The measurement of immunosuppressive drugs bymass spectrometry and immunoassay in a SouthAfrican transplant setting

Abstract

Objectives: Liquid chromatography tandem mass spectrometry (LC-MS/MS) is the gold standard for measurement of immunosuppressive drugs (ISDs), but is technically demanding and less accessible in resource-limited countries. Immunoassays can also measure ISD concentrations, but may be limited by cross-reactivity. We evaluated the performance of the Roche electrochemiluminescence immunoassay (ECLIA) for cyclosporine, everolimus and sirolimus against LC-MS/MS in an African population for the first time.

Methods: Bias for ECLIA was estimated by comparing ECLIA-measured ISD concentrations to those obtained by LC-MS/MS in 42, 43 and 47 patient samples for cyclosporine, everolimus and sirolimus, respectively. Precision was assessed by performing replicate measurements of quality control materials.

Results: Deming regression analysis for all ISDs showed strong correlation between ECLIA and LC-MS/MS with a Pearson's r of >0.94. The slopes for cyclosporine, everolimus and sirolimus were 0.94 [95 % CI: 0.87-1.03], 1.35 [95 % CI: 1.23-1.44] and 0.96 [95 % CI: 0.85-1.15] with y-intercepts of 31.60 μg/L [95 % CI: 2.02-57.63], 0.23 μg/L [95 % CI: 0.21 - 0.72] and 2.61 μg/L [95 % CI: 1.30-3.56], respectively. Difference plots showed a median bias of 2.07 % [95 % CI: 1.42 - 6.99 %], 41.2 % [95 % CI: 34.9-51.8 %] and 34.9 % [95 % CI: 28.4-47.3 %] for cyclosporine, everolimus and sirolimus, respectively.

Conclusions: The cyclosporine ECLIA yielded results comparable to LC-MS/MS while poorly comparable results were obtained for everolimus and sirolimus, which may be explained by ISD metabolite cross-reactivity, amongst other factors. The poor comparability, although not unique, is noteworthy and the clinical consequences of these differences require further investigation.

Keywords: Immunoassay; Immunosuppressants; Mass spectrometry; Therapeutic drug monitoring.

Fig. 1

Deming regression analyses for (A) cyclosporine, (B) everolimus, (C) sirolimus. The dashed line is the identity (1:1) line and the grey shaded area represents the 95 % confidence interval for each drug.

Fig. 2

Difference plots (% difference and absolute difference for (A) cyclosporine, (B) everolimus and (C) sirolimus. The black solid line indicates the median value. The inner dashed lines indicate the confidence interval for the median difference. The outer dashed lines indicate the 95 % limits of the percentage and absolute difference data sets for each drug.