BACKGROUND: Clinical laboratories performing semi-quantitative autoantibody testing face challenges in developing quality systems to assure performance. Careful consideration is required to evaluate assay imprecision, including quality control (QC) matrix, autoantibody titer targets, and performance goals. The objective of this study was to evaluate the performance of a novel patient-based QC solution relative to vendor-provided QC for nine autoantibodies over a four-year period at a tertiary pediatric care centre. METHODS: Internal QC data were extracted over a 4-year period for nine autoantibodies measured via chemiluminescent immunoassays (anti-double stranded DNA (dsDNA), anti-ribonuclear protein, anti-Ro52, anti-Ro60, anti-La, anti-Smith, anti-proteinase 3, anti-myeloperoxidase, and anti-tissue transglutaminase IgA). QC evaluated during study period included vendor-based QC and third-party patient-based QC using defibrinated plasma specimens from autoantibody positive single donors. Lot-specific QC coefficients of variation (CV), titer means, and standard deviations were calculated and compared across QC matrices. RESULTS: Approximately 500 QC values per autoantibody were evaluated. Mean CV across lots ranged from 8.2 to 14.4% for negative vendor-based QC, 8.2-14.5% for positive vendor-based QC, and 9.8-17.8% for positive patient-based QC across evaluated autoantibodies. Imprecision estimates were similar between positive vendor and patient-based QC for all autoantibodies with the exception of anti-dsDNA (vendor: 4.8-13.2%, patient-based: 14.3-21.6%) and anti-Ro52 (vendor: 6.9-12.4%
patient-based: 9.7-18.9%). CONCLUSIONS: This study characterizes imprecision across different QC matrices for nine autoantibodies measured semi-quantitatively using chemiluminescent immunoassays. Results may serve as a benchmark tool to assess imprecision goals for commonly measured autoantibodies in clinical laboratories and as a framework for utilizing third-party patient-based QC solutions.