Method-specific differences in plasma fibroblast growth factor 23 measurement using four commercial ELISAs

Background: There is growing interest in measuring plasma fibroblast growth factor 23 (FGF23) concentration in a number of clinical settings. However, data comparing current commercial intact and C-terminal FGF23 assays is lacking. Methods: We used plasma samples collected from a cohort of healthy adults and patients undergoing chronic haemodialysis therapy (n=67) to compare the precision, recovery, linearity and pre-analytical stability characteristics of four commercial FGF23 assays from Kainos, Millipore and Immutopics Inc. Method agreement was evaluated using Passing-Bablok regression and difference plot analysis. Results: Both Millipore and Immutopics intact FGF23 kits demonstrated marked negative proportional bias relative to Kainos assay readout, particularly in the haemodialysis group, and poor recovery of purified FGF23 standard at high spiking concentrations. Dilution of high-reading samples with saline as recommended by the Immutopics kit resulted in significant deviation from linearity. Immutopics C-terminal FGF23 concentrations displayed a strong association with intact FGF23 concentrations determined with all three intact assays in the haemodialysis group, but showed no significant correlation within the physiological range. Only intact FGF23 measurements made with the Immutopics assay demonstrated evidence of significant instability 8 h after venepuncture. Conclusions: Current ELISA kits for plasma intact FGF23 measurement show poor analytical agreement, and cannot be used interchangeably. This is mainly due to differences in calibration. Harmonisation of available assays using a common international standard would facilitate more meaningful interpretation of data from studies using different kits. Discordance between intact and C-terminal FGF23 assay measurements is more marked at physiological concentrations than in patients undergoing haemodialysis.