Relative efficacy of interventions in the treatment of second-line non-small cell lung cancer: a systematic review and network meta-analysis.

Background: Locally advanced or metastatic non-small cell lung cancer (NSCLC) that has progressed after first-line treatment has a poor prognosis. Recent randomized clinical trials (RCTs) have demonstrated survival benefits of alternative treatments to docetaxel. However, information is lacking on which patients benefit the most and what drug or regimen is optimal. We report a systematic review and network meta-analysis (NMA) of second-line treatments in all subgroup combinations determined by histology, programmed death ligand 1 (PD-L1) expression, and epidermal growth factor receptor (EGFR) mutation.

Methods: MEDLINE, PubMed, EMBASE, Biosciences Information Service (using the Dialog Platform), Cochrane Library, and abstracts from scientific meetings were searched for RCTs published up to September 2015. Key outcomes were overall survival (OS) and progression-free survival (PFS). Bayesian hierarchical exchangeable NMAs were conducted to calculate mean survival times and relative differences for eight subgroups, using docetaxel as the reference comparator. For OS, the NMA was based on hazard ratios applied to a first-order fractional polynomial model fitted to the reference treatment. For PFS, a second-order fractional polynomial model was fitted to reconstructed patient-level data for the entire network of evidence.

Results: The search identified 30 studies containing 17 different treatment regimens. Docetaxel plus ramucirumab was associated with a significant improvement in OS and PFS, relative to docetaxel, regardless of patient type. Docetaxel plus nintedanib showed similar efficacy to docetaxel plus ramucirumab in the nonsquamous populations. EGFR tyrosine kinase inhibitors (TKIs) erlotinib and gefitinib showed superior levels of efficacy in EGFR mutation-positive populations and the one PD-1 immunotherapy (nivolumab) studied showed superior efficacy in the populations exhibiting high PD-L1 expression.

Conclusions: In the absence of head-to-head comparisons, we performed a mixed-treatment analysis to synthesize evidence of the efficacy of each treatment. Benefits are optimized by targeting specific treatments to individual patients guided by histology, PD-L1 expression, and EGFR mutation status.

Systematic Review Registration: This review is registered in PROSPERO (registration number: CRD42014013780 available at www.crd.york.ac.uk/PROSPERO ).