Serum erythropoietin levels in 696 patients investigated for erythrocytosis with JAK2 mutation analysis

Mutational Landscape of Patients Referred for Elevated Hemoglobin Level

Background: Since the identification of JAK2 V617F and exon 12 mutations as driver mutations in polycythemia vera (PV) in 2005, molecular testing of these mutations for patients with erythrocytosis has become a routine clinical practice. However, the incidence of myeloid mutations other than the common JAK2 V617F mutation in unselected patients referred for elevated hemoglobin is not well studied. This study aimed to characterize the mutational landscape in a real-world population of patients referred for erythrocytosis using a targeted next-generation sequencing (NGS)-based assay. Method: A total of 529 patients (hemoglobin levels >160 g/L in females or >165 g/L in males) were assessed between January 2018 and May 2021 for genetic variants using the Oncomine Myeloid Research Assay (ThermoFisher Scientific, Waltham, MA, USA) targeting 40 key genes with diagnostic and prognostic implications in hematological conditions (17 full genes and 23 genes with clinically relevant "hotspot" regions) and a panel of 29 fusion driver genes (>600 fusion partners). Results: JAK2 mutations were detected in 10.9% (58/529) of patients, with 57 patients positive for JAK2 V617F, while one patient had a JAK2 exon 12 mutation. Additional mutations were detected in 34.5% (20/58) of JAK2-positive patients: TET2 (11; 19%), DNMT3A (2;3.4%), ASXL1 (2; 3.4%), SRSF2 (2; 3.4%), BCOR (1; 1.7%), TP53 (1; 1.7%), and ZRSR2 (1; 1.7%). Diagnosis of PV was suspected in 2 JAK2-negative patients based on the 2016 World Health Organization (WHO) diagnostic criteria. Notably, one patient carried mutations in the SRSF2 and TET2 genes, and the other patient carried mutations in the SRSF2, IDH2, and ASXL1 genes. Three JAK2-negative patients with elevated hemoglobin who tested positive for BCR/ABL1 fusion were diagnosed with chronic myeloid leukemia (CML) and excluded from further analysis. The remaining 466 JAK2-negative patients were diagnosed with secondary erythrocytosis and mutations were found in 6% (28/466) of these cases. Conclusion: Mutations other than JAK2 mutations were frequently identified in patients referred for erythrocytosis, with mutations in the TET2, DNMT3A, and ASXL1 genes being detected in 34.5% of JAK2-positive PV patients. The presence of additional mutations, such as ASXL1 mutations, in this population has implications for prognosis. Both the incidence and mutation type identified in patients with secondary erythrocytosis likely reflects incidental, age-associated clonal hematopoiesis of indeterminate potential (CHIP).

High platelet-to-lymphocyte ratio may differentiate polycythemia vera from secondary polycythemia

Discriminating polycythemia vera (PV) from secondary polycythemia (SP) is crucial due to the inherent risk of thrombosis in PV and different treatment approaches. The majority of PV patients have subnormal serum erythropoietin levels and harbor Janus kinase 2 (JAK2) mutations; however, serum erythropoietin levels may be normal in approximately one third of PV patients and mutational analysis is costly and requires access to specialized laboratories. Recently, neutrophil-to-lymphocyte (NLR) and platelet-to-lymphocyte ratios (PLR) emerged as rapidly available biomarkers to identify PV patients under an increased risk of thrombosis and death. This multicenter retrospective study investigated whether these two biomarkers may also be used to differentiate PV from SP. A total of 207 subjects were included (103 PV and 104 SP) with both baseline NLR (median 4.33 vs. 1.89) and PLR (median 259.12 vs. 81.11) being significantly higher in PV than in SP (p < 0.001 for both analyses). According to the receiver operating curve analysis, PLR (area under the curve, AUC 0.936, the optimal cut-off value of > 138.1 had 82.5% sensitivity and 91.67% specificity for the detection of PV) outperformed other tested variables (NLR, total leukocytes, neutrophils, lymphocytes and platelets) and its cut-off values with 100% specificity and sensitivity were able to confirm (PLR > 224.56; 31% patients) and to exclude (PLR < 68.8; 13% patients) the highest proportions of PV patients. Therefore, PLR may represent a cheap and a rapidly available biomarker with valuable diagnostic and prognostic properties. This information may be particularly useful in resource-limited settings; however, our results need validation on larger datasets.