Johns Hopkins Kimmel Cancer Center investigators have developed an artificial intelligence technique that could revolutionize how clinicians monitor pancreatic cancer treatment effectiveness. The method, called ARTEMIS-DELFI, detects DNA fragments shed by tumors circulating in a patient's blood and provides rapid feedback on whether therapies are working.
The research, published May 21 in Science Advances, demonstrates that this blood-based approach could help clinicians make faster, more informed decisions about continuing or changing treatment plans for pancreatic cancer patients.
AI-Powered Blood Test Outperforms Traditional Monitoring Methods
Researchers tested ARTEMIS-DELFI in blood samples from patients participating in two large clinical trials of pancreatic cancer treatments. The results showed that the AI-powered test could identify therapeutic responses more effectively than conventional methods.
"Time is of the essence when treating patients with pancreatic cancer," explains senior study author Victor E. Velculescu, M.D., Ph.D., co-director of the cancer genetics and epigenetics program at the Johns Hopkins Kimmel Cancer Center. "Many patients with pancreatic cancer receive a diagnosis at a late stage, when cancer may progress rapidly."
Currently, clinicians rely on imaging tools to monitor cancer treatment response and tumor progression. However, these tools often produce delayed results and can be less accurate for patients receiving immunotherapies, where interpretation becomes more complex.
Two Approaches Tested in Clinical Trials
The research team evaluated two different approaches to monitoring treatment response in patients participating in the phase 2 CheckPAC trial of immunotherapy for pancreatic cancer:
-
WGMAF (tumor-informed plasma whole-genome sequencing): This method analyzed DNA from tumor biopsies as well as cell-free DNA in blood samples to detect treatment response.
-
ARTEMIS-DELFI (tumor-independent genome-wide cfDNA fragmentation profiles and repeat landscapes): This technique used machine learning to scan millions of cell-free DNA fragments only in the patient's blood samples, without requiring tumor tissue.
While both approaches could detect which patients were benefiting from therapies, ARTEMIS-DELFI proved superior because it worked with more patients and was logistically simpler. Not all patients had tumor samples available, and many tumor samples contained only a small fraction of cancer cells compared to normal tissue, which complicated the WGMAF test.
The team validated ARTEMIS-DELFI's effectiveness in a second clinical trial called PACTO, confirming that it could identify responding patients as early as four weeks after therapy initiation.
A "Fast-Fail" Approach for Personalized Treatment
"The 'fast-fail' ARTEMIS-DELFI approach may be particularly useful in pancreatic cancer where changing therapies quickly could be helpful in patients who do not respond to the initial therapy," says lead study author Carolyn Hruban, who conducted the research as a graduate student at Johns Hopkins and is now a postdoctoral researcher at the Dana-Farber Cancer Institute.
Hruban highlighted the practical advantages of the new technique: "It's simpler, likely less expensive, and more broadly applicable than using tumor samples."
Future Directions and Broader Applications
The next step for the research team will be prospective studies to test whether the information provided by ARTEMIS-DELFI helps clinicians more efficiently find effective therapies and improve patient outcomes. The researchers believe a similar approach could also be used to monitor other types of cancer.
Earlier this year, members of the same team published a study in Nature Communications showing that a variation of the cell-free fragmentation monitoring approach called DELFI-TF was helpful in assessing colon cancer therapy response.
"Our cell-free DNA fragmentation analyses provide a real-time assessment of a patient's therapy response that can be used to personalize care and improve patient outcomes," Velculescu concludes.
Addressing an Urgent Need in Pancreatic Cancer Care
Pancreatic cancer remains one of the most challenging malignancies to treat, with limited therapeutic options and often poor outcomes. The disease is frequently diagnosed at advanced stages when treatment options are more limited and less effective.
With a growing number of experimental therapies becoming available for pancreatic cancer, tools that can quickly determine treatment efficacy are increasingly valuable. ARTEMIS-DELFI represents a significant step forward in the ability to monitor treatment response in real-time, potentially allowing oncologists to pivot to alternative therapies sooner when initial treatments prove ineffective.
The study was supported by multiple foundations and grants, including funding from the National Institutes of Health, the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation, and Stand Up to Cancer, among others.
