ABSTRACT
Standard phase II studies are used to screen new regimens for activity and to decide which ones should be tested further. To screen regimens efficiently, the decisions generally are based on single-arm studies using short-term endpoints, typically tumor response in cancer studies, in limited numbers of patients. The problem is formulated as a test of the null hypothesis H0: p p0 vs. the alternative hypothesis HA: p pA, where p is the probability of response, p0 is the probability that if true would mean that the regimen was not worth studying further, and pA is the probability that if true would mean it would be important to identify the regimen as active and to continue studying it. Typically, p0 is a value at or somewhat below the historical probability of response to standard treatment for the same stage of disease, and pA is typically somewhat above.
