AI Blood Test Shows Promise in Largest Real-World NHS Evaluation

An artificial intelligence (AI) blood test that uses routine blood results to assess cancer risk has shown promising results in the largest real-world NHS evaluation of its kind.

The evaluation assessed the PinPoint blood test, a machine learning tool designed to help identify patients who may be at higher or lower risk of cancer after being referred on an urgent suspected cancer pathway.

Researchers evaluated the technology using data from 16,481 patients across 170 GP practices and five NHS trusts in England. Rather than requiring a new blood sample, the system analyses results from routine blood tests that many patients already have as part of their assessment.

The study found that the technology demonstrated potential clinical value across several cancer pathways, including gynaecological, upper gastrointestinal, lung, head and neck, and lower gastrointestinal cancers. Researchers reported that the tool could help prioritise patients at the highest risk for faster investigation while reducing unnecessary invasive testing for those assessed as being at lower risk.

According to the published evaluation, prioritising the 10% of patients identified as being at the highest risk could significantly improve the efficiency of cancer investigations by increasing the likelihood of detecting cancer among those investigated first.

The researchers emphasised that the technology is intended to support, rather than replace, clinical decision-making. Because the system relies on existing laboratory results, it could be introduced without additional blood tests or new laboratory equipment.

While further implementation work and ongoing evaluation will be needed before wider adoption, the findings highlight how AI could help improve cancer triage, reduce unnecessary investigations and support earlier diagnosis for patients who need urgent assessment.

Source: Mayo Clinic Proceedings: Digital Health – Real-World Validation of PinPoint Blood Tests in the NHS: Multivariable Machine Learning to Predict Cancer Risk in Primary Care Urgent Referrals.

Stay informed with the stories shaping health and social care.