Six versus fewer planned cycles of first-line platinum-based chemotherapy for non-small-cell lung cancer: a systematic review and meta-analysis of individual patient data

Background Platinum-based chemotherapy is the standard first-line treatment for patients with advanced non-small-cell lung cancer. However, the optimum number of treatment cycles remains controversial. Therefore, we did a systematic review and meta-analysis of individual patient data to compare the efficacy of six versus fewer planned cycles of platinum-based chemotherapy. Methods All randomised trials comparing six versus fewer planned cycles of first-line platinum-based chemotherapy for patients with advanced non-small-cell lung cancer were eligible for inclusion in this systematic review and meta-analysis. The primary endpoint was overall survival. Secondary endpoints were progression-free survival, proportion of patients with an objective response, and toxicity. Statistical analyses were by intention-to-treat, stratified by trial. Overall survival and progression-free survival were compared by log-rank test. The proportion of patients with an objective response was compared with a Mantel-Haenszel test. Prespecified analyses explored effect variations by trial and patient characteristics. Findings Five eligible trials were identified; individual patient data could be collected from four of these trials, which included 1139 patients—568 of whom were assigned to six cycles, and 571 to three cycles (two trials) or four cycles (two trials). Patients received cisplatin (two trials) or carboplatin (two trials). No evidence indicated a benefit of six cycles of chemotherapy on overall survival (median 9·54 months [95% CI 8·98–10·69] in patients assigned to six cycles vs 8·68 months [8·03–9·54] in those assigned to fewer cycles; hazard ratio [HR] 0·94 [95% CI 0·83–1·07], p=0·33) with slight heterogeneity between trials (p=0·076; I2=56%). We recorded no evidence of a treatment interaction with histology, sex, performance status, or age. Median progression-free survival was 6·09 months (95% CI 5·82–6·87) in patients assigned to six cycles and 5·33 months (4·90–5·62) in those assigned to fewer cycles (HR 0·79, 95% CI 0·68–0·90; p=0·0007), and 173 (41·3%) of 419 patients assigned to six cycles and 152 (36·5%) of 416 patients assigned to three or four cycles had an objective response (p=0·16), without heterogeneity between the four trials. Anaemia at grade 3 or higher was slightly more frequent with a longer duration of treatment: 12 (2·9%) of 416 patients assigned to three-to-four cycles and 32 (7·8%) of 411 patients assigned to six cycles had severe anaemia. Interpretation Six cycles of first-line platinum-based chemotherapy did not improve overall survival compared with three or four courses in patients with advanced non-small-cell lung cancer. Our findings suggest that fewer than six planned cycles of chemotherapy is a valid treatment option for these patients. Funding None. Platinum-based chemotherapy is the standard first-line treatment for patients with advanced non-small-cell lung cancer. However, the optimum number of treatment cycles remains controversial. Therefore, we did a systematic review and meta-analysis of individual patient data to compare the efficacy of six versus fewer planned cycles of platinum-based chemotherapy. All randomised trials comparing six versus fewer planned cycles of first-line platinum-based chemotherapy for patients with advanced non-small-cell lung cancer were eligible for inclusion in this systematic review and meta-analysis. The primary endpoint was overall survival. Secondary endpoints were progression-free survival, proportion of patients with an objective response, and toxicity. Statistical analyses were by intention-to-treat, stratified by trial. Overall survival and progression-free survival were compared by log-rank test. The proportion of patients with an objective response was compared with a Mantel-Haenszel test. Prespecified analyses explored effect variations by trial and patient characteristics. Five eligible trials were identified; individual patient data could be collected from four of these trials, which included 1139 patients—568 of whom were assigned to six cycles, and 571 to three cycles (two trials) or four cycles (two trials). Patients received cisplatin (two trials) or carboplatin (two trials). No evidence indicated a benefit of six cycles of chemotherapy on overall survival (median 9·54 months [95% CI 8·98–10·69] in patients assigned to six cycles vs 8·68 months [8·03–9·54] in those assigned to fewer cycles; hazard ratio [HR] 0·94 [95% CI 0·83–1·07], p=0·33) with slight heterogeneity between trials (p=0·076; I2=56%). We recorded no evidence of a treatment interaction with histology, sex, performance status, or age. Median progression-free survival was 6·09 months (95% CI 5·82–6·87) in patients assigned to six cycles and 5·33 months (4·90–5·62) in those assigned to fewer cycles (HR 0·79, 95% CI 0·68–0·90; p=0·0007), and 173 (41·3%) of 419 patients assigned to six cycles and 152 (36·5%) of 416 patients assigned to three or four cycles had an objective response (p=0·16), without heterogeneity between the four trials. Anaemia at grade 3 or higher was slightly more frequent with a longer duration of treatment: 12 (2·9%) of 416 patients assigned to three-to-four cycles and 32 (7·8%) of 411 patients assigned to six cycles had severe anaemia. Six cycles of first-line platinum-based chemotherapy did not improve overall survival compared with three or four courses in patients with advanced non-small-cell lung cancer. Our findings suggest that fewer than six planned cycles of chemotherapy is a valid treatment option for these patients.