Real-world investigation of germline predisposition to myelodysplastic syndromes

Follow-up of potential germline variants revealed by diagnostic targeted sequencing (TS) for myelodysplastic syndromes (MDS) has been proposed by clinical guidelines. However, their feasibility and clinical yield in routine MDS practice remain uncertain. We evaluated real-world applicability of systematic germline follow-up in an unselected cohort of 716 patients (median age 74 years) evaluated for MDS. Their diagnostic TS data was analyzed to identify variants in CEBPA, DDX41, ETV6, GATA2, and RUNX1 with a variant allele frequency ≥35%. In total, 98 variants were identified in 87 (12.2%) patients. Germline investigation was possible for 62 patients; for the remaining 25 without biobanked material, medical charts were reviewed. We identified pathogenic/likely pathogenic (P/LP) germline variants in 19 patients (2.7%). Most P/LP variants were found in DDX41 (73.7%), followed by 10.5%, 10.5%, and 5.3% in RUNX1, GATA2, and ETV6, for germline conversion rates between 5.4% and 93.3%. Patients with P/LP variants had a median age at diagnosis of 73 years, with marked male predominance (3.75:1). Most patients were diagnosed with MDS-EB1 or MDS-EB2 (68.4%) and had normal cytogenetics (95%). A review of medical charts in younger patients (<50 years) revealed germline predisposition in other genes in additional five patients, revisiting the prevalence of predisposition to 3.4% overall and to 16.7% in those under 50. Our results show that systematic germline follow-up after diagnostic TS succeeds in identifying predisposition in nearly 3% of patients at a typical adult MDS clinic. Germline testing for genes not routinely assessed by diagnostic TS is required in younger patients.