194 | A NOVEL APPROACH FOR CLASSIFICATION OF SPLENIC B CELL LYMPHOMA USING DNA METHYLATION REVEALS DIVERSE BIOLOGICAL DISEASE MECHANISMS

Résumé fourni par la source

Introduction: Patients within the category of splenic B cell lymphomas (SBL) often exhibit heterogenous, overlapping features, and include hairy cell leukemia (HCL), splenic diffuse red pulp lymphoma (SDRPL), splenic marginal zone lymphoma (SMZL), and HCL-variant (HCLv), which has recently been reassigned into the new WHO entity, splenic B cell lymphoma/leukemia with prominent nucleoli (SBLPN). Diagnosis and classification of SBLs is often challenging and is further obscured by conflicting classification systems. Here we uniformly applied genome-wide DNA methylation profiling and performed unbiased molecular subclassification revealing novel biological SBL subgroups with distinct pathobiological features. Methods: Samples from 396 patients diagnosed with HCL, HCLv, SDRPL and SMZL were obtained from multiple institutions. DNA isolated from FACS-purified lymphoma cells from spleen and/or blood was analyzed by 450/850K Illumina DNA methylation arrays. Genetic mutations were assessed by whole-exome or targeted sequencing. The 1000 most variable CpG methylation sites were clustered using k-means; recursive feature elimination/random forest algorithms were used to develop a classifier. Results: Unsupervised clustering of 396 patients diagnosed with HCL, HCL-V, SDRPL or SMZL revealed 6 distinct DNA methylation (M) subgroups with distinct pathobiological features. M-SBL subgroups displayed discrete molecular features, including BRAF-V600E, MAP2K1, TP53, and CREBBP, mutations, immunophenotype, immunogenetics, and MAPK pathway activation. Based on known features in SBL patients we termed the novel subgroups M-HCL, M-HCLv1/2, M-SDRPL1/2 and M-SMZL. Interestingly, we observed half of M-SRPL2 patients exhibited TCL1A rearrangements not previously described B cell malignancy, recurrently juxtaposing the E(mu) enhancer to TCL1A. We also found TERT-promoter mutations in M-HCLv-1 not previously known in SBL patients. To further investigate the relationship between current SBL classification systems and DNA methylation-based groups, we constructed a methylation-based SBL classifier and validated on an independent cohort of 97 SBL patients. We found that HCLv and SDRPL diagnoses were variably assigned across M-SBL subgroups. A subset of SMZL patients were reassigned to other M-SBL while sharing molecular and phenotypic features with M-HCLv or M-SRPL. Transcriptome and motif enrichment revealed that TCF4 and TBET transcription factors play important roles in aberrant epigenetic programming of SMZL versus other SBLs, respectively. DNA methylation patterns and M-SBL calls were virtually identical between PBMC and spleen-derived tumor cells, as well as longitudinally including with intervening therapy, further supporting the diagnostic potential of DNA methylation in SBL. Conclusion: We have developed a DNA methylation-based approach that identifies novel SBL subgroups providing novel insight into the diverse pathobiology of these rare patients. Keywords: genomics, epigenomics, and other -omics; pathology and classification of lymphomas No potential sources of conflict of interest.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.