Abstract Background: The optimal treatment of early primary myelofibrosis (PMF) – prefibrotic PMF (prePMF) or low/intermediate-1 (int-1) risk overt PMF (oPMF) – is not well established. Most clinical trials have focused on higher risk oPMF patients (pts) but efforts are also needed in identifying treatments to prevent progression of early PMF and improve overall survival (OS). Interferon alfa (IFN) is one of few therapies that has long been investigated for early PMF treatment (Silver, Semin Hematol,1990). Studies of IFN thus far have demonstrated tolerability and promising efficacy, including in hematologic, molecular, and marrow responses (Gill et al. Blood,2023; Silver et al. Blood,2011). However, IFN has not yet been compared to other treatments or observation alone in early PMF. It also remains unknown if IFN improves OS or progression-free survival (PFS). In this comparative cohort study, we report the long-term outcomes associated with IFN use in early PMF. Methods: Medical records were queried and manually reviewed for pts with the diagnosis (dx) of prePMF and oPMF as per 2022 WHO and ICC criteria. The cohort was limited to pts with prePMF or low or int-1 risk oPMF by the Dynamic International Prognostic Scoring System Plus (DIPSS+). Demographic, clinical, lab, treatment, and outcome data were collected. Overall response rate (ORR) included hematologic, anemia, or spleen response as defined by the IWG-MRT 2013 response criteria. Pts that received ≥90 days of IFN therapy were assigned to the IFN arm and all others to the control group (NoIFN). Progression was defined as prePMF to oPMF advancement, accelerated or blast phase transformation, or PMF requiring bone marrow transplantation. OS and PFS were estimated using Kaplan-Meier methods. Cox proportional-hazards models were used for univariable and multivariable analysis (MVA) of mortality and progression risk. Results: We identified 92 pts with early PMF (median age, 59 years yrs at dx), of whom 46 (50%) received IFN. The driver mutation was JAK2 in 53(58%), CALR in 26 (28%), MPL in8 (9%), and triple negative in 4 (4%). There were no statistically significant differences in age, sex, race, driver mutation, blood counts, and DIPSS+ between groups, though the IFN group tended to be younger (56 vs 61 yrs, p=0.07) and included more pts with oPMF (30 65% vs 22 48%, p=0.09) and int-1 risk (15 33% vs 9 20%, p=0.2). Median time from dx to IFN initiation was 0.6 yrs; median IFN duration was 3 yrs. In the NoIFN group, 17 (37%) were on observation only, and others received one or more cytoreductive therapy including ruxolitinib (26%), hydroxyurea (28%), or other (30%). Median follow-up duration was similar in both groups (8.5 yrs IFN vs 10.4 yrs non-IFN, p=0.5). ORR with IFN was 70%; median response duration of 2.4 yrs. Hematologic response was achieved in 31 pts, anemia response from transfusion-independent baseline in 1, and spleen response by palpation in 3. Median OS was longer in the IFN group (18 vs 14 yrs, p0.01), with 5-, 10-, and 15-yr OS rates of 100%, 100%, 93% vs 97%, 72%, 40% for IFN vs NoIFN, respectively. The OS advantage with IFN was confirmed in MVA: time on IFN (per yr) was associated with a significantly lower mortality risk (HR 0.85, CI 0.77-0.95, p=0.003), independent of dx age and DIPSS+ features. 6 IFN pts died vs 24 NoIFN pts, of whom 4 and 15, respectively, were disease-related (due to progression in 3/4 and 6/15, respectively, and non-progression events in others). Median PFS, however, was not significantly different (12 yrs IFN vs 14 yrs NoIFN, p=0.86). Conclusion: In our retrospective analysis of 92 pts with early PMF, treatment with IFN yielded superior OS. MVA confirmed that this survival benefit was independent of age and DIPSS+ lab parameters, supporting the use of IFN for early PMF. Randomized controlled trials are needed to validate and further assess the advantages of IFN on both short and long-term outcomes. The upcoming HOPE-PMF phase 3 trial evaluating ropeginterferon alfa-2b vs observation in early PMF pts (Abu-Zeinah et al. Ann Hematol,2024) will provide essential data on the effects of IFN in a multi-center, randomized, double-blinded, prospective design. Such studies are critical in improving the treatment paradigm and bettering outcomes for the often-overlooked early PMF population.
Rob et al. (Mon,) studied this question.