Breast Cancer Now is investing £800,000 in two research projects using AI to investigate why some oestrogen receptor-positive breast cancers return years after treatment. The work could help identify biological signs of recurrence, but it is research, not a new clinical test or treatment.
Watch Desk analysis
What happened
Pharmafile reports that one project, led by Professor Maria Secrier at University College London with Professor Luca Magnani’s laboratory at the Institute of Cancer Research, will use AI models trained on cancer-cell data to study how cancer cells become dormant and what prompts them to start growing again. Researchers will also examine how those cells interact with their surroundings, including immune cells and blood vessels.
A second project, led by Dr Maggie Cheang at the Institute of Cancer Research, will analyse tumour samples from the POETIC clinical trial. The trial followed patients for more than ten years after breast cancer surgery. AI and computational analysis will be used to look for patterns linked to treatment resistance and later recurrence.
Why it matters
Around 80% of breast cancers in the UK are oestrogen receptor-positive, according to Pharmafile. Some cancer cells can survive treatment in a dormant state and become active again years later. Finding signals associated with that risk could eventually help clinicians tailor treatment or follow-up.
The researchers hope some indicators might be detectable within two weeks of diagnosis. That is a research aim, not a proven timeframe for a usable prediction tool: further work would be needed to establish whether the findings are reliable enough to change patient care.
Our read
The valuable part here is the focus on a difficult, long-term question, using both cancer-cell data and samples followed over years. AI may help researchers find patterns in those data; it does not, by itself, tell clinicians which patient’s cancer will return. The next test is whether any candidate signal holds up beyond the research that identifies it.
What to watch
- Whether either project identifies reproducible indicators of later recurrence.
- Whether the findings are validated in further patient data.
- Whether researchers can show that any eventual tool improves treatment or follow-up decisions.
Discussion spark: If a future AI tool could flag a higher risk of late recurrence, what evidence should clinicians need before using it to change someone’s follow-up or treatment?
Sources and evidence
- Breast Cancer Now commits £800,000 to AI research into late breast cancer recurrence – Pharmafile (28 October 2026, 09:00 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.