Prospective Evaluation of the World Health Organization Criteria for the Diagnosis of Polycythemia Vera,

The World Health Organization (WHO) published consensus criteria (2007) for the diagnosis of polycythemia vera (PV), which have never been objectively evaluated prospectively. Concern regarding limitations of these criteria and treatment algorithms based upon them has been documented. The WHO diagnosis for PV includes 2 major criteria, (1) increased red cell volume (RCV), now most often measured by an increased hemoglobin value, and (2) JAK2 mutation, and 3 minor criteria, hypercellular bone marrow with trilineage hyperplasia, low serum erythropoietin (EPO) level, and endogenous erythroid colony (EEC) formation in vitro. Diagnosis of PV requires that either both major criteria and 1 minor criterion, or increased RCV and 2 minor criteria must be met. Increased red cell volume was arbitrarily defined as: hemoglobin concentration (Hgb) >18.5 g/dL in men and >16.5 g/dL in women, or hematocrit value (Hct) >99^th percentile of institutional normal range, or Hgb >17g/dL in men and >15 g/dL in women if a true sustained increase ≥2 g/dL from baseline exists, or elevated red cell mass (RCM) value >25% above mean normal predicted value according to Pearson, et al.

We prospectively evaluated 30 patients with a clinical phenotype of PV using the WHO criteria for diagnosis. For measuring RCV, simultaneous Cr⁵¹ RCM and I¹²⁵ plasma volume analyses were performed. The RCM of overweight patients (BMI >25 kg/m²) was determined by using a BMI of 22 kg/m² (20% fat) for adjusted weight. JAK2^V617F allele burden was determined both qualitatively by ARMS PCR assay and quantitatively by pyrosequencing. Bone marrow biopsy slides were prepared routinely and stained for reticulin and collagen. The Manoharan scale was used for evaluating fibrosis. Marrow interpretation by 2 observers was initially “blinded”. EEC assays were not performed. All patients have been followed regularly at our institution for a median of 4 years to allow confirmation of diagnosis.

Of the 30 patients, 28 were diagnosed as PV because of an increased RCM and JAK2 positivity. The median phlebotomy requirement during the first year was 4 (range 2–13). The remaining 2 patients met the WHO criterion of JAK2 positivity and an increased hematocrit value but had a normal RCM and were diagnosed as essential thrombocythemia (ET), subsequently confirmed clinically by no increase in Hct or Hgb and no need for phlebotomy in the succeeding years. Eighteen PV patients, 4 women and 14 men, met major criterion #1 by elevated RCM only, but not by the WHO criteria for increased Hgb. The median Hgb count was 15.2 g/dL for the 4 women (range: 14.4–16.4) and 17.2 g/dL for the 14 men (range: 15.6–18.1). All 28 patients had marrow findings consistent with PV; the marrows of the 2 ET patients were interpreted as nonspecific myeloproliferative disease. Only 21 patients had a low EPO level (<5 mU/mL), as did 1 ET patient. There was no correlation between RCM and white blood cell, red blood cell, or platelet counts, or between RCM and JAK2^V617F %, age, phlebotomy requirement, spleen size, or grade of myelofibrosis. Although in general, RCM increased with increasing Hgb and Hct, it was not possible to correlate RCM with a unique Hgb or Hct.

We conclude, as have others, that Cr⁵¹ RCM determination remains critical for distinguishing absolute erythrocytosis and the subsequent diagnosis of PV. Using Hgb as a single criterion, 18 patients (64.3%) would not have been diagnosed as PV and using Hct as a single criterion, 9 patients (32.1%) would have not. We confirm marrow evaluation is highly valuable when performed by skilled hematopathologists. However, serum EPO levels are not diagnostic. EEC assays are rarely performed except in experimental laboratories. The frequency of performing RCM studies has decreased due to a reliance on hemoglobin concentration, but this is incorrect. Informal review of hematocrit ranges shows institutional variability. We recommend immediate revision of the WHO criteria for the diagnosis of PV patients with “early stage” disease, especially to distinguish it from ET. Criteria for the unequivocal diagnosis of PV should be increased RCM, as opposed to any form of increased RCV, and JAK2 mutation. In the rare cases of increased RCM and JAK2 negativity, marrow morphology plus other minor criteria to be discussed can be diagnostic. Since the therapies for myeloproliferative disorders differ, our data has major clinical implications.

No relevant conflicts of interest to declare.