Recently, Liu et al1 and Kralovics et al2 reported in this journal the possible diagnostic value of some biologic markers in chronic myeloproliferative disorders such as polycythemia vera (PV) and essential thrombocythemia (ET). Briefly, they described a strong correlation between endogenous erythroid colony growth in the absence of added erythropoietin (Epo) and polycythemia rubra vera 1 (PRV-1) expression in PV patients.
Our experience demonstrates very similar results using comparable methodologies. We studied the correlations among endogenous growth of circulating erythroid burst-forming unit (eBFU-E) and megakaryocytic colony-forming unit (eCFU-MK) progenitors, clonality by human androgen receptor–polymerase chain reaction (HUMARA-PCR) assay, and expression of PRV-1 in females with PV (n = 11) and ET (n = 17). All patients fulfilled the diagnostic criteria of the Polycythemia Vera Study Group (PVSG). At the time HUMARA and PRV-1 assays were performed, 5 of 17 ET and 2 of 11 PV patients were receiving myelosupressive/platelet-lowering agents. BFU-E and CFU-MK were determined at diagnosis, except for 2 ET patients who were receiving anagrelide.
In vitro cultures were performed as previously reported.3 HUMARA assay was analyzed on granulocytes and CD3+ lymphocytes. PRV-1 expression was quantified by real-time reverse-transcriptase (RT)–PCR in RNA from granulocytes.
In PV patients, 11 (100%) of 11 showed eBFU-E and 4 (36%) of 11, eCFU-MK. Clonality was demonstrated in 5 (50%) of 10 patients, and 1 patient was homozygous. PRV-1 overexpression was detected in 10 (91%) of 11 patients. The correlation among the 3 assays gave a concurrent positive result in 5 patients, but considering the 3 assays separately, we found a positive result in all PV patients (Table 1).
In ET patients, 5 (29%) of 17 had eBFU-E, 7 (41%) of 17 had eCFU-MK, and 9 (53%) of 17 had eBFU-E and/or eCFU-MK. Clonality was demonstrated in 6 (40%) of 15 patients, and the other 2 were homozygous. PRV-1 overexpression was observed in 10 (59%) of 17 patients (Table 1). Correlation between eBFU-E and/or eCFU-MK and overexpression of PRV-1 were observed in 7 (41%) of 17 patients. When we compared the 3 assays, a coinciding positive result was observed in 2 patients, but if they were considered separately, a positive result was found in 15 (88%) of 17 ET patients.
In our preliminary results, the best and most reliable correlation was observed between eBFU-E and PRV-1 expression in PV (91% of patients). In ET, a lower correlation (41%) between the 2 assays was found; but if we consider all assays separately, patients with a positive result increase to 88%. Cultures and PRV-1 results are in agreement with Liu et al1 and Kralovics et al2 except for eCFU-MK, which was not studied. Clonality by HUMARA gene is lower than that obtained by Liu et al1 using G6PD, IDS, MPP1, BTK, and FHL1 genes.
We conclude that endogenous BFU-E remains the single most useful technique for diagnosis of PV and, together with PRV-1 expression, the most reliable biologic markers for PV. In ET, eBFU-E and eCFU-MK with PRV-1 expression are useful diagnostic tests in more than 70% of ET patients. We suggest, in line with others,1,2 that the concomitant implementation of in vitro cultures of BFU-E and/or CFU-MK and PRV-1 expression should be applied in PV and ET for their high diagnostic yield and that, whenever possible, HUMARA assay should be added.
Supported by grants FIS 01/1424 and SAF-2001-4947 (Spain).
The authors wish to express their gratitude to Dr M. Albitar (University of Texas M. D. Anderson Cancer Center, Houston, TX) for his valuable advice in the development of the HUMARA assay.
