PUK5 Systematic Identification of Potential Treatment Effect Modifiers in First-Line Advanced Renal Cell Carcinoma

Abstract

Imbalance in treatment effect modifiers across studies may compromise the validity of a network meta-analysis (NMA) assessing progression- free survival (PFS) and overall survival (OS). Therefore, preceding the NMA, evaluation of differences in potential treatment effect modifiers across studies must be assessed. Following the ISPOR-AMCP-NPC Good Practice Task Force guidelines, this study aimed to identify all potential treatment effect modifiers in first-line (1L) advanced renal cell carcinoma (aRCC) by analyzing subgroup efficacy results reported in literature. A systematic literature review (SLR) identified all available randomized controlled trials in 1L aRCC up to September 18, 2019 through MEDLINE, MEDLINE-IN-PROCESS, EMBASE, and Cochrane library. The relative treatment effects (hazard ratios with confidence intervals) were extracted for progression-free survival (PFS) and overall survival (OS) within subgroups. Subgroups reported by at least three publications were selected for further investigation. The presence and direction of a statistically significant or a considerable, but not significant difference (i.e., point estimate for the HR outside the CI of the other group), in relative treatments effects between subgroups were assessed. The SLR identified 23 studies from 109 publications that reported relative treatment effects of PFS or OS, covering 25 different subgroup variables. Only two studies reported a statistically significant difference in the relative treatment effect between subgroups: one study by liver metastases (OS) and one by PD-L1 status (PFS). Considerable differences were reported for 51% (PFS) and 33% (OS) of the subgroup comparisons for ECOG performance status, gender, IMDC/MSKCC risk factor, liver metastasis, number of metastatic sites, prior nephrectomy, PD-L1 status, race, and geographic region. The review indicated that a number of factors are potential treatment effect modifiers in 1L aRCC. Clinical expert consultations are necessary to determine the clinical relevance of these potential effect modifiers.