A provocative analysis of first-line strategies in advanced urothelial carcinoma: Reconstructed individual patient data comparisons of EV–pembrolizumab, avelumab maintenance, and nivolumab–cisplatin–gemcitabine from their pivotal phase 3 trials.

Abstract

751 Background: The treatment of advanced urothelial carcinoma has been revolutionized by immune checkpoint inhibitors and antibody–drug conjugates, notably enfortumab vedotin (EV). Phase III trials established the efficacy of EV plus pembrolizumab (P), avelumab maintenance and nivolumab combined with cisplatin–gemcitabine (Nivo-CG). However, direct comparisons among these strategies are lacking. Methods: Individual patient data (IPD) were reconstructed from published Kaplan–Meier curves of JAVELIN Bladder 100, EV-302, and CheckMate 901. Reconstruction quality was assessed with stringent, preplanned criteria to ensure high concordance with original survival data. To emulate the JAVELIN design, analyses involving avelumab applied a 5-month landmark approximating completion of platinum induction, excluding early events. Comparative analyses included Cox regression, piecewise HRs, Royston–Parmar models for time-varying hazard ratios [HR(t)], and restricted mean survival time (ΔRMST). Results: Reconstructed curves closely matched published outcomes. EV–P conferred a strong progression-free survival (PFS) advantage over avelumab maintenance (HR 0.55, 95% CI 0.45–0.67, p<0.001; ΔRMST +4.8 mo at 24 mo, p<0.001). For overall survival (OS), EV–P vs avelumab showed early HR(t) benefit at 6 mo (HR 0.67, 95% CI 0.47–0.96, p=0.03) but minimal ΔRMST difference (<1 mo at 24 mo). In cisplatin-eligible patients, EV–P showed clear OS superiority over Nivo–CG (HR 0.59, 95% CI 0.44–0.78, p<0.001; ΔRMST +3.0 mo at 28 mo). By contrast, avelumab maintenance vs Nivo–CG showed early OS differences favoring avelumab, likely reflecting patients who had progressed but remained alive at the 5-mo landmark. HR(t) approached 1 beyond 12 mo and remained stable thereafter, while ΔRMST indicated modest but consistent survival gains for avelumab (+1 to +2 mo at 12–24 mo). Conclusions: This exploratory and hypothesis generating reconstructed-IPD analysis demonstrates the clear OS superiority of EV–P over concomitant chemo-immunotherapy and its robust PFS advantage versus maintenance immunotherapy. However, the OS gain over sequential platinum–avelumab is modest, suggesting comparable long-term clinical outcomes in selected patients. Notably, avelumab maintenance achieved OS comparable to Nivo–CG indicating that maintenance immunotherapy may offer similar benefit without upfront chemo-immunotherapy.