A new project seeking to transform the drug delivery process and take projects out of “the valley of death” between aspiration and accomplishment is getting help from La Jolla-based medical research institute Sanford Burnham Prebys.

The U.S. Department of Health and Human Service’s Advanced Research Projects Agency for Health, or ARPA-H, recently announced a $31.7 million multi-institution contract to fund a project called PREDICTS.

PREDICTS, which stands for Pharmacological Research and Evaluation through Digital Integration and Clinical Trial Simulation, will involve groups developing computational models to predict toxicity in humans.

The overarching mission, according to Sanford Burnham Prebys, is to trim years off the process of drug development. Dr. David Brenner, the institute’s president and chief executive, described it as “a transformational goal.”

“Success could potentially fast-track getting drugs to market, reduce the use of animals in toxicology and preclinical studies and help reduce potential liabilities earlier in the drug development process,” Brenner said in a statement.

Bringing a drug to market usually takes 10-15 years and $2 billion, according to Sanford Burnham Prebys. As a result, 90% of such efforts fail.

Steven Olson, executive director of medicinal chemistry at the institute’s Center for Therapeutics Discovery, said this effort comes as the U.S. Food and Drug Administration is moving away from stringent animal-testing requirements and embracing what it calls new approach methodologies, or NAMs, including organoids and computational modeling based on artificial intelligence.

What may appear safe in animal models may not necessarily be effective for people and in some cases can even be toxic for humans, according to the NAMs approach. Now groups are being tasked with proving the efficacy of alternatives.

“The big picture is, can we replace that?” Brenner said. “And the best way to do that is to get hard-core scientific evidence doing the computational version vs. the traditional version.”

This lays the groundwork for PREDICTS, part of a larger ARPA-H program called Computational ADME-Tox and Physiology Analysis for Safer Therapeutics, or CATALYST, which aims to create “human physiology-based computer models” that evaluate investigational drug candidates. San Francisco-based biotech company Deep Origin leads the effort.

Sanford Burnham Prebys will be part of a consortium of research institutes working to develop the first “virtual human safety platform” that can predict how small molecules and drug candidates behave in the body.

The PREDICTS platform is based on three components: a cloud repository to develop “rich datasets”; training machine learning models to identify safety liabilities; and evaluating how machine learning models stack up with others.

“As a scientist, we think it’s really exciting, but obviously it has to work,” Olson said. “If it does work, I think it could completely change how we do drug discovery. It’ll accelerate our discoveries, help us put drugs into people and help people much faster and be more accurate.” ♦

RevContent Feed