High-dimensional spectral cytometry identifies follicular regulatory CD8⁺ T cells in diffuse large B-cell lymphoma

Diffuse large B-cell lymphoma (DLBCL) constitutes 30-40% of non-Hodgkin lymphoma cases. Despite therapeutic advances, persistence of relapsed cases has been linked to the complex tumor microenvironment (TME) and its interactions with lymphoma cells. In particular, characterising T-cell subsets, including rare cell types, and their interplay with the remaining TME is crucial for unravelling DLBCL pathogenesis and refining therapeutic strategies.

Using flow and spectral cytometry with unsupervised analysis, we investigated T-cell subpopulations across DLBCL biopsies and control lymph nodes (LN). We also inferred communication pathways between T cells and other immune cells in the TME based on the correlation of ligand-receptor expression.

Our analysis revealed a higher frequency of CD8⁺ follicular regulatory T (Tfr) cells in DLBCL biopsies compared to control LN. These cells exhibited an effector-memory phenotype (CD45RA^- CCR7^-), expressed follicular markers (PD-1⁺ CXCR5⁺) and had a regulatory profile (CD127^- CD25⁺) along with an activation/co-stimulatory signature (HLA-DR⁺, ICOS⁺, CD95⁺). Correlation analysis highlighted a co-stimulatory interaction between lymphoma B cells and CD8⁺ Tfr cells through the ICOS/ICOSL pathway, which may contribute to a protumor effect. Validation in independent scRNAseq and flow cytometry datasets confirmed the notable prevalence of CD8⁺ Tfr cells in DLBCL biopsies.

Our study highlights the utility of high-dimensional computational cytometry in elucidating T-cell subpopulations, including an increased frequency of CD8⁺ follicular regulatory T cells and their communication patterns within the DLBCL TME. This unbiased approach sheds light on novel cellular mechanisms in DLBCL, uncovering potential targets and biomarkers for immunotherapy.