AI Method Speeds Up Cancer Tracking Using Blood Tests

Current techniques for detecting cancer DNA in the bloodstream, referred to as circulating tumor DNA (ctDNA), often depend on complex and costly DNA sequencing to identify common mutations associated with cancer. However, because these mutations differ widely among individuals, test outcomes can be inconsistent, making it challenging for physicians to effectively monitor treatment progress through blood-based diagnostics. Now, researchers have developed a new artificial intelligence (AI)-driven approach that simplifies and speeds up cancer tracking via blood tests. By analyzing the size of DNA fragments in a small blood sample, this method can distinguish cancer DNA from healthy DNA, enabling more precise and frequent monitoring of treatment response.

This new approach, known as “Fragle,” was developed by researchers at the ASTAR Genome Institute of Singapore (ASTAR GIS, Singapore). It employs AI to examine fragment sizes of DNA circulating in the blood. Since cancer DNA exhibits size patterns that differ from those of normal DNA, the Fragle model can detect these subtle differences even in very small DNA samples. As a result, this method facilitates more affordable and quicker cancer tracking. Research findings published in Nature Biomedical Engineering showed that Fragle produced reliable and accurate results across hundreds of blood samples from cancer patients spanning various cancer types. The method is also compatible with the DNA profiling tools already in use across hospitals and commercial laboratories, making integration seamless.

Fragle presents a quicker and more cost-effective solution for monitoring cancer via blood samples, requiring minimal DNA input. While traditional commercial tests may exceed SGD $1000 in cost, Fragle’s estimated cost is under SGD $50. Its compatibility with commonly used DNA profiling technologies in both clinical and commercial settings allows for easy adoption within current diagnostic frameworks. Fragle can also detect minimal residual disease (MRD)—small amounts of cancer that may remain after treatment, thus enabling early identification of potential relapse. Researchers are now working to further improve the sensitivity of Fragle, aiming to detect even lower levels of ctDNA for earlier relapse detection in cancer patients. In parallel, the team is establishing collaborative partnerships to identify clinical use cases and facilitate application in real-world settings.

Going forward, the researchers plan to evaluate how Fragle can be implemented in local healthcare systems to improve outcomes for cancer patients. As part of a current study involving over 100 participants in clinical trials, Fragle is being used to measure ctDNA levels bi-monthly during treatment, with the aim of identifying relapse indicators earlier than standard imaging scans. Additionally, the study is investigating whether early shifts in ctDNA levels can help predict which patients are more likely to respond well—or poorly—to treatment. Ultimately, the research aims to determine the value of incorporating ctDNA blood testing into regular cancer patient monitoring throughout the treatment journey.

“We are very excited about the potential Fragle brings, to help our healthcare professionals detect and track cancer more accurately and monitor treatments more effectively, leading to better cancer care for patients,” said Dr. Wan Yue, Executive Director at A*STAR GIS. “It is our hope that our genomic research can be translated to benefit population health not only in Singapore, but worldwide.”

Related Links:
A*STAR GIS