Phase II trial of intravesical nanoparticle albumin bound paclitaxel for the treatment of nonmuscle invasive urothelial carcinoma of the bladder after bacillus Calmette-Guérin treatment failure.

ABSTRACT

PURPOSE: Response rates to current second line intravesical therapies for recurrent nonmuscle invasive bladder cancer range between 10% and 30%. Nanoparticle albumin bound (nab-)paclitaxel has increased solubility and lower toxicity compared to other taxanes. Results of the phase I intravesical trial of this compound demonstrated minimal toxicity during dose escalation. We now report the results of a phase II trial to assess efficacy. MATERIALS AND METHODS: This study was an investigator initiated, single center, single arm, phase II trial investigating the use of nab-paclitaxel in patients with recurrent Tis, T1 and Ta urothelial carcinoma in whom at least 1 prior regimen of intravesical bacillus Calmette-Guérin failed. Patients received 500 mg/100 ml nab-paclitaxel administered in 6 weekly intravesical instillations. Efficacy was evaluated with cystoscopy, biopsy, cytology and imaging. If complete response was achieved, patients were treated with full dose monthly maintenance treatments for 6 months. RESULTS: A total of 28 patients were enrolled in the study. Of these patients 10 (35.7%) exhibited a complete response after initial treatment. At 1 year all of these responses remained durable after maintenance therapy. At a mean followup of 21 months (range 5 to 47) 19 of 28 (67.8%) patients retained their bladders without progression or distant metastases. A single patient had progression to muscle invasive disease at radical cystectomy. Treatment related adverse events were noted in 9 of 28 (32.1%) patients and were limited to grade 1 or 2. CONCLUSIONS: Intravesical nab-paclitaxel has minimal toxicity and a 35.7% response rate in patients with nonmuscle invasive bladder cancer and previous bacillus Calmette-Guérin failure. Complete response remained durable at 1 year followup in this heavily pretreated patient population.