AI Breakthrough Predicts NHS Cancer Drug Response, Aims to Spare Patients Side Effects
Scientists have unveiled a groundbreaking artificial intelligence tool designed to predict how cancer patients will respond to a targeted drug available on the NHS. This innovative development could transform treatment for advanced bowel cancer by identifying individuals most likely to benefit, thereby sparing others from unnecessary and potentially harmful side effects.
Targeting Treatment with Precision
The AI model, named PhenMap—short for phenotype mapping—was created by researchers at the Institute of Cancer Research (ICR) in London and the RCSI University of Medicine and Health Sciences in Dublin. It focuses on bevacizumab, a drug approved in December for treating advanced bowel cancer on the NHS. While bevacizumab can slow cancer growth, it is effective for only a small subset of patients and carries risks such as high blood pressure, gastrointestinal issues, and blood clots.
Nearly 10,000 cases of advanced bowel cancer are diagnosed annually in England, with a concerning rise among young adults. Currently, scientists group cancers into a limited number of subtypes, but PhenMap can detect more intricate patterns, refining these groups and placing patients on a scale from one to 100 based on their likely response.
How PhenMap Works
In research published in the journal Scientific Reports, the team studied 117 European patients treated with bevacizumab and chemotherapy. PhenMap integrates complex data on tumour genetics with clinical information like gender, age, and tumour location. This allows it to uncover new biological signals—patterns relevant to patient response that are otherwise invisible to human analysis.
Anguraj Sadanandam, professor in stratification and precision medicine at the ICR, explained: "Once bowel cancer spreads to other parts of the body, there are very few treatment options available for patients. It is therefore positive that patients can now access the targeted drug bevacizumab on the NHS." He added, "However, we know that the majority of patients won’t benefit from the drug, meaning thousands of people in England could be facing unpleasant side effects unnecessarily. Until now, we haven’t been able to identify these patients."
Revolutionising Cancer Research
Professor Sadanandam emphasised that PhenMap uses advanced AI to analyse large, complex datasets, spotting patterns "otherwise impossible for a human to see" and revealing clues hidden within tumours. The findings, however, require validation in larger cohorts to ensure applicability across all patients.
Professor Kristian Helin, chief executive of the ICR, highlighted the broader impact: "The approval of new drugs to treat cancers is a significant milestone, but we must recognise that one drug won’t work for everyone—understanding why certain patients won’t benefit from the treatment is crucial to improving outcomes." He noted that AI has revolutionised cancer research by enabling rapid analysis of complex data and predicting treatment responses.
"This research is a powerful example of how the ICR is leveraging AI to develop smarter, kinder therapies and deliver them to patients sooner," Professor Helin stated. He also suggested the method could be explored for other cancer types and targeted therapies, potentially expanding its utility across oncology.
Funding and Future Prospects
The project received funding from the EU Horizon 2020, Research Ireland, the Ian Harty Charitable Trust, and the ICR. As AI continues to advance, tools like PhenMap promise to enhance personalised medicine, ensuring that NHS patients receive the most effective treatments while minimising risks and improving overall cancer care outcomes.
