Dual transfusion independence and spleen volume reduction is associated with overall survival in patients with myelofibrosis treated with momelotinib: Post hoc analyses of SIMPLIFY-1 and MOMENTUM

Abstract

Abstract Introduction: Myelofibrosis (MF) is a progressive, life-limiting hematologic malignancy characterized by splenomegaly, anemia, and constitutional symptoms. Improvements in both anemia and splenomegaly have been linked to improved overall survival (OS) in MF; although spleen volume reduction ≥35% (SVR35) is the benchmark for spleen response and has been accepted as an endpoint for drug approval, its validity as the optimal indicator for survival remains uncertain. In a previous analysis of JAK inhibitor (JAKi)–naive patients with MF and baseline hemoglobin <10 g/dL in the phase 3 SIMPLIFY-1 trial (Palandri F, et al. EHA 2025. Poster P1829), transfusion independence (TI) at week 24 was associated with longer OS with the JAK1/JAK2/ACVR1 inhibitor momelotinib, with (hazard ratio [HR], 0.40; 95% CI, 0.18-0.89) or without (HR, 0.19; 95% CI, 0.06-0.65) achievement of SVR35. To better understand the impact of dual SVR and TI on OS in JAKi-naive and -experienced patients with MF, these post hoc analyses further explores the relationship between TI, different SVR thresholds, and OS in the SIMPLIFY-1 and MOMENTUM trials. Methods: SIMPLIFY-1 (NCT01969838) and MOMENTUM (NCT04173494) were randomized, double-blind, phase 3 trials in JAKi-naive and -experienced patients, respectively. In these post hoc analyses of the overall safety population randomized to momelotinib in each trial (SIMPLIFY-1, n=214; MOMENTUM, n=130), OS was compared between dual SVR+TI responders at week 24 vs nonresponders across various SVR thresholds (≥35%, ≥25%, ≥15%, and ≥10%). OS was analyzed using the Kaplan-Meier method, with median OS, HRs, and nominal log-rank P values reported. As all patients in both trials received open-label momelotinib after week 24, this crossover design precludes evaluation of outcomes beyond week 24 in the comparator arms. Results: In SIMPLIFY-1, JAKi-naive patients who had TI and achieved SVR35 at week 24 with momelotinib regardless of baseline anemia status had longer OS, which is consistent with the previous analysis in baseline anemic patients. Median OS was not estimable (NE) in both dual responders (n=53; 95% CI, NE-NE) and nonresponders (n=161; 95% CI, 44.09 mo-NE), but the difference in favor of responders was nominally significant (HR, 0.51; 95% CI, 0.27-0.99; P=.0425). Nominally significant OS benefits of dual TI and SVR response were observed with all other SVR thresholds evaluated: (1) SVR25+TI responders (n=77) vs nonresponders (n=137): median, NE (95% CI, NE-NE) vs 46.2 mo (95% CI, 41.53 mo-NE); HR, 0.46 (95% CI, 0.26-0.82; P=.0068); (2) SVR15+TI responders (n=103) vs nonresponders (n=111): median, NE (95% CI, NE-NE) vs 44.1 mo (95% CI, 35.5 mo-NE); HR, 0.33 (95% CI, 0.19-0.57; P< .001); and (3) SVR10+TI responders (n=113) vs nonresponders (n=101): median, NE (95% CI, NE-NE) vs 41.5 mo (95% CI, 30.6 mo-NE); HR, 0.28 (95% CI, 0.16-0.47; P< .001). In JAKi-experienced patients in MOMENTUM, there were no deaths among dual SVR+TI responders at any evaluated threshold at week 24; thus, OS HRs vs nonresponders could not be derived, and differences refer to comparisons of the survival distributions between responders and nonresponders. Median OS was NE in dual SVR35+TI responders (n=15; 95% CI, NE-NE) vs nonresponders (n=115; 95% CI, 12.5 mo-NE), a difference that did not reach significance (P=.0543). However, OS was significantly longer in dual SVR and TI responders vs nonresponders at all other SVR thresholds: (1) SVR25+TI responders (n=23) vs nonresponders (n=107): median, NE in both groups (95% CI, NE-NE vs 12.5 mo-NE; P=.0109); (2) SVR15+TI responders (n=31) vs nonresponders (n=99): median, NE in both groups (95% CI, NE-NE vs 11.3 mo-NE; P=.0014); and (3) SVR10+TI responders (n=33) vs nonresponders (n=97): median, NE in both groups (95% CI, NE-NE vs 11.3 mo-NE; P=.0009). Conclusions:In both JAKi-naive and -experienced patients, dual TI and SVR response at week 24 was associated with longer OS at all SVR thresholds evaluated; curve separation was more pronounced at lower SVR thresholds (≥10% and ≥20%), suggesting that OS benefits can be achieved with more modest SVRs in patients who are also TI. Overall, these analyses highlight the ability of momelotinib to comprehensively address both splenomegaly and anemia in MF, benefits that may translate into meaningful OS improvements.