Immune microenvironment-related gene mapping predicts immunochemotherapy response and prognosis in diffuse large B-cell lymphoma

Abstract

Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin's lymphoma (NHL). The R-CHOP immunochemotherapy regimen is the first-line treatment option for DLBCL patients and has greatly improved the prognosis of DLBCL, making it a curable disease. However, drug resistance or relapse is the main challenge for current DLBCL treatment. Studies have shown that the tumor microenvironment plays an important role in the onset, development, and responsiveness to drugs in DLBCL. Here, we used the CIBERSORT algorithm to resolve the composition of the immune microenvironment of 471 DLBCL patients from the GEO database. We found that activated memory CD4+ T cells and γδ T cells were significantly associated with immunochemotherapy response. Weighted gene co-expression networks (WGCNA) were constructed using differentially expressed genes from immunochemotherapy responders and non-responders. The module most associated with these two types of T cells was defined as hub module. Enrichment analysis of the hub module showed that baseline immune status was significantly stronger in responders than in non-responders. A protein-protein interaction (PPI) network was constructed for hub module to identify hub genes. After survival analysis, five prognosis-related genes (CD3G, CD3D, GNB4, FCHO2, GPR183) were identified and all these genes were significantly negatively associated with PD1. Using our own patient cohort, we validated the efficacy of CD3G and CD3D in predicting immunochemotherapy response. Our study showed that CD3G, CD3D, GNB4, FCHO2, and GPR183 are involved in the regulation of the immune microenvironment of DLBCL. They can be used as biomarkers for predicting immunochemotherapy response and potential therapeutic targets in DLBCL.

Keywords: Bioinformatics; Diffuse large B-cell lymphoma; Immune microenvironment; Immunochemotherapy response; Prognosis.