British researchers have developed an artificial intelligence system capable of identifying bowel cancer patients who are unlikely to respond to a newly approved NHS drug, potentially saving the healthcare system millions in ineffective treatments. The algorithm analyzes tumor scans and clinical data to predict which patients won’t benefit from the targeted therapy, according to sources familiar with the ongoing trial.
The technology comes as the UK’s National Health Service faces mounting pressure to control costs of precision cancer drugs, some exceeding £50,000 per patient annually. Health analysts note the AI could help prioritize patients for alternative treatments while reducing unnecessary side effects from ineffective therapies.
“This represents a pragmatic approach to precision medicine,” said one oncology researcher involved in the project who spoke on condition of anonymity. “By ruling out non-responders early, we can redirect resources to patients more likely to benefit.”
NHS officials declined to comment on specific implementation timelines but confirmed AI-assisted treatment pathways are under active evaluation. The system builds on previous machine learning applications in cancer diagnostics, now expanding into therapeutic response prediction.
Medical ethicists caution that such technologies require rigorous validation to avoid denying effective care. The developers emphasize the AI would be used alongside, not instead of, clinician judgment. If validated, the approach could be adapted for other high-cost cancer drugs across the NHS formulary.