Acquired Aplastic Anemia (aAA) develops from the immune-mediated destruction of hematopoietic stem cells (HSCs) by auto-reactive T cells. Although the exact mechanisms remain incompletely defined, it is hypothesized that auto-reactive T cells target auto-antigens presented by human leukocyte antigens (HLA), leading to HSC destruction. Acquired genetic alterations resulting in clonal hematopoiesis may therefore emerge in patients with aAA as a means of evading this immune response. Well-recognized patterns of clonal cellular evolution in aAA include myelodysplastic syndrome (MDS) and paroxysmal nocturnal hemoglobinuria (PNH). Recent studies have identified acquired copy-number neutral loss of heterozygosity on the short arm of chromosome 6 (6p CN-LOH) as another recurrent abnormality. The HLA loci map to 6p, and deletions or LOH in this region result in the loss of alleles from one parental HLA haplotype. 6p CN-LOH may allow for immune escape and restoration of hematopoiesis in aAA by deleting the HLA haplotype to which the HLA-restricted auto-reactive T cell response is targeted. The effect of 6p CN-LOH on clinical outcome and further clonal evolution in aAA is not well described. Thus, the aim of our study was to investigate the frequency, evolution and clinical significance of clonal 6p CN-LOH in our institutional cohort of adult and pediatric patients diagnosed with aAA.
Bone marrow (BM) or peripheral blood (PB) samples were collected for genome wide single nucleotide polymorphism (SNP) arrays (Illumina Human OMNI1 or 850K BeadChip), and analyzed for the presence of copy number abnormalities and loss of heterozygosity, with specific reference to 6p CN-LOH. Clinical outcome data were extracted from patient charts. High Resolution HLA typing information was obtained from patient charts or performed on a research basis. HLA allele frequency in our cohort was compared to that of published control populations from the National Marrow Donor Program. Quantitative comparison of HLA haplotype allele frequency in patients with 6p CN-LOH was performed by next generation sequencing to determine the identity of clonally deleted parental alleles.
Our cohort was comprised of 71 patients with aAA and related conditions including PNH and hypoplastic MDS (AA (n=64), PNH (n=5), MDS (n=2); median age 14 years, range 8 months to 68 years; 52% females). A total of 68 patients (96%) had a SNP array performed. 6p CN-LOH (n=8) or acquired clonal deletion of 6p (n=1) was detected in 13% of these patients, with the prevalence significantly higher in adult versus pediatric patients (29% versus 8.3%, p < 0.045). In all patients, clones were first identified at least 6 months after initial diagnosis. The majority of patients with 6p CN-LOH (n=7) received immunosuppressive therapy (IST). Of these, 43% failed IST requiring additional therapy while the rest achieved only a partial response. Importantly, no patient with 6p CN-LOH receiving IST alone achieved complete remission at the time of analysis. The characteristics of 6p CN-LOH varied considerably in terms of the number of clones with different sized regions of LOH, the percentage of cells with LOH, and the specific HLA loci included in the LOH region. 6p CN-LOH patients in whom follow-up SNP analyses were performed exhibited stable clone size, with no new clonal abnormalities. Quantitative analysis of HLA alleles in 6p CN-LOH patients revealed many distinct functionally deleted HLA alleles. While no specific allele was recurrently lost in the 6p CN-LOH clones, several of these deleted alleles occurred at increased frequency in our overall aAA cohort compared to ethnicity-matched control populations.
In conclusion, the results from our institutional aAA cohort confirm the prevalence rates of clonal 6p CN-LOH described previously in ethnically distinct populations, while demonstrating an increased frequency of 6p CN-LOH in adult versus pediatric aAA patients. None of the aAA patients with 6p CN-LOH in our cohort achieved a complete response to IST alone, suggesting that 6p CN-LOH may be a risk factor or marker predicting poor response to IST. Taken together, our findings underscore the critical importance of understanding the mechanisms underlying clonal evolution in aAA and their effects on disease pathogenesis and response to therapy. Future multi-institutional studies are needed to define optimal therapeutic approaches for aAA patients with 6p CN-LOH.
Disclosures
No relevant conflicts of interest to declare.