Straightforward Identification of Masked Polycythemia Vera Based on Proposed Revision of World Health Organization Diagnostic Criteria for BCR-ABL1-Negative Myeloproliferative Neoplasms

Dear Editor Some cases of polycythemia vera (PV) may mimic essential thrombocythemia (ET) owing to prominent thrombocytosis [1]. A recently proposed revision of the WHO diagnostic criteria for BCR-ABL1-negative myeloproliferative neoplasms (MPNs) has highlighted the diagnostic pitfalls of these cases and reintroduced the clinical entity of “masked” PV, which has been identified in patients with anemia or patients whose polycythemia was initially masked by thrombocytosis [2-4]. We speculated that a small number of masked PV patients might have been diagnosed initially as having ET on the basis of such findings as thrombocytosis, presence of JAK2 mutation, elevated megakaryocyte count, and normal serum erythropoietin (EPO) level. We reviewed the hematologic data of 84 patients diagnosed as having ET between January 2003 and April 2013, who were available for assessment of JAK2, MPL, and CALR mutational status. We found 11 (13.1%) patients with Hb or Hct levels between those in the WHO 2008 criteria and those in the proposed 2014 criteria (Hb >16.5 g/dL to 18.5 g/dL or Hct >49% for men, and Hb >16 g/dL to 16.5 g/dL or Hct >48% for women; Table 1). All 11 patients had the JAK2 V617F mutation but neither the MPL nor the CALR mutation, which suggested that they might have the features of PV. We then reviewed bone marrow (BM) histomorphology and identified two (2.4%) patients who fulfilled the proposed revision of criteria for PV. Patient 1 was a 36-yr-old man. His laboratory data included the following: leukocytes, 12.3×10/L; Hb, 17.6 g/dL; platelets, 577×10/L; and normal serum EPO level. The difference between baseline (16.5 g/dL) and pathologic Hb levels was less than 2 g/dL. He had been diagnosed as having ET and received anagrelide therapy. Two months later, he was also diagnosed as having non-ST-segment elevation myocardial infarction (NSTEMI) and underwent off-pump coronary artery bypass. Patient 2 was a 63-yr-old man who had received a drug-eluting coronary stent for NSTEMI approximately two years earlier. He had thrombocytosis one year earlier. His laboratory data included the following: leukocytes, 19.3×10/L; Hb, 17.9 g/dL; and platelets, 1,577×10/L. A serum EPO level was unavailable. The difference between baseline (16.8 g/dL) and pathologic Hb levels was less than 2 g/dL. He had been diagnosed as having ET and received hydroxyurea and antithrombotic therapy. At the time of publication, patients 1 and 2 have been stable for two