Researchers at the A*STAR Genome Institute of Singapore (A*STAR GIS) have created a groundbreaking method using artificial intelligence (AI) to identify even the smallest traces of tumor DNA in blood samples. This new AI technology, known as 'Fragle,' aids doctors in monitoring the effectiveness of cancer treatments.

Anders Skanderup, PhD, a Senior Principal Scientist at A*STAR GIS Laboratory of Computational Cancer Genomics and the lead researcher, likened the process to tracking COVID-19 outbreaks through viral particles in wastewater. By analyzing DNA fragments in blood, Fragle enables precise monitoring of cancer treatment response and early detection of relapse.

Published in the journal Nature Biomedical Engineering, the study titled 'A Deep-Learning Model for Quantifying Circulating Tumor DNA From the Density Distribution of DNA-Fragment Lengths' showcases how the AI model can swiftly and accurately evaluate the size of DNA fragments in blood samples. As circulating tumor DNA (ctDNA) typically differs in size from healthy DNA, the AI model can effectively quantify ctDNA levels and assess a patient's response to cancer treatment.

By utilizing whole-genome sequencing data from various cancers and healthy individuals, the researchers confirmed the effectiveness of the AI model. In their publication, the team noted that Fragle surpassed existing methods in terms of accuracy and detection limits.

Costing only $39 (SGD $50) per test, Fragle offers a more affordable alternative to conventional cancer detection tests, which can cost around $780 (SGD $1000). The AI technology can seamlessly integrate into current workflows in hospitals and commercial laboratories, facilitating easy adoption.

This innovative AI method enables more frequent monitoring of cancer patients during treatment, enabling healthcare providers to assess treatment efficacy promptly. Additionally, it aids in the early detection of cancer relapse post-surgery or treatment, as it can identify even the smallest traces of cancer in the bloodstream.

Collaborating with the National Cancer Centre Singapore, the A*STAR GIS researchers are actively monitoring over 100 clinical trial participants every two months using Fragle to analyze ctDNA levels during cancer treatment. They are also investigating how changes in ctDNA levels can help identify patients who may respond differently to cancer therapy.

Looking ahead, the researchers aim to conduct further studies and introduce their AI technology for cancer screening in hospitals, with the goal of enhancing patient care. Wan Yue, PhD, Executive Director at A*STAR GIS, expressed enthusiasm about the potential of Fragle to improve cancer detection and treatment monitoring, emphasizing its impact on global healthcare.