Radioimmunoassay of immunoreactive erythropoietin as a clinical tool for the classification of polycythaemias.

Radioimmunoassay of erythropoietin (EPO) has been used for evaluating its clinical usefulness in distinguishing polycythaemia vera (PV) from pure erythrocytosis (PE). A normal log distribution (13.40 mU/ml +/- 2.45) was observed in the 66 reference samples. Similar results were observed in 29 pure thrombocytaemias (13.23 +/- 5.19). In PV patients, whether in clinical remission or in active phase, the EPO titer was lower than normal values (respectively 7.32 +/- 3.63 and 6.59 +/- 2.75), without any correlation with the haematocrit (44 to 51% and 52 to 71%). In the anaemic cases (excessive therapy, spent phase, myelofibrosis), a slight excess of EPO titer was observed, but less than expected when taking the hematocrit into account. In contrast, pure erythrocytosis is a very heterogeneous group, with the highest EPO values in the well-defined secondary cases, and normal or slightly excessive values in most of the cases, but with some low values similar to those observed in PV cases. In term of PV diagnosis, a 90% predictive value is observed when a "cut-off" value of 11 mU/ml is chosen. When the "cut-off" is 16 mU/ml, no PV case was observed beyond this value. We conclude that EPO assay is useful in myeloproliferative diseases from a practical point of view.