Add us as a preferred source
This month’s medical research from British universities – from AI use in cancer therapy to toxic waste build-up in the brain.
AI reveals cancer therapy’s effectiveness
Artificial intelligence developed by researchers at UCL has shown that a combination of cancer treatments could significantly improve survival rates for patients with locally advanced rectal cancer. Adding the chemotherapy drug irinotecan to standard chemoradiotherapy substantially boosts outcomes for individuals whose tumours contain a high density of cancerous cells.
The effectiveness of this combined approach was uncovered using an AI model trained to measure cancer cell density on biopsy samples taken at diagnosis. For patients identified as having high tumour cell concentrations before starting treatment, adding irinotecan reduced the risk of cancer recurrence by approximately 43% and halved the overall risk of death over a five-year period compared to standard therapy alone. Patients with lower concentrations of cancer cells, however, saw no added benefit, highlighting the importance of precise patient selection.
Because combining treatments can intensify severe side effects – such as debilitating diarrhoea and low white blood cell counts – clinicians avoid prescribing aggressive regimens without clear evidence of benefit. Manually assessing cell density across large sample sizes has historically been too time-consuming for clinical workflows. To address this, the UCL team automated the process and launched an online tool called Octopath, which allows clinicians to upload digital biopsy slides for rapid automated analysis.
“While the original trial showed little benefit from adding irinotecan, by using artificial intelligence we found that we could distinguish patients who actually benefitted from those who did not. This demonstrates how AI can reveal tumour biology that is difficult to measure consistently by conventional means, and hopefully can be used to reveal additional insights that could lead to future treatments,” said Zhuoyan Shen, lead author from UCL Medical Physics & Biomedical Engineering.
Clinical test can predict risk in patients with rare blood cancer
Researchers at King’s College London have developed a clinical risk scoring system that can predict disease progression in patients with myelodysplastic neoplasms (MDS), a rare form of bone marrow cancer. A routine diagnostic test called flow cytometry can be leveraged to assess a patient’s risk of mortality or transformation into acute myeloid leukaemia.
MDS occurs primarily in older adults when mutated blood stem cells impair normal blood production, leading to severe anaemia, frequent infections, and bleeding. While advanced molecular testing is increasingly used to assess disease trajectory, these tests are expensive, take days to complete and remain unavailable in many healthcare centres. To overcome these barrier limitations, researchers analysed flow cytometry data from over 500 MDS patients across the UK, Germany and the Netherlands. By applying machine learning models to 55 initial measurements – incorporating immune cell variables alongside abnormal blood stem cell markers – the team successfully streamlined the scoring system down to just six essential parameters.
The final six-parameter score strongly predicted patient survival and outperformed existing flow cytometry frameworks. By accounting for complex immune system interactions within the bone marrow, the model allows clinicians to capture subtle biological variations between patients and deliver refined risk stratification at diagnosis more accurately.
“Molecular testing has transformed how we assess risk in MDS, but it is not equally accessible everywhere. Our findings suggest that a simple flow cytometry-based score, using a test already performed routinely in many laboratories, could provide rapid and clinically useful prognostic information at the time of diagnosis,” said Shahram Kordasti, professor of systems cancer immunology at King’s College London.
Potential new therapeutic targets for preeclampsia
Unusual cellular behaviour seen in both maternal and foetal tissues could offer novel therapeutic targets for preeclampsia, according to a study led by researchers at UCL and University College London Hospitals (UCLH).
The research suggests that a combination of stressed placental cells, impaired blood vessel function and an overactive immune response drives the severe pregnancy complication, which affects 2% to 4% of pregnancies globally.
