Introduction: Approximately 10% of adults with myelodysplastic neoplasms (MDS) or acute myeloid leukemia (AML) carry germline variants in genes predisposing to hematologic neoplasms (HNGP). Clonal hematopoiesis (CH), whether as a pre-malignant event or as compensating mechanism, is a focus of interest in the field of germline predisposition in adults. However, systematic and longitudinal studies on HNGP patients' healthy relatives, critical to understand how CH and germline predisposition interact, remain limited. Our study aimed to analyze the prevalence, dynamics, and impact of CH in relatives of HNPG adult patients who carry the causative germline abnormality. Methods: In 2018, the Catalan public healthcare system established criteria for germline testing when diagnosing a hematological neoplasm. These criteria, updated annually, include early-onset, syndromic phenotype, family history of thrombocytopenia or hematological neoplasms, and the presence of a potential germline variant in the initial HN workout. For each patient, a HNG-driven NGS virtual panel was performed using DNA extracted from hair follicles. Additionally, if the physical examination revealed characteristics related to Telomere Biology Disorders (TBD), telomere length was assessed by flow-FISH. First and second-degree relatives of germline variant carriers were offered predictive testing, using Sanger sequencing on peripheral blood (PB). In cases of TBD without an identified molecular cause, flow-FISH was used to select those at-risk relatives. Family carriers underwent NGS for genes with recurrent acquired mutations in myeloid malignancies in PB at the time of diagnosis and after a three-year follow-up period. Results: From 2019 to 2024, we studied 97 patients who met the criteria for germline testing and identified 21 HNPG index cases (21.6%). HNPG patients were defined by the presence of a pathogenic/likely pathogenic variant in ETV6 (n=3), GATA2 (n=2), DDX41 (n=6), ELANE (n=1), CHEK2 (n=1), RUNX1 (n=1), ATM (n=1), TP53 (n=1) or a TBD diagnosis (n=5). Among these last, in 2 we identified a causative variant (RTEL1 and TINF2) and the remaining 3 reached the diagnosis due to a telomere length <1%. These 21 index cases were diagnosed as follows: 11 MDS with multilineage dysplasia, 2 MDS with ring sideroblasts, 2 hypoplastic MDS, 2 MDS with excess blasts, and 4 AML. The median age at diagnosis was 48 years (range: 19-76). Thirty-four out of 60 studied relatives (56.6%) carried the germline variant or telomere abnormality. Twenty-seven (79.4%) were studied using a somatic NGS panel in PB. Among these cases, CH was detected at first time-point in 7 of the carrier relatives (26%) across 4 different families. One family was notable, as 3 of the studied healthy ETV6 carriers had developed CH: a 71-year-old male with normal CBCs and variants in TP53 (VAF 4%; c.394A>G) and SRSF2 (VAF 1.5%;c.284C>T), a 76-year-old female with mild thrombocytopenia and variants in JAK2 (VAF 16%;c.1849G>T), NFE (VAF 10%;c.768_769dup), DNMT3A (VAF 3.2%;c.2693T>A) and PTPN11 (VAF 5%;c.785T>A), and a 48-year-old female with normal CBCs and one variant in CBL (VAF 2.3%; c.1150T>C). Two relatives from one TBD family presented CH: a 57-year-old male with no CBC alteration and a variant in PPMD1 (VAF 4.8%; c.1451T>A) and a 45-year-old female with normal CBC and a variant in U2AF1 (VAF 4.7%; c.101C>T). Another TBD family carrier had a variant in DNMT3A (VAF 1.7%; c.1228G>A). One family carrier from a DDX41 family, a 79-year-old male with normal CBCs, presented a variant in DNMT3A (VAF 3.1%; c.1015-3C>G). We performed a 3 year follow-up NGS in 8 relatives, of which 5 (62%) showed clonal evolution, evidenced by an increase in VAF (n=2) or the acquisition of new variants (n=3). In a second ETV6 family, a heathy carrier relative acquired a KRAS variant VAF (1.3%, c.34G>C) not detected at the initial study. No carrier relatives developed a hematologic neoplasm during the follow-up period. Conclusion: In one of the largest series studied to date of HNGP relatives, we found a 26% of CH development in carriers of the causative predisposition abnormality. That percentage is approximately 10x higher than in the age-adjusted population group, prompting us to extend the longitudinal follow-up of these families. We describe 2 ETV6 families, in which CH was defined in 4 healthy relatives by genes associated with myeloproliferative diseases, a molecular pattern to be confirmed.
Read full abstract