FCRL4 and exhausted B cell signature in tumor infiltrating B cells from lung cancer

Our understanding of the heterogeneity of tumor-infiltrating B cells (TIBs) evolves in parallel with the advances in high-dimensional flow cytometry and single-cell RNA sequencing (scRNA-seq). Subpopulations of TIBs have been described with diverse and even opposite roles in tumor control, based on surface molecule, cytokine and transcriptional factor expression. Furthermore, recently the predictive role of IgA and IgG expression in tumors was shown for melanoma and KRAS-mutated, but not KRAS wild type, lung adenocarcinoma. This may imply that B cells producing different immunoglobulin isotypes also acquire different functional characteristics that have unequal effects on tumor microenvironment (TME) and differentially affect tumor progression. To address this question, we performed comparative bulk transcriptome analysis of tumor-infiltrating surface-IgA+ (sIgA+) and sIgG+ memory B cells from lung adenocarcinoma (LUAD) tumors. We identified FCRL4, PDCD1 and RUNX2 among genes overexpressed in IgA+ B cells, which suggests exhausted, chronically antigen-stimulated phenotype. Using public scRNA-seq data, we showed that FCRL4-expressing memory TIBs form a distinct cluster of cells with transcriptomic profile characterized by upregulation of genes involved in IFNγ and IFNα responses. sIgG+ B cells, on the other hand, overexpress IL5RA, consistent with the role of IL-5 in class-switch recombination to IgG isotype. Moreover, based on the analysis of the TCGA LUAD cohort, we found that FCRL4 to CD20 expression ratio correlates with lower overall survival, further strengthening the view on FCRL4 as a marker of exhausted dysfunctional TIBs. Our findings support the initial hypothesis that IgA/IgG ratio in tumors is not merely an indication of cytokine environment in the TME, but also reflects the difference in the functional characteristics of the respective B cell populations.