Computational models for studying physical instabilities in high concentration biotherapeutic formulations

Abstract

Computational prediction of the behavior of concentrated protein solutions is particularly advantageous in early development stages of biotherapeutics when material availability is limited and a large set of formulation conditions needs to be explored. This review provides an overview of the different computational paradigms that have been successfully used in modeling undesirable physical behaviors of protein solutions with a particular emphasis on high-concentration drug formulations. This includes models ranging from all-atom simulations, coarse-grained representations to macro-scale mathematical descriptions used to study physical instability phenomena of protein solutions such as aggregation, elevated viscosity, and phase separation. These models are compared and summarized in the context of the physical processes and their underlying assumptions and limitations. A detailed analysis is also given for identifying protein interaction processes that are explicitly or implicitly considered in the different modeling approaches and particularly their relations to various formulation parameters. Lastly, many of the shortcomings of existing computational models are discussed, providing perspectives and possible directions toward an efficient computational framework for designing effective protein formulations.

Characterizing Experimental Monoclonal Antibody Interactions and Clustering Using a Coarse-Grained Simulation Library and a Viscosity Model

“…Molecular simulations for estimating S ( q ) across relevant length scales at a high concentration involve thousands of mAbs, which is computationally infeasible for models of PPI that reflect atomistic detail . This limitation has been addressed by adopting models that represent PPI in a coarse-grained (CG) manner. − Commonly, CG model PPI assume that each mAb can be described by a small number of connected beads spatially arranged to mimic the protein’s three-dimensional shape.…”

“…Molecular simulations for estimating S ( q ) across relevant length scales at a high concentration involve thousands of mAbs, which is computationally infeasible for models of PPI that reflect atomistic detail . This limitation has been addressed by adopting models that represent PPI in a coarse-grained (CG) manner. − Commonly, CG model PPI assume that each mAb can be described by a small number of connected beads spatially arranged to mimic the protein’s three-dimensional shape. Each bead represents a collection of amino acids and has its own set of lumped interaction parameters. ,,− ,, One implementation assigns the same tunable short-range (van der Waals-like) attraction parameter to all beads, augmented by bead-specific electrostatic interactions to reflect amino acid partial charges.…”

“…This limitation has been addressed by adopting models that represent PPI in a coarse-grained (CG) manner. − Commonly, CG model PPI assume that each mAb can be described by a small number of connected beads spatially arranged to mimic the protein’s three-dimensional shape. Each bead represents a collection of amino acids and has its own set of lumped interaction parameters. ,,− ,, One implementation assigns the same tunable short-range (van der Waals-like) attraction parameter to all beads, augmented by bead-specific electrostatic interactions to reflect amino acid partial charges. Using such an approach, interaction parameters for a six-bead and a 12-bead model were chosen to match low-concentration B 22 data, and the resulting simulations accurately predicted high-concentration S (0) for net repulsive and weakly attractive mAbs .…”

SUMO –In SilicoSequence Assessment Using Multiple Optimization Parameters

“…In analogy to Lipinski’s rule of five to prioritize the selection of small molecules for entry into clinical development [9], several metrices have been implemented for an in silico developability assessment of antibody sequences [10, 11]. One of the first structure-based approaches is the “spatial aggregation propensity” (SAP) [12] score, that was later combined with the antibodies net charge into the Developability Index [13].…”

Predicting Experimental B22 Values and the Effects of Histidine Charge States for Monoclonal Antibodies Using Coarse-Grained Molecular Simulations

“…Coarse-grained modeling has been used in several works to predict protein–protein interactions from low to high concentrations. − A recently developed coarse-grained (CG) model termed 1bC/D (one-bead-per-charged-site-and-per-domain) was used to calculate B 22 values using a Monte Carlo (MC) algorithm that is deliberately biased to improve the efficiency and sampling of relevant molecular configurations . The simulation method uses the Mayer sampling with overlap sampling (MSOS) algorithm, which calculates virial coefficients with respect to a reference state, − which for the present work was set to be a steric-only-interactions reference state.…”