Stanford Medicine researchers secure ARPA-H contracts

Note: This story was updated Dec. 10 to include two more contracts.

Stanford Medicine researchers have been awarded five Advanced Research Projects Agency for Health contracts.

ARPA-H, an agency within the U.S. Department of Health and Human Services, was established by U.S. Congress and President Joe Biden in 2022 to make investments in breakthrough solutions with the potential to transform critical aspects of medicine and health that cannot readily be accomplished through traditional research or commercial activity.

A group led by Katherine Ferrara, PhD, will develop innovative methods to target and treat cancer. Lisa Knowlton, MD, co-leads a team advancing techniques for visualization and identifying critical structures during surgery. Artificial intelligence investigators Roxana Daneshjou, MD, PhD, and Akshay Chaudhari, PhD, are optimizing large language model medical chatbots to assist patients with tasks such as interpreting imaging results and understanding discharge instructions. Sylvia Plevritis, PhD, is heading a team that’s developing data dashboards to inform cancer treatment decisions. Ophthalmologist Jeffrey Goldberg, MD, PhD, leads research into optic nerve regeneration, with the goal of restoring vision through transplantation.

Advancing precision cancer imaging and therapy

The use of molecularly specific radiolabeled drugs to detect and treat cancer — termed “theragnostics” — is experiencing unprecedented growth. However, only a few molecular targets are addressed by the handful of existing radiopharmaceuticals, and these treatments are not widely available.

A team of researchers, led by Ferrara, division chief of the Molecular Imaging Program and professor of radiology, has been awarded a federal contract of up to $35 million to innovate precision radiopharmaceutical imaging and therapy. The Biden-Harris administration announced awardees as part of President Biden’s Cancer Moonshot Initiative.

The Micro-Radiolabeling for Imaging and Therapy research team includes scientists at Stanford Medicine; the University of California, Los Angeles; and DropletPharma Corp. The project will focus on ovarian and head and neck cancers, neither of which have screening tests although premalignant lesions are common.

In one aim, a moonshot challenge addressed by MiRIT is to use computational techniques to design molecules that bind to disease-specific changes in proteins. The hope is that the AI-driven design process will disrupt conventional drug design. “We have identified key cancer cell targets and, through collaboration with Possu Huang, PhD, assistant professor of bioengineering, have designed and radiolabeled peptides against these targets. We will now expand the portfolio of targets and ligands and seek to translate the resulting drugs through the ARPA-H program,” Ferrara said.

In addition, MiRIT is expected to reduce the costly, complex infrastructure required for personalized treatments through miniaturized radiolabeling technology. One device will be capable of producing very small doses of radiopharmaceuticals on-site, and a second device will permit parallelized labeling of dozens of compounds to facilitate ligand optimization.

Better anatomical visualization in surgery

Knowlton is the co-recipient of an up to $22.3 million award to improve intraoperative anatomy visualization and critical structure identification. Knowlton, associate professor of surgery and an acute care surgeon, will lead the five-year contract with co-principal investigator Jeremy Li, PhD, founding CEO of CisionVision, a medical device company in Mountain View, California.

“A real-time decision aid that provides surgeons with immediate critical structure anatomical labeling has the potential to save lives, reduce serious complications, improve surgical training and lower health care costs,” Knowlton said. “We are honored to collaborate with a multidisciplinary team of experts advancing the frontier of surgical imaging technology.”

Stanford Medicine will be one of three clinical sites, along with Johns Hopkins University and Endeavor Health Northshore, conducting surgical implementation and evaluation. Teams from the Massachusetts Institute of Technology and the University of Notre Dame will contribute computer science, engineering and artificial intelligence expertise.

Chatbots helping the patient experience

Daneshjou, an assistant professor of biomedical data science and of dermatology, and Chaudhari, assistant professor of radiology, were awarded a contract of up to $4.8 million to develop large language model-driven chatbots to assist in patient communication.

The goal of the project is to develop a fully automated and scalable approach for automated prompting and clinician-grade evaluation of large language model medical chatbots for patient-facing use cases. Immediate use cases include the evaluation of common chatbot to help patients understand imaging results and a chatbot to aid patients in understanding instructions after hospital discharge.

“I’m very excited that Akshay Chaudhari and I will be leading a superstar team to tackle prompt engineering and automatic evaluation of patient-facing chatbots,” Daneshjou said.

“I’m looking forward to co-leading this ARPA-H project alongside Roxana. The ARPA-H funding mechanism allows for the creation of inter-disciplinary teams that consists of biomedical researchers, computer scientists, radiologists and hospitalists, which would otherwise be challenging with other funding programs,” Chaudhari said.

Insight into cancer care

An award of up to $8.9 million to the Stanford Department of Biomedical Data Science will fund the development of AI-augmented support tools for cancer tumor boards, in which physicians review and discuss complex cases to devise treatment plans. The goal is to establish a data marketplace and to create a dashboard for clinicians so they can quickly gain insights on patients’ cases to inform treatment decisions.

“Currently, AI plays a small role in cancer tumor boards, but its use can be transformative,” said Plevritis, chair of the Department of Biomedical Data Science and the William M. Hume Professor in the School of Medicine. “With the use of AI, clinical decision-making can be enhanced by integrating and more effectively analyzing patient data from multiple sources, including the electronic health record, radiology and pathology images, and reports and molecular tests.”

The award will also facilitate collaboration between Stanford Medicine, ARPA-H and other entities to lower the barriers associated with large-scale data collection and integration and to improve data usability. The tools and approaches developed at Stanford will be shared broadly to enhance national cancer care capabilities.

A vision for restoring sight

The Byers Eye Institute at Stanford Medicine has received an award, of up to $56 million, with the goal of whole eye transplants that restore vision. The award will bring together more than 40 scientists, doctors and industry experts from around the country. Goldberg, the Blumenkranz Smead Professor and chair of ophthalmology at the institute, will serve as principal investigator.

“This group of people have been working for decades on figuring out how to promote optic nerve regeneration and retinal neuron survival in glaucoma and other blinding diseases,” Goldberg said. “That positions this group of collaborators to be the best situated to take on optic nerve regeneration and neuronal cell survival in the context of eye transplant.”

Annually, more than 70,000 people in the United States donate their eyes, allowing for life-improving and vision-saving transplants of the cornea, the clear, outermost layer of the eye. In fact, the cornea was one of the first tissues to be successfully transplanted. But those transplants don’t address the most common causes of irreversible vision loss, including glaucoma, macular degeneration and diabetic retinopathy.

While whole eye transplants are the north star of the three-phase, six-year project, the effort will undoubtedly bring with it more breakthroughs, and that is just as exciting, Goldberg said. “As we develop a series of new technologies that could be vision restorative in the many patients with glaucoma and other eye diseases, we’ll leverage all the proper channels to ensure new drugs, gene therapies and devices can be accessible to all.”