Diagnostic Utility of PRV-1 mRNA Quantification in Polycythemia Vera and Essential Thrombocythemia.

In recent reports, granulocyte PRV-1 mRNA has been shown to be over-expressed in the majority of polycythemia vera (PV) patients and with a variable frequency in essential thrombocythemia (ET) patients. Here we evaluate PRV-1 mRNA quantification in the diagnosis of PV and ET. Following patients were included in the study: 23 PV patients; 5 secondary erythrocytosis (SE) cases, 4 by lung disease and 1 by a hemoglobinopathy (Hb San Diego); 4 relative erythrocytosis (RE) cases; and 13 ET patients. PV and ET diagnosis was made according to the WHO criteria. At time of sampling 43% of PV patients and 54% of ET patients were receiving myelosuppressive therapy. To establish a cut-off value 28 healthy persons (15 female, 13 male) were also analysed. RNA was extracted from whole-blood leucocytes and quantified by real-time TaqMan PCR. ABL was used to control the mRNA quality (Ct ABL < 27.0 – mean 24.7) and to normalise the PRV-1 values. PRV-1 primer and probe sequences originated from Brohée D et al, Blood, 2002, 100:3146a and ABL primer and probe sequences from Beillard E et al, Leukemia, 2003, 17:2474. The results were expressed as a ratio PRV-1/ABL which could be calculated using the ΔΔCt method as the efficiencies of both PCR reactions were almost equal (for PRV-1 93% and ABL 94%). Within-run and between-run reproducibility of a normal sample had a CV of 10.1 % and 28.6% and of a highly expressing sample a CV of 9.5% and 21.6%. The values of the healthy controls had a normal distribution and a mean of 0.28 with a SD of 0.35. A cut-off level was determined as the upper 95% percentile and was 0.98. Results of the different patient groups are summarised in the table. The differences between PV vs healthy controls, PV vs SE/RE and ET vs healthy controls were significant with a p-value of respectively <.001, <.001 and =.006. Three of the 23 PV patients had a PRV-1 result below the cut-off level. Thus we found a diagnostic sensitivity of 87%. Two of the 3 negative patients had no detectable PRV-1 expression, we hypothise resulting from a congenital deficiency. None of the 9 SE and RE patients had a PRV-1 expression above the cut-off level. Thus in our small series of polycythemic patients the specificity of PRV-1 over-expression for PV is 100%. Five of the 13 (38%) ET patients had a PRV-1 expression above the cut-off level. Patients with secondary thrombocythemia have to be studied to establish the specificity of PRV-1 over-expression for ET. The over-expressing ET patients had a significantly lower level of PRV-1 than the PV patients (p =.001). None of the 5 positive (as neither the other 8) ET had thrombotic complications nor developed PV. In conclusion PRV-1 mRNA quantification on whole-blood leucocytes is easy, reproducible and highly sensitive and specific for PV. The test should facilitate the diagnosis of this disease. For ET more cases of secondary thrombocythemia have to be studied to establish the diagnostic specificity of the assay.

PRV-1/ABL mRNA ratio in healthy controls, PV, SE/RE and ET