The determinatio of the number of patients required in a preliminary and a follow-up trial of a new chemotherapeutic agent.

Abstract Double sampling plans have been used to determine sample sizes for the conduct of a preliminary and a follow-up trial of a new drug. The initial sample is determined so that, if all the patients do not respond, the drug will be dropped with a known chance of a rejection error. If one or more of the patients in the first sample respond, the size of the second sample is determined so that the true effectiveness of the drug is estimated with approximately a specified precision, as defined by the standard error. The probabilities of having a second sample are calculated for drugs with the assumption of various true percentages of effectiveness. Some discussion is given of where such sampling plans might be useful.