̽��ֱ�� of Cambridge - Wellcome

Mouse study suggests a common diabetes drug may prevent leukaemia

cjb250 — Wed, 16 Apr 2025 07:59:00 +0000

Around 3,100 people are diagnosed with acute myeloid leukaemia (AML) each year in the UK. It is an aggressive form of blood cancer that is very difficult to treat. Thanks to recent advances, individuals at high risk of AML can be identified years in advance using blood tests and blood DNA analysis, but there’s no suitable treatment that can prevent them from developing the disease.

In this study, Professor George Vassiliou and colleagues at the ̽��ֱ�� of Cambridge investigated how to prevent abnormal blood stem cells with genetic changes from progressing to become AML. ̽��ֱ��work focused on the most common genetic change, which affects a gene called DNMT3A and is responsible for starting 10-15% of AML cases.

Professor Vassiliou, from the Cambridge Stem Cell Institute at the ̽��ֱ�� of Cambridge and Honorary Consultant Haematologist at Cambridge ̽��ֱ�� Hospitals NHS Foundation Trust (CUH) co-led the study. He said: “Blood cancer poses unique challenges compared to solid cancers like breast or prostate, which can be surgically removed if identified early. With blood cancers, we need to identify people at risk and then use medical treatments to stop cancer progression throughout the body.”

̽��ֱ��research team examined blood stem cells from mice with the same changes in DNMT3A as seen in the pre-cancerous cells in humans. Using a genome-wide screening technique, they showed that these cells depend more on mitochondrial metabolism than healthy cells, making this a potential weak spot. ̽��ֱ��researchers went on to confirm that metformin, and other mitochondria-targeting drugs, substantially slowed the growth of mutation-bearing blood cells in mice. Further experiments also showed that metformin could have the same effect on human blood cells with the DNMT3A mutation.

Dr Malgorzata Gozdecka, Senior Research Associate at the Cambridge Stem Cell Institute and first author of the research said: “Metformin is a drug that impacts mitochondrial metabolism, and these pre-cancerous cells need this energy to keep growing. By blocking this process, we stop the cells from expanding and progressing towards AML, whilst also reversing other effects of the mutated DNMT3A gene.”

In addition, the study looked at data from over 412,000 UK Biobank volunteers and found that people taking metformin were less likely to have changes in the DNMT3A gene. This link remained even after accounting for factors that could have confounded the results such as diabetes status and BMI.

Professor Brian Huntly, Head of the Department of Haematology at the ̽��ֱ�� of Cambridge, Honorary Consultant Haematologist at CUH, and joint lead author of the research, added: “Metformin appears highly specific to this mutation rather than being a generic treatment. That specificity makes it especially compelling as a targeted prevention strategy.

“We’ve done the extensive research all the way from cell-based studies to human data, so we’re now at the point where we have a made a strong case for moving ahead with clinical trials. Importantly, metformin’s lack of toxicity will be a major advantage as it is already used by millions of people worldwide with a well-established safety profile.”

̽��ֱ��results of the study, funded by Blood Cancer UK with additional support from Cancer Research UK, the Leukemia & Lymphoma Society (USA) and the Wellcome Trust, are published in Nature.

Dr Rubina Ahmed, Director of Research at Blood Cancer UK, said: “Blood cancer is the third biggest cancer killer in the UK, with over 280,000 people currently living with the disease. Our Blood Cancer Action plan shed light on the shockingly low survival for acute myeloid leukaemia, with only around 2 in 10 surviving for 5 years, and we urgently need better strategies to save lives. Repurposing safe, widely available drugs like metformin means we could potentially get new treatments to people faster, without the need for lengthy drug development pipelines.”

̽��ֱ��next phase of this research will focus on clinical trials to test metformin’s effectiveness in people with changes in DNMT3A at increased risk of developing AML. With metformin already approved and widely used for diabetes, this repurposing strategy could dramatically reduce the time it takes to bring a new preventive therapy to patients.

Tanya Hollands, Research Information Manager at Cancer Research UK, who contributed funding for the lab-based screening in mice, said: “It's important that we work to find new ways to slow down or prevent AML in people at high risk. Therefore, it’s positive that the findings of this study suggest a possible link between a commonly-used diabetes drug and prevention of AML progression in some people. While this early-stage research is promising, clinical trials are now needed to find out if this drug could benefit people. We look forward to seeing how this work progresses.”

Reference
Gozdecka, M et al. Mitochondrial metabolism sustains DNMT3A-R882-mutant clonal haematopoiesis. Nature; 16 Apr 2025; DOI: 10.1038/s41586-025-08980-6

Adapted from a press release from Blood Cancer UK

Metformin, a widely used and affordable diabetes drug, could prevent a form of acute myeloid leukaemia in people at high risk of the disease, a study in mice has suggested. Further research in clinical trials will be needed to confirm this works for patients.

