Head-to-head effectiveness and safety of pembrolizumab plus axitinib vs. nivolumab plus ipilimumab in metastatic renal cell carcinoma in the United States.

Abstract

4532 Background: Two main first-line combination therapies for metastatic renal cell carcinoma (mRCC) are pembrolizumab plus axitinib (P+A) and nivolumab plus ipilimumab (N+I). No head-to-head clinical trial comparing these treatments has been performed, and the available real-world data (RWD) on their effectiveness is limited. Here, we used RWD to compare the overall survival and posttreatment hospitalization rates of mRCC patients receiving P+A vs. N+I. Methods: We included mRCC patients treated with N+I or P+A between 6/2018 and 3/2022, utilizing data from Elevance Health administrative claims and clinical data from AIM Cancer Care Quality Program. Inverse probability of treatment weighting (IPTW) was used to adjust for age, sex, socioeconomic deprivation index (SDI), ECOG performance status (PS), Charlson Comorbidity Index (CCI), and metastatic sites. Real-world overall survival (rwOS) was defined from first treatment until death or last claim and was estimated using weighted Kaplan-Meier method. Favorable risk was defined based on PS (<2) and with a treatment initiated less than 1 year from diagnosis. Results: We identified 764 mRCC patients, of them 483 were treated with N+I (63%). The P+A group was slightly older (p=0.08), while other features were similar in both groups. Survival analysis using IPTW showed a borderline significant advantage of P+A vs. N+I (p-value=0.07). Inspection of the Kaplan-Meier curves suggested that P+A provides short-term survival benefit: 12-month rwOS was 86% for P+A vs. 77.8% for N+I (p<0.001), while at 24-months, where power was lower, the curves almost converged (rwOS: 68% in P+A vs. 65% in N+I; p=0.52). In a subset of patients with favorable risk (N=69 P+A, N=68 N+I), the P+A survival improvement extended to at least 24-months (p=0.06). Finally, the N+I group had significantly higher rates of posttreatment hospitalizations (3-months after treatment initiation) compared to the P+A group (31.9% vs. 19.6%; p-value<0.001). Conclusions: Our retrospective cohort study suggests that P+A may provide short-term survival benefits over N+I, and this benefit is extended in patients with favorable risk. P+A is also associated with fewer post-treatment hospitalizations. We argue that the higher toxicity of N+I may explain its lower short-term effectiveness and may also explain its lower effectiveness in patients with favorable risk. [Table: see text]