Researchers at the University of Chicago Medicine Comprehensive Cancer Center aim to explore ways to slow or stop the growth of tumors that don’t respond to drugs using artificial intelligence, machine learning and computing capabilities high performance found at Argonne National Laboratory.

The Comprehensive Cancer Center will receive $6 million as part of a project of up to $15 million that will leverage advanced AI/machine learning approaches to mine large data sets and discover patterns that can lead to development new treatments for drug-resistant cancer. Funding is provided by the Advanced Research Projects Agency for Health (ARPA-H), an agency of the U.S. Department of Health and Human Services created in 2022 to accelerate transformative biomedical research.

“The drug discovery process is long, inefficient and expensive, with the majority of new drugs failing in clinical trials,” said Kunle Odunsi, AbbVie Foundation Distinguished Professor of Obstetrics and Gynecology at the University of Chicago; Director of the Comprehensive Cancer Center; and dean of the Department of Oncology in the Division of Biological Sciences. “Cancer patients don’t have time to wait for new treatments. There is therefore an urgent need to reduce the time to drug discovery and we aim to achieve this through new, synergistic approaches that leverage Argonne’s supercomputing capabilities and its strengths in chemistry and technology. cancer biology at the University of Chicago.

Cancer drug discovery is a complex and resource-intensive process, typically taking up to 15 years and more than $2 billion to move a drug from target discovery to FDA approval . Furthermore, although more than 4,500 human proteins are estimated to be druggable, less than 10% of them are currently targeted by approved drugs. For example, it is estimated that only 14% to 28% of patients with gynecologic cancers could be assigned to medications based on the molecular profile of the tumor, meaning that many patients do not have many effective treatment options.

The joint project, entitled “Integrated AI and experimental approaches for targeting Iintrinsically DisorderedProtEins in design Acancer Ligands” (IDEAL), will use cutting-edge technology and experimental approaches to narrow down research to the most promising compounds that can be translated into better treatments.

The UChicago Medicine team will be led by Odunsi as co-principal investigator and includes Christopher Weber, Savas Tay and Bryan Dickinson. The Argonne team will be led by Dan Schabacker; Rick Stevens, Arvind Ramanathan and Thomas Brettin, with support from Andrzej Joachimiak.

“With Argonne’s cutting-edge AI expertise and the University of Chicago’s exceptional cancer research capabilities, we are uniquely positioned to solve complex cancer science challenges, one of the most pressing health problems,” said co-researcher Thomas Brettin, strategic programs manager in Argonne’s Computing, Environment and Life Sciences department.

Researchers will utilize Argonne’s unprecedented computational and experimental facilities: the Argonne Leadership Computing Facility’s Aurora exascale supercomputer and the ultra-bright X-rays of Argonne’s Advanced Photon Source. These technologies will allow researchers to screen billions of possible molecules (including all currently available drugs) in hours and simulate thousands of complexes in days.

The IDEAL team will test this new model on targets known to be relevant in ovarian cancer, the deadliest of gynecological cancers and notoriously resistant to treatment. Although the pilot focuses on ovarian cancer, the accelerated pipeline is intended to be applicable to any target for any cancer type.

“This project brings together exceptional computing capabilities, the best resources in structural biology, a world-class cancer center and some of the best scientists,” Odunsi said. “I believe this ‘dream team’ has the potential to revolutionize the cancer drug discovery timeline and shift the paradigm for patients who currently have a poor prognosis and little hope of a cure.

Adapted from an article first published by UChicago Medicine Comprehensive Cancer Center.