PD-1 Targeted Antibody Discovery Using AI Protein Diffusion.

Area of Science:

• Immunology
• Oncology
• Computational Biology

Background:

• Programmed cell death protein 1 (PD-1) is a critical immune checkpoint protein that suppresses T cell responses against cancer.
• Blocking PD-1 enhances the immune system's ability to clear tumor cells, making it a key therapeutic target in oncology.
• Existing anti-PD-1 therapies like nivolumab and pembrolizumab have shown success, but new therapeutics are needed.

Purpose of the Study:

• To discover novel anti-PD-1 antibody fragments using computational methods.
• To evaluate the binding potential of these computationally derived fragments through an in silico pipeline.

Main Methods:

• Utilized protein diffusion for computational derivation of antibody fragments.
• Employed a scalable in silico pipeline for evaluating predicted binding performance.
• Generated synthetic Fv structures targeting PD-1.

Main Results:

• Successfully derived multiple anti-PD-1 antibody fragments computationally.
• Identified nine synthetic Fv structures with desirable predicted binding performance.
• These structures are suitable for empirical testing of anti-PD-1 activity.

Conclusions:

• Computational protein design can effectively generate novel antibody fragments targeting immune checkpoints.
• The presented synthetic Fv structures represent promising candidates for developing new anti-PD-1 immunotherapies.
• Further empirical validation is warranted to confirm the therapeutic potential of these novel fragments.