Comparison of the everolimus concentrations measured in whole blood with everolimus QMS or sirolimus CMIA

Abstract

Few years ago, it was proposed that everolimus blood levels could be determined with the commercially available sirolimus chemiluminescence magnetic microparticle immunoassay (CMIA). More recently, a highly specific microsphere system (QMS) has been approved by FDA for therapeutic drug monitoring in humans. Aim of the present study was to compare the results of everolimus assay performed with everolimus QMS and with sirolimus CMIA. The two methods were compared with Passing-Bablok regression and Bland-Altman plot analysis. The Passing-Bablok regression analysis showed that although the results obtained with the two techniques were significantly correlated, CMIA-measured differed from QMS-measured everolimus concentrations by both a systematic and a proportional error. Specifically, at blood levels lower than 5 ng/mL CMIA were lower than QMS-measured everolimus concentrations. On the opposite, at everolimus blood concentrations higher than 10 ng/mL CMIA-estimated values became progressively higher than QMS-measured everolimus concentrations. The analysis of the Bland Altman plot showed a less than optimal agreement of the two tests (5.59% of the data point outside the ±1.96 SD interval). Moreover, the relationship between the difference between Evero and Evero and their average was clearly concentration dependent with positive and negative values at concentration values lower and higher than 5 ng/mL respectively. In conclusion, our finding showed that the values of everolimus concentrations measured with sirolimus CMIA differ from those detected with the FDA-approved everolimus QMS further suggesting that sirolimus CMIA should not be used anymore for everolimus therapeutic drug monitoring.

Analytical Performance of the New Siemens Affinity Chrome-Mediated Immunoassay Everolimus Assay and Its Interchangeability With the Thermo Quantitative Microsphere System for Routine Therapeutic Drug Monitoring of Patients After Solid Organ Transplantation

“…This might explain why unlike previous cheating assays, no negative systematic bias (related to metabolites) was observed with EVL-specific comparators. 25 Nevertheless, there are some limitations to consider in this study. A major limitation is the obvious lack of comparison with the analytical gold standard, LC-MS/MS, which allows estimation of actual bias in IAs.…”

“…This might explain why unlike previous cheating assays, no negative systematic bias (related to metabolites) was observed with EVL-specific comparators. 25…”

“…12 This strategy of quantifying EVL by "cheating" IAs developed for SRL is not new and has undergone in-depth investigations in the past, producing conflicting results on its reliability. [20][21][22][23][24][25] Indeed, under most favorable conditions (ie, calibration with EVL), comparison with EVL-specific IAs is characterized by a high linear correlation, proportional bias of 30% (ie, slope = 1.3), and lack of individual agreement. 25 Therefore, our findings agree with the effects of low specificity toward EVL that characterize the "cheating" assays, and this should raise awareness of the actual limits to the interchangeability of ACMIA/EVRO despite its excellent analytical performance.…”