Ruxolitinib versus allogeneic stem cell transplantation for patients with myelofibrosis according to donor availability: A prospective multicenter phase II trial

Abstract Introduction: Myelofibrosis is a chronic myeloproliferative neoplasm characterized by bone marrow fibrosis, splenomegaly, and debilitating constitutional symptoms, often resulting in significant morbidity and mortality. Allogeneic hematopoietic stem-cell transplantation (SCT) remains the only potentially curative treatment but is associated with considerable risks. Ruxolitinib, a JAK1/2 inhibitor, improves splenomegaly and symptoms but does not offer a cure. The primary objective of this study was to compare event-free survival (EFS) between patients who underwent SCT after 3 months of ruxolitinib induction therapy and those who continued ruxolitinib due to lack of a suitable donor (HLA-identical sibling or 10/10 matched unrelated donor). Methods: This prospective, multicenter clinical trial (RuxoAllo study, NCT03333187) enrolled 87 myelofibrosis patients, all of whom began with ruxolitinib induction. Major inclusion criteria comprised primary or secondary myelofibrosis with intermediate-2 or high-risk disease, or intermediate-1 risk with high-risk cytogenetics, transfusion dependency, or thrombocytopenia; all patients were ruxolitinib-naïve and aged ≥18 years. After screening, 73 were assigned to either transplantation (SCT group, n=57) or continued ruxolitinib therapy (ContRuxo group, n=16). Efficacy endpoints included EFS and overall survival (OS). Safety endpoints included incidence of acute and chronic graft-versus-host disease, disease-related mortality, and non-relapse mortality. EFS was defined as survival without relapse, disease progression, death, or change in therapy. Safety was evaluated through adverse events, laboratory parameters, and ECOG performance status. Quality of life (QoL) was assessed before, during, and after treatment. Results: At baseline, the median patient age was 58 years, with 71% having primary myelofibrosis and 63% being male. In the Full Analysis Population, SCT showed significantly improved EFS, with a median EFS not reach for the SCT group vs. 18 months for the ContRuxo group (P<0.001). SCT was associated with 84% reduced risk of death or progression, with a hazard ratio of 0.16 (95% CI: 0.06–0.46). Multivariable adjustment with patient- and disease-specific variables confirmed the independent benefit of SCT, with a hazard ratio of 0.18 (P<0.001). Subgroup analyses confirmed consistent benefits of SCT across age and DIPSS risk groups. Three-year OS was 84% for the SCT group vs. ContRuxo 83%. Spleen responses were notable after ruxolitinib induction but did not differ significantly between arms thereafter. The cumulative incidence of grade II–IV acute graft-versus-host disease by day 100 was 7% (95% CI: 2–15), while chronic graft-versus-host disease occurred in 63% (95% CI: 49–75%) at 3 years, with 32% classified as moderate or severe. Disease-related mortality was lower in the SCT group (9% vs. 17%). Relapse/progression rates at 3 years were significantly higher in the ContRuxo group (56% vs. 18%, p=0.02), whereas non-relapse mortality was higher in the SCT group (0% vs. 7%, p=0.04). Severe adverse events were more frequent with SCT (50%) compared to ContRuxo (31%), including infections, cytopenias, and graft-versus-host disease. Ruxolitinib was mainly associated with hematological toxicities. QoL initially declined following SCT but improved substantially over time, with a 45% reduction in symptom burden and an 8% increase in overall well-being by 24 months. In contrast, QoL in the ContRuxo group remained more stable, with only modest symptom relief (22% reduction) and a 5% decline in functional QoL. Conclusion: The RuxoAllo prospective study demonstrates that SCT following ruxolitinib induction significantly improves EFS and disease control compared to continued ruxolitinib in patients with myelofibrosis, particularly in those with intermediate- or high-risk disease. Despite a relatively low non-relapse mortality of 7%, SCT remains associated with substantial risks, including graft-versus-host disease and other toxicities, underscoring the need for careful patient selection. Ruxolitinib remains valuable for symptom management and can serve as an effective bridge to transplant. These findings support SCT as the preferred curative approach for eligible myelofibrosis patients and highlight the importance of strategies to mitigate transplant-related morbidity.