Polycythemia vera (PV) is a chronic myeloproliferative neoplasm characterized by uncontrolled red blood cell production and a heightened likelihood of thrombosis.¹ PV is traditionally classified into low- and high-risk categories, with the latter indicated for cytoreductive therapy, primarily with hydroxyurea (HU).¹ The revised European LeukemiaNet (ELN) criteria expands on 6 clinical signs and symptoms (CSSs) to guide the initiation of cytoreductive therapy, yet their impact on thrombotic risk across conventional risk categories remains underexplored.¹ This study evaluated the incidence of thrombotic events among patients with PV receiving HU and investigated the role of revised ELN criteria for cytoreductive therapy start (CTS) in identifying high-risk phenotypes.¹ Findings from the multicenter PV-ARC study were presented by Dr. Francesca Palandri at the ASH Annual Meeting & Exposition 2024.¹

This retrospective analysis included 1,162 patients diagnosed with PV based on WHO 2022 criteria.¹ Of these, 739 patients treated with HU were assessed for revised CTS, which was reflective of real-world clinical practice and included persistent/ progressive leukocytosis (100% increase if white blood cells (WBC) 10×109/L at baseline; WBC >15×109/L for at least 3 months), extreme thrombocytosis (platelet >1,000×109/L), progressive splenomegaly (an increase of >5cm from last year pre-HU), inadequate hematocrit (Hct) control (Hct ≥50% for ≥1-year pre-HU) or excessive phlebotomy requirements (≥6 phlebotomy/year for 2 years or intolerant to phlebotomy), relevant cardiovascular risk factors (≥4 following factors: smoking, hypertension, dyslipidemia, diabetes, overweight), and severe itching (score ≥5/10 and/or refractory to medications).¹ Thrombosis-free survival (TFS) and incidence rate ratios (IRRs) were calculated, with multivariate Cox regression identifying independent predictors of thrombotic risk.¹

Among the HU-treated patients, 18.5% were low-risk (LR: age <60 years, no prior thrombosis), and 81.5% were high-risk (HR-AGE: age >60 years, 70.4%; HR-THRO: prior thrombosis, 29.6%).¹ At least 1 CSS was present in 60.4% of the cohort which was more frequently detected in LR patients (69.3%) compared to HR (58.3%).¹ CSSs were equally detected in HR-AGE vs. HR-THRO (p=0.32) and between HR-THRO due to age and previous thrombosis history vs. HR-THRO due to previous thrombosis history only (p=0.69).¹ Additional reasons for CTS were identified in all LR patients and 71.3% of HR patients, which reflect the revised ELN criteria but with lower intensity thresholds.¹ The IRR of thrombotic events during HU treatment was 1.7 per 100 patient-years.¹ HR-THRO patients exhibited the highest thrombotic rates (3.0 per 100 patient-years) compared to LR (1.1 per 100 patient-years, p=0.006) and HR-AGE (1.3 per 100 patient-years, p=0.002).¹ CSSs were associated with worse TFS across all risk groups, with significantly higher thrombosis rates observed in patients with CSS vs. those without (2.2 vs. 0.7 per 100 patient-years, p<0.001).¹ CSSs were associated with worse 10-year TFS outcomes within each risk group: low-risk (with vs. without CSS: 86.4% vs. 100%, p=0.007), HR-AGE (91.4% vs. 97.8%, p=0.003), and HR-THRO (79.5% vs. 88.1%, p=0.02).¹ Multivariate analysis identified progressive splenomegaly (HR=6.10; 95% CI: 1.70-21.88; p=0.005), inadequate hematocrit control (HR=2.20; 95% CI: 1.36-3.54; p=0.001) and relevant cardiovascular risk factors (HR=2.56; 95% CI: 1.09-6.00; p=0.03) as predictors of thrombotic risk.¹

Revised ELN CSSs effectively stratified patients with PV into distinct thrombotic risk profiles, independent of conventional risk categories.¹ These findings underscore the need to integrate CSSs into existing prognostic models to refine treatment strategies.¹ Notably, a significant proportion of LR patients required HU due to CSSs, highlighting the necessity for tailored management approaches and further research into long-term outcomes of HU therapy._¹