Phase III bladder cancer chemoprevention trial with fenretinide.

A148

Purpose : To evaluate the efficacy and toxicity of fenretinide in preventing tumor recurrence in patients with transitional cell carcinoma (TCC) of the bladder. Experimental Design : We conducted a multi-center phase III randomized placebo-controlled trial in patients with stage Ta, Tis or T1 non-muscle-invasive bladder TCC treated with transurethral resection and, adjuvant intravesical bacillus Calmette-Guerin for intermediate- or high-risk tumors. Patients were randomized to 12 months of fenretinide (200 mg orally/day) or matched placebo. Cystoscopy and cytology were performed at 3 month intervals for 12 months (or until histologically confirmed recurrence), and then until the study end every three months by phone to determine bladder cancer status and survival. The primary clinical endpoint was the one-year recurrence rate. Study power was 80% to detect a recurrence rate of 25% (fenretinide) versus an event rate of 44.5% (placebo) in a sample size of 160 patients (80 per arm). Results : The Data Safety and Monitoring Board recommended early closure because of interim data showing a low likelihood that the results could become significantly different if the original accrual goal and follow-up of currently enrolled patients were completed. 149 Patients were enrolled from all participating sites--69% with solitary and 23% with multifocal non-BCG-treated Ta G1, 2 tumors and the remainder with intermediate- or high-risk BCG-treated tumors. Of 134 patients evaluable for the one-year recurrence rate, this rate was 33% in the placebo arm, 32% in the fenretinide arm. The median times to recurrence were 1225 days (3.4 years) (placebo) and not reached (fenretinide) (log rank p = 0.94; Wilcoxon p = 0.73). The hazard rate for recurrence in the fenretinide arm was 0.99 (95% CI 0.74-1.32; p = 0.94). Grade 3 and 4 toxicities occurred in 6 patients on fenretinide versus 5 on placebo. Conclusions: Although well tolerated, fenretinide was not effective in reducing the recurrence rate of bladder TCC patients. [The first two authors contributed equally to this report.]