Progressive disease rate as a surrogate endpoint of phase II trials for non-small-cell lung cancer

Background: Although the potential activity of anticancer agents has been t raditionally assessed by the response rate (RR) in phase II trials, there i s an increasing need to identify alternative endpoints to evaluate the effi cacy of novel types of antineoplastic agents such as cytostatic agents. How ever, none of the proposed alternatives have been validated. Design: RR, rate of progressive disease (PD), and median survival time (MST ) were obtained from 44 treatment arms in 42 single-agent phase II trials f or non-small-cell lung cancer (NSCLC). Correlations between these parameter s and their significance in selection of promising drugs were evaluated. Results: The median (range) RR and PD rate per treatment arm were 17% (0%-4 0%) and 41% (8%-93%), respectively. The PD rate correlated more closely wit h MST (correlation coefficient (r) = 0.80, P < 0.001) than did the RR (r = 0.62, P < 0.001). The RR of active agents against NSCLC ranged broadly from 7% to 40%, whereas their PD rates were all 50% or less. In addition, all t reatment arms with a PD rate over 50% had a poor MST of six months or short er. Conclusions: The PD rate was potentially as good an endpoint as RR, and it may be a good candidate for the primary endpoint of phase II trials for nov el types of anticancer agents.