New AI Method Analyses DNA Fragments in Blood to Track Cancer More Efficiently

New Delhi: A new artificial intelligence (AI)-based method developed by researchers in Singapore may significantly improve how doctors monitor cancer in patients using simple blood tests.

The method, named 'Fragle', was developed by a team from the A*STAR Genome Institute of Singapore (A*STAR GIS). It uses AI to analyse the size of DNA fragments in the blood, which helps distinguish between DNA from healthy cells and cancer cells.

Published in Nature Biomedical Engineering, the study explains how Fragle works with just a small blood sample. It identifies unique patterns in the size of circulating DNA fragments, allowing doctors to track cancer treatment responses more accurately and frequently.

"Just as scientists tracked Covid-19 outbreaks by detecting viral particles in wastewater, Fragle analyses DNA fragments in blood to monitor cancer treatment response and detect relapse early," said lead author Dr. Anders Skanderup, Senior Principal Scientist at A\*STAR GIS Laboratory of Computational Cancer Genomics.

Current methods for monitoring cancer DNA in the blood, known as circulating tumour DNA (ctDNA), typically rely on expensive and complex DNA sequencing. These tests usually screen for specific cancer-related mutations. However, because these mutations can vary significantly between patients, results may not always be consistent, making it harder to assess treatment progress.

Fragle offers an alternative by using AI to focus on DNA fragment size instead of mutations. According to the researchers, cancer DNA tends to break into different sizes compared to healthy DNA. The Fragle model can detect these patterns using minimal DNA quantities, making the method both faster and more cost-effective.

The team reported that Fragle produced accurate results across hundreds of samples from patients with various cancer types. The approach is also compatible with widely used DNA profiling methods in hospitals and commercial labs.

In an ongoing clinical study involving over 100 patients, the A\*STAR GIS and National Cancer Centre Singapore (NCCS) team is using Fragle to monitor ctDNA levels every two months. The goal is to detect relapse earlier than routine scans would typically show.

The researchers are also investigating whether early shifts in ctDNA levels can help identify which patients are more likely to respond well-or poorly-to treatment. The study aims to evaluate the potential of ctDNA testing as a regular tool in monitoring cancer patients during therapy.