Cancer
AI-powered blood test first to spot earliest sign of breast cancer
A new screening method that combines laser analysis with a type of AI is the first of its kind to identify patients in the earliest stage of breast cancer, a study suggests.
The fast, non-invasive technique reveals subtle changes in the bloodstream that occur during the initial phases of the disease, known as stage 1a, which are not detectable with existing tests.
The researchers say their new method could improve early detection and monitoring of the disease and pave the way for a screening test for multiple forms of cancer.
Standard tests for breast cancer can include a physical examination, x-ray or ultrasound scans or analysis of a sample of breast tissue, known as a biopsy. Existing early detection strategies rely upon screening people based on their age or if they are in at-risk groups.
Using the new method, researchers were able to spot breast cancer at the earliest stage by optimising a laser analysis technique – known as Raman spectroscopy – and combining it with machine learning, a form of AI.
Similar approaches have been trialled to screen for other types of cancer, but the earliest they could detect disease was at stage two, the team says.
The new technique works by first shining a laser beam into blood plasma taken from patients. The properties of the light after it interacts with the blood are then analysed using a device called a spectrometer to reveal tiny changes in the chemical make-up of cells and tissues, which are early indicators of disease.
A machine learning algorithm is then used to interpret the results, identifying similar features and helping to classify samples.
In the pilot study involving 12 samples from breast cancer patients and 12 healthy controls, the technique was 98 per cent effective at identifying breast cancer at stage 1a.
The test could also distinguish between each of the four main subtypes of breast cancer with an accuracy of more than 90 per cent, which could enable patients to receive more effective, personalised treatment, the team says.
Implementing this as a screening test would help identify more people in the earliest stages of breast cancer and improve the chances of treatment being successful, the team says. They aim to expand the work to involve more participants and include tests for early forms of other cancer types.
Blood samples used in the study were provided by the Northern Ireland Biobank and Breast Cancer Now Tissue Bank.
Dr Andy Downes, of the University of Edinburgh’s School of Engineering, who led the study, said: “Most deaths from cancer occur following a late-stage diagnosis after symptoms become apparent, so a future screening test for multiple cancer types could find these at a stage where they can be far more easily treated. Early diagnosis is key to long-term survival, and we finally have the technology required.
“We just need to apply it to other cancer types and build up a database, before this can be used as a multi-cancer test.”
An early PET scan after one cycle of chemotherapy may help predict how aggressive breast cancer responds to treatment, a study suggests.
Research led by The Institute of Cancer Research, London and King’s College London suggests that an early scan taken after one cycle of chemotherapy could help predict how well a patient’s cancer will respond to treatment.
The study focused on patients with triple-negative breast cancer (TNBC), an aggressive form of the disease in which cancer cells lack receptors for the hormones oestrogen and progesterone, as well as the HER2 protein.
Patients with TNBC are usually treated with chemotherapy prior to surgery. While many respond well, residual disease at surgery, typically around six months later, is associated with a significantly poorer prognosis. Identifying people sooner who are unlikely to respond remains a major clinical challenge.
The research explored whether using PET imaging shortly after treatment begins, rather than relying only on MRI scans later in the treatment process, could provide earlier insight into how a patient’s cancer is responding. Twenty-two patients were recruited, with fourteen undergoing FDG-PET scans before treatment and after the first cycle of chemotherapy.
The findings, published in Clinical Cancer Research, showed that changes seen on PET scans after just one cycle of chemotherapy were strongly associated with subsequent response, including whether there was no detectable cancer, known as a complete response, by the end of treatment. Importantly, early PET response showed stronger associations with treatment outcomes than standard mid-treatment MRI scans in this study.
Being able to identify patients who are not responding well at an early stage could allow clinicians to adjust treatment sooner or consider alternative approaches. These findings may also support future strategies to better tailor treatment intensity to individual patients.
The study also compared two types of PET tracers, FDG and FLT, to determine which was most suitable. While both met the study’s technical criteria, FDG-PET was selected for further evaluation due to its better image quality, greater consistency and wider use in clinical practice.
The research also explored how imaging changes after just one cycle of chemotherapy relate to the body’s immune response to treatment. Biopsies taken before and after the first cycle of chemotherapy showed that an increase in immune cells within the tumour was strongly associated with both early PET changes and improved treatment outcomes.
The researchers emphasise that these findings now need to be validated in larger studies. Future work will aim to confirm these results in broader patient groups and explore more accessible imaging approaches, such as ultrasound, alongside PET and MRI.
Sheeba Irshad, professor of cancer immunology at King’s College London and lead of the Breast Cancer Now KCL Research Unit, said:
“In patients who had PET scans both before treatment and after the first cycle, we found that this early scan could predict whether they were likely to achieve a complete response by the end of treatment. These findings highlight the potential of early imaging to guide treatment decisions, and now need to be validated in larger, modern clinical trials.”
Andrew Tutt, professor of breast oncology at The Institute of Cancer Research, London, said:
“Research that helps us determine early who is already benefitting from standard neoadjuvant chemotherapy and who might benefit from clinical trials to find better treatments is vital. This study shows that FDG-PET may have great value in this regard. We hope to be able to design studies that further investigate and validate these findings.”
The study was supported by funding from King’s College London and Guy’s and St Thomas’ NHS Foundation Trust, Breast Cancer Now, Cancer Research UK, and Guy’s and St Thomas’ Charity.