Preeclampsia is a primary cause of maternal and foetal mortality, causing high blood pressure that restricts blood flow to the foetus and threatens life if left untreated. Using advanced genomic testing, the team analysed individual cells across 20 pregnant women – comparing healthy controls with severe preeclampsia cases. Researchers examined the placenta alongside surrounding tissues where maternal and foetal cells interact, including the myometrium and chorioamniotic membranes. In preeclamptic pregnancies resulting in premature birth during the third trimester, placental cells exhibited signs of stress, low oxygen and abnormal energy usage. Crucially, the analysis uncovered hyperactive immune responses and blood vessel dysfunction extending beyond the placenta into adjacent tissues and maternal blood, explaining why the disease impacts the entire body.
Because there is currently no cure, identifying these distinct biological processes opens up opportunities for targeted, early-stage treatments to prevent premature births and safeguard maternal health.
“We studied individual cells from both the mother and the baby to see how their activity changes in healthy pregnancies compared with preeclampsia. This helped us to confirm some changes already suspected in the condition and also discover new ones,” said senior author Sara Hillman, associate professor at UCL EGA Institute for Women’s Health.
World-first perfused pancreas transplant
The University of Oxford has carried out the world’s first successful transplant of a perfused human pancreas. This addresses a bottleneck in diabetes treatment where up to half of donated pancreases are currently discarded due to damage sustained during standard ice-box storage.
The Hypothermic Oxygenated Pancreas Perfusion (HOPP) study, conducted at Oxford University Hospitals NHS Foundation Trust, is evaluating a preservation technique that circulates oxygenated cold fluid through donor organs instead of relying on cold storage. Because the pancreas is delicate, traditional cold preservation slows deterioration but cannot prevent cellular decay. Pumping cold, oxygen-enriched fluid through the organ supplies oxygen while removing toxic metabolic by-products and protects tissue integrity before surgery. In the trial, each donor pancreas is perfused for two hours while the recipient’s simultaneous kidney transplant is underway, after which surgical teams complete the pancreas transplant as normal.
Researchers will compare clinical outcomes across 30 trial participants against a historical control group of 60 patients whose organs were stored using conventional static methods. By measuring graft function, post-transplant complications, and hospital stays, the team aims to prove that oxygenated perfusion reduces organ injury and significantly expands the pool of usable donor organs.
“It is an exciting time to be part of this research project, and although it is early days, having such a positive outcome for the first patient is very promising. Without the generosity of families who support donation, this study would not be possible, and lifesaving transplants would not happen,” said James Hunter, chief investigator and associate professor at the University of Oxford’s Nuffield Department of Surgical Sciences.
Hormone differences point to endometriosis blood test
Researchers at the University of Edinburgh have developed a promising non-invasive alternative to treat endometriosis by looking at the pattern of their blood hormones.
Endometriosis occurs when tissue similar to the womb lining grows elsewhere in the body. While the condition has traditionally been viewed as driven exclusively by the female hormones oestrogen and progesterone, the research team examined blood samples from 159 women with confirmed endometriosis and 57 healthy controls to analyse the role of androgens – so-called male hormones. They found a distinct hormone fingerprint in patients with the condition, characterised by elevated levels of an adrenal-derived androgen known as 11-ketotestosterone. Using this specific biomarker signature, researchers identified more than 95% of endometriosis patients.
By establishing the role of adrenal androgens in the disease’s development, the breakthrough provides a dual benefit: laying the foundation for a rapid blood test while pointing toward novel non-hormonal treatment targets. Working alongside Edinburgh Innovations – the university’s commercialisation service – the team is now seeking industry partners to help develop and scale the diagnostic test for clinical use.
“These findings mark a significant breakthrough in our understanding of endometriosis. Traditionally seen as an oestrogen-driven disorder, our research challenges this view by showing different androgen levels in the condition. We are optimistic that this new insight will lead to earlier diagnosis and the development of innovative new treatments for those affected by endometriosis,” said study lead Douglas Gibson from the Centre for Reproductive Health at the University of Edinburgh.
Novel tech could help targeted lung cancer treatment
Researchers at the University of Edinburgh and NHS Lothian have developed an imaging and artificial intelligence platform that predicts key genetic mutations in lung cancer directly from untreated biopsy tissue.
