Performance of Prognostic Models in Patients With Metastatic Urothelial Carcinoma Treated With Enfortumab Vedotin After Platinum‐Based Chemotherapy and Immune Checkpoint Inhibitors: YUSHIMA Study

OBJECTIVES: Enfortumab vedotin monotherapy is now a standard treatment for metastatic urothelial carcinoma. Although several prognostic models have been established, their applicability to patients treated with enfortumab vedotin remains unclear. This study aimed to evaluate the prognostic performance of four existing prognostic models (Bajorin, Bellmunt, Khaki, and MD Anderson Cancer Center score) for their ability to predict clinical outcomes and to refine these models for improved prognostic accuracy. METHODS: This multicenter, retrospective study included 115 patients with metastatic urothelial carcinoma who received enfortumab vedotin following platinum-based chemotherapy and immune checkpoint inhibitors. Prognostic model scores were calculated using clinical variables assessed before initiating enfortumab vedotin. The discriminative ability for survival outcomes was evaluated using Harrell's concordance index. A new prognostic model was developed based on independent prognostic factors identified through multivariable analysis. RESULTS: All four models significantly stratified survival outcomes (log-rank p ≤ 0.001). Among these models, the MD Anderson Cancer Center score demonstrated superior discriminative ability for predicting both progression-free survival and overall survival, with concordance indices of 0.689 and 0.738, respectively. The newly developed model-comprising Eastern Cooperative Oncology Group performance status > 0, serum albumin < 3.5 g/dL, and the presence of liver metastasis-demonstrated favorable discriminative ability, with concordance indices of 0.709 for progression-free survival and 0.732 for overall survival. CONCLUSIONS: Existing prognostic models, particularly those incorporating nutritional markers, remain clinically relevant for patients treated with enfortumab vedotin monotherapy. Their application may assist clinicians in optimizing patient selection and guiding treatment decisions.