A new AI tool that can predict whether an advanced bowel cancer drug will work for a patient before they start taking it could spare thousands of people from harsh side effects each year.
According to a Guardian report, researchers at the Institute of Cancer Research in London and RCSI University of Medicine and Health Sciences in Dublin developed the tool – called PhenMap – to address a painful gap in bowel cancer treatment.
The NHS approved bevacizumab, a targeted drug that slows tumour growth by starving cancer cells of the proteins they need, in December. But while the drug offers real hope, it only works for a minority of patients. Everyone else faces serious side effects, from blood clots to gastrointestinal problems, with no benefit.
PhenMap – a portmanteau of “phenotype” (an organism’s observable traits) and “mapping” – analyses the complex genetic makeup of a patient’s tumour and identifies patterns in how they’re likely to respond. It’s the kind of work that would be impossible for a human to do by hand, given the sheer volume and complexity of the data involved.
“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,” said Anguraj Sadanandam, a professor in stratification and precision medicine at the ICR.
“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.”
The stakes are high. Bowel cancer has the second highest mortality rate of any cancer in the UK, behind only lung cancer, and nearly 10,000 advanced cases are identified each year. Diagnoses among young adults are on the rise. When the disease is caught early, survival rates can reach 98%, but for advanced cases the five-year survival rate can fall as low as 10% – making the right treatment at the right time all the more important.
In the study, the team tracked 117 European patients who had been treated with chemotherapy and bevacizumab. PhenMap allowed researchers to “integrate complex data on the genetic makeup of the tumour,” revealing patterns in how different patients responded to the drug and pinpointing a group who shared the same gene mutation and were at high risk of negative reactions.
Sadanandam said the AI’s ability to process vast amounts of data at speed is what sets it apart. “In our research, we have shown that this allows us to identify the patients least likely to respond to treatment with bevacizumab,” he said.
The findings are encouraging, though Sadanandam stressed the tool still needs to be tested on a larger group of patients before it can be validated for clinical use. The team is already looking to expand its sample size and explore whether the approach could apply to other types of cancer.
“In future, I hope this approach will lead to a test that can be used by clinicians, to ensure patients receive personalised care that has the highest chance of working against their cancer,” he said.
