Prognostic factors in patients enrolled in clinical trials of second-line chemotherapy for advanced non-small cell lung cancer (aNSCLC): A pooled analysis of 11 randomized trials

Abstract

8082 Background: Prognostic factors for newly diagnosed aNSCLC have been extensively studied, but less information is available on factors associated with outcome of 2nd-line treatment. Better knowledge of prognostic factors could be helpful to better select patients for further chemotherapy after failure of 1st-line treatment. Methods: The analysis is based on individual patient data collected for 2 meta-analyses on 2nd-line treatment of aNSCLC, one comparing weekly vs every 3 week docetaxel (Di Maio, J Clin Oncol 2007; 25:1377), and one comparing single-agent vs doublet chemotherapy (Di Maio, J Clin Oncol in press). Primary endpoint was overall survival (OS). All analyses were stratified by trial. Results: Out of 11 trials, 2 trials were excluded due to missing information on baseline characteristics considered in multivariate analysis. Of the other 9 trials (1239 pts), 1197 pts (97%) had complete information: 78%/22% males / females, 83%/17% younger / older than 70, 28%/59%/13% Performance Status (PS) 0 / 1 / 2, 18%/82% stage IIIB / IV, 32%/47%/21% squamous / adenocarcinoma / other histology. 84% were pretreated with platin and 44% had obtained objective response (OR) to 1st line treatment. With 956 deaths, median OS was 7.4 months. At multivariate analysis, prognosis was significantly influenced by gender (worse in males vs females, Hazard Ratio [HR] 1.23 [95%CI 1.04–1.45], p=0.01), by PS (worse in PS1 vs PS0, HR 1.36 [1.16–1.59], p=0.0001 and in PS2 vs PS0, HR 3.01 [2.41–3.76], p<0.00001), by tumor histology (better in adenocarcinoma vs squamous, HR 0.85 [0.73–0.99], p=0.04 and worse in other histology vs squamous, HR 1.27 [1.05–1.52], p=0.01), by stage (worse in stage IV vs IIIB, HR 1.28 [1.07–1.53], p=0.007), by type of previous treatment (worse for pts pretreated with platin vs pts not pretreated with platin, HR 1.49 [1.14–1.93], p=0.003), and worse for pts not obtaining OR vs pts obtaining OR during 1st line (HR 1.25 [1.10–1.44], p=0.001). Conclusions: In addition to patient-related (gender, PS) or tumor-related factors (histology, stage), prognosis of pts eligible for 2nd line treatment of aNSCLC is significantly conditioned by previous use of platin and response to 1st line treatment. No significant financial relationships to disclose.