Purpose: A method for efficiently screening reagent pairs for use in ligand-binding assays on multiple platforms has been developed to increase the speed of Bioanalytical method development. As part of a new Assay Development Kit” (ADK) concept, multiple pairs of conjugated antibody reagents are screened in parallel across 3 main platforms for optimal performance with respect to sensitivity, quantitative range, and robustness.
Methods: Liquid handling for the screening runs is accomplished using the Perkin Elmer Janus and Tecan Evo platforms for speed, flexibility, and reproducibility. Users may select up to 12 combinations of (capture reagent - concentration - buffer) x (detection reagent - concentration - buffer) per plate, and up to 3 plates per run. These combinations are tested using the Gyros, MSD (electrochemiluminescence), and chemiluminescent ELISA platforms that are employed in the group. The automation design is also amenable to buffer or concentration screening for a pair of reagents (screening DOE), or reagent concentration optimization DOE.
For each combination, a proprietary algorithm is used to estimate quantitative range, based on curve fit and tolerances for precision and minimum signal to noise. The estimated quantitative range of the method is calculated using loose, regular, and strict criteria that may be applied depending on the intended use of the method (qualified vs. validated). Additionally, runs can be simulated to represent future potential method performance, and overall ranges of performance are used in the comparison of different reagent pairs for selection in the ultimate assay conditions.
Results: This process has been applied to more than 6 programs in nonclinical and clinical development for developing PK methods over the past 2 years. An app written in R/Shiny has been developed to rapidly generate results, figures, and reports for speed and reproducibility. This has resulted in method development time decreasing from up to 50 runs down to approximately 3-10.
Conclusion: Through the implementation of this process to rapidly evaluate, screen, and optimize assay platform, as well as model future method performance, turnaround time has decreased dramatically and ruggedness has improved, with very minimal need for further development or troubleshooting throughout the method lifecycle. In concert with the rest of the reagent generation and characterization efforts in our department, we have established a solid foundation for ensuring the fastest path to data generation to support decision-making and delivering our medicines to patients.
Jonathan Haulenbeek– Princeton, New Jersey
Scott Robotham– Princeton, New Jersey
Adela Buzescu– Princeton, New Jersey
Billy Akinsanya– Princeton, New Jersey
Jon Peterson– Princeton, New Jersey
Renuka Pillutla– Executive Director, Bristol-Myers Squibb Company, Princeton, New Jersey