Imetelstat Treatment Results in Clinical Benefits, Including Improved Overall Survival, in Patients with Higher-Risk Triple-Negative Myelofibrosis Relapsed/Refractory to Janus Kinase Inhibitors (JAKi)

Background: Triple-negative (TN) is a high-risk molecular signature in MF patients (pts) associated with a higher incidence of leukemic transformation and approximately 3y overall survival (OS) from diagnosis vs pts carrying a mutation in JAK2, CALR or MPL. Allogeneic hematopoietic stem cell transplantation (alloHSCT) is the only potentially curative treatment for MF, but TN MF pts also have worse prognosis and non-relapse mortality vs non-TN pts after alloHSCT. New agents with novel mechanisms of action beyond JAKi are needed to treat TN pts. IMbark (MYF2001; NCT02426086) was a 2-dose, randomized, single-blinded, phase 2 study of imetelstat (telomerase inhibitor) that enrolled R/R intermediate-2/high-risk MF pts, including TN pts. 32% symptom response rate and median OS of 29.9 mo were reported in the overall population on the 9.4 mg/kg arm, with acceptable safety (Mascarenhas ASH 2018 #685). Here we report data from enrolled TN pts.

Aims: To analyze clinical outcomes, including OS, in TN pts enrolled in the IMbark study and determine if this molecularly defined subset, associated with poor outcomes, benefits from imetelstat.

Methods: DNA extracted from blood collected at baseline was analyzed for driver mutations by next-generation sequencing. Telomere length (TL) was assessed by quantitative fluorescence in situ hybridization and human telomerase reverse transcriptase (hTERT) level by Taqman RT-PCR. Bone marrow fibrosis was assessed by central pathologist.

Results: Sequencing data were available for 105 pts. JAK2V617F, CALR, or MPL driver mutation was observed in 61%, 8.5% and 5.7% of pts, respectively; 24.8% of pts were TN. On the 9.4 mg/kg arm, spleen response (≥35% SVR) rate was higher in TN (18.8%) vs non-TN pts (7.3%); and symptom response (≥35% SVR) was higher in TN (50%) vs non-TN (24.4%). Higher rate of fibrosis improvement (ie, decrease in fibrosis by ≥1 grade per central review) was noted in TN (50%) vs non-TN (35%) despite TN pts having worse fibrosis at study entry (91.7% TN vs 65.2% non-TN had grade 3 fibrosis). All correlative analyses were not statistically significant, possibly due to small sample size.

Figure 1 showed that imetelstat treatment at 9.4 mg/kg resulted in longer median OS of 35.9 mo for TN pts (95% CI 23.2, NE) vs 24.6 mo for non-TN pts (95% CI 19.6, 29.9) with HR=0.45 (95% CI 0.19, 1.03, p=0.05). The prolonged OS in TN pts R/R to JAKi in the study is improved to the reported OS from diagnosis in TN pts.

To elucidate biological evidence supporting clinical benefits in TN pts, relationships between TN status with baseline TL and hTERT level were explored. A higher proportion of TN pts had shorter baseline telomeres (78.6% vs 45.9% for TN vs non-TN) and higher levels of hTERT (60% vs 47.5% for TN vs non-TN). Furthermore, a trend for higher spleen and symptom response rates in pts with shorter TL or higher hTERT level at baseline was also observed, which may explain why the telomerase inhibitor imetelstat provides improved clinical benefits in TN pts.

Conclusions: TN MF pts are a subpopulation associated with poor prognosis. In this study, in MF R/R to JAKi, TN pts treated with imetelstat 9.4 mg/kg had better clinical outcomes and OS vs non-TN pts, suggesting that treatment with imetelstat may overcome the poor outcomes expected with TN pts. Furthermore, TN pts had shorter TL and higher hTERT levels at baseline, representing a suitable target population for telomerase inhibitor treatment. These data warrant further investigation of imetelstat in a clinical trial in TN pts who have limited treatment options.