The detection of anti-erythropoietin antibodies in human serum and plasma Part II.: Validation of a semi-quantitative 3H-thymidine uptake assay for neutralizing antibodies

Neutralizing antibodies to erythropoietin (EPO) can cause a loss of response to recombinant human EPO (rHuEPO) and lead to rare cases of sudden, unexplained, severe anemia in chronic renal failure patients treated with rHuEPO. An assay for neutralizing anti-EPO antibodies has been validated that is based on the inhibition of proliferation of human UT-7/EPO cells, an immortalized cell line, by neutralizing antibodies in serum test samples using H-3-thymidine as a marker for proliferation. The dependence of the human cell line on EPO for growth and proliferation in a concentration-dependent manner enabled the validation of a rHuEPO standard curve for cell proliferation that can be used to determine the presence of neutralizing anti-EPO antibodies in serum samples. Proliferation of the cells increases with increasing concentrations of EPO, forming an S-shaped standard curve, which is fit with a 4-parameter logistic model, between 2.5 and 50 mU/mL rHuEPO, with a percent coefficient of variation (% CV) from 8.7% to 22.1% and a % accuracy of 103.5% to 109.5%. Anti-EPO antibodies and serum with anti-EPO antibodies neutralize UT-7/EPO proliferation by 10 mU/mL rHuEPO in a concentration- or dilution-dependent manner with <= 25% CV. Percent neutralization is calculated by determining the amount of EPO recovered from the original 10 mU/mL added using the formula [((10 -concentration recovered)/ 10) x 100%]. Stem cell factor (SCF) stimulated cell proliferation, but not as effectively as rHuEPO. Antibodies to SCF were not able to inhibit the proliferative response induced by EPO and vice versa, confirming the specificity of the assay for antibodies to EPO. High EPO levels can impact both the radiomirminoprecipitation and neutralization assays to produce a false negative result. However, the impact can be mitigated by the large dilutions used in the neutralization assay. (c) 2005 Elsevier B.V. All rights reserved.