Enrichment of rare pathogenic variants in common cancer predisposition genes in lymphatic malignancy: A comprehensive analysis of 2,138 cases

Abstract Introduction: The link between rare, inherited pathogenic variants (PVs) and an increased risk of developing specific malignancies, as with BRCA1/BRCA2 variants and breast and ovarian cancers, is well established. In hematologic malignancies, PVs in DDX41 are a recognized risk factor for the development of myeloid neoplasms. However, insights into the implications of such variants in lymphatic malignancies (LMs) are lacking, as is the understanding of the role of PVs in known cancer predisposition genes in the risk of developing LMs and its different subtypes. This study uses a large in-house dataset of LM cases to investigate the association between PVs in 18 cancer predisposition genes that are regularly included on multi-gene panel testing based on NCCN guidelines and the LM risk. Methods: Our study population consisted of 2,138 LM individuals, with a median age of 65 years (range: 0.1 – 93.9 years). All cases were diagnosed according to the World Health Organization (WHO) classification. Whole genome sequencing (WGS) was performed on DNA samples extracted from bone marrow or whole blood, alternative material as germline surrogate was not used. To focus on potentially inherited PVs, rare variants (gnomAD allele frequency ≤0.2%) with a variant allele frequency of 40-60% and >90% in 18 LM predisposition genes (ATM, BRCA1, BRCA2, CHEK2, PALB2, BARD1, BRIP1, CDKN2A, MLH1, MSH2, MSH6, NBN, NF1, PTEN, RAD50, RAD51C, RAD51D, TP53) were extracted. Rare PVs were defined as loss-of-function variants (e.g. nonsense, frameshift or splice site mutations) or variants classified as ‘pathogenic’ or ‘likely pathogenic’ in the ClinVar database. Additionally, all cases were analyzed for 66 genes, which are frequently mutated somatically within lymphatic neoplasms. Results: Among the 2,138 LM cases studied, the most frequent lymphatic subtypes were mature B-cell neoplasms (853/2,138, 40%), acute lymphoblastic leukemia (454/2,138, 21%), and plasma cell dyscrasia (402/2,138, 19%). A total of 8% (171/2,138) of lymphatic cases were found to carry at least one PV across the 18 common cancer predisposition genes. Notably, older patients (≥60 years) were significantly more likely to carry PVs, with 9% affected, compared to 6% of patients under 60 years (Wilcoxon rank sum test p< 0.05). Of the cases with PVs, 22% (38/171) carried multiple variants, 32% of which exhibited PVs in at least two different genes. Additionally, 16% of the identified PVs were homozygous. The highest frequency of PVs was observed in TP53 (4.2%, 89/2,138), ATM (1.4%, 31/2,138), CHEK2 (0.6%, 12/2,138) and PTEN (0.5%, 11/2,138). Compared to healthy controls from gnomAD (n = 141,456) and the UK Biobank (UKBB, n= 389,731), PVs were significantly more frequent in our LM cohort. Specifically, variants in TP53 (odds ratio [OR] = 105.7), ATM (OR = 5.4) and CHEK2 (OR = 3.6) were associated with a significantly higher risk of LM than in UKBB (median age 56 years) and gnomAD controls. There was no significant difference in the frequency of LM subtypes between individuals with PVs and the overall cohort. Within the 66 genes frequently mutated in lymphatic neoplasms, in LM cases with PVs, patients carried a median of two additional somatic mutations (range: 1-7), compared to a median of one somatic mutation (range: 1–7) in patients without PVs. No significant difference was found between the two groups (Wilcoxon rank sum test, p = 0.58). There was no correlation beween a PV and a specific somatic gene mutation. Moreover, no correlation was identified between the number of additional somatic mutations and the age of the LM cases studied.Conclusion: Our findings suggest an enrichment of rare PVs in established cancer predisposition genes among LM patients, indicating a genetic predisposition potentially distinguishing these individuals from the general population. Thus, our findings demonstrate that rare inherited variants add another layer of genetic variation in the development of LM. Identifying these variants may help improve risk assessment models for LM and may thus be incorporated as genetic screening into routine diagnostics of LM.