iwfal and ocyanblue and any stat experts, some questions:
The premise, as I (mis?)understand it, is that in a life threatening disease in a clinical trial using an event based Kaplan Meier log rank all cause survival analysis, the designed power of a trial assumes a predicted treatment effect and estimates the number of deaths at a given power that may be required to discern the treatment effect in the ITT arm vs. the placebo or standard of care in a control arm. As a result, an interim look with a substantially lower number of events is not expected to be statistically significant, but if the sponsor desires an interim look (for example, to allow a cost avoiding termination of a trial for futility) and continues the trial, an alpha allocation is deducted from the normal statistical significance to account for having a second look, even though bias, post the interim look, should not be a problem in a survival trial. If the treatment effect proves to be statistically significant at the interim look, wouldn't this be possibly due to a greater treatment effect that was used in the power calculation. Theoretically, why would the FDA have any problem allowing earlier termination and the filing of a BLA /NDA due to an unexpectedly efficacious treatment effect?
Another question wrt statistically significant evidence of a combination therapy vs. a standalone therapy where post therapy use of a combination agent was not controlled. At 2006 ASCO, Dr. Small presented an analysis for 9901 that showed that the control patients as a group received some 15% more post Provenge docetaxel than the ITT group, presumably to support the hypothesis that post Provenge docetaxel was not a factor in the increased median survival of the ITT arm as compared to the control arm. Dr. Petrylak’s presentation on 11/10/06 seemingly refutes this ,showing that when compared to the Halabi nomogram reference, the 51 of 147 9901 and 9902a ITT patients receiving some combination of Provenge followed by docetaxel had increased median survival of 13.6 months compared to 5.8 months for the actual over predicted Halabi derived median survival of 5.8 months as reported by Dr. Small for the 9901 ITT group, suggesting that the increase in median survival of the 96 Provenge only 9901 and 9902a ITT patients per the Halabi nomogram might be less than 2 months. The dose and duration of post Provenge docetaxel was not reported. In the lower dosed 9901 and 9902a crossover control group, 31 of the 59 crossover patients who received docetaxel after Provenge appear to have accounted for all the increased median survival. As far as I can see, the Petrylak analysis was not included in the Briefing Material for the FDA Advisory Committee meeting for Provenge. The AC panel was, of course, asked whether there was substantial evidence of Provenge efficacy , rather than substantial evidence of the efficacy of the combination of Provenge and docetaxel. Given the aversion of many AIPC / HRPC patients to docetaxel therapy, one can imagine some patients hesitating to commit to the combination therapy. OTOH, a projected 13.6 months increase in median survival for the combination might persuade more men to start the combination therapy when asymptomatic since there is no evidence that the combination would be effective for sicker symptomatic patients. Given your experience in statistics, how would a statistician decide whether the standalone use of Provenge in asymptomatic AIPC/HRPC would be approved by the FDA vs. the combination therapy of Provenge plus Taxotere, or both or neither? Could this have been one of the reasons for the FDA approvable letter?
AI
