A combined proof of concept and dose finding study with multiple endpoints: A Bayesian adaptive design in chronic prostatitis/chronic pelvic pain syndrome

: There is a need for identifying effective drugs or terminating ineffective drugs as early as possible to optimize efficient and cost effective drug development. The aim of the proposed trial was to simultaneously establish Proof of Concept (PoC) and dose finding (DF) for a new drug with a novel mode of action in a new indication. We simulated and executed an adaptive allocation design to investigate the effects of a drug on male patients suffering from chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS). This manuscript describes the clinical trial simulations and primary analysis results. A Bayesian adaptive allocation procedure was employed to allocate patients to treatment using a normal dynamic linear model. The study was to stop early for efficacy if the probability of a clinically significant difference between an experimental arm and placebo was at least 90%. The study was to stop for futility if the probability that the maximum effective dose was better than placebo by at least the futility difference was less than 20%. During the execution phase the study was stopped early, that is 32% less than planned maximum sample size, due to futility. The final results confirmed that the predefined stopping rules were met. In conclusion, the simulations showed that, if the drug was effective, this adaptive design could accomplish both the goals of PoC and DF. However, the study stopped early for futility in line with the simulation predictions for stopping. This resulted in the early stopping of a trial recruiting patients on ineffective treatment.