New AI method makes cancer tracking faster and easier using blood tests
Peer-Reviewed Publication
Updates every hour. Last Updated: 7-Jul-2025 21:11 ET (8-Jul-2025 01:11 GMT/UTC)
Singapore – Scientists from the A*STAR Genome Institute of Singapore (A*STAR GIS) have developed a new artificial intelligence (AI)-based method called "Fragle" that makes tracking cancer easier and faster using blood tests. Requiring only a small blood sample, this method analyses the size of DNA fragments in the blood to reveal distinct patterns that differentiate cancer DNA from healthy DNA, helping doctors track cancer treatment response more accurately and frequently. The research was published in Nature Biomedical Engineering in March 2025.
In recent years, the regulatory role of the gut microbiota in the initiation and progression of colorectal cancer (CRC) has attracted growing attention. Among the key microbial contributors, Fusobacterium nucleatum (Fn) has been identified as a critical pathogenic factor in CRC. As an oral anaerobic commensal, Fn is rarely found in the lower gastrointestinal tract of healthy individuals. However, under pathological conditions, it can ectopically colonize the gastrointestinal tract. Once enriched in the colorectal environment, mounting evidence suggests that Fn is involved in multiple aspects of CRC pathogenesis, including initiation, progression, metastasis, and resistance to conventional therapies such as chemotherapy, radiotherapy, and immunotherapy. A recent review by Wei Wei and Diwei Zheng's team at the Institute of Process Engineering systematically outlines the pathogenic mechanisms of Fn in CRC and summarizes both current and emerging strategies for its therapeutic targeting. Furthermore, the authors propose potential approaches to overcome existing challenges in Fn modulation, aiming to facilitate more effective therapeutic interventions and improve clinical outcomes.
Not all DNA looks like the familiar twisted ladder. Sometimes, parts of our genetic code fold into unusual shapes. One such structure, the G-quadruplex (G4), looks like a knot. These knots can play important roles in turning genes on or off. But if not untangled in time, they can harm our genome. Now, researchers from the Knipscheer Group at the Hubrecht Institute, in collaboration with the Karolinska Institute, have uncovered a surprising mechanism that keeps these knots in check. Their work, published in Science on June 12th, could lead to new ways to treat diseases like cancer.
Continued surveillance and potential new options
—
Regulatory authorities issued precautionary recommendations following the ORAL Surveillance trial, which demonstrated an elevated risk of cancer with tofacitinib, compared to TNF inhibitors (TNFi).1 While these findings have influenced clinical guidance, there is still limited real-world evidence on malignancy risks associated with the janus kinase inhibitors (JAKi), and they remain part of the recommendations for rheumatoid arthritis (RA) from EULAR – The European Alliance of Associations for Rheumatology – although with the caveat that pertinent risk factors must be taken into account, with clinicians advised to consider age, smoking, and other key medical factors when intending to prescribe a JAKi.2 But there remains a need for more evidence.
Cancer cells have evolved numerous strategies to suppress immune cells like NK cells, even when these cancer cell are producing the immune boosting factor IL-15. An obvious solution is to supply cancer patients with drugs that trigger the IL-15 receptor on immune cells, however these approaches have proven too toxic for patients because they boosts the activity of immune cells in every tissue, not just in the tumour, resulting in severe side-effects.
Until now.
A team of researchers at Monash University and oNKo-Innate in Melbourne, Australia have found a gene that can be switched off in NK cells which makes them extremely sensitive to the body’s own IL-15, opening the way for the development of a new therapy to treat cancer.