|
Abstract INTRODUCTION: Mass spectrometry (MS)is an emerging technology that finds its way into medical laboratories for quantitation of proteins and proteoforms. Due to the complexity of bottom up proteomics quality requirements should be met during both method development and implementation as a lab-developed-test in order to guarantee that MS-based tests are fit-for-clinical-purpose(1). We previously developed a multiplexed LC-MRM-MS test for serum apolipoproteins (apo’s)(2). To maintain stable test performance, test quality has to be managed. Here we present long-term analytical performance data of an MS-based apo test in the real world setting of a medical lab.
METHODS: An LC-MRM-MS method was developed for multiplex quantitation of apo’s (a), A-I, A-II, A-IV, B, C-I, C-II, C-III and E in agreement with CLSI C62A recommendations. The apo method makes use of automated tryptic digestion of serum proteins and subsequent detection of at least 2 proteotypic peptides/protein and 3 MRM transitions/peptide. Five value-assigned, native serum samples were used as external calibrators. Test results were traceable to WHO-IFCC reference materials for apoA-I, B, and (a) and to manufacturer’s working calibrators for the other apo’s. A System Suitability Test comprising of synthetic peptides was established and two serum pools were used as IQC material. CLSI-EP15 validation demonstrated that all nine apo’s could be measured with overall CVs between 2.5 and 7.3 %. Both IQC samples were analyzed (3x) on each of the 96-well plates (n=70 runs) since July 2015 (3). The MS instrument was replaced after 12 runs, while an SPE-step was introduced after 37 runs to enable inclusion of medium abundant apo’s (a), A-II and A-IV. Trending of IQC results was monitored in Levey- Jennings plots.
RESULTS: IQC measurements remained within desirable analytical performance specifications over almost four years, independent of different technical operators, reagent lot variations or trypsin batches. Replacing the MS instrument did not significantly alter the results, nor did inclusion of an SPE-step. For the 23 peptides, measured CVs varied between 3.3 % and 11.5 % in all 70 runs.
CONCLUSION: This study indicates that long-term robustness of MS-based protein quantitation can be achieved through thoughtful method development in combination with stringent quality procedures as demanded by ISO15189:2012. This requires also properly trained staff, documentation of procedures, optimal maintenance of equipment, system suitability testing and checks for both precision and long term accuracy through IQC (and EQA when available). We demonstrate that MS-based tests for quantitation of proteins and their proteoforms perform stable in the setting of a medical laboratory and can be run with similar analytical performance as immunoassay-based tests.
1.Smit et al. TranslProteomics 2013;2,1.
2.van den Broek et al. Clin Chem 2016;62,188.
3.Ruhaak et al. Clin Chem 2018;64,747. |
= Discovery stage.
= Translation stage.
= Clinically available.
