Phase II study of biweekly gemcitabine as first line therapy in CIS of the bladder: What does an aborted trial tell us?

OBJECTIVE: The objective of this study was to challenge the activity of a promising intravesical drug (gemcitabine), administered at an intensive regimen (2,000 mg twice a week for 6 weeks) in treatment naïve CIS of the bladder and to observe side-effects. The statistical design was conceived to provide sufficient information through the enrolment of a low number of patients. MATERIAL AND METHODS: Primary, secondary, and concurrent CIS with no prior intravesical therapy were eligible. Treatment schedule: 2000 mg gemcitabine in 50 ml saline, unbuffered, for 1 hour twice a week for 6 weeks. Complete response (CR) = negative cytology and negative cystoscopy + bladder mapping at 3 months. Failure (FA) = all other situations. Side-effects were recorded according to Common Terminology Criteria for Adverse Events (CTCAE). Study design: A 3-stage design. After testing the drug on 11 patients in the first stage, the trial had to be terminated if 3 or fewer CRs. After testing the drug on 21 patients in the first and second stages, the trial had to be terminated if 7 or fewer CRs. After testing the drug on 32 patients in all 3 stages, the drug was considered active in case of >12 CRs. Survival data up to 4 years from trial closure were collected. RESULTS: The study proceeded to stage II since 5/11 responded at stage I but it was stopped after including 18 patients due to side-effects; 6/18 had primary CIS, 7 had secondary CIS, and 5 concomitant CIS; 6/18 (33.3%) had grade 3 side-effects (4 G3 cystitis, 3 G3 leucopenia), leading to stopping the treatment in all 6 cases. CRs were observed in 8/18 patients (44.4%), FA in 10/18 (55.5%). Median overall survival (OS) was 44 months with a 4-year cancer-specific survival of 100%. CONCLUSION: Biweekly gemcitabine as first line treatment for CIS led to excess toxicity and suboptimal activity. Due to the peculiar statistical design, a negative response was generated enrolling a low number of patients. The absolute 4-year CSS suggests that no window of opportunity for disease cure may have been lost by assessing a new, non standard, treatment for CIS.