BACKGROUNDAlthough a myriad of novel treatments entered the treatment paradigm for advanced melanoma, there is a lack of head-to-head evidence. We conducted a network meta-analysis (NMA) to estimate each treatment's relative effectiveness and safety.
METHODSA systematic literature review (SLR) was conducted in Embase, MEDLINE, and Cochrane to identify all phase-III randomised controlled trials (RCTs); timeframe: January 1, 2010 to March 11, 2019. We retrieved evidence on treatment-related grade 3/4 adverse events, progression-free survival (PFS), and overall survival (OS). Evidence was synthesised using a Bayesian fixed-effect NMA. Reference treatment was dacarbazine. In accordance with RCTs, dacarbazine was pooled with temozolomide, paclitaxel, and paclitaxel plus carboplatin. To increase homogeneity of the study populations, RCTs were only included if patients were not previously treated with novel treatments.
RESULTSThe SLR identified 28 phase-III RCTs involving 14,376 patients. Nineteen (seventeen) treatments were included in the effectiveness (safety) NMA. For PFS, dabrafenib plus trametinib (hazard ratio [HR] PFS: 0.21) and vemurafenib plus cobimetinib (HR PFS: 0.22) were identified as most favourable treatments. Both had, however, less favourable safety profiles. Five other treatments closely followed