Real‐World Characteristics and Early Outcomes of Enfortumab Vedotin Plus Pembrolizumab in Advanced Urothelial Carcinoma
Takuya Tsujino, Keita Tamura, Rikiya Taoka, Makito Miyake, Keita Kobayashi, Yutaro Sasaki, Kohei Ogawa, Shuichi Tatarano, Satoshi Katayama, Minoru Kato, Yohei Abe, Yuki Oda, Koichiro Wada, Junya Furukawa, Motoo Araki, Hideki Enokida, Koji Shiraishi, Keiji Inoue, Teruo Inamoto, Haruhito Azuma,ABSTRACT
Background
Enfortumab vedotin (EV) plus pembrolizumab (EVP) has become a standard first‐line regimen for unresectable locally advanced or metastatic urothelial carcinoma. Further characterization of clinical heterogeneity and toxicity may complement existing Japanese real‐world data.
Methods
We conducted a multicenter prospective study across 11 Japanese institutions. Consecutive patients who initiated EVP for unresectable locally advanced or metastatic urothelial carcinoma between October 2024 and December 2025 were included. Patient characteristics, treatment exposure, response, adverse events, and short‐term survival outcomes were descriptively analyzed.
Results
A total of 172 patients were included. Median age was 75 years, 73.8% had creatinine clearance < 60 mL/min, and 76.7% were classified as cisplatin‐unfit using modified Galsky criteria. Upper tract urothelial carcinoma accounted for 35.5% of cases. Median follow‐up was 7.1 months. Early EV dose reduction was recorded in 44.8% of patients, and discontinuation of at least one agent occurred in 61.0%. Among 158 response‐evaluable patients, the best overall objective response rate was 62.7% and the disease control rate was 82.3%; corresponding all‐treated estimates were 57.6% and 75.6%, respectively. Skin reaction was the most common adverse event, while pneumonitis was a major cause of treatment discontinuation.
Conclusions
In Japanese real‐world practice, EVP was frequently used in older, renally impaired, cisplatin‐unfit patients. Response was clinically meaningful among evaluable patients, but interpretation requires caution because of short follow‐up and denominator‐dependent response estimates.