Evaluation of highly sensitive immunoassay technologies for quantitative measurements of sub-pg/mL levels of cytokines in human serum

Journal of Immunological Methods | August 19, 2016

David Yeung et. al.
Journal of Immunological Methods
DOI: 10.1016/j.jim.2016.08.003

Download Paper

Abstract:  A comprehensive cross-platform and cross-assay evaluation using nine technology platforms and four cytokine immunoassays (IL-6, TNFα, IL-17a, IL-2) was performed by comparing assay precision, sensitivity, parallelism and data correlation between platforms. The precision was acceptable for most evaluated assays. In addition to comparing the analytical assay sensitivity using a spiked recombinant analyte in buffer, forty serum samples from both normal controls and multiple sclerosis patients were used to measure the frequency of endogenous analyte detection (FEAD) as a parameter of each assay's ability to detect the endogenous analyte. The highest FEAD measurements were observed on the Simoa™, Erenna®, Milliplex® and Imperacer® platforms. However, only Simoa and Erenna results showed a high correlation across all evaluated cytokine assays, followed by a more moderate correlation of results across platforms for the V-plex™, high sensitivity ELISA and the Ella™ IL-6 and TNFα assays. In contrast, results from the evaluated cytokine assays on the Milliplex, AMMP™ ViBE® and Imperacer platforms did not correlate to each other nor to other evaluated assays. Acceptable parallelism was observed for the Simoa, Erenna, V-plex and Ella assays but not for the Milliplex, AMMP ViBE and Imperacer assays. In conclusion, the Simoa, Erenna,V-plex and Ella platforms performed well in one or more evaluated cytokine assays. Among those, the Simoa and Erenna assays had the highest sensitivity for detection of cytokines present at sub-pg/mL levels in human serum. In addition, the cross-platform and cross-assay comparisons demonstrated that different immunoassays may yield different results, which underscores the importance of performing such comparative evaluations, especially in the absence of reliable reference standards for the quantitative assessments of biomarkers in immunoassays.