Loizos G. Loizou, Head of Cyprus International Action Plan for Children with Cancer at ELPIDA Foundation for Children with Cancer, shared on LinkedIn:
“Cancer Detection and Artificial Intelligence (AI): Promising Developments in Practice.
New AI-powered blood tests detect cancer recurrence months before it can be seen on CT and MRI scans.
The Problem:
After surgery or chemotherapy, monitoring the progression of the cancer is based on CT scans and PET scans. However, these tests have limits: A tumor must grow into millions of cells to become visible. In this ‘blind spot’, minimal residual disease (MRD) can develop silently.
The Solution:
Recent scientific publications in the journals Nature Medicine and The Lancet Oncology combine full-genome liquid biopsies with artificial intelligence (AI) algorithms. When cancer cells are destroyed, they release small fragments of DNA (ctDNA) into the bloodstream. AI models isolate real cancer DNA from the body’s normal ‘noise’ by detecting it at extremely low concentrations.
The Benefits:
- Early Warning: Detects molecular recurrence 8 to 12+ months before the tumor becomes visible on a CT or PET scan.
- Non-Invasive Method: Requires only a simple blood draw during regular re-examinations.
- High Accuracy: The AI filters the sample, drastically reducing false positives.
Practical application in clinical practice:
- Is it available today? Yes. Liquid ctDNA biopsy tests are already being used in oncology practice, while new AI-enhanced models are rapidly being integrated into clinical protocols.
- What do doctors do today if it comes out positive? Although large clinical trials still confirm the survival benefits of immediate treatment change, a positive ctDNA test allows oncologists to thicken imaging intervals, schedule targeted screening, and enroll the patient in early treatment studies long before visible recurrence occurs.
Bibliography/Source:
Nature Medicine/The Lancet Oncology. AI-enhanced whole-genome sequencing for ultra-sensitive minimal residual disease (MRD) detection in circulating tumor DNA (ctDNA).”
You can also read:
Artificial Intelligence and Digital Pathology in Leukemia Diagnosis
