Dr. Eli Larson
Affiliation: Senior Scientist, Merck, NJ
Talk Title: Multifactor Modeling of Biotherapeutic Separations to Enable In Silico Robustness Assessment
Bio: Eli began studying separations while completing his undergraduate work at Gustavus Adolphus College in the lab of Professor Dwight Stoll, studying the impact of the dimensional interface in two-dimensional liquid chromatography systems. He completed his doctorate at the University of Wisconsin-Madison under the mentorship of Professor Ying Ge, developing new front-end separations for top-down proteomics with emphasis in antibody-drug conjugate characterization by Fourier transform ion cyclotron resonance mass spectrometry and trapped ion mobility spectrometry. In 2023, Eli joined the Analytical Enabling Capabilities group at Merck as a Senior Scientist based in in Rahway, NJ, where he focuses on development and deployment of new assays for biotherapeutics and vaccines.
Abstract: Ensuring chromatographic method robustness is critical for method validation and deployment in a QC environment. Traditionally, robustness assessment is conducted via extensive design-of-experiments (DoE) studies using parameters known to impact the assay. Determining the appropriate range of input values for traditional DoE is challenging, and it is often difficult to ensure success during the first set of experiments. In silico modeling has been extensively used to map the separation landscape of multicomponent mixtures and can be leveraged to optimize methods and assess robustness as a function of various factors, while greatly reducing the experimental burden relative to DoE approaches. Although this strategy has been demonstrated with great success for small molecules, its utility for biologics has been very limited until recently due to the complexity of biotherapeutic separations. Here, we introduce a new approach to enable three-dimensional (3D) in silico method robustness assessment using the ACD/Labs LC Simulator to model a hydrophobic interaction chromatography (HIC) method for determining the drug-to-antibody ratio (DAR) of a high DAR antibody–drug conjugate. We further extend this approach to create a 3D robustness model of an ion exchange chromatography (IEX) method for measuring charge variants of a monoclonal antibody. Finally, we report a new strategy to study the impact of column lot variability and enable four factor visualization of method robustness using a custom Python script to compare 3D models. This work demonstrates the strong potential of in silico robustness studies to accelerate method validation, ensure best science first (ab initio), and increase the depth of data understanding through new data visualization approaches.