(T1030-01-01) Selectivity of Human Cytokine Quantification in Human and Cyno Serum Using a Fully Automated Microfluidic Immunoassay System

Purpose: The use of biomarkers in biotherapeutic development to monitor drug pharmacodynamics, safety, and efficacy has been steadily increasing in an FDA-supported effort to guide decision making and improve the success of new therapeutics in clinical studies. Context of use for these assays demands high performance and reliability metrics, which typically points to single-analyte assays over multiplex screening assays to enable straightforward validation and assay robustness. Cytokine quantification requires discrimination the analyte of interest in biological matrices containing other biomolecules. In this poster, the detection of human IL-4, IL-6, IL-10, IFN-gamma, and TNF-alpha in single-analyte assays using an automated microfluidic immunoassay platform is evaluated. Human and cynomolgus monkey (cyno)serum samples containing all 5 cytokines to evaluate the selectivity and robustness of the assays in the presence of the other cytokines were used to determine assay ranges, spike recovery, and dilutional linearity to assess the assay performance for each of the 5 cytokine assays.

Methods: Cytokine analysis in an automated, CD-based format was performed on the Gyrolab microfluidic platform with commercially available Gyrolab cytokine kit reagents. Recombinant human cytokines IL-4, IL-6, IL-10, IFN-gamma, and TNF-alpha were spiked into human and cyno serum samples with low endogenous cytokine levels. Three-step sandwich immunoassays with biotinylated anti-human cytokine monoclonal antibody capture reagent and an Alexa Fluor® 647 labeled anti-human cytokine monoclonal antibody detection reagent with the Gyrolab Bioaffy 4000 CD. Each cytokine assay was run on a separate CD in a continuous manner.

Results: The standard curves for the Gyrolab human cytokine assays for IL-4, IL-6, IL-10, IFN-gamma, and TNF-alpha show a 3-log dynamic range and low- or sub-pg/mL levels. Excellent spike recovery and dilutional linearity were shown for the experimental samples prepared as a mixture of all 5 cytokines for all the cytokine assays. One sample, cyno serum sample 2 (IL-4 quantification with spike mix 4), showed poor recovery of 54%, while the other samples in the same dilution series showed complete recovery (80-130%). One data set, quantification of IL-4 in spike combination 2, produced recovery >120%. Excluding the data point above, all other data across all assays showed a maximum of 129% recovery and minimum of 80%, with a mean recovery of 102% and a standard deviation (SD) of 10. This mean and SD recovery demonstrates excellent matrix tolerance and assay robustness. The assay results were exceptionally precise, with a mean %CV of 3.9, SD 3.9. Except for one %CV of 21.5, all other precision data was 17.5% or below.

Conclusion: The automated microfluidic cytokine assays demonstrated high selectivity and robustness, demonstrating the suitability of these reagents and platform for cytokine quantification.


Assay performance for quantification of 5 human cytokines in human serum using an automated microfluidic immunoassay platform


Dilutional linearity for 5 human cytokine assays in human and cyno serum


Spike recovery and precision for 5 human cytokine assays using a microfluidic immunoassay platform

Acknowledgements: Note: this is new data that will be published in an application note soon. But as of the abstract submission date it's not been published anywhere.