Performance evaluation of human cytokines profiles obtained by various multiplexed-based technologies underlines a need for standardization

Background: Multiplexed methods permit simultaneous quantification of multiple cytokines. As several manufacturers offer reagents to quantify the same cytokines on a single instrument, comparison of the distribution should be made to determine whether these data are comparable from one assay to another. Methods: We performed the quantification of cytokines in serum samples with three commercially available assays: Cytometric Bead Array (CBA), Protein Biochip Array Technology (PBAT), and Luminex Technology analysis. Using detection limit and reference range of the three commercial multiplex technologies, we evaluated: 1) the overall distribution of cytokines; and 2) the clinical impact. Results: The three cytokines, IL-1 beta, IL-1 alpha and IL-4, cannot be measured by these methods because of the high number of non-detected data (>50%). By contrast, four cytokines as IL-8, VEGF, MCP-1 and EGF exhibited a low percentage of non-detected data whatever method was used. The comparison of the percentage of samples with values higher than the respective reference range of each method reported an absence of clinical concordance (Cohen's kappa-test <0.40). Conclusions: Our results highlight the lack of transferability between the three commercially available multiplex methods evaluated (CBA, PBAT and Luminex Technology). Analytical performances are adequate for longitudinal studies using a same methodology but caution should be used for comparisons between results obtained with different methods underlying a need for standardization.