A role of multimodality bladder-preserving therapy in patients with muscle-invasive bladder cancer plus hydronephrosis with or without pelvic nodal involvement.

ABSTRACT

BACKGROUND/PURPOSE: To retrospectively evaluate the failure patterns of multimodality bladder-preserving therapy in patients with muscle-invasive bladder cancer. METHODS: Patients with muscle-invasive bladder cancer underwent maximal transurethral resection of bladder tumor and induction chemotherapy, followed by concurrent chemoradiotherapy (CCRT). Radiotherapy was given with 45 Gy to the pelvis, 50.4 Gy to the bladder, and 64.8 Gy to the tumor bed. Three protocols of trimodality treatment were used Protocol A, three cycles of cisplatin and fluorouracil (CF), followed by CCRT with 6 weekly cisplatin; Protocol B, three cycles of weekly paclitaxel plus CF, followed by CCRT with 6 weekly paclitaxel and cisplatin; Protocol C, three cycles of gemcitabine and cisplatin, followed by CCRT with 6 weekly cisplatin. Interval cystoscopy confirmed complete response (CR) after induction chemotherapy and 40-50 Gy of radiotherapy. Patients without CR were referred for salvage cystectomy. RESULTS: A total of 60 patients were enrolled, including 11 patients with unfavorable factors defined as hydronephrosis and/or pelvic nodal involvement. After a median follow-up of 86.7 months, the 5-year overall, progression-free, and bladder preservation-specific survival rates were 76.3%, 62.9%, and 71.5%, respectively. Three patients underwent salvage cystectomy for invasive bladder recurrence. Of 45 surviving patients, 42 patients (93.3%) retained functioning bladders. Patients with unfavorable factors had significantly lower metastasis-free survival (p=0.002), but not bladder preservation-specific survival (p=0.25). CONCLUSION: With trimodality treatment involving visually complete transurethral resection of bladder tumor, cisplatin-based induction chemotherapy, and CCRT, patients with unfavorable factors maintained satisfactory bladder preservation but not systemic control.