The study shows that fluorescence lifetime imaging microscopy (FLIM) combined with AI algorithms can accurately identify epidermal growth factor receptor (EGFR) mutations without conventional laboratory gene sequencing or tissue staining.
Lung cancer remains the leading cause of cancer-related mortality globally and specific genetic alterations such as EGFR mutations determine whether patients respond to targeted therapies. By analysing natural light signals emitted by untreated tissue samples, the non-destructive FLIM approach predicts EGFR status with high precision and successfully differentiates between the two most common mutation variants that guide clinical decisions. Crucially, because the technology leaves biopsy material intact, samples remain preserved for further analysis.
As expanded national screening programmes identify higher volumes of early-stage lung cancers, pathology services face pressure to deliver rapid diagnostic results from smaller tissue yields. The research team is now working toward clinical validation and extend the imaging platform to other cancer types.
“This approach has the potential to take processes that currently cost thousands of pounds and require weeks of lab work and reduce them to something that takes minutes and costs hundreds. That is a step change in what is clinically achievable, particularly for centres and health systems where access to complex molecular testing is limited,” said study co-lead Qiang Wang from the Institute for Regeneration and Repair at the University of Edinburgh.
An unexpected role for a fertility protein in cancer
Researchers at the University of Liverpool have found that a protein previously believed to function solely during the production of sperm and eggs can be reactivated in cancer cells to help tumours survive and grow. The fertility protein SYCP1 is hijacked by cancer cells to perform an entirely different role than its standard reproductive function.
Instead of assisting chromosomes to pair during meiosis, SYCP1 enters the nucleus of cancer cells, binds directly to DNA, and controls genes responsible for cell division and DNA repair. Researchers found that removing SYCP1 made cancer cells significantly more sensitive to chemotherapy drugs that induce DNA damage. This suggests that tumours exploit the re-expressed protein to repair treatment-induced damage, allowing them to resist therapies and continue proliferating.
The findings challenge the long-held assumption that fertility-specific proteins are biologically irrelevant outside the reproductive system. By demonstrating how cancers evolve by repurposing developmental programmes, the study positions SYCP1 as a promising target for future precision cancer treatments.
“Our findings show that cancer cells can hijack proteins that normally exist only in reproductive tissues and give them completely new jobs. Understanding these unexpected functions opens up exciting opportunities to develop new treatments that make existing cancer therapies more effective,” explained Urszula McClurg, lecturer in biochemistry, cell and systems biology at the University of Liverpool.
Toxic waste’ build-up in the brain may be driving dementia and ALS
Scientists at The University of Manchester have found that a build-up of a common body waste product in the brain could be helping to drive two neurological conditions. The study identified unusually high levels of urea – a waste chemical normally excreted in urine – in post-mortem brain tissue from individuals with frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS).
FTD affects areas of the brain linked to behaviour and personality, while ALS attacks nerve cells controlling movement. Up to 15% of patients with ALS also develop FTD, and both conditions are incurable and fatal. Using sensitive laboratory techniques, the team measured urea levels across different brain regions in post-mortem tissue donated for research. In FTD cases, elevated urea was detected widely across both severely damaged and relatively spared brain regions. In contrast, the waste build-up in ALS was mainly concentrated in regions controlling motor function.
The findings build on previous work by the research group showing similar urea accumulation in five other neurodegenerative conditions, including Alzheimer’s and Parkinson’s-related dementia. The results suggest the brain may struggle to eliminate waste properly, potentially poisoning nerve cells over time and revealing a shared underlying mechanism across multiple brain diseases.
“This is exciting because it suggests we might be looking at a common problem underlying several different brain diseases, rather than separate conditions with completely different causes. If we can work out why this waste is building up and how to clear it, we may be able to slow or even stop these diseases, opening the door to new treatments for conditions that currently have very few options,” said Sasha Philbert, lead author and Alzheimer’s Society postdoctoral fellow at the University of Manchester.
