Abstract: 【Abstract】 Objective To assess the efficacy of immune checkpoint inhibitors and targeted therapy in the treatment of resectable melanoma by Bayesian network meta-analysis. Methods PubMed, Embase and Cochrane databases were searched for randomized controlled trials on adjuvant therapy of resectable melanoma. Based on hazard ratios, a network meta-analysis of relapse-free survival was performed using a Bayesian fixed-effect model to assess therapeutic effect of adjuvant therapy on resectable melanoma. Data were comprehensively analyzed by using StataSE 15 and OpenBUGS 3.2.3 softwares. Results Six eligible articles involving 5 587 patients assigned to 7 treatment regimens were included. There were 5 019 patients in the stage Ⅲ subgroup, 2 085 in the ulcerated subgroup, 2 629 in the non-ulcerated subgroup, and 2 054 in the BRAF-mutated subgroup; the 7 treatment regimens included surgery + observation or placebo, surgery + adjuvant dabrafenib plus trametinib, surgery + adjuvant nivolumab, surgery + adjuvant ipilimumab, surgery + adjuvant pembrolizumab, surgery + adjuvant bevacizumab, and surgery + adjuvant vemurafenib. In the network meta-analysis, surgery + adjuvant dabrafenib plus trametinib （HR = 0.47, 95% CI: 0.39 - 0.57）, surgery + adjuvant nivolumab （HR = 0.49, 95% CI: 0.36 - 0.65）, and surgery + adjuvant pembrolizumab （HR = 0.57, 95% CI: 0.43 - 0.75） were more effective for the improvement of relapse-free survival than surgery alone; the subgroup analysis of stage Ⅲ and ulcerated resectable melanoma showed the same results as the above-mentioned network meta-analysis. In the subgroup analysis of non-ulcerated resectable melanoma, surgery + adjuvant vemurafenib （HR = 0.48, 95% CI: 0.29 - 0.79）, surgery + adjuvant dabrafenib plus trametinib （HR = 0.48, 95% CI: 0.33 - 0.70）, and surgery + adjuvant nivolumab （HR = 0.50, 95% CI: 0.31 - 0.79） could significantly prolong the relapse-free survival compared with surgery alone, but surgery + adjuvant pembrolizumab （HR = 0.69, 95% CI: 0.45 - 1.06） was not superior to surgery alone. In the subgroup analysis of BRAF-mutated melanoma, surgery + adjuvant bevacizumab （HR = 0.60, 95% CI: 0.43 - 0.85）, surgery + adjuvant dabrafenib plus trametinib （HR = 0.47, 95% CI: 0.38 - 0.57）, surgery + adjuvant pembrolizumab （HR = 0.59, 95% CI: 0.38 - 0.92） and surgery + adjuvant vemurafenib （HR = 0.65, 95% CI: 0.50 - 0.85） could significantly prolong the relapse-free survival compared with surgery alone. The network meta-analysis was carried out for ranking of these adjuvant treatments, and adjuvant dabrafenib plus trametinib was most likely to rank first in the network meta-analysis and subgroup analysis of stage Ⅲ, ulcerated or BRAF-mutated resectable melanoma, while adjuvant vemurafenib was most likely to rank first in the subgroup analysis of non-ulcerated resectable melanoma. Conclusion For patients with ulcerated or BRAF-mutated resectable melanoma, dabrafenib plus trametinib may be the optimal adjuvant therapy; for those with BRAF-unknown or wild-type resectable melanoma, nivolumab may be the optimal adjuvant therapy.

Key words: Melanoma, Molecular targeted therapy, Meta-analysis, Immune checkpoint inhibitors, Resected melanoma