Diagrams of phase II/III designs. Design A: including a concurrent control in both phases, using a short-term binary endpoint (eg, ORR) in phase II to select the optimal dose and using a long-term time-to-event endpoint (eg, PFS or OS) in phase III to evaluate the therapeutic effect of the treatment. In this example, dose 2 is selected as the optimal dose at the end of phase II, and patients are randomly assigned to dose 2 and a concurrent control arm. Design B: a modification of design A, including a control only in phase III. Design C: similar to design A but the short-term endpoint (eg, ORR) is used for phase II dose selection and phase III final analysis. Design D: a simplified version of design C without control in both phases. ORR = objective response rate; OS = overall survival; PFS = progression-free survival.