New England Section of the American Urological Association

NEAUA Home NEAUA Home Past & Future Meetings Past & Future Meetings

Back to 2025 Abstracts


Modeling survival outcomes of CHECKMATE-274 with contemporary standard of care control-arm treatment: A simulation study
Keervani Kandala, BS1, Syed N. Rahman, MD2, Spencer L. James, MD, MPH3, Fady Ghali, MD2.
1Yale University School of Public Health, New Haven, CT, USA, 2Yale University School of Medicine, New Haven, CT, USA, 3Docere.ai, Seattle, WA, USA.

Background: CHECKMATE-274, reporting improved overall survival(OS) with adjuvant checkpoint inhibitor(CPI) in high-risk urothelial carcinoma(UC) suffers from a common challenge: rapidly evolving standards of care (SOC) resulting in obsolete comparisons, limiting the applicability of trial outcomes. We sought to simulate CHECKMATE-274 using contemporary SOC treatments to assess adjuvant CPI in UC more accurately. Methods: Using our validated methodology, Gated Counterfactual Simulation, reconstructed individual patient data from Kaplan-Meier curves (K-M) as published in CHECKMATE-214, EV-302, and EV-301 were extracted. Attributable survival benefit between treatment arms within EV302 and EV301 were calculated. Survival benefit estimates from EV302 were incorporated to control-arm progressors of CHECKMATE-274 in a weighted fashion - these measures estimate counterfactual OS outcome for the CHECKMATE-274 control group using EV302 outcomes to approximate survival had they received contemporary SOC upon progression. This was repeated with adjustment of the CHECKMATE-274 treatment-arm using EV301 estimates. Results: Within CHECKMATE-274, 39.3% of control-arm patients were not reported to have received CPI or EV upon progression. When the trial is simulated with contemporary SOC for the control-arm, osHR was 0.99(CI0.78-1.25,p=0.90,Figure1A). Within the treatment arm of CHECKMATE-274, 21.2% did not receive enfortumab vedotin (EV). When adjusting both control- and treatment-arms, osHR was 0.89(CI0.70-1.13,p=0.34,Figure1B). Conclusions: In a simulation study of adjuvant nivolumab in UC, we find that when the control arm is treated with a contemporary standard of care and when the treatment is similarly adjusted, no survival benefit is observed with the intervention. Our findings suggest that survival benefits of adjuvant nivolumab may be limited in the current therapeutic era.



Back to 2025 Abstracts