Sample size calculation for a proof of concept study

Sample size calculation is vital for a confirmatory clinical trial since the regulatory agencies require the probability of making Type I error to be significantly small, usually less than 0.05 or 0.025. However, the importance of the sample size calculation for studies conducted by a pharmaceutical company for internal decision making, e.g., a proof of concept (PoC) study, has not received enough attention. This article introduces a Bayesian method that identifies the information required for planning a PoC and the process of sample size calculation. The results will be presented in terms of the relationships between the regulatory requirements, the probability of reaching the regulatory requirements, the goalpost for PoC, and the sample size used for PoC.