Signal- and Bead-Dependent Inter-Operator Variability in Single-Center Luminex-Based HLA Antibody Testing

Abstract

Background: Luminex-based single antigen bead assays are widely used for HLA antibody assessment in transplantation, but their semi-quantitative fluorescence outputs may be affected by technical, signal-dependent, and bead-specific factors. This single-center study evaluated inter-operator variability under routine laboratory conditions. Methods: Twenty serum samples were tested for each assay format: Class I SAB, Class II SAB, and screening. For each assay type, the same serum set was independently tested by four trained operators using the same Luminex FlexMap 3D instrument and xPONENT software version 4.3. Sera were EDTA treated, and exported trimmed mean fluorescence intensity values were analyzed at the sample–bead level. Variability was assessed using log-transformed inter-operator spread, threshold-based discordance, variance decomposition, direct item-level operator deviations, and bead-level variability profiling. Results: Overall variability was limited, with median item-level log-range values of 0.185, 0.172, and 0.181 in Class I, Class II, and screening assays, respectively. The analytical item accounted for most total variance. However, global discordance of 1.84–5.57% increased markedly near fixed thresholds, reaching 58.7–88.9% in Class I, 26.6–96.2% in Class II, and 57.1–90.9% in screening assays. Variability was signal dependent and unevenly distributed across beads. Conclusions: Luminex-based HLA antibody assays were globally reproducible but showed localized, signal- and bead-dependent vulnerability, particularly in threshold-sensitive regions.

Keywords