The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to provide antibody engineers with a user-friendly computational platform to assist in the design and engineering of therapeutic antibodies. Antibody-based drugs are becoming the treatment of choice for diverse diseases such as multiple sclerosis, rheumatoid arthritis, and several types of cancers. The success of therapeutic antibodies with over forty approved or in review in the United States and Europe is due to their exquisite specificity, high potency, stability, solubility, clinical tolerability and relatively inexpensive manufacturing process in comparison with other biologic drugs. Thus, with a market forecast of over $50 billion in sales for 2015, the interest in developing faster and more efficient methods for discovering and optimizing antibodies for therapeutic applications has been gaining momentum. A user-friendly computational platform that generates more predictable and robust antibody designs should significantly reduce the attrition rate in late development, and result in significant cost savings.
This SBIR Phase I project proposes the development of an automated antibody humanization platform. Humanization methods are key to engineering nonhuman antibodies for human therapy, and their development has been driven by three main goals: (1) Increase human content to minimize immunogenic reactions; (2) preserve binding profile to retain potency and minimize costs in further engineering processes, e.g., affinity maturation; and (3) secure intellectual property. The proposed platform aims to maximize all three of these goals as follows. First, knowledge-based rules will be used to select human germline genes as framework region (FR) donors for nonhuman Complementary-Determining Regions (CDRs) grafting. It is expected that the selected FRs preserve binding and facilitate further development of the therapeutic antibody. Second, the CDRs will be humanized. CDR grafting protocols typically do not modify the CDRs; however, based on work published by the PI, it is anticipated that the number of nonhuman residues in the CDRs can be reduced significantly without impacting binding. If so, the final product will be indistinguishable from human antibodies isolated from transgenic mice or phage display libraries. As proof of concept, the biochemical and biophysical profiles of designs generated by our computational platform will be assessed experimentally.
