How AI Can Predict Pancreatic Cancer Risk 3 Years Before Diagnosis

While any cancer diagnosis is devastating, pancreatic cancer is among the most difficult. Because it's hard to detect in its early stages, it has the lowest survival rate of any cancer1. In 80% of cases, patients are diagnosed at stage 41 and given one year to live. More and more families are now being dealt this blow2—and researchers estimate over 64,000 Americans will be diagnosed with pancreatic cancer this year.

For a study published earlier this month, researchers at Harvard Medical School (HMS) and the University of Copenhagen gave an AI tool access to 9 million patient records3 across the Danish medical system and U.S. VA hospitals. They trained to tool to read diagnostic codes and identify patterns between cancer diagnosis and other preexisting diagnoses. Then, it gave the tool a new set of medical records and asked it to identify each patient's risk of pancreatic cancer within three months, six months, one year, two years, and three years.

When assessing short-term risk, the tool flagged more obvious diagnostic codes, like unspecified jaundice, diseases of biliary tract, abdominal and pelvic pain, weight loss, and neoplasms of digestive organs, which researchers say could actually be symptoms of already present cancer. But when asked to assess long-term risk, the tool identified diagnoses that are not directly related, like Type 2 and insulin-independent diabetes.

Researchers believe the tool is more accurate than current population-wide estimates and at least as accurate as genetic testing, which is currently given only to those already identified as high risk.

Our current system relies on individual doctors to refer patients for a series of expensive and invasive tests (think a CT scan, MRI, biopsy, or endoscopic ultrasound). People usually undergo testing because of a family history of genetic mutations or symptoms that appear only once the disease has progressed.

This new tool provides a relatively inexpensive (and entirely noninvasive) way for patient files to be automatically flagged as high-risk, providing the opportunity for patients—some of whom may not have been identified as high-risk by their doctors—to be tested. Chris Sander, Ph.D., a faculty member in the Department of Systems Biology in the Blavatnik Institute at HMS, said in a news release that such a system could lead to earlier detection and treatment and ultimately, lengthen patients' lives.