Adjuvant vs switch-maintenance checkpoint inhibition in urothelial carcinoma: A systematic review and meta-analysis of randomized trials through July 2025.

732 Background: Immune checkpoint inhibitors (ICIs) are used in urothelial carcinoma (UC) as adjuvant therapy after radical cystectomy or nephroureterectomy and as switch-maintenance after response or stable disease to platinum chemotherapy. Randomized controlled trials (RCTs) show setting-specific benefits, but pooled evidence is limited. Methods: Following PRISMA guidelines, PubMed, Embase, Cochrane CENTRAL, and ASCO/ESMO/ASCO Genitourinary (GU) proceedings were searched through July 31, 2025. Eligible studies were phase II/III RCTs of programmed death-1 (PD-1) or programmed death-ligand 1 (PD-L1) inhibitors in adjuvant or switch-maintenance UC with placebo, observation, or best supportive care (BSC) controls. Outcomes were disease-free survival (DFS; adjuvant) and overall survival (OS; both). Grade ≥3 adverse events (AEs) were summarized. Random-effects meta-analyses were performed. Results: Five RCTs (N=3,028) were included: adjuvant nivolumab (CheckMate 274), pembrolizumab (AMBASSADOR), atezolizumab (IMvigor010), and switch-maintenance avelumab (JAVELIN Bladder 100) and pembrolizumab (HCRN GU14-182). Adjuvant pooled DFS favored ICIs (HR 0.77, 95% CI 0.67–0.90). Pooled OS was not significant (HR 0.87, 95% CI 0.75–1.01): CheckMate 274 trended positive; AMBASSADOR and IMvigor010 were neutral. In maintenance, pooled OS favored ICIs (HR 0.71, 95% CI 0.59–0.87), driven by JAVELIN. Biomarkers showed PD-L1 enrichment in CheckMate 274 (DFS HR 0.52) and JAVELIN (OS HR 0.56), DFS benefit regardless of PD-L1 in AMBASSADOR, and circulating tumor DNA (ctDNA) stratification in IMvigor010 (ctDNA-positive OS HR 0.59). Grade ≥3 AEs were more common with ICIs. Conclusions: Adjuvant ICIs reduce recurrence but OS benefit remains inconsistent, likely due to design, cross-over, and later ICI use. Switch-maintenance provides a consistent OS advantage and is standard of care. PD-L1 and ctDNA may refine selection. These findings highlight differing maturity of evidence between perioperative strategies and guide sequencing in UC. RCTs of adjuvant and switch-maintenance ICI in UC. Trial Setting N Intervention Comparator Primary Endpoint(s) Key Outcomes CheckMate 274 Adjuvant 709 Nivolumab Placebo Disease-free survival (DFS; ITT and PD-L1 ≥1%) DFS HR 0.70; interim overall survival (OS) HR 0.76 AMBASSADOR / A031501 Adjuvant 702 Pembrolizumab Observation Co-primary: DFS and OS DFS HR 0.73; interim OS HR 0.98 IMvigor010 Adjuvant 809 Atezolizumab Observation DFS DFS HR 0.89; OS HR 0.91 (ITT); circulating tumor DNA (ctDNA)-positive OS HR 0.59 JAVELIN Bladder 100 Switch-maintenance 700 Avelumab Best supportive care (BSC) OS (overall and PD-L1+) OS HR 0.69; PD-L1+ OS HR 0.56 HCRN GU14-182 (phase II) Switch-maintenance 108 Pembrolizumab Placebo Progression-free survival (PFS) PFS HR 0.65; median OS 22.0 vs 18.7 mo