Rational optimization of a monoclonal antibody for simultaneous improvements in its solution properties and biological activity

Abstract

Developability considerations should be integrated with lead engineering of antibody drug candidates in interest of their cost effective translations into medicines. To explore feasibility of this imperative, we have performed rational mutagenesis studies on a monoclonal antibody (MAB1) whose development was discontinued owing to manufacturability hurdles. Seven computationally designed variants of MAB1 containing single point (V44K, E59S, E59T and E59Y) and double (V44KE59S, V44KE59T and V44KE59Y) mutations in its light chain were produced in Chinese Hamster Ovary (CHO) cells and purified by using platform processes employed during commercial scale production of monoclonal antibodies. MAB1 and its variants were formulated in the same platform buffer and subjected to a battery of experiments to assess their solution behaviors, and biological activities. Five of the seven (71%) variants of MAB1 demonstrated improved biophysical attributes in multiple experimental testings. Contrary to the commonly expressed reservations about potential biological activity loss upon developability optimizations, the improvements in solution behavior of MAB1 also increased its biological activity up to ~180%. In particular, concentrate-ability and apparent solubility of V44KE59S improved to ~150% and ~160%, respectively. Its diffusion interaction parameter (kD) reduced to 28% and viscosity at ~100 mg/ml decreased to less than half of the corresponding values for MAB1. V44KE59S is also slightly more active and its transfections in CHO cells were more productive. It also degraded slower than MAB1 in three month long 25°C and 40°C formulation stability studies. These results open doors to an exciting realm of structure-based biologic drug design where developability and biological activity can be simultaneously optimized at the molecular engineering stages.

Engineering Stability, Viscosity, and Immunogenicity of Antibodies by Computational Design

“…On the other hand, a chargeneutralizing mutation of a negative surface residue was able to reduce viscosity while simultaneously maintaining conformational stability and antigen-binding capability. In another study, Kumar et al 167 have also used a homology model of the same antibody as Nichols et al 166 and have designed 7 variants based on free energy change upon mutation, as assessed by the residue-scan module in MOE2014.09, to improve the physicochemical properties. The actual improvements were experimentally verified in 5 out of 7 cases.…”

Charge and hydrophobicity are key features in sequence-trained machine learning models for predicting the biophysical properties of clinical-stage antibodies

“…The use of these in silico candidate screening techniques accelerates the biotherapeutic development process, through the identification of high value leads and new engineering targets (Shan et al, 2018), and in some cases even improving biological activity (Kumar et al, 2018), However, the development of these tools is reliant on the availability of high quality experimental datasets and is thus heavily dependent on the progress of experimental techniques. Notably, the recent release of antibody biophysical characterisation datasets (Goyon et al, 2017;Jain et al, 2017a) has allowed the development of further theoretical tools to predict, assess and understand the physicochemical properties that are correlated with the successful development of a therapeutic antibody, on a scale previously unattainable to academic researchers.…”

Validation of a de‐immunization strategy for monoclonal antibodies using cynomolgus macaque as a surrogate for human

“…Differential scanning calorimetry (DSC) was used to examine the conformational stability of all of the engineered antibodies. A decrease in thermal stability of a mAb might indicate decreased physical stability and increased propensity to aggregate (Kumar et al, ; Moise et al, ; West et al, ). Representative DSC thermograms are shown in Supplemental Figure 1.…”