Comparison of cyclosporin A measurement in whole blood by six different methods

In a multicentre trial, we analyzed and compared the cyclosporin concentrations in whole blood specimens (trough values) from renal and bone marrow transplant patients using six different methods: high-performance liquid chromatography (n-butyl reversed-phase column) as reference, and 5 immunoassay procedures using monoclonal specific or non-specific antibodies, or polyclonal antibodies, for measuring cyclosporin with and without its cross-reacting metabolites. Results obtained by the 2 specific RIAs show good correlations (r-values of 0.945 and 0.921) versus the HPLC method, with average assay ratios of: 1.52 for 3H-RIA/HPLC and 1.26 for 125I-RIA/HPLC. In contrast, ratios for non-specific immunoassay/HPLC show multiple-fold overestimations of cyclosporin with very wide variations: 4.58 for 3H-RIA/HPLC, 3.97 for 125I-RIA/HPLC and 3.87 for fluorescence polarization immunoassay (FPIA)/HPLC. These findings indicate 1) that the 'specific' 3H- and 125I-RIAs can be used for cyclosporin measurements in whole blood using normal therapeutic ranges established by HPLC, 2) that the important disparity and variable overestimations by 'non-specific' RIA or fluorescence polarization immunoassay, compared with HPLC, require adjustments in therapeutic ranges, and 3) that the available 'specific' and 'non-specific' immunoassays should enable establishment of within-house reference 'therapeutic/toxic' ranges for cyclosporin in individual centres, based on these 'specific' versus 'non-specific' assays.