Only one per cent of women can name all five gynaecological cancers, new research suggests, as 21 women in the UK die every day of the diseases.
The report also found that 31 per cent of women have put off or avoided seeking medical advice for gynaecological symptoms.
It also found that 43 per cent of women invited for cervical screening said barriers had put them off attending, while 18 per cent of respondents aged 25 to 34 who had been invited had never attended.
The five main gynaecological cancers are womb, also called uterine, ovarian, cervical, vulval and vaginal cancer.
The Lady Garden Foundation said that, while progress has been made since the UK government’s 2022 Women’s Health Strategy aimed to improve gynaecological cancer care, significant challenges remain.
John Butler, medical director and trustee at the Lady Garden Foundation, said: “The fact that only one per cent of the population can name the diseases that directly affect half of us underscores a significant awareness gap, impacting individuals’ ability to recognise vital signs and symptoms or seek timely medical help.
“Addressing this isn’t just about awareness; it’s a critical public health priority. Our collective efforts are essential to ensure the latest commitments announced by this government translate into tangible change that saves lives.”
The report said key reasons for delaying medical advice included difficulty making appointments, embarrassment and, for cervical screening, fear of pain or previous bad experiences.
Women also reported challenges within healthcare interactions, including feeling “not taken seriously”, “dismissed” or “not believed” when seeking gynaecological advice.
Jenny Halpern Prince, chief executive and charity co-founder, said: “We frequently hear reports of women feeling ‘not taken seriously,’ ‘dismissed,’ or ‘not believed’ when seeking gynaecological advice.
“These experiences highlight crucial areas where we can improve patient support and trust within our healthcare system, ensuring women receive the empathetic and effective care they need.”
The Lady Garden Foundation said it aims to increase awareness of both the charity and the five gynaecological cancers.
It also aims to serve as a primary entry point for reliable, stigma-free information, helping people understand their bodies, recognise symptoms and overcome barriers to accessing care.
Its Silent No More Garden was unveiled at the RHS Chelsea Flower Show 2026. Designed by Darren Hawkes, the garden serves as a national call to action, using five sculptures to spark conversations, break long-standing taboos and encourage open dialogue about symptoms and preventative care.
Butler said: “Continued focus and collaborative action are essential to progress.
“The ongoing commitment from the government, alongside societal efforts to break down taboos surrounding gynaecological health, are crucial.
“The Lady Garden Foundation is dedicated to being a beacon of information and support, empowering women with the knowledge they need. We urge everyone to learn the signs, speak up, and help us save lives.”
AI could transform ovarian care by personalising cancer and fertility treatment, but more clinical validation is needed before routine use.
A systematic review and meta-analysis found AI models showed high diagnostic accuracy for ovarian cancer when combining data such as ultrasound scans and blood test results.
Across 81 studies, AI models correctly identified ovarian cancer in around nine out of 10 cases, with pooled rates of 89 to 94 per cent.
They were also highly accurate at ruling out ovarian cancer when it was not present, with specificity of 85 to 91 per cent.
The analysis also found that explainable AI tools could predict complete surgical cytoreduction in advanced ovarian cancer.
Complete surgical cytoreduction means removing all visible cancer during surgery, which can be an important goal in treatment planning.
The tools achieved a pooled AUC of 0.87. AUC is a measure of how well a model distinguishes between different outcomes, with higher scores showing stronger performance.
In reproductive medicine, AI algorithms helped physicians optimise ovarian stimulation protocols and predict follicular growth during IVF.
Ovarian stimulation is the use of hormones to encourage the ovaries to produce eggs, while follicles are the small sacs in the ovaries where eggs develop.
The review found AI could reliably model ovarian response in IVF with a pooled AUC of 0.81.
However, researchers said challenges remain in translating promising research findings into routine clinical practice.
They identified substantial variation across studies, driven by retrospective study designs, variable AI systems and a lack of standardised validation.
Only 22 per cent of analysed studies reported prospective, multicentre external validation, where models are tested forward in time across multiple healthcare settings.
The authors called for rigorous validation to help close the gap between research and routine clinical practice, alongside standardised methodological and reporting frameworks, smooth integration with clinical workflow and robust governance to support responsible and ethical AI use.
They concluded: “Artificial intelligence is a transformative force in the management of ovarian conditions.
“In gynaecologic oncology, AI enhances every phase of care, from early detection and accurate diagnosis to prognostic stratification and surgical planning.”
In reproductive medicine, AI personalises ovarian stimulation and refines the diagnosis of heterogenous endocrine disorders such as PCOS.
PCOS, or polycystic ovary syndrome, is a hormonal condition that can affect periods, skin, weight and fertility.
News4 weeks agoWomen’s digital health market set to reach US$5.28 billion in 2026 – report
Insight4 weeks agoWhy the UK’s fertility rate keeps falling – and what it means if you’re trying now
Events4 weeks agoWomen’s HealthX unveils Northwell Health, Corewell Health, Biogen & more to headline Chronic Disease stage
Opinion3 weeks agoWhat Maternal Mental Health Month reveals about where postpartum support actually breaks down
Fertility4 weeks agoToxins and climate harms having ‘alarming’ effect on fertility, research warns
News3 weeks agoNIH Grant terminations disproportionately impact minority scientists, research finds
Events2 weeks agoWUKA brings Period-Positive Pool Party to London Aquatics Centre to keep girls swimming through puberty
Fertility4 weeks agoResearcher explores weight loss jab impact on PCOS