AI Predicts Pancreatic Cancer Risk Up to Three Years Early

Author: Harvard Medical School
Published: 8 May 2023 - Updated: 22 Oct 2026
Publication Details: Peer-Reviewed | Research, Study, Analysis

Table of Contents:
Synopsis - Definition - Overview - FAQs - Insights, Updates - Related Content

Synopsis

This research, peer reviewed and published in Nature Medicine, describes a deep learning model developed by Harvard Medical School and the University of Copenhagen that flags people at the highest risk of pancreatic cancer up to three years before diagnosis using nothing more than their routine medical records. The model was trained on 6.2 million Danish health records covering 41 years and then tested on nearly 3 million U.S. Veterans Health Administration records, learning from the order and timing of diagnosis codes rather than from genetic tests or family history. That matters because there is currently no population-wide screening for pancreatic cancer, only targeted testing for people with known inherited risk, and most cases are found after the cancer has already spread. For older adults and people living with chronic conditions or disabilities, who often have long medical histories but may face real barriers to specialist care, a tool that works from data already on file could help doctors decide who should be sent for CT, MRI or endoscopic ultrasound first.*

At a Glance

• 1 - Pancreatitis was a strong warning sign, predicting pancreatic cancer within an even shorter two-year window.
• 2 - The pancreas is nicknamed the angry organ because it sits deep in the abdomen and is easily inflamed by biopsy.
• 3 - About 44 percent of people diagnosed at an early stage survive five years, yet only about 12 percent of cases are caught that early.
• 4 - Accuracy dropped when the Danish-trained model was applied to U.S. veterans, but improved once it was retrained on U.S. data, showing the value of local training data.

Topic Definition

AI-Based Pancreatic Cancer Risk Prediction

AI-based pancreatic cancer risk prediction is the use of machine learning models to estimate a person's likelihood of developing pancreatic cancer within a set period, typically months to a few years, by analyzing data already held in electronic health records. Instead of relying only on family history or inherited gene mutations, these models look for patterns in the type, order and timing of past diagnoses, such as gallstones, anemia, type 2 diabetes or pancreatitis, that tend to precede the disease. The output is a risk score, not a diagnosis, and it is intended to help clinicians decide which patients should be referred for more sensitive tests like CT, MRI or endoscopic ultrasound, so that a cancer known for late detection can be found at an earlier and more treatable stage.

Overview

"A Deep Learning Algorithm to Predict Risk of Pancreatic Cancer From Disease Trajectories" - Nature Medicine.

An artificial intelligence tool has successfully identified people at the highest risk for pancreatic cancer up to three years before diagnosis using solely the patients' medical records, according to new research led by investigators at Harvard Medical School and the University of Copenhagen, in collaboration with VA Boston Healthcare System, Dana-Farber Cancer Institute, and the Harvard T.H. Chan School of Public Health.

The findings, published May 8, 2023, in Nature Medicine, suggest that AI-based population screening could be valuable in finding those at elevated risk for the disease and could expedite the diagnosis of a condition found all too often at advanced stages when treatment is less effective and outcomes are dismal, the researchers said. Pancreatic cancer is one of the deadliest cancers in the world, and its toll projected to increase.

Currently, there are no population-based tools to screen broadly for pancreatic cancer. Those with a family history and certain genetic mutations that predispose them to pancreatic cancer are screened in a targeted fashion. But such targeted screenings can miss other cases that fall outside of those categories, the researchers said.

"One of the most important decisions clinicians face day to day is who is at high risk for a disease, and who would benefit from further testing, which can also mean more invasive and more expensive procedures that carry their own risks," said study co-senior investigator Chris Sander, faculty member in the Department of Systems Biology in the Blavatnik Institute at HMS. "An AI tool that can zero in on those at highest risk for pancreatic cancer who stand to benefit most from further tests could go a long way toward improving clinical decision-making."

Applied at scale, Sander added, such an approach could expedite detection of pancreatic cancer, lead to earlier treatment, and improve outcomes and prolong patients' life spans.

"Many types of cancer, especially those hard to identify and treat early, exert a disproportionate toll on patients, families and the healthcare system as a whole," said study co-senior investigator Søren Brunak, professor of disease systems biology and director of research at the Novo Nordisk Foundation Center for Protein Research at the University of Copenhagen. "AI-based screening is an opportunity to alter the trajectory of pancreatic cancer, an aggressive disease that is notoriously hard to diagnose early and treat promptly when the chances for success are highest."

In the new study, the AI algorithm was trained on two separate data sets totaling 9 million patient records from Denmark and the United States. The researchers "asked" the AI model to look for telltale signs based on the data contained in the records. Based on combinations of disease codes and the timing of their occurrence, the model was able to predict which patients are likely to develop pancreatic cancer in the future. Notably, many of the symptoms and disease codes were not directly related to or stemming from the pancreas.

