New research shows value of machine learning in heading off prostate cancer

Artificial intelligence tools can predict the recurrence of prostate cancer more than a decade ahead, according to a new paper in the prestigious science journal Nature Cancer.

The study found that artificial intelligence could detect genetic divergence and morphological diversity in a tumour which indicated a tumour’s ability to adapt and survive.

The researchers, from the Institute of Cancer Research (ICR) and the Royal Marden NHS Foundation Trust in the UK, used “machine learning” to analyse 2,500 tissue samples from prostate cancer patients who had received radiotherapy. This provided integrated information about each patient’s cell genomics and morphology, along with their outcomes over more than a decade.

The artificial intelligence analysis of this data showed that how much cells evolved was a strong predictor of recurrence, with the combination of the two measurements identifying a subgroup of patients who experienced disease recurrence twice as fast as the rest.

Professor David Dearnaley, Emeritus Professor at ICR, said: “We believe that our findings will make it possible to identify the patients with high-risk localised cancer who are most likely to benefit from early treatment with life-extending medications.”