Optimizing the Aggregation Propensity of Therapeutic Monoclonal Antibodies against Cancer and Autoimmune Diseases: A Computational Study

Abstract

Monoclonal antibodies (mAbs) represent the most promising and rapidly growing class of therapeutic compounds for treating a wide variety of human chronic and acute diseases.  Despite their benefits, a major drawback in the exploitation of antibodies is their tendency to form aggregates.  As a result, severe immunological reactions in patients have been recorded. In this study, we investigated the susceptibility of a set of therapeutic monoclonal antibodies to form aggregates. We selected antibodies that have all been approved by the U. S. Food and Drug Administration and are indicated for the treatment of various types of cancer and autoimmune diseases. The AMYLPRED 2 consensus method was used to predict ‘aggregation-prone’ regions on the surface of these proteins. These regions are conserved and observed in almost all monoclonal antibodies commercially available. Considering the amino acid sequences of these antibodies, common groups of ‘aggregation-prone’ regions were identified, called clusters. We successfully reduced or even fully eliminated ‘aggregation-prone’ groups (clusters) by specific ‘mutations’ of the amino acids with exposed side chains. This information may be useful in future studies of monoclonal antibodies by improving existing therapeutic products or by designing novel ones.