PF213 THE GENETIC LANDSCAPE OF FAMILIAL MDS/AML; RECURRING MUTATIONS IN THE RNA HELICASE DHX34 LEADING TO DEFECTS IN NONSENSE‐MEDIATED RNA DECAY

Abstract

Background:The inclusion of familial myeloid malignancies as a separate disease entity in the revised WHO classification has renewed efforts to improve both the recognition and the management of this group of at risk individuals. Inherited forms account for <5% of new diagnoses and their genetic basis can be separated into 3 groups: where approved testing exists (group 1); genes associated with other inherited haematological syndromes or emerging from basic research and requiring validation (group 2); and cases in which the specific gene mutations have not been established (group 3).Aims:Characterisation of novel genes mutated in familial AML/MDS.Methods:Familial AML/MDS references families where two or more family members were diagnosed with a haematological malignancy (predominantly AML, MDS, aplastic anaemia) or a prodromal cytopenia (e.g. thrombocytopenia) but at least one case was MDS or AML. Bone marrow or peripheral blood samples were collected from at least one affected individual from 86 families together with the corresponding laboratory and clinical information.Results:A primary genetic lesion associated with groups 1 & 2 was detected in 48 families (56%). WES analysis in the remaining 38 families rationalised 147 germline variants in 60 candidate genes. These genes were divisible into 3 discrete groups based on association with rare haematological syndromes, loci with relevant functions mutated in >3 families, or loci overlapping with other series of potential familial MDS/AML or IBMF syndromes. Germline variants in 22 of the selected 60 genes (37%) have been implicated across a broad range of inherited disorders including T cell‐negative, B cell‐negative, natural killer cell‐negative severe combined immunodeficiency (ADA, p.Gln3∗ and p.Glu88Argfs∗34), Platelet‐type bleeding disorder‐11 (GP6, p.Arg159∗ and p.Met430Lys) and Congenital Dyserythropoietic Anemia Type II (SEC23B, p.Pro542Leu and p.Glu361Asp). Among the non‐syndromic genes, we noted 4 families with variants in TET2, an epigenetic regulator and additional families harboured variants in DHX34, an RNA helicase that has been shown to function in the Nonsense‐mediated RNA decay (NMD) pathway and to co‐purify with the spliceosome, suggesting its potential involvement in pre‐mRNA splicing. All 4 DHX34 variants (p.Tyr515Cys, p.Arg897Cys, p.Asp638Asn and p.Glu444Asp), compromised NMD activity, as shown by their inability to promote phosphorylation of UPF1, which is essential for NMD activation. Cross referencing our WES data with the familial MDS/AML candidate loci documented by other groups highlighted the candidacy of 15 additional loci, including variants in DNAH9 in 3 families, a gene recurrently mutated in sporadic AML (2% of cases), one family with a variant in SH2B3 (p.Ala49Thr) and a germline duplication detected in one additional family NAPRT1 (p.Val216_Phe219dup).Summary/Conclusion:Familial AML/MDS represent a challenge where our genetic capability has raced ahead of the capacity to bring to bear much of our new knowledge into the clinical arena. Herein we propose several new recurrent predisposing genes in familial MDS/AML, validate the significant overlap with the genetics of inherited BMF syndromes and rare hematological syndromes, and demonstrate that heterozygous mutations in DHX34, detected in 4 families, individually impact activity of the NMD pathway. Overall these new findings are not only important for the respective families but offer novel insights into the aetiology of acute leukaemia.