Applications and challenges in designing VHH-based bispecific antibodies: leveraging machine learning solutions

Area of Science:

• Biotechnology
• Immunology
• Computational Biology

Background:

• Bispecific antibodies require diverse binding domains for therapeutic efficacy.
• Single-domain antibodies, especially VHH, are valuable for bispecific antibody construction due to their unique properties.

Purpose of the Study:

• To review the application of VHH domains in multispecific antibody development.
• To explore challenges and advancements in creating optimized VHH-based bispecific antibodies.

Main Methods:

• Review of traditional VHH antibody development approaches.
• Integration of machine learning (ML) in VHH antibody engineering.
• Application of ML for structural prediction, lead identification, optimization, and humanization.

Main Results:

• VHH domains provide modularity and favorable biophysical properties for bispecific antibody design.
• Machine learning significantly enhances various stages of VHH antibody development.
• ML integration addresses challenges in creating high-quality, optimized multispecific antibodies.

Conclusions:

• VHH domains are versatile building blocks for advanced bispecific antibody therapeutics.
• Machine learning is crucial for overcoming hurdles in VHH antibody optimization and humanization.
• Optimized VHH-based bispecific antibodies hold significant therapeutic potential.