Acquired aplastic anemia (aAA) is an acquired deficiency of early hematopoietic cells, characterized by inadequate blood production, and a predisposition to myelodysplastic syndrome (MDS) and leukemia. Although its exact pathogenesis is unknown, aAA is thought to be driven by human leukocyte antigen (HLA)–restricted T cell immunity, with earlier studies favoring HLA class II-mediated pathways. Using whole-exome sequencing (WES), we recently identified 2 patients with aAA with somatic mutations in HLA class I genes. We hypothesized that HLA class I mutations are pathognomonic for autoimmunity in aAA, but were previously underappreciated because the major histocompatibility complex (MHC) region is notoriously difficult to analyze by WES. Using a combination of targeted deep sequencing of HLA class I genes and single nucleotide polymorphism array (SNP-A) genotyping, we screened 66 patients with aAA for somatic HLA class I loss. We found somatic HLA loss in 11 patients (17%), with 13 loss-of-function mutations in HLA-A*33:03, HLA-A*68:01, HLA-B*14:02, and HLA-B*40:02 alleles. Three patients had more than 1 mutation targeting the same HLA allele. Interestingly, HLA-B*14:02 and HLA-B*40:02 were significantly overrepresented in patients with aAA compared with ethnicity-matched controls. Patients who inherited the targeted HLA alleles, regardless of HLA mutation status, had a more severe disease course with more frequent clonal complications as assessed by WES, SNP-A, and metaphase cytogenetics, and more frequent secondary MDS. The finding of recurrent HLA class I mutations provides compelling evidence for a predominant HLA class I-driven autoimmunity in aAA and establishes a novel link between immunogenetics and clonal evolution of patients with aAA.