Acquired Aplastic Anemia: New Genetics, New Genomics

Clinical and laboratory data have established that aplastic anemia is an immune-mediated disease in which T lymphocytes destroy hematopoietic stem and progenitor cells. Immunosuppressive therapies are effective in the majority of patients. Addition of the thrombopoietin mimetic eltrombopag increases the overall and complete response rates to anti-thymocyte globulin and as a single agent can rescue patients refractory to immunosuppressive therapy (IST). Multi-lineage robust blood count improvements with increased marrow cellularity suggest activity of eltrombopag on the hematopoietic stem cell. Recently, genetic factors have been identified that increase the risk of bone marrow failure: adults may present in the clinic with late manifestations of a pediatric syndrome (dyskeratosis congenita), but more typically there are no physical anomalies, often no family history, and earlier blood counts may be normal. In the telomeropathies, which in later life are almost always due to mutations in TERT (which encodes the telomerase) or TERC (which encodes the RNA template), there may be personal or familial multi-organ involvement, especially of liver and lung; early hair greying is a useful clinical clue. Detection of extremely short telomeres of leukocytes, accompanied by gene sequencing, is used to establish the diagnosis. In the syndrome defined by GATA2 mutations, there may be a history of unusual or persistent mycobacterial and viral infections, and monocytopenia preceding pancytopenia. Diagnosis requires sophisticated interpretation of gene sequencing. Large genomic data sets are now available acquired (somatic) mutations in aplastic anemia. For almost 300 NIH aplastic anemia patients treated with IST, candidate gene sequencing of myeloid cells obtained six months after treatment revealed somatic mutations in about one-third of cases. PIGA was most frequently abnormal, followed by BCOR1, DNMT3A, and ASXL1. Initial variant allele frequency was usually low. Mutations in a subset of genes predicted poor survival and progression to myelodysplastic syndrome and acute myeloid leukemia, while mutations in BCOR and PIGA correlated favorably with outcomes. When we also examined a subset of patients at the time of progression to monosomy 7 with pancytopenia and/or incipient leukemia by whole exome sequencing.DNMT3A and ASXL1 were implicated in only two cases, RUNX1 in another, and there were no novel recurring mutations. Telomere attrition was markedly accelerated in these monosomy 7 cases. A good model of evolution from bone marrow failure to myeloid malignancy centers on haploinsufficiency of specific genes, by mutations or chromosome loss, both of which would be favored in a stressed, regenerative environment. The failed marrow environment may select for defective cells, as, for example, in differentiation capability. Malignant evolution should be predictable in the clinic and amenable to therapeutic intervention.