Diagnosis and Treatment, Featured, Pancreatic Cancer, Research and Clinical Trials

Beyond the Code: Advancing early pancreatic cancer diagnoses

By Nicole Ferrara • April 15, 2025

Estimated reading time: 4 minutes

By Mayo Clinic staff

Pancreatic cancer is one of the most challenging cancers to treat. "If you think of all the known cancers, it is the most deadly, the least survivable of all cancers that we know of. And that's what makes it such an awful disease to deal with," says Mark Truty, M.D., a Mayo Clinic surgical oncologist.

Michelle Williamson, a Mayo Clinic physician assistant, says pancreatic cancer takes an emotional toll on both patients and care teams. "They understand that they do not have a very good chance of beating this disease. So, we're trying to give them hope," she says.

In this video from BBC StoryWorks' online series "Beyond the Code," Dr. Truty and other Mayo Clinic healthcare professionals share insights from their experiences caring for people with pancreatic cancer and their research. They highlight the importance of early diagnosis and the role of artificial intelligence (AI) in transforming patient outcomes:

Pancreatic cancer is often diagnosed at a late stage.

Pancreatic cancer treatment depends on the stage of the cancer, its location and a person's overall health and preferences. It may include surgery, radiation, chemotherapy or a combination of these. When the cancer is advanced, treatment may focus on relieving symptoms and maintaining quality of life for as long as possible.

When Dr. Truty was in college, his father was diagnosed with the disease, which quickly spread to his liver and abdomen. "Ultimately, he died in my arms. This is something that, unfortunately, continues to happen on a daily basis in the United States and globally. That's probably my inner drive; I don't want this to happen to anyone else," he says.

Pancreatic cancer is often diagnosed at a late stage when the cancer has spread beyond the pancreas. Dr. Truty says he suspects even patients diagnosed with what appear to be earlier stages of the disease likely already have cancer spread. "If we follow them long term, we eventually see things show up, meaning it's already there. It's just too small to see with current technology. If we can identify a patient with pancreas cancer at a much earlier stage, those patients are much more likely to have a curative procedure for long-term cure," he says.

Finding pancreatic cancer earlier with AI.

Along with Dr. Truty, Ajit Goenka, M.D., a radiologist and nuclear medicine specialist, is developing an AI tool to find pancreatic cancer in more people at an earlier stage. "AI is really good at quantifying those very subtle changes that happen on the images that human beings cannot pick up due to the inherent limitations of their eyesight. We want AI to help experts overcome those limitations so that we can have the best combination of machine and expert," says Dr. Goenka.

"We started looking at patients who were seen here with pancreas cancer — who had a previous scan within one to two or three years before their diagnosis — to see if we could leverage AI technology to identify that lesion prior to being aware of it," says Dr. Truty.

Dr. Goenka says the AI models the Mayo Clinic team has trained can review a patient scan for cancer in a fraction of a second, with more accuracy than a trained healthcare professional. "What we have seen is that AI can help us diagnose pancreas cancer almost a year before its clinical presentation," he says.

Earlier diagnosis means treatment will be more successful for more people with pancreatic cancer.

AI offers hope for other cancers, too.

Dr. Truty says this type of AI technology could also be used to diagnose other cancers, allowing care teams to screen higher-risk patients to detect cancer earlier and intervene with treatment before symptoms appear. "We can take survivals that are currently measured in months to only a couple of years to — hopefully — many years and decades and ultimately a cure."