The researchers tested different versions of the AI models for their ability to detect people at elevated risk for disease development within different time scales - 6 months, one year, two years, and three years. Overall, each version of the AI algorithm was substantially more accurate at predicting who would develop pancreatic cancer than current population-wide estimates of disease incidence - defined as how often a condition develops in a population over a specific period of time. The researchers said they believe the model is at least as accurate in predicting disease occurrence as are current genetic sequencing tests that are usually available only for a small subset of patients in data sets.

The Angry Organ

Screening for certain common cancers such as those of the breast, cervix, and prostate gland relies on relatively simple and highly effective techniques - a mammogram, a Pap smear, and a blood test, respectively. These screening methods have transformed outcomes for these diseases by ensuring early detection and intervention during the most treatable stages.

By comparison, pancreatic cancer is harder and more expensive to screen and test for. Physicians look mainly at family history and the presence of genetic mutations, which, while important indicators of future risk, often miss many patients. One particular advantage of the AI tool is that it could be used on any and all patients for whom health records and medical history are available, not just in those with known family history or genetic predisposition for the disease. This is especially important, the researchers add, because many patients at high risk may not even be aware of their genetic predisposition or family history.

In the absence of symptoms and without a clear indication that someone is at high risk for pancreatic cancer, clinicians may be understandably cautious to recommend more sophisticated and more expensive testing, such as CT scans, MRI or endoscopic ultrasound. When these tests are used and suspicious lesions discovered, the patient must undergo a procedure to obtain a biopsy. Positioned deep inside the abdomen, the organ is hard to access and easy to provoke and inflame. Its irritability has earned it the moniker "the angry organ."

An AI tool that identifies those at the highest risk for pancreatic cancer would ensure that clinicians test the right population, while sparing others unnecessary testing and additional procedures, the researchers said.

About 44 percent of people diagnosed in the early stages of pancreatic cancer survive five years after diagnosis, but only 12 percent of cases are diagnosed that early. The survival rate drops to 2 to 9 percent in those whose tumors have grown beyond their site of origin, researchers estimate. UPDATE Oct 20, 2026: The American Cancer Society's figures are about 44 percent for early-stage cancer, 16 percent once it has spread nearby, 3 percent once it has spread to distant organs, and 13 percent overall.

Update Oct 20, 2026: Current American Cancer Society figures put five-year relative survival at about 44 percent for early-stage pancreatic cancer, 16 percent once it has spread nearby, 3 percent once it has spread to distant organs, and 13 percent overall.

"That low survival rate is despite marked advances in surgical techniques, chemotherapy, and immunotherapy," Sander said. "So, in addition to sophisticated treatments, there is a clear need for better screening, more targeted testing, and earlier diagnosis, and this is where the AI-based approach comes in as the first critical step in this continuum."

Previous Diagnoses Portend Future Risk

For the current study, the researchers designed several versions of the AI model and trained them on the health records of 6.2 million patients from Denmark's national health system spanning 41 years. Of those patients, 23,985 developed pancreatic cancer over time. During the training, the algorithm discerned patterns indicative of future pancreatic cancer risk based on disease trajectories, that is, whether the patient had certain conditions that occurred in a certain sequence over time.

For example, diagnoses such as gallstones, anemia, type 2 diabetes, and other GI-related problems portended greater risk for pancreatic cancer within 3 years of evaluation. Less surprisingly, inflammation of the pancreas was strongly predictive of future pancreatic cancer within an even shorter time span of two years. The researchers caution that none of these diagnoses by themselves should be deemed indicative or causative of future pancreatic cancer. However, the pattern and sequence in which they occur over time offer clues for an AI-based surveillance model and could prompt physicians to monitor those at elevated risk more closely or test accordingly.

Next, the researchers tested the best performing algorithm on an entirely new set of patient records it had not previously encountered - a US Veterans Health Administration data set of nearly 3 million records spanning 21 years and containing 3,864 individuals diagnosed with pancreatic cancer. The tool's predictive accuracy was somewhat lower on the US data set. This was most likely because the US dataset was collected over a shorter time and contained a somewhat different patient population profile - the entire population of Denmark in the Danish data set versus current and former military personnel in the Veterans' Affairs data set. When the algorithm was retrained from scratch on the US dataset, its predictive accuracy improved. This, the researchers said, underscores two important points: First, ensuring that AI models are trained on high quality and rich data. Second, the need for access to large representative datasets of clinical records aggregated nationally and internationally. In the absence of such globally valid models, AI models should be trained on local health data to ensure their training reflects the idiosyncrasies of local populations.

Authorship

Co-authors on the study were Davide Placido, Bo Yuan, Jessica Hjaltelin, Chunlei Zheng, Amelie Haue, Piotr Chmura, Chen Yuan, Jihye Kim, Renato Umeton, Gregory Antell, Alexander Chowdhury, Alexandra Franz, Lauren Brais, Elizabeth Andrews, Debora Marks, Aviv Regev, Siamack Ayandeh, Mary Brophy, Nhan Do, Peter Kraft, Brian Wolpin, Michael Rosenthal, and Nathanael Fillmore.

