Dynamics of Telomere Length Reflect the Clonal Suppression Seen with the Telomerase Inhibitor Imetelstat in Patients with Essential Thrombocythemia

Background/Aims:

Myeloproliferative neoplasms (MPN), including essential thrombocythemia (ET), polycythemia vera (PV) and myelofibrosis (MF), are clonal disorders characterized by unlimited proliferation of hematopoietic cells. Besides the dominant clone, in the majority of cases defined by a driver mutation (JAK2V617F, CALR or MPL mutation), sub-clones carrying additional mutations (e. g. in the genes ASXL1, CBL, DNMT3A, EZH2, IDH1, U2AF1, SF3B1, TET2, TP53) and/or normal polyclonal hematopoiesis coexist. Due to the loss of telomere sequences with each cell division, the length of telomeres can be used to describe clonal dynamics of hematopoietic cells. In neoplastic cells with a high mitotic rate shorter telomeres are typically found compared to the telomere length (TL) in normal counterparts. In the ET phase 2 study with the telomerase inhibitor imetelstat we demonstrated rapid and durable hematologic and molecular responses (Baerlocher et al. N Engl J Med 2015) and suppression of clones with non-driver mutations (Oppliger Leibundgut et al, ASH 2015).

Here, our aims were 1) to evaluate the TL in subsets of leukocytes from patients (pts) with MPN compared to healthy individuals and 2) to study the TL dynamics in ET pts treated with imetelstat and to correlate those with hematologic and molecular responses.

Patients and Methods:

Blood samples from 63 standard of care (SOC)-treated or untreated pts with MPN (15 ET, 36 PV, 12 MF) diagnosed according to WHO 2008 criteria as well as from 14 ET pts treated with imetelstat after failure or intolerance to ≥1 prior therapy were analyzed for TL. TL percentiles from over 400 healthy individuals served as reference. Leukocytes were extracted from the peripheral blood and TL was measured in subsets of leukocytes by automated multicolor flow-FISH (Baerlocher et al, Nat Protoc. 2006).

Results:

81% of the 62 MPN pts (11/15 ET, 32/35 PV, 7/12 MF) carried a JAK2 mutation, 6 (1 ET, 5 MF) a CALR mutation, 1 ET pt had a MPL mutation, 2 ET pts were triple negative and 3 PV pts were JAK2 wild type. 23 pts were treated with phlebotomy, 32 with hydroxyurea, 3 with anagrelide and 3 with IFN-alpha. The median age at study entry was 62 yrs (23-89, ET/PV/MF 64/58/66 yrs) and the median time since diagnosis was 2.7 yrs (0.3-14.2) in the ET SOC cohort. In the imetelstat cohort, 9/14 ET pts carried a JAK2V617F mutation and 5 a CALR mutation. The median age at study entry was 60.5 yrs (21-80) and the median time since diagnosis was 6.7 yrs (0.3-21). The telomere length values (TLV) in granulocytes from the 63 MPN pts on SOC were around the 10^th percentile for pts with ET and PV and below the 1^st percentile for most pts with MF. In the imetelstat cohort, 8 of 14 ET pts with a median of 2 prior therapies demonstrated TLV in granulocytes <1^st percentile before the imetelstat treatment, which is significantly lower than in ET pts on SOC (p < 0.01). In 6 of 9 ET pts on imetelstat the TLV were higher after 9 months. In addition, this difference in TL correlated significantly with the maximum reduction of the JAK2V617F mutational burden (p = 0.0137). Of interest, the 3 pts with lower or steady TLV after 9 months of treatment had the highest number of additional mutations before imetelstat treatment and despite a good suppression of the clone with the driver mutation these sub-clones were only partially responsive to imetelstat (i.e. 3 to 5 additional mutations in ASXL1, CBL, DNMT3A, EZH2, TET2, TP53, SF3B1 and U2AF1). Overall, pts with high level additional mutational burden had shorter TLV (age-adjusted mean difference from 50^th percentile ±STD -3.6 ± 0.5 kb) at baseline than pts with no or low level additional mutational burden (age-adjusted mean difference from 50^th percentile ±STD -2.9 ± 1 kb), and in both sets of pts mean TLV were higher after 9 months of imetelstat treatment.

Conclusions:

The lower TLV found in granulocytes of pts with MPN and especially with MF compared to healthy individuals reflect the higher mitotic history of malignant clones. In the ET imetelstat cohort, the very low TLV found at baseline might result from longer disease duration and resistance and/or intolerance to prior therapies. The higher TLV observed after 9 months of treatment with imetelstat and correlation with the reduction in the driver mutational burden suggests that imetelstat treatment in ET pts may suppress neoplastic clones and, in the absence of a high additional mutational burden, favor recovery of normal hematopoiesis.