- Using fluorescence polarization immunoassay for determination of erythrocyte methotrexate polyglutamates, a quick and easy test?
Using fluorescence polarization immunoassay for determination of erythrocyte methotrexate polyglutamates, a quick and easy test?
The folate antagonist methotrexate (MTX) is the anchor drug in the treatment of rheumatoid arthritis. The therapeutic effects of MTX are attributed to the intracellular levels of MTX, present in the cell as polyglutamates (MTX-PGs). We aimed to validate an immunoassay for the measurement of MTX-PG in erythrocytes. Samples were analyzed by an adapted fluorescence polarization immune assay (FPIA) method on the FLx analyzer (Abbott). Cross-reactivity was determined in both plasma and erythrocyte pellet. In erythrocyte pellet, the imprecision, linearity, and lower limit of quantitation were determined. The method was compared with our in-house liquid chromatography tandem mass spectrometry (LC-MS/MS) method for total MTX-PG. For the adapted FPIA method, a linear range of 25-1000 nmol/L (R = 0.993) was obtained for total MTX-PG in erythrocytes. A coefficient of variation of <17% for interday and <8% for intraday imprecision was found and average recovery was 91%. Lower limit of quantitation was determined at 50 nmol/L total MTX-PG with a coefficient of variation of 15%. There was no significant proportional bias of the FPIA assay compared with our in-house LC-MS/MS method, but a (nonsignificant) constant positive bias was present [FPIA = 1.00 (95% confidence interval: 0.60-1.95) × LC-MS/MS + 31.00 nmol/L (95% confidence interval: -11.83 to 61.00)]. Results could be very different for individual patients as reflected in the poor R of 0.419. The FPIA method can be used to measure total MTX-PG in erythrocytes. Although there was no significant bias detected compared with the LC-MS/MS method, the FPIA method showed constant positive bias, probably because of interference from folates and MTX metabolites 2,4-diamino-N10-methylpteroic acid and 7-hydroxy-MTX. The correlation between both methods was average and resulted in large differences in individual patients, most likely because of problems during sample preparation.