Systematic literature review (SLR) and network meta-analysis (NMA) of first-line therapies (1L) for locally advanced/metastatic urothelial carcinoma (la/mUC).

Abstract

570 Background: Standard of care (SOC) for 1L la/mUC is gemcitabine plus cisplatin (GC) or carboplatin (GCa), but the landscape is evolving with new therapies emerging. To compare outcomes of other approved/investigational 1L regimens with SOC in the context of recently published data on newer therapies, we updated a previously reported SLR/NMA of phase 2/3 randomized control trials. Methods: The SLR was conducted in line with PRISMA and NICE guidelines (01/2000-05/2020; updated 06/2020-06/2021). Three networks were formed: cisplatin (cis)-eligible/mixed eligibility; cis-ineligible (strict; studies including cis-ineligible patients only); and cis-ineligible (wide; expanded to also include study arms with an investigator’s choice of carboplatin in KEYNOTE-361, IMvigor130, and DANUBE). Comparative efficacy and safety were assessed under a Bayesian framework. Overall survival (OS) and progression-free survival (PFS) with 1L la/mUC regimens vs SOC (GC/GCa) are reported. Results: Among 2,312 citations identified, 55 unique trials were selected for data extraction. Of these, the NMA included 11 studies in the cis-eligible/mixed, 6 in the cis-ineligible (strict), and 8 in the cis-ineligible (wide) network. The NMA excluded therapies that were not effective or adopted in clinical practice; 6 maintenance trials were excluded due to differences in design precluding comparisons. Median OS in the SOC arms was 13.2 mo (95% confidence interval [CI] 12.4-14.0) for cis-eligible/mixed, 9.7 mo (95% CI 6.7-12.8) for cis-ineligible (strict), and 12.0 mo (95% CI 10.4-13.5) for cis-ineligible (wide); median PFS was 6.6 mo (95% CI 6.3-6.9) for cis-eligible/mixed and 5.6 mo (95% CI 5.0-6.3) for both cis-ineligible strict and wide. OS and PFS were similar to SOC across therapies in each network: hazard ratios (HR) ranged 0.7-1.4 for OS for cis-eligible/mixed, 0.9-1.4 for cis-ineligible (strict), and 0.8-1.4 for cis-ineligible (wide) (Table); HR for PFS ranged 0.5-1.6 for cis-eligible/mixed and 0.8-1.1 for both cis-ineligible strict and wide networks; all credible intervals (CrI) crossed or were close to 1. Conclusions: In this updated SLR/NMA, survival outcomes were similar and remained poor among established and emerging 1L la/mUC therapies, despite inclusion of recent trial data. This further highlights the unmet need in this population.[Table: see text]