Disrupting the Timeline: Harnessing AI for the Early Detection of Pancreatic Cancer
Artificial intelligence is helping researchers identify pancreatic cancer earlier, giving more patients the opportunity to benefit from today's breakthrough treatments.
We are investing in a future where pancreatic cancer is found earlier, treated more effectively, and ultimately survived.
.avif)
Led by researchers at The University of Texas MD Anderson Cancer Center, one of the world's leading comprehensive cancer centres and a global leader in pancreatic cancer research, this Pancreatic Cancer North America-sponsored initiative is exploring how artificial intelligence and blood-based biomarkers can help detect pancreatic cancer months, and potentially years, before conventional diagnosis.
Most patients are still diagnosed at stage IV, after the disease has spread. Without an early-stage detection method, patients are still being diagnosed too late to fully benefit from targeted therapies or even qualify for clinical trials. By the time symptoms appear and the disease is found, opportunities for intervention have often already been lost. Without early detection, even the most promising breakthroughs cannot fully deliver on their potential.
The Breakthrough We Are Building
For many years, research into the early detection of pancreatic cancer has focused primarily on blood biomarkers. What makes this project unique is that it combines blood biomarkers with pre-diagnostic CT scans, using advanced artificial intelligence to identify subtle signs of pancreatic cancer before conventional diagnosis.
Blood-based biomarkers
Researchers are validating promising blood biomarkers that may help identify people at higher risk of developing pancreatic cancer.
Artificial intelligence imaging
Researchers are using advanced imaging and artificial intelligence to analyze routine CT scans for subtle changes that cannot be detected using conventional techniques.
Supported by seed funding from The Warren Y. Soper Charitable Trust, this research is designed around tools already used throughout today's healthcare systems, creating a realistic pathway toward future clinical adoption.
"We are training artificial intelligence to detect subtle signs of pancreatic cancer years before diagnosis - signals invisible to the human eye. This could fundamentally change how we identify the disease, allowing us to intervene while it is still curable."
Dr. Eugene Koay
Scientific Director, The Ahmed Centre for Pancreatic Cancer Research at MD Anderson Cancer Center
Why This Approach Matters
Early detection represents the greatest opportunity to transform survival rates.
These technologies may identify pancreatic cancer months, and potentially years, before conventional diagnosis. Earlier detection means more patients are eligible for surgery, and more patients eligible for surgery means more patients with a chance to survive.
Built for real-world adoption.
By building on routine blood tests and CT scans already used throughout healthcare systems, this approach is intentionally designed to move quickly from research into future clinical practice.
Powered by an unprecedented foundation of data.
The study leverages one of the largest repositories of pre-diagnostic blood samples and CT images ever assembled for pancreatic cancer research, helping accelerate validation and future clinical adoption.
The Questions Driving This Research
Can artificial intelligence recognize invisible warning signs?
Researchers are studying whether advanced AI can identify subtle structural changes in routine CT scans before pancreatic cancer becomes clinically apparent.
Can combining blood biomarkers and imaging improve accuracy?
The team is evaluating whether combining both approaches creates a more accurate way to identify people at higher risk while reducing unnecessary procedures.
A North American Research Collaboration
Dr. Eugene Koay and Dr. Johannes Fahrmann at The University of Texas MD Anderson Cancer Center are leading this collaboration, bringing together researchers from across North America to accelerate the development of an integrated approach to early detection.
