NHS pilots AI blood test to reduce invasive womb cancer checks

Several NHS hospitals in England are preparing to use an AI-powered blood test to assess women referred with suspected womb cancer and reduce the need for invasive checks such as transvaginal ultrasound. The test analyses routine blood markers and gives a risk score to help clinicians decide whether hospital procedures are needed.

The test was developed by Leeds-based PinPoint Data Science and examines about 30 blood markers. It classifies patients as low, elevated or high risk. PinPoint says the test can be used within existing referral routes, costs about £30 and can be run on routine blood samples.

The tool was evaluated in a trial of 16,481 patients referred through urgent suspected cancer pathways across Yorkshire. The trial included women whose symptoms prompted concern about womb or other gynaecological cancers. The company reported that the test flagged 99.1% of cancers as elevated or high risk and produced a negative predictive value of 99.8% for the lowest-risk group. Trial data also showed around one in 10 women referred because of heavy bleeding had cancer.

Under current practice, women referred for suspected reproductive system cancers usually have a pelvic examination and a transvaginal ultrasound to measure womb lining thickness. If clinicians remain concerned, patients can be offered hysteroscopy or biopsy. PinPoint estimates the blood test could spare about one in five women referred for suspected womb cancer from needing an immediate transvaginal scan, which would be roughly 18,000 fewer scans a year in England.

Mid Yorkshire NHS Teaching Trust plans to use the test across six types of gynaecological and upper gastrointestinal cancer referrals. Leeds Teaching Hospitals NHS Trust intends to deploy it for gynaecological referrals. PinPoint has also applied the tool in lung, upper and lower gastrointestinal, and head and neck cancer pathways.

Professor Sean Duffy, chief medical officer at PinPoint and a former NHS England national clinical director for cancer, described the test’s role as identifying women at very low risk so invasive tests can be avoided. Dr Jacinta Walsh, a GP in Normanton, West Yorkshire, noted some patients currently need multiple GP visits before cancer is ruled out and said a reliable blood test could shorten that process. Tracy Jackson, consultant gynaecologist and cancer unit lead at Leeds Teaching Hospitals NHS Trust, pointed out that most women referred do not have cancer and that the investigations can be uncomfortable; she said a risk score can help prioritise patients who need faster assessment.

Cancer Research UK described the test as promising and called for further research to understand its effects on patient outcomes and referral patterns. A spokesperson for the charity said early detection saves lives but that more evidence is needed on how the blood test would change diagnoses and service use.

The PinPoint deployment is part of wider NHS activity to introduce artificial intelligence into clinical pathways. Other systems in use include an infection-risk tool at Kent and Canterbury Hospital, an AI triage function being rolled out via the NHS App, and AI-powered chest X-ray software used in suspected lung cancer assessment. The government has committed £20 million to extend chest X-ray AI tools to all NHS trusts in England by 2029; the technology is already available in around half of trusts and has supported assessment for more than four million patients investigated for lung cancer.

Each year about 90,000 postmenopausal women in England are referred by GPs for checks after heavy vaginal bleeding. Approximately 10,000 women are diagnosed with womb cancer annually and about 2,700 die from the disease. Regulators and clinicians say further evaluation will be needed to confirm how a blood-based AI test affects patient outcomes, waiting times and diagnostic capacity across the NHS.