Telomere Length Is Associated with Specific Mutations and Mutation Classes in Patients with Acute Myeloid Leukemia

It is unknown if telomere length (TL) is associated with clinical outcome or molecular profile in acute myeloid leukemia (AML). We collected tumor samples from 67 AML patients treated at Memorial Sloan Kettering Cancer Center. DNA extraction was performed using viably frozen peripheral blood and bone marrow mononuclear cells. RainDance was used to amplify all exons of a set of 30 genes commonly mutated in AML. Capture was followed by next-generation sequencing; mutations were called if the variant was supported by >5% of the total number of reads (minimum >10 reads). TL was measured as mean telomere content by qPCR. Patients were assessed for FLT3 and NPM1 mutations, cytogenetics, and outcomes by chart review.

Results

In our 67 patient AML cohort, median TL was 5.22 kb (range 3.73-8.76). Median age was 64.1 years (range 26.2-84.4). While in healthy individuals TL shortens with age, in our cohort there was no association (R²=0.043). There was no difference in TL between newly diagnosed (ND) and relapsed/refractory (RR) patients or between de novo and secondary AML patients. In the 45 ND patients, there was a trend of early improved survival following sample collection in the longest TL tertile compared to the middle and shortest TL tertiles (86.7% vs. 60.0% and 33.3% at 6 months); however, this association was not statistically significant (p=0.662) (Fig 1). In the 22 RR patients, there was also a trend toward improved OS in the longest TL tertile (60.0% vs. 11.1% and 25.0%, p=0.284). In ND patients, there was no association between TL and primary induction failure or relapse-free survival.

Targeted sequencing data for 30 myeloid genes were available in 62 of the 67 patients. Analysis of single mutation correlation with TL showed that patients with IDH1 mutations had significantly longer TL than those without (p=0.02) (Table 1). Moreover, mutations in a set of genes associated with epigenetic functions (IDH1/2, ASXL1, DNMT3A, and TET2) also correlated with longer TL when examined together as a group (p=0.073).FLT3-ITD mutations were associated with shorter TL (p=0.084). The median ages of patients with IDH1 or FLT3 mutations were not different from the rest of the cohort. Of note, FLT3-mutated patients did have a higher WBC than FLT3 wild-type patients (p<0.001), suggesting that increased proliferative rate may be associated with shorter TL. Patients with RUNX1 mutations, t(8;21), or inv(16) also had a non-significant trend toward shorter TL (Table 1), and when we examined FLT3, RUNX1, t(8;21), and inv(16) together as a group, there was an association with shorter TL (p=0.026). There was no association between TL and NPM1 mutations. There was no difference in TL in patients with normal (n=35) vs. abnormal karyotype (n=31).

Conclusion

There was a trend of early improved survival for patients with the longest TL, suggesting that longer TL may be associated with better response rates to chemotherapy. However, the analysis was limited by the relatively small size of our cohort, andlarger studies are needed to further assess this association. We also demonstrated that IDH1 mutations are associated with longer TL (p=0.02), and that TL in general may be associated with specific classes of AML mutations. For example, mutations in transcription factors or receptor tyrosine kinases conferring a proliferative advantage may be associated with shorter TL, while mutations in epigenetic modifiers appear to be associated with longer TL. This is a novel and intriguing finding that warrants further study of TL, mutational profile, and epigenetic alterations in AML.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal