Comparison of Treatment Modalities in Frontline Therapy of Stages I&IIE NK/T-Cell Lymphoma (ENKTL): An Updated Network Meta-analysis

Abstract

Despite many studies reporting on the optimal therapies for ENKTL in the front-line setting, this area continues to be subject to a significant controversy regarding single modalities versus combination or sequential approaches. Several studies with limited samples compared such approaches to one another. The results were often inconclusive and occasionally conflicting. In the absence of direct head-to-head randomized controlled trials in this particular clinical setting, a network meta-analysis was conducted to compare these therapeutic approaches and their respective impact on 5-year survival. A review of the medical literature was conducted using online databases. Inclusion criteria consisted of (i) English language, (ii) diagnosis of stage I/IIE ENKTL, (iii) treatment with chemotherapy (CT), radiation (RT), sequencing individual modalities (iSEQ), sequencing combined modalities (cSEQ), and chemoradiotherapy (CRT), (iv) comparative studies (v) studies that reported 5-year survival rates. Studies that reported on mixed samples of early and advanced ENKTL were excluded. A frequentist network meta-analysis was conducted using the netmeta package and random-effects model. Fourteen studies comprising a total of 2308 participants were included. Our network meta-analysis revealed that upfront CRT and RT, unlike iSEQ and cSEQ, were each significantly superior to CT. Based on the pair-wise and network meta-analyses, CRT was ranked as the most effective first-line treatment approach followed by RT, iSEQ, cSEQ, and CT in decreasing order. Inconsistency analysis did not reveal any significant differences between direct and indirect estimates. This is the first network meta-analysis to compare all commonly utilized upfront treatment modalities in stages I/IIE ENKTL. It confirms the superiority of CRT and the inferiority of CT. Both iSEQ and cSEQ ranked as intermediates in efficacy. This may be the result of small numbers and wide confidence intervals or a warning signal of excessive related toxicity resulting in delays or compromised dosing of RT. Adequately powered randomized trials are warranted.