Impact of gemcitabine (G) on the activity of first-line chemotherapy in non-small cell lung cancer (NSCLC): A meta-analysis

Abstract

7138 Background: A two-drug platinum-based regimen in which cisplatin or carboplatin is combined with a third-generation agent (i.e. paclitaxel, vinorelbine, docetaxel, or G) is the standard first-line treatment for NSCLC patients with good performance status. Encouraging results have recently been reported for nonplatinum regimens composed of two third-generation drugs. Methods: To assess the impact of G on the activity of first-line chemotherapy in NSCLC, we carried out a meta-analysis on data from 4,362 NSCLC patients who were enrolled in 11 randomized trials comparing a G-containing vs. G-free new generation regimens. We constructed 2x2 tables using response to treatment data. For trials with more than one eligible G-free comparator arm, individual comparisons between the G-based treatment arms and each of the comparator arms were analyzed. A general variance-based method was used to estimate the pooled odds ratio (OR) and 95% confidence interval (CI). We assessed for heterogeneity among the trials based on standard methods. Results: Sixteen comparisons contributed to this analysis. G-containing regimens included: G+cisplatin (894), G+docetaxel (565), G+paclitaxel (200 patients), G+vinorelbine (157), G+carboplatin (49). G-free regimens included: vinorelbine+cisplatin (866), carboplatin+paclitaxel (539), docetaxel+cisplatin (494), cisplatin+paclitaxel (439), vinorelbine+carboplatin (159). Comparing G-containing vs. G-free regimens, the OR for progression was 0.867 (CI 95% 0.770–0.977; p = 0.019), with heterogeneity chi-square 11.639 (p = 0.71). No significant difference was observed for complete (OR 0.909, CI 95% 0.556–1.487; P = 0.707) and overall (complete + partial) response (OR 0.987, CI 95% 0.881–1.106; P = 0.819). Conclusions: These data demonstrate that the progression of disease is more likely in patients treated with G-free doublets. Further analyses are required to address whether disease control (objective response + stable disease) is associated with a survival benefit and may therefore be used as a surrogate end point for survival in chemotherapy trials of NSCLC. No significant financial relationships to disclose.