This retrospective analysis combined the patient-level data from three phase 3 randomized controlled trials that compared the efficacy of different third generation platinum- and non-platinum based doublets. Unadjusted median survival times and Cox covariate-adjusted treatment hazard ratio (HR) estimates were calculated. Overall results and subgroups by histological type were reported.
This combined analysis consisted of 3467 patients. In the overall analysis, adjusted HRs favored pemetrexed (HR <1.0) to each of the other 5 regimens, though none of these HRs were statistically significant. Among patients with non-squamous histology, pemetrexed-cisplatin produced favorable HRs to each of the other regimens, achieving statistical significance when compared with vinorelbine-cisplatin (HR = 0.67; 95%confidence intervals [CI]: 0.50, 0.91) and gemcitabine-cisplatin (HR = 0.85; 95%CI: 0.75, 0.97). Among patients with squamous histology, 4 of the 5 comparison regimens produced favorable HRs (HR >1.0) when compared with pemetrexed-cisplatin, with only the comparison with gemcitabine-cisplatin achieving statistical significance (HR = 1.23; 95%CI: 1.00, 1.51).
In the absence of randomized clinical trial data comparing pemetrexed-cisplatin to commonly used doublets in advanced NSCLC other than gemcitabine-cisplatin, this combined analysis of multiple trials provides estimates for such comparisons.