A molecular immunology approach to antibody humanization and functional optimization

Area of Science:

• Immunology
• Biotechnology
• Computational Biology

Background:

• Antibody humanization is crucial for reducing immunogenicity in therapeutic antibodies.
• Current methods like CDR grafting have limitations in achieving optimal humanness and retaining binding affinity.

Purpose of the Study:

• To introduce a novel antibody humanization method based on human string content (HSC).
• To demonstrate improved immunogenicity and antigen binding compared to traditional methods.

Main Methods:

• Developed a new metric, human string content (HSC), to quantify antibody humanness at the epitope level.
• Utilized HSC as an optimization goal to sample diverse human germline sequences.
• Applied the method to humanize four antibodies with varying specificities.

Main Results:

• Generated antibody variable domains that are immunologically more human than those from CDR grafting.
• Achieved superior or comparable antigen binding affinity without affinity maturation.
• Demonstrated that humanness is derived from multiple discrete germline sequences.

Conclusions:

• The HSC-based humanization method offers a more refined approach to antibody engineering.
• This method enhances antibody properties beyond immunogenicity, including antigen affinity and stability.
• The approach allows for the generation of multiple, diverse, and highly humanized antibody candidates.