Clinical Impact of Telomere Shortening in Normal and Leukemia Cells in Chronic Lymphocytic Leukemia

Introduction: In chronic lymphocytic leukemia (CLL), short telomere length in the leukemia cells predicts poor prognosis. However, it is not known whether telomere length in normal tissues also predicts patient outcome and can improve the prognostic value of CLL telomere length. Prognosis in CLL is heterogeneous with the primary cause of death being cancer and infections, thought related to immunosuppression. The elderly, males and those with comorbidities have a particularly poor relative survival. Whether this reflects differences in the biology of CLL in these patient groups, or to extrinsic factors such as increased immunosuppression or fragility, is unknown. In normal individuals, the telomeres in somatic cells (ie, buccal cells (BC)) shorten with aging and prior comorbidities; short telomeres being predictive of early mortality related to infections, cancer and cardiovascular disease. In the present study, we have evaluated BC telomere length in patients with CLL, to determine if this is predictive of survival, and can enhance the predictive value of leukemia cell telomere length.

Methods: The Manitoba CLL tumor bank contains 235 samples from newly diagnosed CLL patients between 2007-2011; with a median follow-up of 2 years. One-quarter of these patients required chemotherapy and one-tenth have died. Genomic DNA was extracted from purified CLL cells and buccal cells (BC) collected at diagnosis. Telomere length was established by multiplex quantitative real-time PCR. Telomere/standard (t/s) ratio was calculated using the beta-globulin gene as the standard. Statistical analysis was performed using Statistical Analysis Software (SAS) and Prism software.

Results: The median adjusted telomere length was much shorter in CLL cells than in BCs being 0.53 and 2.01, respectively. In BCs, telomere length significantly shortened with increasing age (OR 1.04, CI 95% (1.00-1.078)) and at this short follow-up time, correlation with second malignancies was approaching significance (p=0.06). However, BC telomere length was not reflective of the number of comorbidities or survival. In contrast, telomere length in CLL cells was independent of age (p=0.44) and sex (p=0.75) confirming that these factors did not influence the cellular biology of the disease. Telomere length also correlated with other biological markers with short telomeres correlating with unmutated IgHV status (p<0.0001), Zap-70 positivity (p=0.05), and CD38 positivity (p=0.003). These patients with short telomeres also had clinical markers of poor prognosis including short lymphocyte doubling time (p=0.004), higher Rai stage (p=0.02) and an earlier time to treatment (p<0.0001). The prognostic value of CLL telomere length was not enhanced by the addition of BC telomere length.

Conclusions: These results demonstrate that BC telomere length in CLL patients shorten with age, and short telomeres may predict subsequent second malignancies. Telomere length in CLL cells correlates with biological and clinical markers of aggressive disease, but it does not explain the poor prognosis seen in the elderly and male patients. While BC telomere length does not seem to influence the initial clinical course of CLL, ongoing studies are evaluating whether it is predictive of the long-term risk of infections and second malignancies. In addition, whether BC telomere length in CLL is shorter than in the normal population and whether shortening correlates with specific comorbidities is presently being determined.