Flow-FISH evaluation of telomere length in Philadelphia-negative myeloproliferative neoplasms

Telomeres are repeated DNA sequences localized at the end of chromosomes and involved in maintaining chromosomal stability and integrity.¹ Telomere length (TL) is considered a reliable marker of cell turnover, as telomeres progressively shorten with aging and cell proliferation. In various hematologic malignancies, hematopoietic cells have been shown to be characterized by shortened telomeres.¹ So far, a few studies have evaluated the role of TL in myeloproliferative neoplasms (MPN).^2–6

We analyzed TL in a cohort of MPN patients in order to investigate its relationship with biological and clinical parameters. We studied 139 consecutive MPN patients, including 42 cases of polycythemia vera (PV), 63 essential thrombocythemia (ET), 18 primary myelofibrosis (PMF), 12 post-PV MF, 4 post-ET MF followed at the Department of Hematology Oncology, Fondazione IRCCS Policlinico San Matteo, University of Pavia, Italy. Results were compared with findings in a cohort of 65 healthy subjects. This study was approved by the Ethics Committee, Fondazione IRCCS Policlinico San Matteo, Pavia, and the procedures followed were in accordance with the Helsinki Declaration of 1975, as revised in 2000.

Demographic and hematologic characteristics of patients are listed in Table 1. TL was measured in both peripheral blood (PB) granulocytes and lymphocytes by flow-FISH, as previously described,⁷ with at least 30,000 events acquired for each sample. The age-corrected value of TL in granulocytes (deltaTEL_gran) was obtained by subtracting the individual patient’s measurement from the linear regression line established for healthy subjects: consequently, the shorter the patient’s TL value, the more negative the deltaTEL_gran value.

Table 1.

Demographic and hematologic characteristics of 139 MPN patients at the time of telomere length evaluation.

As a first step, we evaluated telomere fluorescence in PB granulocytes and lymphocytes, expressed as Molecular Equivalent of Soluble Fluorescence (MESF), from MPN patients and compared it to that of control subjects. In healthy individuals a significant negative linear correlation was found between telomere fluorescence of both granulocytes and lymphocytes and age (Spearman’s rho −0.90; P<0.001), whereas in MPN patients this relationship was retained in lymphocytes (r −0.30; P<0.001) but lost in granulocytes (r −0.07; P=0.40) (Figure 1A), suggesting that in MPN patients the impact of a clonal hematopoiesis exceeds the expected effect of demographic parameters on telomere shortening.

Figure 1.

Dynamics of telomere shortening in patients with myeloproliferative neoplasm (MPN) and in healthy subjects. (A) In healthy controls Spearman’s rank test revealed a significant negative linear correlation between telomere fluorescence, expressed as Molecular Equivalent of Soluble Fluorescence (MESF), and age both in peripheral blood (PB) granulocytes and lymphocytes (P<0.001). In contrast, in MPN patients this relationship was present in PB lymphocytes (P<0.001) but not in granulocytes (P=0.40). (B) Age-corrected telomere length (DeltaTEL_gran) is significantly shorter in MPN patients (median value −11.14, range −23.99 to 9.78) than in healthy individuals (−0.05, −4.74 to 5.75; Kolmogorov-Smirnov’s test P<0.001).

When considering the age-corrected value, patients with MPN showed a significantly lower deltaTEL_gran (median value −11.14, range −23.99 to 9.78) than healthy individuals (−0.05, − 4.74 to 5.75; P<0.001) (Figure 1B). In keeping with published data,^4,5 we did not find any significant difference in deltaTEL_gran among MPN subtypes (P=0.249): median deltaTEL_gran in PV patients was –11.09 (25^th percentile −16.90, 75^th percentile −7.30), in ET −11.43 (−17.04 to −7.01) and in PMF, post-PV MF and post-ET MF −11.44 (−16.16 to −1.18). Moreover, no significant difference was observed between deltaTEL_gran in chronic phase of PV and ET and in their relative fibrotic evolution (P=0.64 and P=0.84, respectively). The wide variability in TL observed in PMF patients might reflect the heterogeneity of biological and pathogenetic features of this disease, even if further data from a larger cohort of patients are needed to investigate this observation thoroughly. In contrast, disease duration significantly affected deltaTEL_gran (Spearman’s rho −0.16; P=0.049), also when adjusted by MPN subtypes in multivariate analysis (P=0.01).

Overall, 97 of 139 (70%) patients carried the JAK2 (V617F) mutation, evaluated with a quantitative AS-PCR.^8,9 The deltaTEL_gran was similarly shortened in JAK2 (V617F)-positive and JAK2 (V617F)-negative patients. This result was confirmed both by analyzing MPN patients as a whole (P 0.81) and according to MPN subtype (ET P=0.69, PMF P=0.44). In contrast, no significant relationship was observed between deltaTEL_gran and percentage of mutant alleles, either when considering this latter as a continuous variable (Spearman’s rho 0.03; P=0.73), or when comparing patients with less than 50% or 50% or more mutant alleles (P=0.56). Furthermore, we did not found any difference in deltaTEL_gran between patients carrying an MPL mutation¹⁰ (6 out of 99 evaluable patients, 6%) and patients with wild-type MPL (P=0.46). Taken together, these data confirm that patients with MPN present shortened telomeres irrespective of their JAK2 or MPL mutational status, implying that different molecular mechanisms may affect stem cells in the same way, and suggest that telomere assessment may be of potential diagnostic value in the work up of suspected MPN.

Finally, we investigated the effect of cytoreductive therapy on telomere shortening. At the time of TL evaluation, 87 of 139 patients were on cytoreductive therapy, mainly hydroxyurea. Our analysis showed that deltaTEL_gran was not affected by cytoreductive therapy per se (P=0.93) nor by its duration (Spearman’s ρ −0.07; P=0.42). Future evaluation of the impact of interferon α₁₁ or innovative drugs¹² on TL in MPN might be of interest.