(T0930-06-34) Assessing Parallelism of IFN-beta and IL-29 in Human K2EDTA Plasma Post Stimulation with LPS and PHA

Purpose: Parallelism is a critical parameter that should be assessed in biomarker method development and validation. The main reason for determining parallelism in ligand binding biomarker assays is to demonstrate that the calibrator material recognizes the endogenous analyte and is suitable for determining the analyte concentration in matrix. Parallelism data also informs appropriateness of surrogate matrix, the minimum sample dilution, and matrix interference or selectivity. Parallelism experiments are usually conducted early in method development and the results are confirmed in assay validation where samples with high endogenous analyte are serially diluted and assessed for parallelism relative to the calibrator dilution curve. Identifying samples with appropriate levels of endogenous biomarker for use in these experiments can be challenging when using normal matrices containing low or even undetectable levels of certain endogenous analytes. Here, we describe a method for generating parallelism samples for two such analytes, IFN-beta and IL-29. During method development, human K2EDTA plasma was stimulated with Lipopolysaccharide (LPS) and Phytohemagglutinin (PHA) to increase the levels of these cytokines, allowing for assessment of parallelism in the assay.

Methods: Briefly, blood was collected from multiple normal healthy donors with two separate K2EDTA tubes per individual. The tubes from each donor were labeled as “unstimulated” or “stimulated”. Unstimulated tubes were subjected to the same procedure as stimulated but without the addition of LPS and PHA. The processing procedure was started within 90 minutes of blood draw where the K2EDTA whole blood was transferred to a 1mL tube. 50uL of a 1mg/mL LPS solution in PB-MAX™ Karyotyping Medium containing PHA was added to the “stimulated” tubes. The tubes were capped and mixed gently by tapping. The tubes were incubated at 37°C for 24 hours. At the end of the incubation, 30uL of 100 mM ATP was added to the stimulated tubes and incubated at 37°C for 30 minutes. After the final incubation, all tubes were centrifuged for 7 minutes at 600 x g at room temperature. Plasma supernatant was collected from all tubes, aliquoted into labeled tubes, and stored at -80°C until analyzed in the S-Plex IFN-beta (cat#K151ADRS) and U-Plex IL-29/IFN-λ1 (cat# K151WDK) Meso Scale Discovery assay kit and components in accordance with the method SOPs.

Results: The S-Plex IFN-beta assay was previously validated in human K2EDTA plasma with a calibration curve range of 47.85 to 196,000 fg/mL. In order to determine parallelism, 20 human K2EDTA plasma samples were screened in the assay, where all samples had no detectable level of IFN-beta in the assay (< LLOQ). After employing the stimulation method with LPS and PHA with multiple donors, the signal of IFN-beta in the stimulated samples was increased 2-7.5 fold compared to the unstimulated samples. Parallelism was further tested in the stimulated samples by diluting the plasma samples from 2-512 fold dilutions and determining the %CV across back-calculated dilutions. The %CV across dilutions was < 20% for a majority of samples tested indicating parallelism between the endogenous analyte and the calibrator material. Similarly, for IL-29, the calibration curve range of the assay was determined to be 2.08 to 8540 pg/mL in the U-Plex IL-29 assay. When 20 human K2EDTA samples were analyzed in the assay, they had results < LLOQ. After stimulation with LPS/PHA, 4 samples showed sufficient detectable levels of IL-29 and allowed for assessment of parallelism with analyzing at least 3 dilutions of the stimulated sample. The %CV across dilutions was < 30% for the four samples tested.

Conclusion: Parallelism is a fundamental parameter that is demonstrated in method development and validation for any ligand-binding biomarker assay. Parallelism is difficult to assess in cases where endogenous analyte is undetectable or below the sensitivity limits of the assay. This challenge is still present in assays that utilize high sensitivity technologies such as the MSD S-Plex which can detect femtogram concentrations of analyte such as shown here for the IFN-beta S-Plex assay. In this presentation, we have described a stimulation method for whole blood using LPS and PHA that increases the concentration of the two endogenous cytokines of interest, allowing for the evaluation of parallelism in the assays using K2EDTA plasma as the final matrix. Stimulation methods have been historically challenging in matrices with chelators such as K2EDTA. This stimulation method allows for demonstrating parallelism in assays that have extremely low concentrations of endogenous analyte. The method is advantageous to understanding parallelism and more relevant than selectivity experiments where recombinant analyte is spiked into the matrix and not a good substitute for understanding comparability of endogenous and recombinant material in the assay.

Parallelism Evaluation of Human K2EDTA Samples Post-Stimulation in IFN-beta Assay

Parallelism Evaluation of Human K2EDTA Samples Post-Stimulation in IL-29 Assay