European LeukemiaNet clinical signs sharpen risk assessment during ruxolitinib in polycythemia vera

Polycythemia vera management still begins with broad categories, but this cohort shows why bedside detail can outperform age alone. In 254 patients treated with second-line ruxolitinib, European LeukemiaNet clinical signs and symptoms identified a subgroup with materially higher thrombotic and progression risk, even when conventional age- and thrombosis-based stratification failed to separate outcomes.

The central message is immediate and clinically familiar. A patient may sit in the “low-risk” box on paper and still carry an active myeloproliferative phenotype that continues to generate vascular and fibrotic danger.

European LeukemiaNet defines these clinical signs and symptoms in polycythemia vera as progressive splenomegaly, leukocytosis, thrombocytosis, inadequate hematocrit control, and cardiovascular risk factors. Those features are more than descriptive clutter. They signal persistent marrow-driven proliferation, inflammatory activation, and blood viscosity stress, all of which can feed thrombosis even after treatment has shifted from hydroxyurea to ruxolitinib.

What the cohort showed

The PV-ARC multicenter cohort included 254 patients receiving ruxolitinib. CSSs were present in 50.4% of the full population, with nearly identical frequency in low- and high-risk patients: 50% versus 50.5%. Conventional thrombotic risk categories did not predict either thrombosis or progression to secondary myelofibrosis or acute leukemia during ruxolitinib therapy.

By contrast, the CSS-defined phenotype separated outcomes cleanly. Thrombosis occurred at an incidence rate ratio of 1.8 versus 0 %p-y in patients with CSSs, with _p_ = 0.05. Disease progression rose to 6.0 versus 2.3 %p-y, with _p_ = 0.02. The overall incidence rate ratio for thrombosis was 1.0, while progression to secondary myelofibrosis or acute leukemia reached 4.2 %p-y in the treated cohort.

Persistent or progressive leukocytosis emerged as the strongest predictor of progression. That finding is biologically plausible. White cell excess in PV is not a passive laboratory abnormality; it reflects ongoing myeloproliferation and can contribute to endothelial injury, microvascular disturbance, and a prothrombotic milieu. When leukocytosis does not settle, the disease remains active even if treatment has improved symptoms or hematocrit control.

The signal was especially important because it did not depend on standard risk bins. Age and prior thrombosis remain useful for treatment planning, but they do not fully capture the current biology of the disease. A younger patient with persistent splenomegaly, rising leukocytes, or inadequate hematocrit control may carry a more aggressive trajectory than the label suggests. In practice, that means the exam room and the full blood count still matter after the formal risk category has been assigned.

Why ruxolitinib does not erase biologic risk

Ruxolitinib suppresses JAK-STAT signaling and often improves spleen size, symptom burden, and inflammatory features in myeloproliferative neoplasms. Yet symptom relief does not always mean disease quiescence. The PV-ARC data suggest that residual clinical signs at treatment start continue to identify patients in whom the malignant clone remains sufficiently active to generate vascular events or disease evolution.

That is the core clinical implication for hematology follow-up. Risk assessment in PV should not stop at age and thrombosis history once a patient is on second-line therapy. Persistent leukocytosis, splenomegaly, thrombocytosis, and inadequate hematocrit control should remain part of active prognostic thinking, because they appear to track the biology that drives later thrombosis and progression.

The study also clarifies what is not useful in this setting. Conventional thrombotic risk categories alone did not predict either endpoint during ruxolitinib treatment. For clinicians, that creates a mismatch between legacy stratification and near-term outcomes. The disease can behave as if it is higher risk long before the standard rubric says so.

Bedside consequences for routine monitoring

Routine PV monitoring already includes counts, symptoms, and spleen assessment, but this analysis gives those familiar data points sharper prognostic meaning. Persistent leukocytosis deserves special attention because it carried the strongest link to progression. Spleen enlargement and inadequate hematocrit control should not be dismissed as treatment residue if they persist at ruxolitinib initiation.

The practical consequence is a more discriminating clinic visit. A hematologist who reviews only age and thrombosis history may miss the patient whose disease biology remains active. A clinician who reads the blood counts and symptom burden in full is more likely to identify who needs closer surveillance for thrombosis, fibrotic evolution, or leukemic transformation.

The findings also fit the broader biology of PV progression. Chronic inflammatory stimulation, continued proliferative signaling, and vascular injury create conditions in which thrombosis and transformation can travel together. CSSs are useful because they condense those biological pressures into bedside-recognizable features.

This multicenter cohort does not overturn existing PV risk models, but it does add a layer they have lacked. CSSs refine prognostication during ruxolitinib and appear to identify patients whose disease course is more event-prone than conventional classification implies. For clinicians managing PV after hydroxyurea failure, that is a signal worth carrying into everyday practice.

The remaining challenge is implementation at the point of care: whether routine PV follow-up can systematically incorporate CSSs into treatment decisions, surveillance intensity, and escalation planning without adding ambiguity to an already complex disease.


Reporting by Dr. Revathy Mohan V.R., MBBS for Medicine World Foundation, with additional data provided by British Journal of Cancer.
PRIMARY CLINICAL EVIDENCE & REGISTRY DATA