NHS pilots AI blood test to reduce invasive womb cancer scans

Several NHS hospitals in England are piloting an AI blood test developed by Leeds-based PinPoint Data Science to assess the risk of womb (endometrial) cancer for patients referred on urgent suspected cancer pathways. The trial in Yorkshire involved 16,481 patients referred with symptoms that raised concern about possible gynaecological cancer.

The test uses machine learning to analyse about 30 blood markers and classifies patients as low, elevated or high risk. PinPoint reports the test costs about £30 and is intended to provide a risk score for clinicians to use within existing referral routes to guide monitoring, further investigation or prioritised assessment.

In the Yorkshire trial roughly one in 10 women referred because of heavy bleeding were diagnosed with cancer. The company reported that 99.1% of cancers were flagged as elevated or high risk and that the negative predictive value for women classified as lowest risk was 99.8%.

Under current diagnostic practice, patients with suspected womb cancer typically have a pelvic examination and a transvaginal ultrasound to measure the thickness of the womb lining; persistent concern can lead to a biopsy or hysteroscopy. Some patients find the ultrasound and follow-up procedures uncomfortable, and GPs have described cases where multiple visits are needed before cancer is ruled out.

PinPoint estimates the blood test could avoid a transvaginal ultrasound for about one in five women referred with symptoms, which would equate to roughly 18,000 fewer scans a year in England. Mid Yorkshire NHS Teaching Trust plans to introduce the test across six gynaecological and upper gastrointestinal cancer pathways, and Leeds Teaching Hospitals NHS Trust intends to use it for gynaecological referrals. The company also reports deployment of the test in pathways for lung, upper and lower gastrointestinal, head and neck and gynaecological cancers.

Professor Sean Duffy, chief medical officer at PinPoint and a former national clinical director for cancer at NHS England, described the test’s value as identifying women at very low risk so that resources can be prioritised for those at higher risk. GP Jacinta Walsh said the test could shorten the diagnostic process and reduce the number of GP visits some patients currently need. Consultant gynaecologist Tracy Jackson noted that most women referred under current routes do not have cancer and that a blood test could help triage patients before hospital-based investigations.

Other AI tools are being introduced across the NHS. An AI system called MEMORI is in use at Kent and Canterbury Hospital to assess infection risk from routine patient data. An AI triage tool in the NHS App is being rolled out and is expected to reach more than 200,000 patients within 12 months, with wider availability planned by April 2028. The government has committed £20 million to expand AI-powered chest X-ray tools to all trusts by 2029; such tools are already available in about half of trusts and have supported assessments for more than four million patients being investigated for lung cancer.

Cancer Research UK described the PinPoint test as promising but said more research is needed to understand how its use would affect patient outcomes, referral decisions and diagnostic capacity. A spokesperson for the charity noted that earlier diagnosis saves lives and that a reliable blood test could rule out endometrial cancer in some women without further invasive investigations.