EEC Growth, JAK2 Mutation, PRV-1 Overexpression and X-Chromosome Inactivation Pattern: Predicting Values on Polycythemia Vera Evolution and Risk of Thrombosis in Idiopathic Thrombocytosis.

We have previously demonstrated that endogenous erythroid colony (EEC) growth could predict PV evolution in idiopathic thrombocytosis (IT) (Blood 83:744, 1994). We also showed that female patients with IT and monoclonal X-chromosome inactivation pattern (XCIP) had a high risk of vascular complications (

1. to determine the correlation among the molecular markers of EEC growth, JAK2 mutation status, PRV-1 level and XCIP, and

2. to assess the predicting values of these biomarkers on PV evolution and risk of thrombosis.

Allele-specific PCR assay was used to detect JAK2V617F in peripheral blood granulocytes (n=178) or bone marrow cells (n=75) from patients with IT. EEC assay was performed on 235 patients in a serum-free culture system at initial diagnosis. PRV-1 mRNA expression in granulocytes from 187 patients was measured by real-time RQ-PCR assay and expressed as the copy number of PRV-1 normalized to GAPDH gene (NCN). Clonality analysis of XCIP with HUMARA-PCR assay was performed on 123 female patients. JAK2V617F was detected in 57.7% of patients with IT. Forty-one percent of IT patients had EEC growth at initial presentation. Of those analyzed for PRV-1 levels, 138 patients had not received platelet-lowering agents at time of analysis; 72 (38.5%) had PRV-1 overexpression with NCN ≥ 1 (NCN of 30 normal subjects, mean±SE: 0.32±0.04, range: 0.04–0.97). Seventy-six female patients had monoclonal XCIP and 28 had polyclonal XCIP; the remaining 19 patients had ambiguous or homozygous pattern of XCIP. EEC growth was strongly correlated with JAK2 mutation (P<0.0001) and PRV-1 overexpression (P<0.0001). PRV-1 overexpression was highly associated with JAK2 mutation (P<0.0001) and monoclonal XCIP (P=0.003). The frequency of JAK2 mutation or EEC growth was not different between monoclonal and polyclonal XCIP. Thirty-four patients subsequently developed PV (IT-PV), ranging from 1.5 months to 90 months. All 34 IT-PV patients had EEC growth and all but one harbored JAK2 mutation at initial presentation. Patients with IT-PV had a significantly higher frequency of EEC(+) (P<0.0001), JAK2(+) (P<0.0001), and PRV-1(+) (P<0.0001) than those without PV evolution, whereas there was no difference in age (P=0.914) or XCIP (P=0.550) between IT with and without PV evolution. One hundred and sixty-nine patients had all biomarkers of JAK2/EEC/PRV-1 analyzed, 57 patients were JAK2(+)/EEC(+)/PRV-1(+) and 55 were JAK2(-)/EEC(-)/PRV-1(-). Eighteen patients with all 3 biomarkers-positive developed PV later compared with none of patients with all 3 biomarkers-negative (P<0.0001). Eighty-two patients experienced thrombotic complications during their disease courses. The occurrence of thrombosis was strongly associated with EEC growth (P=0.0002), JAK2 mutation (P=0.007), PRV-1 level (P=0.004), and XCIP (P=0.034). The present study showed that EEC growth and JAK2 mutation were predictors of PV evolution; EEC growth, JAK2 mutation, PRV-1 overexpression and monoclonal XCIP predicted the risk of thrombosis in patients with IT.