The impact of deficient antigen presentation on the tumour microenvironment of diffuse large B cell lymphoma

Abstract

Introduction: Interactions between tumour cells and the immune system play a critical role in regulating tumour development. Therapeutic strategies that harness the immune system have shown variable responses in diffuse large B cell lymphoma (DLBCL) which suggests that the immune tumour microenvironment (TME) might determine which patients are likely to respond to immune-based therapies. However, the key determinants of the tumour microenvironment and the relative contribution of immune effector and inhibitory cells to lymphomagenesis are poorly understood. The loss of the antigen presentation and immune cell recognition molecules beta-2-microglobulin (B2M) and CD58 by malignant cells is a frequent mechanism of immune evasion identified in DLBCL. The genetic events leading to B2M and CD58 protein loss are not fully understood; mutations and copy number loss are present in only a proportion of cases and a significant number of wild type cases also show protein loss.Aim: This project sought to determine the effect of B2M and CD58 expression on the infiltration and composition of immune cells in the TME of DLBCL and to investigate the mechanisms of B2M and CD58 protein loss.Methods: Ninety-seven de novo systemic DLBCL FFPE tissue biopsies underwent targeted resequencing of the B2M and CD58 exons and 5r and 3r UTRs (Illumina TSCA) in addition to quantitative gene expression of B2M, CD58, immune effectors (CD4, CD8, CD56, CD137) immunosuppressive macrophage markers (CD68, CD163) and immune checkpoints (TIM3, LAG3, PD1, PDL1 and PDL2) (NanoStringt Technologies) and B2M and CD58 copy number analysis by qPCR. B2M promoter methylation by mass array (EpiTYPERr), immunohistochemistry (IHC) for the above markers and high-throughput T-cell receptor b sequencing (Adaptive Biotechnologies) were performed on a subset of cases. Sequencing and gene expression findings were validated against an external whole exome and transcriptome cohort of approximately 1000 DLBCL samples.Results: B2M mutations were detected in 14/97 (14.4%) samples and had significantly lower B2M gene expression compared to B2MWT (p = 0.0094) and all showed B2M protein loss. However, protein loss was also observed in the majority of B2MWT samples (31/40; 77%). The B2M promoter region was not differentially methylated compared to controls. B2M copy number analysis was limited by technical difficulties but did not show an association with gene or protein expression. These findings demonstrate that mechanisms other than mutation, copy number or methylation status contribute to loss of B2M surface expression in DLBCL and that B2M gene expression provides a more robust assessment of intratumoural B2M status. Consistent with this, there were no significant differences between B2MMut/WT and expression of intratumoural immune markers, however B2M gene expression was significantly associated and positively correlated with the gene expression of CD4, CD8, CD56, CD137, CD68, CD163, PD1, PDL1 and LAG3n (all p l0.01) and with the protein expression of CD4, CD8, CD137, PDL1 and LAG3 (all p l0.01). These findings are consistent with a co-ordinately regulated immune response and suggest adaptive immune resistance by malignant cells through the upregulation of checkpoint inhibitors. High B2M gene expression was also significantly associated and correlated with a higher total number of intratumoural T cells (p = 0.0337, Spearman r = 0.34) and less TCR diversity (p = 0.0252, Spearman r = -0.34) compared to low B2M gene expression, suggesting that clonal T cell expansions are more likely with intact antigen presentation. The validation cohort confirmed that B2M gene expression correlated with immune cell infiltration and additionally showed that B2M gene expression positively correlated with the gene expression of HLA Class I//II molecules and a range of regulatory, transport and assembly molecules involved in the antigen presentation pathway.Similar analyses were performed for CD58 and showed no significant associations between CD58Mut/WT and CD58 gene expression or immune effector or inhibitor gene or protein expression. CD58 gene expression was associated with the gene expression of CD4, CD8, CD56, CD68, PD1 and PDL1 (all p l0.01) but not protein expression. These findings were not confirmed by the validation cohort. CD58 IHC data could not be obtained due to technical limitations, which restricts our study on the role of CD58 expression on the TME. On the basis of our limited data and the validation cohort findings, there is no evidence to suggest that CD58 is a key determinant of the TME in DLBCL.Conclusions: Gene expression is a robust measure of B2M quantification in the TME. Our findings indicate that high B2M gene expression reflects an immunologically active or lhotr tumor microenvironment in DLBCL characterised by higher levels of immune cell infiltration and a more clonal T cell infiltrate.n This data suggests that B2M gene expression level could be used as a biomarker of an active intratumoural immune response in DLBCL and may help to stratify the selection of patients in whom immune-based therapies are more likely to be effective.n