EClinicalMedicine. 2026 Oct;100
104193
Trine Alma Knudsen,
Dennis Lund Hansen,
Lukas Frans Ocias,
Ole Weis Bjerrum,
Mette Brabrand,
Sarah Friis Christensen,
Christina Schjellerup Eickhardt-Dalbøge,
Christina Ellervik,
Daniel El Fassi,
Mikael Frederiksen,
Lasse Kjær,
Thomas Kielsgaard Kristensen,
Torben A Kruse,
Morten Kranker Larsen,
Torben Mourits-Andersen,
Sören Möller,
Ulrik Malthe Overgaard,
Marianne Tang Severinsen,
Vibe Skov,
Anders Lindholm Sørensen,
Jesper Stentoft,
Jørn Starklint,
Karin de Stricker,
Mads Thomassen,
Thomas S Larsen,
Hans C Hasselbalch.
Background: Hydroxyurea (HU) is the most commonly used cytoreductive treatment for patients with myeloproliferative neoplasms (MPN), but emerging data on pegylated interferon-alpha2 (pegIFNα) are encouraging. Optimal first-line treatment remains unclear.
Methods: DALIAH was a randomised, open-label, parallel-group, phase 3 trial conducted at nine centres in Denmark comparing HU with pegIFNα in patients with newly diagnosed MPN. Adults (≥18 years) with newly diagnosed or cytoreductive treatment-naïve MPN (essential thrombocythaemia [ET], polycythaemia vera [PV], prefibrotic myelofibrosis [pre-PMF], or primary myelofibrosis [PMF]; according to 2008 World Health Organization [WHO] criteria) and evidence of active disease were eligible, regardless of risk score. Patients aged >60 years were randomised (1:1:1) to either HU, pegIFNα-2a, or pegIFNα-2b; patients aged ≤60 years were randomised (1:1) to either pegIFNα-2a or pegIFNα-2b. The primary endpoint was the proportion of molecular responders (MR; partial or complete) at 18, 36, and 60 months (intention-to-treat [ITT], HU vs pegIFNα). This trial was registered with ClinicalTrials.gov, NCT01387763.
Findings: Between Feb 7, 2012, and July 6, 2015, 203 eligible participants were enrolled in the modified ITT population (HU n = 38 [19%], pegIFNα n = 165 [81%]). In the ITT analysis, MR proportions were similar between HU and pegIFNα at 18 months (19% vs 21%, p = 1·00), 36 months (19% vs 26%, p = 0·64), and 60 months (23% vs 24%, p = 1·00). In the per-protocol (PP) analysis, restricted to patients who remained on therapy, pegIFNα showed higher MR beyond 36 months (36 months: 23% vs 56%, p = 0·01; 60 months: 35% vs 67%, p = 0·03). At month 60, PegIFNα discontinuation was high (65% vs 37% HU, p = 0·0019). No significant difference in grade ≥3 adverse events was observed between groups (HU 58% vs pegIFNα 45%, p = 0·21). Adverse events occurring in ≥10% of patients differed between treatments: dyspepsia, pyrexia, and urinary tract infections were more common with HU, whereas fatigue, influenza-like illness, injection-site irritation, myalgia, and decreased white cell blood count were more frequent with pegIFNα. Anaemia and decreased neutrophil count were the most common haematological events in both groups. Treatment-related adverse events were reported in 77 patients (HU 27% vs pegIFNα 41%, p = 0·14) and led to permanent discontinuation in 13 patients (all pegIFNα).
Interpretation: HU and pegIFNα display distinct response kinetics. Molecular response was similar in the ITT analysis, reflecting that non-response largely resulted from higher pegIFNα discontinuation rather than lack of efficacy. Among patients who tolerated therapy, pegIFNα achieved superior long-term molecular responses. Findings should be considered within the context of the significantly higher treatment discontinuation rate with pegIFNα than with HU. These findings suggest that pegIFNα may offer long-term molecular benefit in selected patients highlighting the need for biomarkers that predict response and long-term tolerability, preferably already at diagnosis.
Funding: OUH-Region Sjaelland Faelles Forskningspulje, Region Sjællands Sundhedsvidenskabelige Forskningsfond (RSSF) 2018, Gangstedfonden, OUH Frie Forskningsmidler, Swedish Orphan, Fonden til Lægevidenskabens Fremme, Ellen og Aage Fausboells Helsefond af 1975, and Desirée og Niels Ydes Fond.
Keywords: Efficacy and safety; Hydroxyurea; Molecular response; Myeloproliferative neoplasms; Pegylated interferon alpha2; Randomised controlled trial