Indirect Treatment Comparisons (ITCs) of the Effect of Fedratinib Versus Ruxolitinib (RUX) on Spleen Volume for Patients with Myelofibrosis (MF) and No Prior RUX Treatment

Context: Since 2011, there have been a limited number of treatment options for patients with MF, including allogeneic stem cell transplantation and RUX, a JAK inhibitor. In 2019, another JAK inhibitor, fedratinib, was approved in the USA. As no RCTs have compared fedratinib with RUX, we performed ITCs to estimate the relative efficacy of fedratinib versus RUX in spleen volume reduction (SVR) in patients with RUX-naive MF. Objective: To explore the relative efficacy of fedratinib versus RUX for SVR in patients with RUX-naive MF. Design: A systematic literature review (SLR) originally conducted in August 2018 (and updated February 2020) examined clinical evidence published in EMBASE®, MEDLINE®, and the Cochrane Library (no time limit). RCTs investigating fedratinib or RUX in RUX-naive patients were considered for ITCs based on comparability of study design, population, intervention, comparators, and outcomes. Bucher ITCs were performed. Matching-adjusted indirect comparisons (MAICs) were explored to adjust for potential treatment effect modifiers (treatment-by-subgroup interaction term p Setting: Multicenter RCTs. Patients or other participants: Patients with RUX-naive MF. Interventions: Fedratinib versus RUX. Main outcome measures: The binary outcome of SVR ≥35% from baseline to Week 24 was investigated. Results: The SLR identified 3 studies: JAKARTA, COMFORT-I, COMFORT-II. The study control arms (placebo, best available therapy) were assumed to have equal efficacy, validated via clinician input and published literature, which enabled use of an anchored ITC. Baseline characteristics were generally comparable. A Bucher ITC indicated a numerically higher proportion of fedratinib-treated patients achieved SVR versus RUX (risk difference [95%CI] 11.0% [-1.4–23.4] comparing JAKARTA subgroup with platelet count ≥100 × 109/L; 9.4% [-2.2–20.9] comparing intention-to-treat [ITT] populations). After further accounting for baseline JAK2 mutation status (potential observed treatment effect modifier) in an MAIC, a significantly higher proportion of patients on fedratinib achieved SVR versus RUX (risk difference [95%CI] 14.7% [2.4–27.1] comparing JAKARTA subgroup with platelet count ≥100 × 109/L; 12.3% [0.6–24.0] comparing ITT populations). Scenario analyses were generally consistent. Conclusions: The MAIC analyses accounting for observable treatment effect modifiers demonstrated that a statistically significantly higher proportion of patients treated with fedratinib had SVR at Week 24 compared with RUX in RUX-naive MF.