Pb2207: neutrophil–to-lymphocyte ratio is an early screening tool for diagnosis of polycythemia vera

Topic: 16. Myeloproliferative neoplasms - Clinical Background: Polycythemia vera (PV) is an inflammatory cancer that belongs to myeloproliferative neoplasm (MPN). It is very important to rapidly and accurately diagnose PV in patients with erythrocytosis and provide appropriate treatment to prevent thrombosis. If there are indications suggesting PV rather than secondary polycythemia (SP) in patients with erythrocytosis, JAK2 mutation testing can be conducted more actively. The neutrophil-lymphocyte ratio (NLR) which is the number of neutrophils divided by the number of lymphocytes is a fast and simple method for assessing inflammatory status, which may be used in many disease for predicting inflammation and mortality. The NLR may be promptly evaluated with hemoglobin using a complete blood count with differential test. In a previous study, we confirmed that the NLR was higher in MPN patients than in the normal population. Particularly in patients with PV, panmyelosis is necessary for diagnosis, and in the presence of panmyelosis, the neutrophil count rises and the proportion of lymphocytes falls, resulting in an increase in NLR. Aims: We analyzed NLR in patients with PV and SP and studied whether it can be a screening tool for diagnosis of PV. Methods: We retrospectively reviewed the medical records of patients with erythrocytosis who underwent JAK2 V617F or exon 12 mutation tests from June 2007 to January 2023 at Soonchunhyang university hospitals (Seoul, Bucheon and Cheonan). Diagnose of PV was definitely verified using 2016 WHO criteria. Levels of EPO and a BM examination are necessary for the diagnosis of PV, although individuals who did not undergo these tests were excluded. Also, patients with a serum C-reactive protein (CRP) level > 1 mg/dL or clinical signs of infection were excluded, because the inclusion of such patients may have influenced the NLR result. NLR in patients with PV were compared to in patients with SP at the time of diagnosis. Results: Of 379 patients, 104 patients had a diagnosis of PV and 275 patients had a diagnosis of SP. The median NLR was significantly higher in PV group than in SP group (5.1 vs 1.84, p < 0.001). There were 160 patients with NLR less than 2, whereas the PV was 2 and the SP was 158. There were 145 patients with NLR 2 or more and 4 or less, whereas the PV was 34 and the SP was 111. There were 74 patients with NLR more than 4, whereas the PV was 68 and the SP was 6 (Figure). In receiver operator characteristic analysis, the best cut-off value of the NLR was 3.30 (sensitivity 77.9%, specificity 92.4%, AUC = 0.936). When the cut-off value of the NLR was reduced to 2, the sensitivity and negative predictive value (NPV) were very high (sensitivity, 98.1%; NPV, 98.8%). When the cut-off value of the NLR was increased to 4, the specificity and positive predictive value (PPV) were high (specificity, 97.8%; PPV, 91.9%). Summary/Conclusion: We found that NLR may be a useful screening method for diagnosis of PV. In particular, the NPV at NLR less than 2 is very high at 98.8%, so patients with erythrocytosis and NLR less than 2 can expect a very low probability of PV.Keywords: Polycythemia vera, Screening