Abstract 1821: Integrative genomic analysis of diffuse large B cell lymphoma (DLBCL) identifies recurrent oncogenic alterations associated with distinct immune-related clusters

Abstract

Abstract Introduction: Genetic alterations in tumor cells shape the cancer immune landscape and influence response to checkpoint blockade therapy (CBT). Despite recent advances in defining DLBCL genomics, the impact of genetic alterations on the DLBCL immune environment is not well-characterized. Here, we aim to define unique immune landscapes in DLBCL and determine the impact of lymphoma-cell intrinsic alterations on shaping immune environments in this disease, which may have implications for CBT response Methods: Gene set variation analysis (GSVA) was used to generate sample-wise enrichment scores for immune-related gene sets in ~500 DLBCLs. A cell-of-origin (COO) signature that classifies DLBCLs based on transcriptional similarity to activated B cells (ABC) or germinal center B cells (GCB) was included to control for COO-related genomic heterogeneity. Samples were hierarchically clustered. Paired exome sequencing was analyzed to identify mutations associated with each cluster. Cas9-mediated gene editing was used to modulate genes of interest in a murine lymphoma model (A20) Results: GSVA revealed 4 DLBCL immune clusters - ABC hot, ABC cold, GCB hot and GCB cold. Concordant with previous work, PD-L1 gene amplifications were highly enriched in ABC hot DLBCLs, while double-hit signature DLBCLs fell within the GCB cold cluster. Mutations and oncogenic pathways associated with each cluster were identified. ABC cold DLBCLs harbored recurrent inactivating mutations in TMEM30A and PRDM1, while mutations in KLHL6 and SPEN were associated with ABC hot DLBCLs. The GBC cold cluster was enriched for mutations in TP53 and FOXO1, and inactivating SOCS1 mutations were prevalent in GCB hot DLBCLs. MYC activation signatures characterized ABC and GCB cold clusters. A20 lymphomas with inducible c-Myc activation (A20MycER) contained fewer CD8+ T cells following tamoxifen versus control treatment, confirming our computational findings. Moreover, as SOCS1 mutations strongly associated with GCB hot DLBCLs and are prevalent in CBT-sensitive lymphomas, the extent to which Socs1-deficient lymphoma cells showed increased sensitivity to the critical T cell effector cytokine IFNγ was examined. Upon IFNγ exposure, A20Socs1-/- cells showed increased pStat1 levels, enhanced expression of PD-L1 and MHC-I, and strikingly diminished proliferation compared to A20WT cells, suggesting that enhanced JAK/STAT signaling via Socs1-/- increases IFNγ sensitivity, rendering lymphoma cells vulnerable to CBT Conclusion: Through computational analyses in a large genomic dataset, we have defined unique immune-related DLBCL clusters and associated recurrent genomic alterations. Through ongoing studies in murine models, we aim to elucidate mechanisms by which particular oncogenic alterations in lymphoma cells shape the DLBCL immune landscape and impact CBT response Citation Format: Sravya Tumuluru, James K. Godfrey, Jovian Yu, Alan Cooper, Michael Leukam, Sonali M. Smith, Xiufen Chen, Justin P. Kline. Integrative genomic analysis of diffuse large B cell lymphoma (DLBCL) identifies recurrent oncogenic alterations associated with distinct immune-related clusters [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1821.