AUTHORS
Aaron T. Gerds, MD, MS, Jose Miguel Torregrosa Diaz, MD2, Kristen Pettit, MD3, Shiyuan Zhang, MSc4, Tom (sizhu) Liu, MS5, Mahshid Azimi, PharmD6, Catherine Ellis, BS7, Sunkyung Kim8, Katherine Mays9, Shirley Wang, MS10, Nicholas Meader11, Adrian Segura Diaz, PhD12, Massimo Breccia, MD13
AFFILIATIONS
1 Cleveland Clinic Cancer Institute, Cleveland, Ohio, United States
2 Centre Hospitalier Universitaire (CHU) de Poitiers, Poitiers, France
3 Division of Hematology/Oncology, University of Michigan, Ann Arbor, Michigan, United States
4 GSK, Collegeville, Pennsylvania, United States
5 GSK, Collegeville, Pennsylvania, United States
6 GSK, Collegeville, Pennsylvania, United States
7 GSK, Collegeville, Pennsylvania, United States
8 GSK, Collegeville, Pennsylvania, United States
9 GSK, Collegeville, Pennsylvania, United States
10 GSK, Collegeville, Pennsylvania, United States
11 GSK, London, United Kingdom
12 H. U. Dr. Negrin de Gran Canaria, Instituto de Investigación Sanitaria de Canarias (IISC), Las Palmas, Spain
13 Sapienza University of Rome, Rome, Italy
ABSTRACT
Introduction: Transfusion-dependent (TD) anemia is a clinically significant complication of myelofibrosis (MF) associated with poor prognosis. Patients (pts) with MF who develop TD anemia while receiving certain Janus kinase inhibitors (JAKi), such as ruxolitinib (RUX), represent a population with substantial unmet medical needs.
Momelotinib (MMB) is a JAK1/JAK2/ACVR1 inhibitor approved for adults with MF and anemia. Luspatercept (LUSPA) is an erythroid maturation agent that is being evaluated in combination with JAKi (RUX or fedratinib) in the phase 3 INDEPENDENCE study in pts with MF and transfusion-requiring anemia (Passamonti F, et al. EHA 2026. S215). No head-to-head randomized clinical trial comparing MMB vs LUSPA+JAKi has been conducted to date. The objective of this indirect comparison was to evaluate transfusion-related outcomes associated with MMB monotherapy vs LUSPA+JAKi in JAKi-experienced pts with MF and TD anemia, using available phase 3 trial data.
Methods: A population-adjusted indirect comparison between MMB and LUSPA+JAKi was performed using published aggregate-level data from pts treated with LUSPA+JAKi in INDEPENDENCE and pooled individual pt-level data from pts treated with MMB in SIMPLIFY-2 and MOMENTUM meeting TD criteria aligned with INDEPENDENCE (4-12 red blood cell [RBC] transfusion units during the 12 wk before randomization with no interval >6 wk without ≥1 RBC transfusion and hemoglobin [Hb] <10 g/dL at baseline). Unanchored matching-adjusted indirect comparison (MAIC) was used to mitigate potential bias due to study population differences. Pt-level weights were derived for the pooled MMB population to match aggregate baseline characteristics in INDEPENDENCE, based on baseline DIPSS risk category (intermediate-1, intermediate-2, or high), MF subtype (primary or secondary), and RBC transfusion burden (median units/12 wk). Weighted MMB outcomes were then compared with published outcomes in INDEPENDENCE. The primary analysis evaluated the proportion of pts in each arm achieving RBC transfusion independence (RBC-TI) ≥12 consecutive wk by wk 24; secondary analyses included additional transfusion-related efficacy outcomes. Safety analyses are in progress and will be presented.
Results: A total of 78 pts in the MMB group were matched to 208 pts in the LUSPA+JAKi group, resulting in an effective sample size of 58.8 in the MMB group. Before matching, more pts in the MMB vs LUSPA+JAKi group were male (70.5% vs 57.2%) and had a high-risk DIPSS score (37.2% vs 12.5%); after matching, key baseline characteristics were balanced.
A favorable trend toward MMB was observed for all RBC transfusion endpoints after matching. RBC-TI ≥12 wk (95% CI) was achieved by 34.8% (22.3-47.2) in the MMB group and 23.1% (17.3-28.8) in the LUSPA+JAKi group (odds ratio [OR], 1.77 [95% CI, 0.93-3.38]). Pts in the MMB group were significantly more likely to achieve the composite endpoint of RBC-TI ≥12 wk + mean Hb increase ≥1 g/dL from baseline (30.4% [95% CI, 18.4-42.5]) vs pts in the LUSPA+JAKi group (15.9% [95% CI, 10.9-20.8]; OR, 2.32 [95% CI, 1.16-4.62]). Trends favored MMB vs LUSPA+JAKi for RBC-TI ≥16 wk (24.8% vs 19.2%; OR, 1.38 [95% CI, 0.68-2.82]), reduction in RBC transfusion burden by ≥50% over any 12-wk period (57.5% vs 52.9%; OR, 1.20 [95% CI, 0.67-2.17]), and reduction in RBC transfusion burden by ≥50% and by ≥4 RBC units from baseline over any 12-wk period (52.3% vs 40.9%; OR, 1.59 [95% CI, 0.88-2.87]).
Due to the impact of the pt subgroup from mainland China on the results of INDEPENDENCE, a sensitivity analysis excluding this region was also conducted using RBC-TI ≥12 wk data from the LUSPA+JAKi arm and baseline characteristics from the LUSPA+JAKi and placebo+JAKi arms, due to unavailability of characteristics for the LUSPA+JAKi arm alone. In the sensitivity analysis, 78 pts in the MMB group were matched to 268 pts in INDEPENDENCE, resulting in an effective sample size of 59.3 in the MMB group. A similar trend favoring MMB vs LUSPA+JAKi for RBC-TI ≥12 wk was observed (34.4% vs 22.5%; OR, 1.81 [95% CI, 0.94-3.52]).
Conclusions: In this MAIC, MMB monotherapy demonstrated greater reduction in transfusion burden vs LUSPA+JAKi in pts with MF and TD anemia, supporting use of MMB before addition of anemia supportive care to another JAKi in this population. These findings also support continued evaluation of the potential anemia-related benefits of an MMB+LUSPA combination in the ongoing ODYSSEY trial (NCT06517875).
DISCLOSURES
Massimo Breccia, MD: AbbVie - Honoraria; Pfizer - Honoraria; Otsuka - Honoraria; Novartis - Honoraria; Incyte - Honoraria; GlaxoSmithKline - Honoraria; AOP Health - Honoraria; Servier - Honoraria; Bristol Myers Squibb/Celgene - Honoraria