Comparison of progression-free survival (PFS) and tumor response as endpoints for predicting overall survival (OS) in untreated extensive-stage small cell lung cancer (ED-SCLC): Findings based on North Central Cancer Treatment Group (NCCTG) trials

Abstract

8085 Background: Historically, tumor response has been the primary endpoint in phase II (P2) trials in ED-SCLC. We investigated the suitability of alternate PFS based endpoints to predict OS as early evidence of efficacy in the P2 setting. Methods: Individual patient (pt) data from 942 pts from 11 previously untreated ED-SCLC P2 and phase III (P3) platinum- or paclitaxel-based treatment trials were pooled. Best response (BR), response confirmed (RC), objective status at 16 weeks (RR16), and PFS rate at 5 and 6 months were considered. Percent agreement (PA) and kappa (k) for PFS5, PFS6, BR, RC, and RR16 with OS at 12 months (OS12) was calculated on a per-pt basis and predictive utility was assessed using the area under the receiver operating characteristic (A- ROC) curve in logistic models. Cox models were used to assess the prognostic impact of the endpoints on subsequent survival, using landmark analysis. Results: The median OS and PFS were 9.6 m and 5.5 m, respectively. PFS5 and PFS6 had the highest PA, k, and A-ROC values, and were predictive of subsequent survival in the landmark analysis (p <0.0001; c-statistics ≥ 0.60). While RR16 and BR were significantly associated with subsequent survival (p<0.0001, c-statistics of 0.61 and 0.57, respectively) the PA, k, and A-ROC values were lower. Conclusions: PFS rate at 5 and 6 months is more predictive of 12-month OS and subsequent survival than tumor response in untreated ED-SCLC. PFS based endpoints should be routinely used as primary endpoints in P2 trials within ED-SCLC. [Table: see text] No significant financial relationships to disclose.