Impact of renal function eligibility criteria in clinical trials and real-world survival outcomes among patients with metastatic renal cell carcinoma.

Abstract

407 Background: Registration trials in metastatic renal cell carcinoma (mRCC) establishing the efficacy of first-line (1L) combination regimens employed strict renal function eligibility criteria (NCT02853331, NCT02811861, NCT03141177, NCT02231749); however, pharmacokinetics and safety of immune checkpoint inhibitors and anti-vascular endothelial growth factor therapies have been shown to be similar regardless of baseline renal function. Therefore, we aim to evaluate the impact of renal function eligibility criteria used in previous trials on real world outcomes. Methods: Data from the US nationwide Flatiron Health electronic health record-derived de-identified database included adult patients (pts) with mRCC who received any 1L systemic treatment on or after 4/16/2018 (first FDA combination regimen approval), had hemoglobin level ≥9 g/dL and ECOG performance status (ECOG) 0-1 at 1L. Renal function eligibility criteria at 1L (-60 days to +7 days) include serum creatinine (≤1.5 vs >1.5 x ULN), eGFR (≥30 vs <30 ml/min/1.73m2) and calculated CrCl (≥40 vs <40 ml/min) based on registration trial protocols. Outcomes include real world progression free survival (rwPFS), time to next treatment (rwTTNT), and overall survival (rwOS) from 1L. Inverse probability of treatment weighted (IPTW) Kaplan Meier methods and Cox proportional hazard models with multiple imputation were applied. Results: Of 2038 pts in this cohort, 1390 (68.2%) met all criteria for renal function (included), 140 (6.9%) met at least one criteria (relaxed), 46 (2.3%) met none (excluded), and 462 (22.7%) had unknown renal function. Compared to included pts, excluded pts were more likely to be older (median age: 74 vs 66), female (35% vs 27%), have recurrent disease (61% vs 53%), less likely to have ECOG 0 (24% vs 40%) and favorable IMDC risk (9% vs 14%). Excluded pts are also more likely to be Non-Latinx Black (13% vs 6%) and have lowest SES quintile (22% vs 15%) compared to included pts. Although, after weighting included pts had higher median rwPFS of 9.3 (95% CI 8.4 - 10.1) vs 6.9 (95% CI 3.7 - 9.9), rwTTNT 12.0 (95% CI 11.0 - 13.1) vs 9.1 (95% CI 5.5 - 16.4) and rwOS 38.5 mo (95% CI 35.9 - 44.5) vs 19.3 mo (95% CI 16.6 - 41.8) compared to relaxed/excluded pts, none of the hazard ratios (HRs) were statistically significant after multiple imputation (Table). Conclusions: In this real world study, we found that although only 68.2% of patients met all renal function eligibility criteria, survival outcomes were not different across pts groups. This suggests that mRCC patients with reduced renal function may benefit similarly to 1L treatment and should not be excluded from clinical trials. [Table: see text]