Randomized Phase II and Pharmacogenetic Study of Pemetrexed Compared With Pemetrexed Plus Carboplatin in Pretreated Patients With Advanced Non-Small-Cell Lung Cancer

Purpose We performed a randomized phase II trial comparing pemetrexed with pemetrexed plus carboplatin (PC) in patients experiencing relapse after platinum-based chemotherapy. Patients and Methods Main eligibility criteria were histologic or cytologic proof of advanced non-small-cell lung cancer (NSCLC), relapse more than 3 months after platinum-based chemotherapy, normal organ function, and Eastern Cooperative Oncology Group performance status 0 to 2. Patients were randomly assigned to pemetrexed 500 mg/m(2) (arm A) or carboplatin area under the curve 5 and pemetrexed 500 mg/m(2) (arm B), both administered intravenously every 3 weeks. Response assessment was performed every 6 weeks; toxicity assessment was performed every 3 weeks. Primary end point was time to progression (TTP); secondary end points were objective response rate (ORR), overall survival (OS), and toxicity. The study was designed to detect a 33% decrease in the hazard of disease progression in the combination arm (alpha = 0.05, two-sided log-rank test). Polymorphisms of thymidylate synthase, the reduced folate carrier, gamma-glutamyl hydrolase, and methylenetetrahydrofolate reductase (MTHF) were investigated in peripheral WBCs of consenting patients. Results Two hundred forty patients were enrolled. Median TTP was 2.8 months for arm A versus 4.2 months for arm B (hazard ratio, 0.67; 95% CI, 0.51 to 0.89; P = .005). Median OS was 7.6 months and 8.0 months and ORR was 4% and 9% for arms A and B, respectively. Subgroup analyses found adenocarcinoma to be associated with favorable outcome. Toxicities in both arms was negligible, with one potential toxic death in arm A. Patients with MTHFR C677T homozygous mutation had increased progression-free survival compared with patients with wild-type or heterozygous mutations (P = .03). Conclusion PC as second-line treatment for relapsed NSCLC resulted in a significant 33% reduction of the hazard of disease progression as compared with pemetrexed alone. J Clin Oncol 27: 2038-2045. (C) 2009 by American Society of Clinical Oncology