Integrating genomic and transcriptome features for characterization and prognostic prediction of angioimmunoblastic T-cell lymphoma.

Abstract

In this study, we conducted integrated molecular analyses of the transcriptome and tumor genome in 24 newly diagnosed patients with angioimmunoblastic T-cell lymphoma (AITL). Gene expression profiling revealed significant enrichment of B cell receptor signaling and innate immune-related pathways in the response group. CIBERSORT-based deconvolution analysis showed that the proportions of tumor-infiltrating B cells and M1 macrophages were significantly higher in the response group compared to the non-response group (B cells: 17.4% vs. 7.8%, P = 0.012; M1 macrophages: 11.3% vs. 6.1%, P = 0.005). The abundance of these immune cells was associated with favorable progression-free survival and overall survival. Conversely, T follicular helper (TFH) cells, which reflect tumor burden, were inversely correlated with these favorable immune subsets. The most frequently mutated genes in this cohort included TET2 (73.9%), RHOA (47.8%), IDH2 (34.7%), and DNMT3A (26.1%). Mutations in RHOA, IDH2, and DNMT3A were negatively correlated with tumor-infiltrating B cells and were associated with poorer survival outcomes. Furthermore, higher variant allele frequencies (VAFs) of TET2 mutations were also negatively correlated with B cell infiltration, while RHOA VAFs were positively associated with TFH cell abundance, suggesting a link between mutational clonality and immune suppression. In conclusion, our study highlights the interplay between tumor genetic alterations and the immune microenvironment in AITL. We identified a favorable immune profile, characterized by increased infiltration of B cells and M1 macrophages, that correlates with chemosensitivity and improved prognosis. In contrast, mutations in RHOA, IDH2, DNMT3A, and high VAFs of TET2 were associated with adverse clinical outcomes and unfavorable immune contexture.