246 Background: Designing clinical trials after BCG failure can be problematic without a reasonable control arm. Thus, there is a need for early identification of BCG failure relevant to clinical trial design which allows one arm to continue on BCG. Here we present a definition of BCG failure based on FISH profile at 6th week of induction BCG that would facilitate such trials. Methods: The definition presented is based on findings from our IRB-approved, prospective clinical trial where Urovysion FISH assays were performed serially during the normal course of BCG therapy (SWOG protocol) at various time points (pre-BCG, 6 weeks, 3 and 6 mo.). Herein we incorporated the FISH analysis at 6 weeks in patients with a negative 3 month cystoscopy and correlated the result with recurrence and progression rates at 24 months. A novel definition of BCG failure was proposed by focusing the analysis on 84 patients with high grade disease (cTa: 33, cT1: 44, cTis 7). The 6 week FISH was selected as this would allow the control arm to proceed with BCG maintenance in a timely fashion, taking into account the time required for obtaining FISH results, registration of patients, and randomization. Results: Of the 36 patients with a positive FISH at 6 weeks (and no tumor at 3 months), 17 recurred (PPV = 47%) and 11 progressed (PPV = 31%), while among those with a negative FISH, 5 recurred (NPV = 90%) and 4 progressed (NPV = 92%) (p<0.001). Kaplan Meier estimates of recurrence free survival with a positive 6 week FISH were 67% and 52%, and for progression free survival were 75% and 52%, at 1 and 2 years respectively. Therefore, if patients with a positive 6 week FISH (and negative 3 month cystoscopy) were considered as "molecular" BCG failures, a prospective clinical trial with 101 patients in the control (i.e. BCG maintenance) and experimental arms to detect a 20% difference in recurrence rates (α=0.05) with 80% power. Conclusions: Using the FISH status of patients at the 6 week interval on BCG therapy would allow a clinical trial design that incorporates one arm with continued BCG therapy, while enriching for events of interest, namely recurrence and progression. This would allow meaningful comparisons while negating any ethical concerns regarding a lack of “standard treatment” in BCG failure studies.
Improved amputation-free survival in unreconstructable critical limb ischemia and its implications for clinical trial design and quality measurement