Funding

The work was supported by the Novo Nordisk Foundation grants NNF17OC0027594 and NNF14CC0001; Stand Up to Cancer/Lustgarten Foundation grant SU2C6180; the National Institutes of Health grants U01 CA210171 and P50 CA127003; with additional support from the Pancreatic Cancer Action Network, the Noble Effort Fund, the Wexler Family Fund, Promises for Purple and the Bob Parsons Fund; the VA Cooperative Studies Program; the American Heart Association (857078); the Department of Defense/Uniformed Services University of the Health Sciences; the Hale Family Center for Pancreatic Cancer Research.

Disclosure

Brunak has ownership in Intomics A/S, Hoba Therapeutics Aps, Novo Nordisk A/S, Lundbeck A/S and ALK Abello and has managing board memberships in Proscion A/S and Intomics A/S. Wolpin has received grant funding from Celgene and Eli Lilly and consulting fees from BioLineRx, Celgene and GRAIL. Regev is a co-founder and equity holder in Celsius Therapeutics, an equity holder in Immunitas and was a scientific advisory board member of Thermo Fisher Scientific, Syros Pharmaceuticals, Neogene Therapeutics and Asimov until July 31, 2020. As of Aug. 1, 2020, Regev has been an employee of Genentech and has equity in Roche. Marks is an advisor for Dyno Therapeutics, Octant, Jura Bio, Tectonic Therapeutic and Genentech and is a co-founder of Seismic Therapeutic. Sander is on the scientific advisory board of CytoReason.

Frequently Asked Questions

NOTE: Researched FAQs by Disabled World (DW)

What are the early symptoms of pancreatic cancer?

Early pancreatic cancer often causes no symptoms at all. When symptoms do appear they can include yellowing of the skin or eyes, dark urine, pale stools, pain in the upper abdomen or back, unexplained weight loss, loss of appetite, itchy skin and new-onset diabetes.

Who is considered at high risk for pancreatic cancer?

People with two or more close relatives who had pancreatic cancer, or who carry inherited mutations such as BRCA1, BRCA2, PALB2, ATM or Lynch syndrome genes, are considered high risk. Smoking, obesity, long-standing diabetes, chronic pancreatitis and older age also raise risk.

Can new-onset diabetes be a warning sign of pancreatic cancer?

Yes. In adults over 50, diabetes that appears suddenly, especially with weight loss rather than weight gain, can sometimes be an early sign of pancreatic cancer. Only a small percentage of people with new diabetes have cancer, but it is a pattern doctors are paying closer attention to.

Is this AI tool available to doctors now?

The study described a research model rather than an approved screening test. Before routine use it would need prospective testing in real clinical settings and validation on local patient populations, since accuracy can change from one health system to another.

What tests are used to screen people at high risk of pancreatic cancer?

People in high-risk surveillance programs are usually monitored with endoscopic ultrasound or MRI, often once a year. CT scans and blood tests such as CA 19-9 may also be used, although CA 19-9 is not reliable enough to screen the general population.

Does AI risk screening replace genetic testing?

No. AI risk prediction and genetic testing answer different questions and work best together. Genetic testing identifies inherited mutations, while AI models look for patterns in medical history that can flag people who have no known genetic or family risk.

Can people lower their risk of pancreatic cancer?

Some risk factors can be changed. Not smoking, keeping a healthy weight, staying active, limiting alcohol to reduce the risk of pancreatitis and managing diabetes well can all help lower risk, although they cannot remove it entirely.

Should I worry if I have gallstones, anemia or diabetes?

Having one of these conditions does not mean you will develop pancreatic cancer, as they are very common. If you also have a strong family history or new symptoms such as jaundice or unexplained weight loss, talk to your doctor about whether further evaluation is needed.

More on this topic: Pancreatic Cancer Publications and AI and Disabilities Publications.

Insights, Analysis, and Developments

Editorial Note: The most useful idea in this study is a practical one: the warning signs of pancreatic cancer may already be sitting in a patient's chart as a sequence of ordinary diagnoses such as gallstones, anemia and type 2 diabetes. None of these conditions points to cancer on its own, and the researchers are clear that the model sorts people by risk rather than diagnosing anyone. Its real value lies in helping clinicians decide who should move ahead to imaging and who can be spared costly, invasive procedures on an organ that does not tolerate them well. The drop in accuracy on veterans' records is also a reminder that an AI model is only as good as the population it learns from, so any health system adopting this kind of screening would need to validate it on its own patients first - including seniors and individuals with disabilities, whose medical histories tend to be longer and more complex.*

Attribution/Source(s): This peer-reviewed publication was selected for publishing by the editors of Disabled World (DW) due to its relevance to the disability community. Originally issued as a press release by Harvard Medical School and published on 8 May 2023. Last updated on 22 Oct 2026. This content may have been edited for style, clarity, or brevity.

* Editorial additions by Ian C. Langtree.