NHS AI blood test could reduce invasive cancer exams for women

Thousands of women could avoid invasive diagnostic procedures for suspected womb cancer under an NHS trial of an AI blood test designed to identify patients at low risk of disease.

The PinPoint blood test, developed by Leeds-based PinPoint Data Science, is being introduced by several NHS organisations following a large-scale evaluation involving 16,481 patients referred by GPs with suspected cancers across 170 practices in Yorkshire.

Leeds Teaching Hospitals NHS Trust and the University of Leeds first collaborated with PinPoint Data Science in 2023 to develop the AI blood test.

According to The Guardian, around 90,000 postmenopausal women are referred each year in England for investigation after experiencing postmenopausal bleeding, a key symptom of womb (endometrial) cancer.

While approximately 10,000 women are diagnosed annually with the disease, current NHS pathways require all women to undergo a transvaginal ultrasound scan, with some subsequently requiring hysteroscopy and biopsy.

The AI-enabled blood test could enable around one in five women referred with suspected womb cancer – approximately 18,000 patients each year – to avoid unnecessary invasive investigations by safely ruling out those at very low risk, The Guardian reported.

The technology uses machine learning to analyse 30 blood biomarkers and categorise patients as being at low, elevated or high risk of cancer. Of the 16,481 patients included in the trial, 3,313 were referred with suspected gynaecological cancer following postmenopausal bleeding.

Professor Sean Duffy, chief medical officer at PinPoint Data Science and former NHS England national clinical director for cancer, said the results were “remarkable by any clinical standards”.

“But equally, its value lies in safely ruling out very low-risk women. This has the potential to spare thousands of patients from painful invasive procedures they do not need,” he told The Guardian.

According to trial findings, the test achieved a 99% accuracy rate for both identifying gynaecological cancers and ruling them out. The findings suggested it outperformed conventional diagnostic approaches.

Mid Yorkshire Teaching NHS Trust plans to introduce the test for six gynaecological cancers and upper gastrointestinal cancers, while Leeds Teaching Hospitals NHS Trust will initially deploy it for suspected gynaecological cancer referrals.

Consultant gynaecologist Tracy Jackson, cancer unit lead at Leeds Teaching Hospitals NHS Trust, said the technology could improve patient triage and reduce pressure on services.

“But the reality is that most women we see do not have cancer and we are acutely aware that the investigations can be uncomfortable and, for some, distressing,” she said.

She added that using the blood test to identify low-risk patients in primary care could reduce unnecessary procedures, shorten waiting lists and enable women with cancer to be diagnosed and treated more quickly.

Samantha Harrison, spokesperson for Cancer Research UK, added: “Spotting cancer early saves lives, but right now patients are not being diagnosed quickly enough.

“This test could help to rule out endometrial cancer in some women, through a simple blood test, without the need for further testing.

“More research is needed to understand the benefits for patients and the NHS, but the results of this study are promising.”