AUTHORS
Shaoyu Chang, MD, MPH, Joseph Shatzel, MD2, Ghaith Abu-Zeinah I, MD3, Ruben Mesa, MD4, Anthony Michael Hunter, MD5, Brandi N Reeves, MD6, Lucia Masarova, MD7, Albert Qin, MD PhD8, Ling-Yu Cai9, Yu-Feng Wei10
AFFILIATIONS
1 PharmaEssentia USA, Burlington, Massachusetts, United States
2 Oregon Health and Science University, Portland, Oregon, United States
3 Weill Cornell Medicine, New York, New York, United States
4 Wake Forest University School of Medicine, Winston-Salem, North Carolina, United States
5 Winship Cancer Institute of Emory University, Atlanta, Georgia
6 UNC School of Medicine, Chapel Hill, North Carolina, United States
7 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
8 PharmaEssentia Corporation, Taipei, Taiwan
9 Vizuro LLC, Burlington, Massachusetts, United States
10 Vizuro LLC, Burlington, Massachusetts, United States
ABSTRACT
Background: Cytoreductive therapy is essential in polycythemia vera (PV) to control hematocrit and reduce thrombosis risk. Multiple agents are now established — hydroxyurea (HU, off-label), peginterferon alfa-2a (peg-IFN, off-label), ruxolitinib, and ropeginterferon alfa-2b (ropeg-IFN) — with ropeg-IFN as the most recently approved. Building on our prior analysis (Reeves et al., Ann Hematol 2025), this study extends the observation period through April 2026 to characterize evolving safety reporting patterns across all four therapies.
Methods: A retrospective disproportionality analysis was conducted using the FDA Adverse Event Monitoring System (AEMS; formerly FAERS), focusing on AE reports from PV patients receiving ropeg-IFN, peg-IFN, ruxolitinib, or HU from January 2022 through April 2026. A total of 6,917 PV-specific reports were identified (68.2% US-based). Age-stratified analysis was not performed due to incomplete age data (available in 25.5% of reports). Outcomes were categorized as fatal, serious (non-fatal), or non-serious. For each therapy, the background comparator was pooled reports of the other three therapies. Safety signals were quantified using Reporting Odds Ratios (ROR) with 95% confidence intervals (CI), analyzed by yearly intervals and cumulative longitudinal pooling via the R package PhViD. Adverse events were categorized at the System Organ Class (SOC) level using the Medical Dictionary for Regulatory Activities (MedDRA).
Results: ROR calculations revealed dynamic year-over-year fluctuations, most notably a transient spike in serious outcome reporting for ropeg-IFN in 2025 (ROR: 2.8, 95% CI: 1.7–4.6). Sensitivity analysis indicated that the 2025 estimate was sensitive to a localized decline in US non-serious ropeg-IFN reporting — consistent with the Weber effect as market familiarity grows — rather than a true safety signal. Cumulative pooled analysis (January 2022–April 2026) showed non-serious outcomes disproportionately represented among ropeg-IFN reports (ROR: 3.9, 95% CI: 3.3–4.6), while serious outcomes were reported less frequently relative to comparators (ROR: 0.3, 95% CI: 0.3–0.4). Serious non-fatal outcomes were disproportionately reported with HU (ROR: 13.8, 95% CI: 10.2–18.7) and peg-IFN (ROR: 3.7, 95% CI: 2.8–5.0), while fatal outcomes were consistently disproportionate in ruxolitinib reports (ROR: 3.8, 95% CI: 2.7–5.2).
Pooled SOC analysis confirmed prior findings and revealed new signals. HU and ruxolitinib showed disproportionate Blood and Lymphatic System Disorders (HU: n=47, ROR: 1.5; ruxolitinib: n=475, ROR: 1.5; anemia, thrombocytopenia, splenomegaly) and neoplasm signals (HU: n=66, ROR: 2.5; ruxolitinib: n=428, ROR: 1.2, 95% CI: 1.0–1.5). Ropeg-IFN's psychiatric (n=163, ROR: 2.1; insomnia, depression, anxiety) and eye disorder (n=66, ROR: 2.2) signals were confirmed; its endocrine signal now meets statistical significance (n=20, ROR: 5.1, 95% CI: 2.7–9.5; thyroid disorders including hypothyroidism and Graves' disease). Ruxolitinib showed vascular signals (n=289, ROR: 1.5; thrombosis, hypertension) and infection signals consistent with JAK inhibitor immunosuppression (n=701, ROR: 1.3; COVID-19, pneumonia, herpes zoster).
Conclusions: Real-world evidence confirms distinct safety reporting profiles across PV cytoreductive therapies. Ropeg-IFN remained consistent with prior observations: non-serious outcomes were disproportionately represented, while serious outcomes were reported less frequently relative to comparators, though this pattern likely reflects reporting dynamics, including the Weber effect and shorter market experience, rather than confirmed differential clinical risk. Established IFN class signals, including psychiatric symptoms, eye disorders, and thyroid dysfunction, were confirmed, with the endocrine signal now reaching statistical significance. HU and peg-IFN showed persistent serious outcome signals. Ruxolitinib's fatal, vascular, and infection signals persisted; channeling bias should be considered given its preferential use in HU-resistant/intolerant or advanced-stage PV. As AEMS lacks exposure denominators and is subject to reporting biases, results cannot be interpreted as comparative incidence or risk. The 2025 ropeg-IFN estimate illustrates how annual RORs are sensitive to reporting composition changes, supporting interpretation alongside cumulative analyses. Continued pharmacovigilance is essential to advance safety understanding across all therapies.
DISCLOSURES
Ghaith Abu-Zeinah I, MD: PharmaEssentia - Other (including travel support); AOP Health - Other (including travel support); Incyte - Other (including travel support); PharmaEssentia - Honoraria; AOP Health - Honoraria; Pharma& - Research Funding; PharmaEssentia - Research Funding; Silence therapeutics - Consultancy (Includes expert testimony); Incyte - Consultancy (Includes expert testimony); PharmaEssentia - Consultancy (Includes expert testimony); Takeda - Consultancy (Includes expert testimony); Silence therapeutics - Membership on a Board/Advisory Committee; Takeda - Membership on a Board/Advisory Committee; Incyte - Membership on a Board/Advisory Committee; PharmaEssentia - Membership on a Board/Advisory Committee