Genetic Characterization in Tissue and Cfdna in Marginal Zone Lymphomas

Introduction: Marginal zone lymphoma (MZL) is a group of non-Hodgkin lymphomas that originate from the marginal zone of lymphoid follicles. The WHO/ICC classifies MZL in primary splenic (SMZL), primary nodal (NMZL) and extranodal lymphoma of the mucosa-associated lymphoid tissue (EMZL). Their diagnosis remains difficult as they do not have pathognomonic features. Thus, integration of all the diagnostic available tools including genetic characterization is crucial. Circulating cell-free DNA (cfDNA) analysis is being incorporated in the study of some lymphoma types, but there are no studies in MZL. Methods: 98 patients were identified between 2014 and 2022 (33 SMZL, 32 EMZL, 6 NMZL, 16 monoclonal non-CLL B cell lymphocytosis (MZ-CBL) and 11 unclassified B-cell lymphoproliferative syndromes (LPS-NOS) with MZL clinico-biological features). DNA for tissue analysis was extracted from the diagnostic samples (54 from mononuclear cells from peripheral blood, 44 from formalin-fixed paraffin embedded tissues). cfDNA was extracted from paired plasma in 34/98 patients (17 SMZL, 6 EMZL, 4 NMZL, 6 MZ-CBL and 1 LPS-NOS). Libraries were prepared using a custom panel covering 31 MZL-associated genes (Qiagen Hilden, Germany) and sequenced with NextSeq (Illumina, San Diego CA). Results: We found mutations in 77% of the tissue-based samples (88% SMZL, 59% EMZL, 100% NMZL, 69% MZ-CBL and 91% LPS-NOS). The most frequently mutated genes in SMZL were KLF2 (27%), DNMT3A (24%), TP53 (21%), TNFAIP3 (18%), KMT2D (18%), ARID1A (18%), CCND3 (12%) and MYD88 (12%); in EMZL TNFAIP3 (19%), TET2 (19%) and KMT2D (9%); in NMZL KMT2D stood out (50%); in MZ-CBL DNMT3A (19%), CCND3 (19%), MYD88 (19%) and TP53 (13%); in LPS-NOS TP53 (36%), MYD88 (27%), CCND3 (18%) and BIRC3 (18%) (Figure 1). KLF2 was overrepresented in SMZL (p < 0.05) and TP53 was underrepresented in EZML (p < 0.05) compared to the other MZL types. Overall, 14/98 patients were TP53mut (8/14 multi-hit, 7 with cooccurrence of 17p deletion and 1 with 3 mutations). When comparing SMZL and MZ-CBL, KLF2 mutations defined SMZL over MZ-CBL (9/33 vs. 0/16, p < 0.05). The ability to find any tissue mutation in cfDNA was 94% in SMZL, 33% in EMZL, 100% in NMZL and 100% in MZ-CBL. Besides, in 76% of SMZL we detected the 100% of the tissue mutations. cfDNA revealed mutations not present in the tissue in 41% of SMZL, 33% of MALT, 100% of NMZL and 0% of MZ-CBL (86%, 50% and 50% of these mutations respectively were found in clonal hematopoiesis (CH) potentially related genes: DNMT3A, TET2, ASXL1 and TP53). The research was funded by: FIS/FEDER PI19/00034, SEHH, SCHH Keywords: Diagnostic and Prognostic Biomarkers, Genomics, Epigenomics, and Other -Omics, Indolent non-Hodgkin lymphoma No conflicts of interests pertinent to the abstract.