ANKRD26-related thrombocytopenia: Hematologic malignancy characteristics, clinical outcomes, and precursor states

Abstract Introduction Heterozygous germline pathogenic variants (PVs) in the 5' untranslated region of ANKRD26 cause inherited thrombocytopenia and predisposition to hematologic malignancy (HM). The estimated lifetime HM risk is ~10% based on families ascertained from an inherited thrombocytopenia cohort. However, the malignancy spectrum, molecular features, precursor states, and overall clinical outcomes remain poorly defined. Methods We established an international, multicenter cohort of individuals with ANKRD26-related thrombocytopenia diagnosed after undergoing genetic evaluation due to a personal and/or family history of thrombocytopenia and/or HM. Clinical, pathologic, and molecular data were collected for 27 individuals with germline ANKRD26 PVs from 21 unrelated families treated at 11 institutions across 5 countries. Statistical analyses were performed using R (v4.4.2). Results Patients were predominantly male (18/27) and white (25/27). Thrombocytopenia was typically identified in childhood (median age 10 years, range 0-48) with a genetic diagnosis made later (median age 53 years, range 2-80). The mean (+/- SD) pre-malignancy platelet count was 45 +/- 24 K/uL (range 5-87). ANKRD26 PVs observed in each family included: c.-126T>C (n=9), c.-128G>A (n=3), c.-134G>A (n=3), c.-116C>T (n=3), c.-118C>T (n=2), and c.-118C>A (n=1). Thirteen HMs were diagnosed in 11 of 27 (41%) patients, including myelodysplastic neoplasm/syndrome (MDS, n=5), acute myeloid leukemia (AML, n=3), chronic myeloid leukemia (CML, n=2), MDS/myeloproliferative neoplasm (MPN) overlap (n=2), and Langerhans cell histiocytosis (n=1). The median age at first HM diagnosis was 63 years (range 49-80). Recurrent somatic features in myeloid malignancies included complex karyotype (4/11, 36%) and TP53 mutations (5/10, 50%). Of the 3 patients with AML, 1 achieved a partial remission (PR) with venetoclax combined with cladribine, idarubicin, and cytarabine but relapsed, 1 achieved a PR with azacitidine but relapsed <100 days after haploidentical allogeneic stem cell transplantation (alloSCT), and 1 had primary refractory disease. Among 5 patients with MDS, 1 was refractory to decitabine, 1 underwent alloSCT but relapsed 2 years post-transplant, 1 remains in remission post-alloSCT with 1 year of follow-up, 1 is undergoing alloSCT evaluation, and 1 was lost to follow-up. Both patients with CML achieved durable major molecular remissions with imatinib. To better estimate HM risk, we assumed 100% penetrance of thrombocytopenia for ANKRD26 PV carriers and included family members with chronic thrombocytopenia who had not had germline genetic testing. In this expanded series (n=62), 16 individuals developed a HM (26%). Additional diagnoses included MDS with progression to AML (n=1), AML (n=1), CML (n=1), and leukemia of unknown type (n=2). Cumulative incidence of HM was 0% by age 40 years, 17% by age 60 years, and 60% by age 80 years. Somatic sequencing of peripheral blood was performed in 5 HM-unaffected ANKRD26 PV carriers. One individual was found to have clonal hematopoiesis driven by TET2 (c.1648C>T, p.Arg550*) with a variant allele frequency (VAF) of 9% at age 55. Bone marrow biopsy showed mild megakaryocytic atypia and 4% blasts without overt evidence of malignancy. To assess the prevalence of ANKRD26-related HM, we assessed the frequency of ANKRD26 PVs in 2 large MDS/AML cohorts. In the ARUP database (9662 AML, 6397 MDS), 2 individuals (0.01%) had an ANKRD26 PV with a VAF consistent with germline origin. In the MSK-IMPACT Heme Germline MDS/AML cohort (n = 284), no ANKRD26 P/LP variants were identified. Conclusions Germline ANKRD26 PVs underlie <1% of MDS/AML diagnoses but may represent a high-risk subtype characterized by complex karyotype, TP53 mutations, and high risk of relapse. Among our series, the incidence of HM was low early in life but increased substantially with age, reaching 60% by 80 years. This exceeds prior estimates, which were derived from individuals ascertained from thrombocytopenia-based registries. Together, our findings suggest that ANKRD26-associated HM may represent a high-risk clinical entity, and that age and family history should inform surveillance strategies. More studies are needed to define genetic and environmental modifiers of HM risk as well as precursor states to enable early detection and disease interception.