1
Robust and Accurate two-year Performance of a Quantitative Mass Spectrometry-Based Apolipoprotein Test in a Clinical Chemistry Laboratory
L. Renee Ruhaak1, Nico P.M. Smit1, Fred P.H.T.M. Romijn1, Mervin M. Pieterse1, Yuri E.M. van der Burgt1, Arnoud van der Laarse1 and Christa M. Cobbaert1
1Department of Clinical Chemistry and Laboratory Medicine, Leiden University Medical Center, Leiden, The Netherlands
To the editor:
Previously, we described a multiplex liquid chromatography-mass spectrometry (LC-MS) test for the quantification of six apolipoproteins (A-I, B, C-I, C-II, C-III and E) (1). Analytical method validation was performed according to CLSI protocols and intraassay CVs were 2.3%–5.5%, and total CVs were 2.5%–5.9%. Moreover, results were interchangeable with immunoturbidimetric assays for apoA-I (Deming slope 1.014) and for apoB-100 (Deming slope 1.016) (1). Our lab-developed test has now been employed intermittently during a two year period in various studies on clinical cohorts. Here we demonstrate the robustness and accuracy of MS-based apolipoprotein quantification using stable-isotope-dilution and multiple reaction monitoring (SID-MRM) on a triple quadrupole for routine practice in clinical laboratories.
To evaluate performance of the LC-MS test with regard to precision and bias of test results, two native serum pools were included as internal quality control (QC) on each of the 96-wells plate batches analyzed (both in triplicate) over this period. The measurements were compared to assigned values to assess bias and precision by analysis of repeated QC observations in comparison to predefined IQC intervals (1). The results for these two QC samples measured between July 2015 and August 2017 for the six serum apolipoproteins are shown in Figure 1 and provide estimates of bias and precision which were not significantly different from those determined at validation (1). Assay performance was unaffected by typical laboratory workflow variation and reagent lot changes. The data represents assays run by five technicians across two instruments with different model numbers (Agilent 6490, 13 runs, n=40 and Agilent 6495, 37 runs, n=123). Two lots of trypsin and three HPLC columns were used in the course of this analyses. Similar results were found for both QC samples and could be confirmed for the second peptide of each protein with CVs<10% for all peptides with one exception (peptide ESLSSYWESAK in apo C-II).
Test development was performed according to CLSI C62-A recommendations (2) and the test was analytically validated against predefined desirable performance criteria based on biological variation (3). Semi-automated sample preparation has been realized using an Agilent Bravo liquid handling system (1). The digestion conditions were optimized for equimolar conversion from protein to signature peptides using two peptides per protein and three transitions per peptide. Stable isotope- labeled peptides were used as internal standards. To ensure standardization, five value-assigned, native serum samples were used as external calibrators. Test results are traceable to WHO-IFCC Reference Materials SP1–01 and SP3–07 for respectively apoA-I and apoB, and to manufacturer’s working calibrators for the other apolipoproteins. These considerations allow for, but do not yet ensure long-term accuracy of multiplex apolipoprotein test results.
2
After analytical validation, the multiplex apolipoprotein test was secured in the ISO 15189:2012 quality system of the department. Proper instrument performance is required for accurate and precise performance. Instrument performance was assessed on days when tests were run through the evaluation of a system suitability sample consisting of 14 synthetic peptides and its SIL- derivatives, all at 0.15 mol/L, which was analyzed before and after each batch to assess ion abundance, which should meet preset values to ensure sensitivity, relative responses with CVs<10%
to ensure precise quantitation, ion ratios that do not deviate >15% from the target value to ensure proper fragmentation, and carry-over <1% to minimize interference from previous samples.
Moreover, HPLC column pressure, column oven temperature and MS vacuum system were monitored prior to each run. Regular cleaning and preventive maintenance were also carried out according to manufacturer’s instructions and comprised replacement of sensitive parts of the LC-MS and annual maintenance by the manufacturer.
The long-term QC precision traces for high abundance serum apolipoproteins provided in this study demonstrate the robustness of multiplex protein quantification on triple quadrupole instrumentation. QC results for apolipoproteins using LC-SID-MRM-MS are stable over 2 years and meet desirable analytical bias and precision requirements for accuracy. The clinical relevance for measuring apoA-I and apoB is overt, whereas the clinical utility of apoCs and apoE needs further exploration. We recently revealed a new apoC-II-deficient phenotype (4) using this test, and the clinical evaluation of this test according to the EFLM Test Evaluation framework (5) is ongoing. We believe that this MS-based multiplex protein quantification test is robust due to careful methodical attention to best practice approaches in development and validation, in combination with a thorough implementation strategy.
References
1. van den Broek I, Romijn FP, Nouta J, van der Laarse A, Drijfhout JW, Smit NP, et al.
Automated multiplex lc-ms/ms assay for quantifying serum apolipoproteins A-I, B, C-I, C-II, C-III, and E with qualitative apolipoprotein E phenotyping. Clin Chem 2016;62:188-97.
2. Lynch KL. CLSI C62-A: A New Standard for Clinical Mass Spectrometry. Clin Chem 2016;62:24- 9.
3. Sandberg S, Fraser CG, Horvath AR, Jansen R, Jones G, Oosterhuis W, et al. Defining analytical performance specifications: Consensus statement from the 1st strategic conference of the european federation of clinical chemistry and laboratory medicine. Clin Chem Lab Med 2015;53:833-5.
4. Hermans MPJ, Bodde MC, Jukema JW, Schalij MJ, van der Laarse A, Cobbaert CM. Low levels of apolipoprotein-CII in normotriglyceridemic patients with very premature coronary artery disease: Observations from the MISSION! Intervention study. J Clin Lipidol 2017;11:1407- 1414.
5. Horvath AR, Lord SJ, StJohn A, Sandberg S, Cobbaert CM, Lorenz S, et al. From biomarkers to medical tests: the changing landscape of test evaluation. Clin Chim Acta 2014;427:49-57.
Formatted: English (United States)
3 Figure legend
Figure 1. Levey-Jennings charts for two QC samples (QC1/QC2) for six signature peptides (AKPAL, apoA-I; FPEVD, apoB; TPDVS, apoC-I; ESLSS, apoC-II; GWVTD, apoC-III and LGPLV, apoE). Days after method validation according to CLSI EP-15 (1) are depicted on the x-axis. The dashed and horizontal lines indicate the target values as determined during method validation of the apolipoprotein concentrations with one, two and three standard deviations. The concentration values on the y-axis are normalized and should read for QC1(SD)/QC2(SD) as 1.48(0.0910)/1.34(0.0539), 0.861(0.0296)/0.993(0.0431), 17.6(0.839)/21.9(1.41), 37.8(3.03)/56.4(3.15), 103(4.41)/169(4.84), 31.4(2.05)/56.0(4.18) for apo A-I (in g/L), B (in g/L), C-I (in mg/L), C-II (in mg/L), C-III (in mg/L) and E (in mg/L), respectively.
