Efficacy of Frontline Chemotherapy for Extranodal Natural Killer/T-Cell Lymphoma: A Systematic Review and Network Meta-Analysis.

Background: Treatment with non-anthracycline (ANT)-based chemotherapy has increased survival in patients with extranodal natural killer/T-cell lymphoma (ENKTCL). However, the relative efficacy of various drug combinations has been contentious. We aimed to identify the most effective chemotherapy regimens for newly diagnosed ENKTCL. Methods: A network meta-analysis was performed to evaluate the differences in survival and treatment responses across various regimens. The primary objective was overall survival (OS), while secondary outcomes included progression-free survival (PFS), objective response rate (ORR), and complete response (CR). We utilized a Bayesian framework to perform the network meta-analysis. Rank probabilities were assessed by the surface under the cumulative ranking curve (SUCRA). Node-splitting method was used to assess the inconsistency. Results: A total of 1,113 patients were enrolled across 10 studies. Chemotherapy regimens were grouped into five modalities, for which six types of direct comparisons were available. We identified the asparaginase (ASP)/gemcitabine (GEM)-based regimens superiority over ANT-based, non-ASP/ANT-based and ASP/methotrexate (MTX)-based regimens on OS. Although no significant differences were observed compared with ASP/not otherwise specified-based, ASP/GEM-based regimens were still the best option chemotherapy for OS. Moreover, the ASP/GEM-based regimens demonstrated advantages in PFS, ORR and CR. Conclusions: According to our network meta-analysis, it appears that ASP/GEM-based regimens could potentially serve as the most effective frontline chemotherapy option for ENKTCL.