AI, machine learning to speed antibody therapy development

The UNC Eshelman School of Pharmacy and the Lampe Joint Department of Biomedical Engineering have received up to $5.6 million from the Computational ADME-Tox and Physiology Analysis for Safer Therapeutics program, through the federal Advanced Research Projects Agency for Health. CATALYST funds will be used to develop artificial intelligence tools that can predict how antibody-based therapies behave in the human body.

The Antibody Intelligence for Modeling of PharmAcokinetics and Toxicity in Humans project is called AIM-PATH. Its leaders are associate professors Yanguang “Carter” Cao and William Polacheck. Cao works in the pharmacy school’s pharmacotherapy and experimental therapeutics division. Polacheck works in the Carolina and NC State University joint biomedical engineering department and in the UNC School of Medicine’s cell biology and physiology department.

The award supports research aimed at improving how scientists model the distribution, activity and potential toxicity of antibody therapeutics — drugs that have transformed treatment for cancer, autoimmune diseases and other serious conditions.

“Antibody drugs are among the most powerful therapies available today, but predicting how they move through the body and interact with tissues remains a major scientific challenge,” said Cao. “We still rely heavily on animal studies, particularly in nonhuman primates, to test antibody drug candidates. Our goal is to combine AI/ML tools and biological data to create models that can more accurately simulate these complex processes, ultimately supporting animal-free Investigational New Drug applications and guiding the design of safer, more effective treatments.”

The AIM-PATH team will integrate machine learning, computational pharmacokinetic modeling, and in vitro tissue-chip systems to predict how antibodies behave across different tissues and patient populations. By using human cell-based systems and computational modeling instead of traditional animal studies, the team aims to replace many preclinical animal tests with approaches more relevant to humans. This strategy is expected to significantly reduce the cost and time required to bring new antibody therapies to market.

“The unique strength of AIM-PATH is how it bridges data-driven models with experimental systems,” said Polacheck. “By uniting computational predictions with our micro physiological tissue models, we can capture a much clearer picture of antibody behavior before clinical testing.”

For Carolina, the award highlights growing leadership in using AI to accelerate drug development.

“This work has the potential to fundamentally change how biologic drugs are developed,” said Cao. “By reducing reliance on animal testing and improving predictive accuracy, we can develop safer medicines faster and at lower cost — benefiting both patients and the research ecosystem.”