Heavy chain-only antibodies (HCAbs) are an emerging therapeutic modality. Traditional HCAb discovery relies on animal immunization, which can produce candidates with developability problems requiring further optimization. To address this, we have developed a novel human, synthetic HCAb discovery platform in yeast. This poster describes the design principles of the platform, the characterization of its output, and how it is applied to generate various multispecific antibody formats. 

Approach and outcomes 

• The synthetic HCAb library was designed around two complementary objectives: naïve libraries prioritized diversity in the CDR H3 region, while optimization libraries focused diversity within CDR H1 and CDR H2. Machine learning augmented the library design, with the goal of reducing predicted hydrophobicity, polyspecificity, and chemical stability liabilities.

• The full workflow, from FACS selections through production, purification, and characterization, delivers HCAbs with the desired specificity profile within 4–5 months of selection start. Adimab's synthetic HCAb libraries have now been applied across more than 20 therapeutic programs. 

• HCAbs delivered across multiple campaigns and targets consistently show high affinity and favorable developability, including low polyspecificity, good HIC retention times, and high melting temperatures. 

• Adimab-developed chain pairing solutions, including chain exchange (ChEx), enabled generation of VHH-based bispecifics and trispecifics across multiple campaigns. These molecules showed high purity by size exclusion chromatography (SEC) across a range of architectures, from 1+1 bispecific to 1+1+1 trispecific formats. 

Why it matters 

Traditional HCAb discovery through animal immunization introduces developability risk at the outset, requiring additional optimization before candidates can advance. Adimab's synthetic, yeast-based platform addresses this by designing libraries from the start with developability in mind, producing HCAbs with high affinity, low polyspecificity, and favorable biophysical properties across multiple targets. The platform's compatibility with ChEx and other chain pairing solutions enables efficient generation of multispecific formats with flexible architectures and high purity, supporting scalable discovery of both monospecific and multispecific antibody therapeutics.