PNNL

Abstract

The enhanced precision and selectivity of liquid chromatography-tandem mass spectrometry (LC-MS/MS), as well as the capability of multiplexing, has made it a technology of choice to replace many immunoassays in clinical laboratories. If robust LC-MS/MS assays developed and validated in one laboratory could be easily transferred to other laboratories, it would enable the community to perform the same tests for research and clinical care. This could also facilitate harmonization of test results and when a reference measurement procedure is available, could result in standardization of results across sites. This, in turn, could help ensure high-quality patient care via optimal clinical laboratory testing. We aimed to evaluate inter-laboratory comparability of an antibody-free multiplexed insulin and C-peptide LC-MS/MS assay. Methods The three laboratories that participated in this study comprise the NIDDK Targeted Mass Spectrometry Assays for Diabetes and Obesity Research (TaMADOR) consortium. The assay was previously validated in one laboratory and demonstrated good correlation with commercially available immunoassays. As a part of the current study, the other two participating laboratories verified the performance of the assay [reproducibility, linearity, and lower limit of quantitation (LLOQ)]. Then, an inter-lab comparison study was performed using shared calibrators, de-identified leftover laboratory samples, and reference materials. Results Total imprecision for the multiplexed assay in the two new laboratories was between 3.72% and 12.74%. Linearity was verified from 4 to 15 ng/mL for C-peptide and 2 to 14 ng/mL for insulin. In each laboratory, the LLOQ was 0.1 and 0.04 ng/mL for C-peptide and 0.03 and 0.03 ng/mL for insulin, respectively. Median imprecision across the three laboratories for C-peptide and insulin measurements of shared samples (N= 84 measurements) was 13.4% (IQR 11.6%) and 22.2% (IQR 20.9%), respectively. When replicate measurements on different days of the sample were averaged, the median imprecision across the three laboratories for the measurement of shared samples (N=60 samples) was10.8% (IQR 8.7%) and 15.3% (IQR 14.9%) for C-peptide and insulin, respectively. When comparing the results obtained for reference materials with the new assay to those of the available reference methods, a robust linear correlation was observed, with a slope of 1.044 and an r2 = 0.99 for C-peptide and a slope of 1.067 and an r2 = 0.99 for insulin. Discussion Our results suggest that the antibody-free assay of C-peptide and insulin by LC-MS/MS is a robust and transferrable workflow and that standardization with a reference measurement procedure could allow accurate and precise measurements across sites. This could be important in diabetes research and might help patient care in the future.