Clinical and Laboratory Associations of JAK2 V617F Variant Allele Frequency in Patients With JAK2 V617F-Positive Erythrocytosis: A Single-Centre Retrospective Analysis.

BACKGROUND: JAK2 V617F variant allele frequency (VAF) reflects clonal burden in myeloproliferative neoplasms, but the clinical associations of low-level VAF in patients with erythrocytosis remain uncertain. OBJECTIVE: The objective of this study is to examine associations between diagnostic JAK2 V617F VAF categories and clinical, laboratory, bone marrow and thrombotic outcomes in patients with JAK2 V617F-positive erythrocytosis. PATIENTS AND METHODS: This single-centre retrospective cohort included 174 adults. Patients were stratified by diagnostic VAF as low (< 5%; n = 62), intermediate (5%-50%; n = 81) or high (> 50%; n = 31). Clinical and laboratory characteristics, bone marrow findings and incident thrombotic events during follow-up were compared. Multivariable logistic regression was used to evaluate factors associated with thrombosis. RESULTS: Follow-up duration was comparable amongst the VAF groups (median, 5.0 vs. 8.0 vs. 6.0 years; p = 0.227). Thrombosis rates increased across the groups (12.9% vs. 22.2% vs. 48.4%; p = 0.001), as did splenomegaly (25.8% vs. 53.1% vs. 64.5%; p < 0.001) and myelofibrosis transformation (1.6% vs. 13.6% vs. 20.0%; p = 0.011). Amongst patients with available measurements, median serum erythropoietin decreased with higher VAF (8.0 vs. 3.1 vs. 2.2 mIU/mL; p < 0.001), whereas the proportion with subnormal erythropoietin increased (17.8% vs. 47.6% vs. 82.6%; p < 0.001). Amongst biopsied patients, polycythemia vera-supportive bone marrow morphology was present in 50.0%, 93.5% and 100% of the groups respectively. A diagnostic VAF > 50% was independently associated with subsequent thrombosis (OR: 3.35; 95% CI: 1.06-10.58; p = 0.039), as was platelet count (OR: 9.92; 95% CI: 1.52-64.63; p = 0.016). CONCLUSIONS: A diagnostic JAK2 V617F VAF > 50% was independently associated with subsequent thrombosis in this cohort. The low-VAF group was diagnostically and clinically heterogeneous. These findings should be considered hypothesis-generating and require prospective validation.