IMPORTANCE Locoregional failure for patients with locally advanced bladder cancer (LABC) after radical cystectomy (RC) is common even with chemotherapy and is associated with high morbidity and mortality. Adjuvant radiotherapy (RT) can decrease locoregional failure but has not been studied in the chemotherapy era. OBJECTIVE To investigate if adjuvant sequential RT plus chemotherapy can improve locoregional recurrence-free survival (LRFS) compared with adjuvant chemotherapy alone. DESIGN, SETTING, AND PARTICIPANTS Arandomized phase 3 trialwas opened to compare adjuvant RT vs sequential chemotherapy plus RT after RC for LABC, but a third armwas added later as a randomized phase 2 trial to compare chemotherapy plus RT vs adjuvant chemotherapy alone, an emerging standard. The intent-to-treat phase 2 trial reported herein enrolled patients from December 2002 to July 2008.Datawere analyzed from August 3, 2015, to January 6, 2016. Routine follow-up and surveillance pelvic computed tomographic (CT) scans every 6 months during the first 2 yearswere performed. The settingwas an academic center. Patients with bladder cancer 70years or younger having 1 or more risk factors (pT3b, grade 3, or positive nodes) with negative margins after radical cystectomy plus pelvic lymph node dissectionwere eligible.Patientshad Eastern Cooperative Oncology Groupper for mancestatusof0to2, noevidence of distantmetastases on CT scan of the abdomen and pelvis or on chest imaging, and adequate renal, hepatic, and hematologic function. Ninety-one percent (109 of 120) had pT3 disease. INTERVENTIONS Chemotherapy plus RT included 2 cycles of gemcitabine (1000mg/m2 intravenously on days 1, 8, and 15) and cisplatin (70mg/m2 intravenously on day 2) before and after RT to 4500 cGy in 150 cGy twice-daily fractions over 3 weeks using 3-dimensional conformal techniques. Chemotherapy alone included 4 cycles of gemcitabine and cisplatin. MAIN OUTCOME AND MEASURE Locoregional recurrence-free survival. RESULTS The chemotherapy plus RT arm accrued 75 patients, and the chemotherapy-alone arm accrued45patients, withaweightedrandomizationtospeedaccrual.Fifty-threepercent(64of120) hadurothelialcarcinoma, and46.7%(56of120)hadsquamouscellcarcinomaorother.Thearmswere balancedexceptforage(median, 52vs55years; P = .04)andtumorsize(mean, 4.9vs5.8cm; P < .01), both favoring chemotherapy plus RT. Two-year outcomes and overall adjusted hazard ratios (HRs) forchemotherapyplusRTvschemotherapyalonewere96%vs69%(HR, 0.08; 95%CI, 0.02-0.39; P < .01) for LRFS, 68%vs 56%(HR, 0.53; 95%CI, 0.27-1.06; P = .07) for disease-free survival, and 71% vs 60% (HR, 0.61; 95% CI, 0.33-1.11; P = .11) for overall survival (OS). Five patients (7%) had RT-associated late grade 3 gastrointestinal tract adverse effects in the chemotherapy plus RT arm. CONCLUSIONS AND RELEVANCE Adjuvant chemotherapy plus RTwas reasonablywell tolerated andwas associated with significant improvements in LRFS and marginal improvements in disease-free survival vs chemotherapy alone in LABC. The addition of adjuvant RT should be considered for LABC. This regimenwarrants further study in phase 3 trials. TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT01734798
