e13004 Background: The traditional endpoint of phase II studies, response rate (RR), may not be appropriate for agents that inhibit growth but do not shrink the tumor. For these agents, early progressive disease (EPD), i.e., progression at first tumor measurement after treatment, may provide an additional measure of drug activity. Using a previously developed set of stopping rules employing both RR and EPD (dual endpoint stopping rule, DESR), the outcomes of completed phase II trials was compared with the conventional rules of Fleming and Gehan. Methods: We previously created a computer model (JCO 2008, 3715-3720) that determined patient RR and EPD thresholds needed to achieve the desired power and alpha error for a single-arm two-stage study. The EPD rate of interest was set as ≥ 0.4 and the EPD rate of disinterest as ≥ 0.6. The performance of this model in assessing agent activity was evaluated using a publically available set of phase II trial results (Dent et al, JCO, 2001, pp785-91), which were originally designed based on either the Fleming (16 trials) or Gehan (23 trials) designs. Results: Of the 10 trials stopped by the Fleming design at stage I, the DESR design stopped four. Of the 6 studies permitted to accrue to stage II by Fleming, one was stopped by investigators, while the Fleming design rejected the null hypothesis (Hnul) at the end of the second stage in 2 of the remaining 5 while the DESR design rejected Hnul in all 5 trials. All differences were due to high EPD. For studies conducted under the Gehan design, eight were stopped at stage I for acceptance of Hnul by both Gehan and the DESR. For the 15 studies accruing to stage II according to the Gehan design, Hnul was rejected in eight by both Gehan and DESR designs. However, the DESR would have stopped the other seven at the first stage on account of high EPD rates. Conclusions: In this study set of trials, the DESR was more likely than the Fleming rule to reject Hnul. Conversely, the DESR rejected Hnul for the same drugs as Gehan but was more likely to recommend stopping at stage I. The EPD rates of interest heavily influence the stopping behavior of the DESR. Alternate EPD values will be assessed. No significant financial relationships to disclose.