Momelotinib for myelofibrosis: 12‐year survival data and retrospective comparison to ruxolitinib

Abstract

To the Editor: Momelotinib is one of the several JAK2 inhibitors used to treat myelofibrosis, with FDA approval so far secured for ruxolitinib, fedratinib, and pacritinib.1 All four JAK2 inhibitors are effective in reducing spleen size and alleviating symptoms. The additional value of momelotinib in combating anemia, ascribed to its inhibitory activity on activin receptor type,1, 2 was first recognized a decade ago3 and more recently confirmed in a controlled setting.4 Our group pioneered the initial pre-clinical5 and early phase clinical3 studies of momelotinib for the treatment of myelofibrosis, which resulted in accrual of 100 Mayo Clinic patients, between November 2009 and March 2011. Details on this initial phase 1/2 core study (NCT00935987), including patient inclusion criteria, treatment efficacy and toxicity, and reasons for drug discontinuation have since been published;6 efficacy-wise, 12-week anemia response was 54%, including 68% of those who were transfusion-dependent, and spleen response was 40%; toxicity-wise, grade-3/4 adverse events included thrombocytopenia 34% and neutropenia 8% while noteworthy grade-1/2 adverse events included diarrhea 48%, nausea 39%, vomiting 24%, dizziness 40%, peripheral neuropathy 30%, and flushing 11%; in addition, all grades increase in liver function tests and pancreatic enzymes were documented in 15%–18% and 11%–13%, respectively. In the current study, we focused on long-term survival data on 79 Mayo Clinic study participants who were JAK2 inhibitor-naïve at time of study entry. A separate cohort of 50 Mayo Clinic patients who participated in another phase 1/2 study of ruxolitinib for myelofibrosis (NCT00509899), accrued between October 2007 and February 2009, was used as a comparator.7 The current study was conducted under an institutional review board approved minimum risk protocol that allowed retrospective collection and analysis of data from Mayo Clinic patients who participated in the above-specified phase-1/2 core studies for momelotinib (NCT00935987) and ruxolitinib (NCT00509899). Study design, drug doses and schedule, and additional details on treatment effect and side effects have previously been published.6, 7 The current study is restricted to analysis of overall survival data, leukemic transformation rates, on-treatment survival (drug retention), and outcome of allogeneic hematopoietic stem cell transplant (AHSCT) in study patients who underwent the procedure after failing treatment with the study drug. Conventional statistical methods were applied using JMP Pro 16.0.0 software (SAS Institute, Cary, NC, USA). Survival was calculated from time of drug initiation for both momelotinib and ruxolitinib, including those who received AHSCT after failing treatment with either momelotinib or ruxolitinib. On-treatment survival is calculated from time of initiation to time of discontinuation of the study drug (momelotinib or ruxolitinib). Table S1 lists patient characteristics at time of study entry, including cytogenetic and mutation information on study patient cohorts for momelotinib (median age 67 years, range 34–89; 54% males) and ruxolitinib (median age 62 years, range 39–78; 74% males). Patients on the ruxolitinib arm were more likely to be younger (p = .002), males (p = .02), JAK2 mutated, (p = .01), affected by constitutional symptoms (p = .002), and harbor very high-risk karyotype (p = .02) (Table S1). Patients on the momelotinib arm were more likely to belong to higher risk category (p = .01), require red cell transfusions (p = .002), display thrombocytopenia <100 × 109/L (p = .002), and harbor CALR type 1/like mutation (p = .01) (Table S1). ASXL1 and SRSF2 mutation distribution and median time from initial diagnosis to study entry were similar between the two study cohorts. All but one patient on the ruxolitinib treatment cohort was followed to death or 2022; median follow-up for living patients was 11.7 years (range 11.4–14.4) for momelotinib and 14.2 years (range 8–14.4) for ruxolitinib (p = .36). During this period, 68 (86%) deaths, 75 (95%) treatment discontinuations, 13 (16%) leukemic transformations, and 7 (9%) AHSCTs were documented in the momelotinib treatment group; the corresponding figures for ruxolitinib-treated patients were 86%, 98%, 18%, and 12% (Table S1). Figure 1 illustrates survival data for both momelotinib and ruxolitinib treatment cohorts; median survivals from initiation of the study drug were 3.5 years (10-year survival 20%) for momelotinib and 4.0 years (10-year survival 23%) for ruxolitinib (p = .32). On-treatment survival was superior for momelotinib, compared to ruxolitinib, with 3/5-year drug discontinuation rate of 68%/84% versus 88%/97%, respectively (p < .001; Figure 2). AHSCT had a favorable survival impact on 13 cases who underwent the procedure after failing momelotinib or ruxolitinib (Figure 1); median survivals were not reached (10-year survival 68%) for transplanted patients versus 3.2 years (10-year survival 15%) for non-transplanted patients (p < .001). The two treatment groups were subsequently combined (n = 129) for ensuring adequate number of informative cases in specific mutation categories; multivariable analysis identified AHSCT (p < .001), younger age (p = .03), presence of type 1/like CALR mutation (p = .03), and absence of ASXL1/SRSF2 mutation (p = .003), as independent predictors of superior survival (Table S1). These risk factors are similar to those generally considered in the presence or absence of JAK2 inhibitor therapy.1 The current study features the longest follow-up data on record concerning myelofibrosis patients treated with JAK2 inhibitors, in general, and momelotinib and ruxolitinib, in particular. In previous publications, we had shown the lack of a significant survival impact from either momelotinib8 or ruxolitinib,9 in patients with myelofibrosis, compared to risk-adjusted patient cohorts not receiving JAK inhibitor therapy. In the current study, we show similar long-term survival data for momelotinib versus ruxolitinib (p = .32), despite significantly different drug retention rates that favored momelotinib (Figure 2; p < .001), possibly related to its salutary effect on anemia. However, this was not a randomized study, and the particular observation is, therefore, not conclusive. What is instead clearly shown is the inability of either one of the two JAK2 inhibitors to secure long-term survival and the relatively high rate of drug discontinuation, often occurring in the first 2 years of treatment. Increasing familiarity with the use of these drugs should improve drug retention but is unlikely to retard disease progression. We are encouraged by the favorable survival impact of AHSCT (Figure 1), which argues for its implementation sooner than later, especially in the presence of high-risk genetic features. Outcome of AHSCT in myelofibrosis is likely to be improved by further refinements of pre- and post-transplant measures to curb graft failure, delayed count recovery, graft versus host disease, infection risk, and disease relapse.10-14 Finally, we noted that a small number (<10%) of patients on momelotinib remained on active therapy for over 10 years, without undergoing AHSCT, an observation that is intriguing and worthy of further investigation. The authors declare no potential conflict of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. Table S1 (Top half) Clinical and laboratory characteristics of 129 patients with myelofibrosis obtained at time of study entry, stratified by treatment with momelotinib (n = 79) or ruxolitinib (n = 50); (bottom half) predictors of inferior survival among all 129 patients treated with either momelotinib or ruxolitinib. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts