontinued treatment and were included in the analysis, of whom 297 (CaboNivo, n = 129; SUN, n = 168) initiated subsequent systemic therapy. Patient characteristics were balanced between ITT and TSE populations. Median (interquartile range) time from randomization to treatment discontinuation was 22.2 (9.92–31.51) months for CaboNivo and 9.6 (3.48–18.62) months for SUN. The TSE-adjusted median (95% confidence interval [CI]) OS was 42.7 (38.4–49.1) months for CaboNivo and 21.3 (18.1–24.9) months for SUN (HR [95% CI] 0.53 [0.43–0.65; p < 0.001]), compared with unadjusted estimates of 46.5 (40.6–53.7) months and 35.5 (29.2–42.8) months, respectively (HR [95% CI] 0.79 [0.65–0.96; p = 0.0196]). Removing the effect of subsequent therapy strengthened the relative OS benefit of CaboNivo versus SUN in CM9ER. Consistent with clinical practice, these results further support first-line CaboNivo as a standard of care for aRCC. Graphical abstract available for this article. NCT03141177. The first choice of treatment for patients with advanced kidney cancer is often combination therapy with a checkpoint inhibitor and a tyrosine kinase inhibitor. The combination slows rapid tumor growth and helps the immune system to fight cancer. Cabozantinib plus nivolumab (“CaboNivo”) is a combination approved for the initial treatment of advanced kidney cancer on the basis of the CheckMate 9ER trial results. People treated with CaboNivo in the trial lived significantly longer than those treated with a tyrosine kinase inhibitor called sunitinib (“SUN”). This finding was based on how long people lived after starting their trial therapy. However, during a trial, some patients might switch from their study therapy to other treatment if their cancer worsens or if they have side effects. This switching can affect their survival. We used a statistical approach to estimate survival in the CheckMate 9ER trial with the effects of switching removed. First, we calculated the effect that switching had on survival. Then we recalculated survival for CaboNivo and for SUN with this effect removed. We included 540 patients who were observed during the trial for up to 5 years. When we removed the effects of switching to another treatment, we found that on average people treated with CaboNivo lived more than twice as long as people treated with SUN; 42.7 months with CaboNivo compared with 21.3 months with SUN. These results reinforce the benefit of CaboNivo versus SUN and support CaboNivo as a first treatment for people with advanced kidney cancer.