UI Health receives two research grants to focus on early detection of lung cancer

Lung cancer is the leading cause of cancer death in the United States, claiming more lives than breast, prostate and colon cancers combined. At UI Health, a team of physicians, nurses and scientists is using artificial intelligence and molecular science to catch lung cancer earlier and better understand who is most at risk.

“At the end of the day, this is about detecting lung cancer early enough to make a difference,” said Dr. Frank Weinberg, who leads the Thoracic Oncology Program at UI Health and a member of the University of Illinois Cancer Center. “At UI Health, we are working as a team to move beyond traditional screening — to predict, prevent and ultimately reduce the number of people who ever develop advanced lung cancer disease.”

A collaborative approach to early detection

The UI Health Lung Cancer Program brings together experts from across disciplines, including Dr. Kevin Kovitz, director of interventional pulmonology; nurse practitioner Mary Pasquinelli, Doctor of Nursing Practice and director of the Lung Cancer Screening Program; and Weinberg, alongside thoracic surgeons, radiation oncologists and population health scientists. Together, they are building a comprehensive, team-based model of care that bridges the clinic, lab and community.

“Forty years ago, we could do very little for lung cancer,” said Kovitz. “The evolution has been phenomenal. The minimally invasive technologies we have to diagnose and to stage people have evolved tremendously as well. If we detect lung cancer early, we will continue to save lives.”

Harnessing AI to solve a critical unknown

UI Health recently received two major research grants to expand this work. Both grants are focused on improving early detection and refining who is at highest risk for lung cancer.

The first grant, $3 million from pharmaceutical company AstraZeneca, was received by UI Health and its partners in the Sybil Implementation Consortium – Massachusetts General Hospital, Baptist Memorial Health Care in Tennessee and WellStar Health System in Georgia. Each of the four organizations was awarded $750,000.

Sybil is an AI model developed by MIT and Massachusetts General Hospital that can predict an individual’s risk of developing lung cancer up to six years after a single low-dose CT scan.

Previously, the Sybil AI model had been tested by the Consortium in mostly white populations. However, more recently, the team at UI Health demonstrated Sybil also performs equally well in racially and socioeconomically diverse populations who have an increased risk for developing lung cancer. In September, Dr. Pasquinelli presented this work at the World Conference on Lung Cancer in Barcelona.

While Sybil has shown strong predictive performance in standard lung cancer screening populations, the new AstraZeneca-funded “Resolve Study” will take that research further by testing whether Sybil can help doctors interpret incidental pulmonary nodules, small spots on the lungs that are often discovered by chance during scans for other medical reasons. These spots can become cancerous.

“By working together across institutions, we are proving that these tools can benefit everyone, not just a subset of patients,” said Pasquinelli, who is also a member of the University of Illinois Cancer Center. “Our mission is to make sure that innovation in lung-cancer screening reaches the populations most at risk.”

Precision screening and prevention

With a second grant, $1 million from pharmaceutical company Lilly, UI Health will develop a precision-based lung cancer screening program that integrates AI, population health and molecular science. The project will create a scalable, community-informed model of screening — one that will lead to more equitable care and can be replicated nationally. The initiative has three aims.

The first aim is to expand access to lung cancer screening so more people who may be at risk can be screened, not just those who fit the criteria that are the national norm. Using what is called the Potter criteria, the new approach looks at anyone with a long history of tobacco use, even if they quit smoking many years ago. This ensures that people who may still be at risk can be screened early, when lung cancer is most treatable.

Pasquinelli emphasizes that lung cancer screening should not just be for people with a history of heavy smoking. “If you have lungs, you are at risk,” she said. “Half of people who get lung cancer today would not even qualify for current screening guidelines. Our goal is to change that.”

Second, UI Health will incorporate Sybil risk scores into real-world practice to better identify individuals who may develop lung cancer and evaluate how AI can be integrated safely and equitably into care.

“If Sybil predicts that an individual is likely to develop lung cancer after a CT scan — essentially pre-diagnosing the disease — then we could potentially intervene with medication to prevent the cancer from developing,” said Pasquinelli.

Lastly, UI Health will integrate molecular blood-based immune and metabolic biomarkers from Weinberg’s laboratory in the University of Illinois Cancer Center to link biomarkers of inflammation and metabolism with cancer risk. These metabolic and immune biomarkers are unique because they reflect the body’s earliest biological “warning signs”— subtle shifts in how cells process energy and respond to inflammation potentially long before a tumor becomes visible on a scan. The identification of these biomarkers will potentially pave the way for future preventative treatments.

“If we can connect what we see on a CT scan with what’s happening at the molecular level, we can identify who’s most at risk and intervene even earlier,” said Weinberg. “That’s how we move from detection to prevention. This is team science in action. We’re uniting data scientists, clinicians and community partners with one goal — to detect lung cancer earlier and save more lives.”