Somatic Mutations Implicated In Myeloid Malignancies Are Frequent In Idiopathic Aplastic Anaemia and Its Relevance To Disease Classification and Treatment- a Comprehensive Analysis Of 150 Patients

Acquired aplastic anemia (AA) is an immune mediated disease, characterized by severe quantitative defects in stem cell number leading to a hypocellular marrow and peripheral blood cytopenias. A major complication of AA following immunosuppressive therapy (IST) is the emergence of clonal hemopoiesis, manifesting as paroxysmal nocturnal hemoglobinuria (PNH), myelodysplastic syndrome (MDS) and acute myeloid leukaemia. The nature of genetic defects in such cases evolving to clonal hemopoiesis has not been addressed hitherto. It is likely that the hematopoietic stress leads to telomere attrition, chromosomal instability and susceptibility to acquire mutations in genes several years after the initial diagnosis. We postulated the existence of mutations leading to clonal haemopoiesis in a subset of AA and thereby helping in early prediction.

We undertook a massively parallel targeted sequencing of 832 genes implicated in regulating hematopoiesis initially in a cohort of 57 patients with acquired AA. We also performed single nucleotide polymorphism-array karyotyping, telomere measurement, PNH flow cytometry and correlated it with clinical characteristics. A second cohort of 93 patients was sequenced at a greater depth for ASXL1 and DNMT3A.

The median age of AA patients was 44 years (range 17-87 years). The diagnosis was non severe AA (NSAA) 64, severe AA (SAA) 52, very severe AA (VSAA) 27 and pure red cell aplasia (PRCA)/amegakaryocytic thrombocytopenia 7. PNH-type cells and cytogenetic aberrations were seen in 52% and 9% of patients respectively. IST was administered in 72% of patients with 55% receiving ATG containing regimens. Allogeneic hematopoietic stem cell transplant was performed in 25% (n=37). Clonal evolution to MDS/AML was observed in 12% of patients.

Somatic mutations in genes usually implicated in myeloid malignancies were detected in 15% (22/150) of AA. No particular sub-group of AA patients had a predilection to harbour somatic mutations, although a slightly higher incidence was noted in SAA and NSAA, compared to VSAA (22% vs. 14%, p=0.1).Thirty-five (23%) AA patients had either morphological (n=18) and/or molecular (n=22) evidence of evolution to MDS. Of the 18 patients with morphological evolution to MDS, ASXL1 and TET2 mutations were observed in 5 patients. ASXL1 (n=11) mutations were the most common somatic variation and correlated with older age, evolution to MDS and the presence of monosomy 7. Although there was no predilection for a particular ‘hot spot’, all mutations occurred in exon 12 and were uniformly heterozygous. DNMT3A (n=7) and BCOR(n=2) were the other frequent mutations. The median allele burden of somatic mutations was 30%. AA with somatic mutations were older (51 vs. 38 years, p<0.07), had a longer disease duration (44 vs. 9 months, p< 0.06), shorter telomere lengths (median T/S length, 1.3 vs. 0.8, p<0.008) and were less likely to achieve complete remission to IST (10% vs. 34%) compared to those without mutations.

Up to 25% of AA patients have morphological or molecular evidence of clonal evolution with a fifth of AA patients harboring mutations in genes implicated in myeloid malignancies, with both diagnostic and therapeutic implications.

No relevant conflicts of interest to declare.