|
Abstract BACKGROUND:
Patients and clinicians expect consistent test results across hospitals, making metrological traceability crucial. Also, medical results used in clinical care should be traceable to higher order reference measurement systems (RMS), consisting of reference measurement procedures (RMPs) and reference materials (RMs). The previous RMSs for Apolipoprotein AI (ApoAI) and Apolipoprotein B (ApoB) are no longer available. Therefore, the IFCC working group on apolipoprotein standardization by mass spectrometry developed and validated a multiplex liquid chromatography–multiple reaction monitoring mass spectrometry (LC-MRM-MS)-based RMP for ApoAI and ApoB quantification in serum. The RMP presented here meets the ISO 15193 requirements and has a calibration hierarchy in line with ISO 17511:2020.
METHODS:
An LC-MRM-MS based RMP was developed for ApoAI and ApoB. Serum samples underwent preparation, including reduction, alkylation, 1 hour LysC pre-digestion (1:700 w/w) and trypsin digestion (1:47 w/w) for 3 hours at 37 °C. The proteotypic peptides are purified using solid-phase extraction (Oasis HLB 3 mg/well, 80% MeOH elution) and analyzed via an Agilent 1290 UHPLC system coupled to a 6495C triple quadrupole MS. Sample preparation involved a semi-automated Bravo liquid handling system (Agilent). Absolute quantification was performed using isotope dilution MS with (13C,15N)-labeled internal standards. Retention times, precursor ion m/z, and 3 fragment ion m/z as well as collision energies were selected per proteotypic peptide. Deidentified serum samples, prepared per CLSI C37A, served as calibrators traceable to WHO/IFCC SP1-01 (ApoAI) and SP3-07 (ApoB). The calibrators and value assignment are still provisional. RMP validity was ensured through internal quality control (IQC) assessment and intrinsic metadata evaluation. The RMP included bilevel IQC, with mean concentrations of 2.1 g/L (QC1) and 1.3 g/L (QC2) for AKPAL, and 0.6 g/L (QC1) and 1.4 g/L (QC2) for FIIPS. In addition, system suitability testing (SST) to ensure instrument performance, ion ratio monitoring to ensure peptide-specificity, and interpeptide agreement between quantifying and qualifying peptides to assess individual sample validity, were implemented. R scripts were used for automated data evaluation and checked SST, IQC limits, and interpeptide agreement.
The method underwent analytical validation following ISO 15193:2009, including evaluation of linearity, matrix effects, carryover, recovery, measurement uncertainty, precision (repeatability and within-lab CV), Limit of Blank (LoB), limit of detection (LoD), and lower limit of quantitation (LLoQ).
RESULTS:
For proteotypic peptide selection of ApoA-I and ApoB using tryptic in silico digestion, we excluded peptides containing tryptophan, methionine, and cysteine. The RMP targets three tryptic proteotypic peptides for ApoAI, with AKPALEDLR (AKPAL) serving as the quantifying peptide and ATEHLSTLSEK and THLAPYSDELR as qualifying peptides. In addition, five tryptic proteotypic peptides were measured for ApoB, with FIIPSPK (FIIPS) as the quantifying peptide and GFEPTLEALFGK, SVSLPSLDPASAK, AAIQALR and TEVIPPLIENR as the qualifying peptides. Interpeptide comparison over 8 runs for analytical validation shows strong concordance, with high correlation coefficients observed between monitored peptides. For ApoAI, ATEHLSTLSEK and THLAPYSDELR compared to AKPAL showed Pearsons R of 0.996 and 0.999 (with slopes of 1.008 and 0.996). For ApoB, correlation coefficients were 0.997 with slope 0.990 (AAIQALR vs. FIIPS), 0.999 with slope 0.989 (GFEPTLEALFGK vs. FIIPS), 0.998 with slope 0.980 (SVSLPSLDPASAK vs. FIIPS), and 0.999 with slope 0.997 (TEVIPPLIENR vs. FIIPS), confirming analytical selectivity of the measurands. Linearity (CLSI EP6) was demonstrated for ApoAI (from 0.9 – 1.98 g/L) and for ApoB (from 0.54 – 1.47 g/L). Serial dilution experiments covered broader concentration ranges, from 0.09 – 2.81 g/L for ApoAI and 0.26 – 3.66 g/L for ApoB. For the quantifying peptides AKPAL and FIIPS, measured concentrations closely aligned with expected values, yielding linear regression slopes of 0.945 and 0.985, respectively, with Pearson correlation coefficients of 0.999 and 1. All concentrations stayed within the ±10% allowable deviation. Carryover assessment (CLSI EP10) showed all peptide differences remained within error limits. This was ±0.1% of the high mean (≤0.003 g/L) for ApoAI and < ±0.3% of the high mean (≤0.010 g/L) for ApoB, confirming robustness with minimal carryover in low-concentration samples. The RMP shows high precision (CLSI EP05) for ApoAI and ApoB across (n=20) clinical serum/IQC samples spread over 15 runs. For ApoAI, within-run imprecision (CVr) ranged from 0.92% to 2.87%, between-run imprecision (CVb) varied from 0.00% to 2.11%, and within-laboratory imprecision (CVwl) ranged from 0.96% to 3.56%. Similarly, for ApoB, CVr ranged from 0.85% to 2.61%, CVb ranged from 0.00% to 2.12%, and CVwl for ApoB ranged from 1.25% to 3.28. For AKPAL, the LoB (CLSI EP17-A2) determined on IgG/Albuman matrix was 0.000 g/L, while for FIIPS, it was 0.0007 g/L. For LoD, samples were measured (n=30) over four test days. The LoD was 0.001 g/L AKPAL, and for FIIPS, it was 0.002 g/L.
DISCUSSION and CONCLUSION:
The IFCC WG APO MS has developed a LC-MRM-MS-based RMP for serum ApoAI and ApoB, a key component of the envisioned Apo RMS. The RMP meets ISO and CLSI standards, showing strong linearity, low imprecision, negligible carryover, and clinically insignificant LoB and LoD. The MS-based RMP allows molecular characterization of the proteins and guarantees analytical selectivity. Its adoption by IVD-manufacturers as the holy grail for future ApoAI and apoB standardization of commercial kits supports global standardization. The RMP will serve as the higher-order RMP in IFCC-endorsed calibration labs, certifying IVD manufacturers to ensure ApoAI and ApoB traceability within allowable uncertainty.
|
= Discovery stage. (53.14%, 2025)
= Translation stage. (22.33%, 2025)
= Clinically available. (24.53%, 2025)
