To analyze the effectiveness of adjuvant polychemotherapy after radical cystectomy for nonorgan-confined transitional cell bladder cancer (Stages pT3b, pT4a, and/or pN1 or pN2). Of 166 consecutive patients undergoing cystectomy at two institutions from 1987 to 1993, 80 received adjuvant polychemotherapy with methotrexate, vinblastine, and cisplatin plus doxorubicin (MVAC) or epirubicin (MVEC), whereas 86 had cystectomy only. The patients were evaluated for relapse-free survival and length of progression-free interval on the basis of follow-up data obtained in 1995 and 1996. Kaplan-Meier analysis revealed a significantly higher progression-free rate for patients after adjuvant chemotherapy ( P=0.0002, log-rank test). With and without adjuvant chemotherapy, prognosis declined in a stepwise manner, depending on the extent of lymph node involvement. Nevertheless, the superior prognosis of the chemotherapy group could be demonstrated at each lymph node stage. Of the 166 patients, 49 had initially entered a prospective trial comparing adjuvant with no adjuvant treatment. That study was discontinued in December 1990 after an interim analysis revealed a significant prognostic advantage in favor of the 26 patients randomized to receive chemotherapy compared with the 23 control patients. Current follow-up data continue to demonstrate a significant improvement in progression-free survival in favor of patients randomized to receive adjuvant chemotherapy ( P=0.0040). The follow-up period of patients living free of disease ranges from 58 to 96 months. Adjuvant chemotherapy with MVAC/MVEC leads to significant prolongation of relapse-free survival and improvement of the definitive cure rate after radical cystectomy for locally advanced transitional cell carcinoma of the urinary bladder.