Clonal Evolution In Aplastic Anemia Is Driven By Chromosomal Instability Rather Than Mutations In Myeloid Malignancy Candidate Gene

The pathophysiology of human aplastic anemia (AA) is immune-mediated destruction of bone marrow stem and progenitor cells. Most patients respond to immunosuppressive therapies (IST), which markedly improved survival in this disease. However, a minority of patients undergoes transformation to malignant hematologic disease, myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), usually accompanied by the cytogenetic abnormality monosomy 7. Clonal evolution in AA confers a poor prognosis in the clinic but is an opportunity to assess early events in oncogenesis in the setting of inflammation and tissue regeneration. We previously reported that low mean telomere length of leukocytes at the time of diagnosis of AA (Scheinberg et al., JAMA 2010) was associated with increased risk of progression to MDS. In the current study, we directly compared acquired mutations in candidate genes and chromosomal instability, as measured by telomere length, in a cohort of AA patients that had progressed to MDS.

Thirteen AA patients who developed monosomy 7 were compared with 30 AA patients who had received similar treatments but did not progress to MDS. Leukocytes telomere content was measured by qPCR in samples obtained at different time points from the diagnosis of AA and chromosome-specific telomere length was assessed by single telomere length assay (STELA) for Xp, Yp, 12q, and 17p. In the AA patients who had evolved to MDS and AML, analysis of acquired mutations in myeloid-specific genes was performed by comparison with control “germline” DNA from purified CD3 lymphocytes by exome sequencing.

Cells from AA patients with clonal evolution showed marked progressive telomere attrition, 419 bp/year during the period preceding development of monosomy 7. Telomere attrition was progressive from the time of diagnosis of AA. By STELA, accumulation of very short telomere fragments was apparent at 6 months after IST. In contrast, for the AA control group, patients whose disease was stable, telomere attrition was not accelerated in serial qPCR determinations, nor was there increased accumulation of short telomere fragments by STELA. A similar pattern of increased telomere attrition was reproduced in vitro by cultivation of bone marrow cells obtained six months after IST in all AA patients who developed MDS, while none of the AA control bone marrow cultured cells developed shorter telomeres.

We examined bone marrow myeloid cells at the time of monosomy 7 for acquired mutations in 125 candidate genes reported to be recurrently mutated in AML and MDS. Exome sequencing was performed using Agilent SureSelect Target Enrichment System. The raw reads were mapped to UCSC Human Genome hg19 by BWA software with default setting. With an average of 111-fold coverage on selected exon regions, somatic mutations were identified between paired samples using SAMtools and Shimmer software for SNP detection. Acquired mutations in myeloid cells were found in two cases. One patient had a heterozygous mutation in DNMT3A (K829T) present since diagnosis of AA. The other patient also had a heterozygous mutation in DNMT3A (P904S) as well as mutations in DOTL1, ASXL1, SETBP1, and STAT3. All these mutations were identifiable after IST as neutrophils recovered. Despite the presence of multiple mutations, this patient had shown a good hematologic response to IST; evolution was manifest as recurrent pancytopenia and stable marrow myeloblasts at about 5% for over 2 years after first detection of monosomy 7. The remaining 11 patients, all of whom lacked candidate gene mutations, had progressive increase in bone marrow myeloblast numbers; the only other three survivors in this cohort had received hematopoietic stem cell transplant.

In conclusion, telomere shortening rather than accumulation of point mutations in hematopoietic cells preceded aneuploidy and malignant transformation at an early stage of oncogenesis in this group of patients. These results from AA may be generalizable to other cancers arising in the setting of inflammation and tissue regeneration in other organs. Identification of critically short telomeres before the development of cytogenetic abnormalities may allow for improved management of patients at risk of clonal evolution, and pharmacologic strategies to increase telomerase activity might mitigate the risk of cancer in these settings