In addition to classical familiar forms of bone marrow failure, some cases of aplastic anemia (AA) have been linked to inherited germ line polymorphism/mutations of telomerase machinery, leading to excessive telomere shortening. Germline telomere maintenance machinery mutations have been also been found in a proportion of acute myeloid leukemia (AML) and Myelodysplastic syndromes (MDS) patients (pts). However, the molecular pathogenesis of adult MDS and AML is complex and determination of genetic risk factors in addition to established familial and congenital syndromes has been difficult. To date targeted sequencing has been used for mutational screens with the inherent limitations of limited exome coverage, empiric bias and labor intensity. New generation (NGS) whole genome approaches prioritize somatic mutations as initial discovery targets, but the availability of sequenced cohorts allows also for detection of germline lesions both in a targeted and an unbiased fashion.

Using NGS we studied 136 pts (mean age, 68.8 years, range 41-85) with MDS and related myeloid neoplasms for the presence of non-synonymous polymorphisms (SNV), which could affect telomerase machinery. These genes included TERT, DKC1, SMG6, NOP10, POT1, WRAP53, NHP2, GAR1, TINF2. No somatic defects of the telomerase complex were detected. We focused on novel sequence alterations or those described in available databases with a population allelic frequency of less than 5%. We identified 45 non-synonymous germline sequence alterations in 39 cases (32%). Most frequent SNV were found in TERT (n=15), DKC1 (n=7), SMG6 (n=6), NOP10 (n=4), POT1 (n=4), WRAP53 (n=4), while observations of NHP2 (n=3), GAR1 (n=1), TINF2 (n=1) were less prevalent. These variants were distributed in an almost mutually exclusive manner. Out of 3 variants in TERT, p.H412Y (n=3) and p.A279T (n=9) were reported to be pathogenic in bone marrow failure syndromes. In addition, p.A999T found in 8 cases in our cohort could also be pathogenic since it is less frequent in healthy controls. Similarly, p.441_442del (n=1), located in the N-terminal region, is a completely novel germline variant not detected in 6500 samples publicly available in ESP6500. In the pAML cohort (TCGA; n=197), the observations of germline variants for these telomerase complex genes were SMG6 (n=21), POT1 (n=19), NHP2 (n=1), NOP10 (n=1) GAR1 (n=1).

Next, we analyzed clinical characteristics, including treatment responsiveness as assessed per modified 2006 IWG response criteria. The mean age of the 39 patients with germline telomerase machinery alterations was 67 years, 24% (9/39) were younger (age<60 years) compared to 12% (12/97) of wild type (WT; p=.12). Of note, 58% of these cases had a family history of solid tumors including breast, gastrointestinal and prostate and 8% (3/36) had a family history of myeloid malignancies. 41% (16/39) of the telomerase mutants had higher-risk MDS/sAML at presentation compared to 23% in WT cases (23/97; p=.19). A higher percentage of mutants also had complex cytogenetics compared to WT (35% vs. 13%; p=.01). Response rates to common therapies, including hypomethylating agents were similar, but we noted that none of the carrier cases (n=16) treated with lenalidomide showed therapeutic responses (0% vs. 37%). The mean overall survival of the carrier cases was lower compared to the WT (36 vs. 39 months, p=.10). When we studied cases with telomerase alterations for the presence of coinciding somatic mutations, using a targeted deep sequencing panel of the 100 most common mutations acquired in pts with germline telomerase complex alterations, we found most common the acquisition of DNMT3 (18% vs. 6%, p.10) and cohesin mutations (13% vs. 4%,p=.11).

In sum, unbiased NGS sequencing approaches in MDS and related myeloid neoplasms allowed for identification of genetic germline alterations in telomerase maintenance machinery at higher rates than previously detected using targeted screening approaches, suggesting that such genetic defects may more frequently than previously thought contribute to cryptic and likely complex genetic predisposition to these diseases.

Disclosures:

Makishima:AA & MDS international foundation: Research Funding; Scott Hamilton CARES grant: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution