nac-VIRADS做為肌肉侵襲膀胱癌接受 Enfortumab vedotin 合併 Pembrolizumab前導性治療後之準確性研究

張家郡¹、張兆祥¹、林維卿²

¹中國醫藥大學附設醫院 泌尿部，²中國醫藥大學附設醫院 醫學影像部

Validation of the Neoadjuvant Chemotherapy Vesical Imaging-Reporting and Data System (nacVI-RADS) in Patients with Muscle-Invasive Bladder Cancer Receiving Neoadjuvant Enfortumab Vedotin and Pembrolizumab

Jia-Jyun Jhang¹, Chao-Hsiang Chang¹, Wei-Ching Lin²

¹Department of Urology, China Medical University Hospital, Taichung, Taiwan; ²Department of Radiology, China Medical University Hospital, Taichung, Taiwan

Purpose: The Neoadjuvant Chemotherapy Vesical Imaging-Reporting and Data System (nacVI-RADS) is increasingly utilized to evaluate treatment response following neoadjuvant chemotherapy or immunotherapy in muscle-invasive bladder cancer (MIBC). While the combination of Enfortumab Vedotin (EV) and Pembrolizumab has demonstrated paradigm shift in metastatic urothelial carcinoma (UC), this regimen is also rapidly being integrated into the neoadjuvant setting. Our study aims to validate the diagnostic accuracy of nacVI-RADS in predicting pathological response for MIBC patients receiving neoadjuvant EV plus Pembrolizumab.

Materials and Methods: We retrospectively evaluated MIBC patients treated with neoadjuvant EV plus Pembrolizumab at a tertiary medical center between September 2023 and March 2026. Multiparametric bladder MRI was performed at two time-points: prior to diagnostic transurethral resection of bladder tumor (TURBT) and following 3 cycles of neoadjuvant therapy. The nacVI-RADS scores were dichotomized using two distinct thresholds for performance assessment: (1) scores 1–2 vs. 3–5 to predict pathological complete response(pCR) (ypT0 vs. ypT≧1), and (2) scores 1–3 vs. 4–5 to predict downstaging to non-muscle invasive disease (ypT0/is/a/1 vs. ypT≧2). Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated by correlating MRI interpretations with the final radical cystectomy (RC) pathology. Exclusion criteria included incomplete imaging protocols or failure to undergo definitive surgery.

Results: 10 patients were included (8 males, 2 females; median age 70 years). At baseline, 3 patients (30%) had cT2 disease, and 7 (70%) had ≥cT3 disease. Patients received a median of 4 cycles of neoadjuvant EV plus Pembrolizumab prior to RC. Pathological assessment revealed pCR(ypT0) in 70% (n=7) of patients, while 20% (n=2) had ypT3 and 10% (n=1) had ypT4 disease. For predicting pCR (threshold nacVI-RADS ≤2), the sensitivity, specificity, PPV, NPV, and accuracy were 71.43%, 100%, 100%, 60%, and 80%, respectively. Notably, for predicting residual muscle-invasive disease (threshold nacVI-RADS ≥4), nacVI-RADS demonstrated 100% sensitivity, specificity, PPV, NPV, and accuracy.

Conclusions: nacVI-RADS demonstrates high diagnostic performance in predicting treatment response for MIBC patients receiving neoadjuvant EV plus Pembrolizumab, particularly in identifying residual muscle-invasive disease. Our findings suggest that nacVI-RADS is a reliable tool for monitoring response to this novel treatment paradigm. Notably, our protocol involved performing the restaging MRI prior to the final cycle of neoadjuvant therapy, which may account for the lower sensitivity and NPV observed in predicting pCR. Further prospective validation in larger cohorts is warranted to optimize imaging timing and clinical integration.