We’ve done the extensive research all the way from cell-based studies to human data, so we’re now at the point where we have a made a strong case for moving ahead with clinical trials

Brian Huntly

̽��ֱ�� of Cambridge

Brown lab mouse on blue gloved hand

̽��ֱ��text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

Genetic study reveals hidden chapter in human evolution

sc604 — Tue, 18 Mar 2025 10:00:00 +0000

Using advanced analysis based on full genome sequences, researchers from the ̽��ֱ�� of Cambridge have found evidence that modern humans are the result of a genetic mixing event between two ancient populations that diverged around 1.5 million years ago. About 300,000 years ago, these groups came back together, with one group contributing 80% of the genetic makeup of modern humans and the other contributing 20%.

For the last two decades, the prevailing view in human evolutionary genetics has been that Homo sapiens first appeared in Africa around 200,000 to 300,000 years ago, and descended from a single lineage. However, these latest results, reported in the journal Nature Genetics, suggest a more complex story.

“ ̽��ֱ��question of where we come from is one that has fascinated humans for centuries,” said first author Dr Trevor Cousins from Cambridge’s Department of Genetics. “For a long time, it’s been assumed that we evolved from a single continuous ancestral lineage, but the exact details of our origins are uncertain.”

“Our research shows clear signs that our evolutionary origins are more complex, involving different groups that developed separately for more than a million years, then came back to form the modern human species,” said co-author Professor Richard Durbin, also from the Department of Genetics.

While earlier research has already shown that Neanderthals and Denisovans – two now-extinct human relatives – interbred with Homo sapiens around 50,000 years ago, this new research suggests that long before those interactions – around 300,000 years ago – a much more substantial genetic mixing took place. Unlike Neanderthal DNA, which makes up roughly 2% of the genome of non-African modern humans, this ancient mixing event contributed as much as 10 times that amount and is found in all modern humans.

̽��ֱ��team’s method relied on analysing modern human DNA, rather than extracting genetic material from ancient bones, and enabled them to infer the presence of ancestral populations that may have otherwise left no physical trace. ̽��ֱ��data used in the study is from the 1000 Genomes Project, a global initiative that sequenced DNA from populations across Africa, Asia, Europe, and the Americas.

̽��ֱ��team developed a computational algorithm called cobraa that models how ancient human populations split apart and later merged back together. They tested the algorithm using simulated data and applied it to real human genetic data from the 1000 Genomes Project.

While the researchers were able to identify these two ancestral populations, they also identified some striking changes that happened after the two populations initially broke apart.

“Immediately after the two ancestral populations split, we see a severe bottleneck in one of them—suggesting it shrank to a very small size before slowly growing over a period of one million years,” said co-author Professor Aylwyn Scally, also from the Department of Genetics. “This population would later contribute about 80% of the genetic material of modern humans, and also seems to have been the ancestral population from which Neanderthals and Denisovans diverged.”

̽��ֱ��study also found that genes inherited from the second population were often located away from regions of the genome linked to gene functions, suggesting that they may have been less compatible with the majority genetic background. This hints at a process known as purifying selection, where natural selection removes harmful mutations over time.

“However, some of the genes from the population which contributed a minority of our genetic material, particularly those related to brain function and neural processing, may have played a crucial role in human evolution,” said Cousins.

Beyond human ancestry, the researchers say their method could help to transform how scientists study the evolution of other species. In addition to their analysis of human evolutionary history, they applied the cobraa model to genetic data from bats, dolphins, chimpanzees, and gorillas, finding evidence of ancestral population structure in some but not all of these.

“What’s becoming clear is that the idea of species evolving in clean, distinct lineages is too simplistic,” said Cousins. “Interbreeding and genetic exchange have likely played a major role in the emergence of new species repeatedly across the animal kingdom.”

So who were our mysterious human ancestors? Fossil evidence suggests that species such as Homo erectus and Homo heidelbergensis lived both in Africa and other regions during this period, making them potential candidates for these ancestral populations, although more research (and perhaps more evidence) will be needed to identify which genetic ancestors corresponded to which fossil group.

Looking ahead, the team hopes to refine their model to account for more gradual genetic exchanges between populations, rather than sharp splits and reunions. They also plan to explore how their findings relate to other discoveries in anthropology, such as fossil evidence from Africa that suggests early humans may have been far more diverse than previously thought.

“ ̽��ֱ��fact that we can reconstruct events from hundreds of thousands or millions of years ago just by looking at DNA today is astonishing,” said Scally. “And it tells us that our history is far richer and more complex than we imagined.”

̽��ֱ��research was supported by Wellcome. Aylwyn Scally is a Fellow of Darwin College, Cambridge. Trevor Cousins is a member of Darwin College, Cambridge.

Reference:
Trevor Cousins, Aylwyn Scally & Richard Durbin. ‘A structured coalescent model reveals deep ancestral structure shared by all modern humans.’ Nature Genetics (2025). DOI: 10.1038/s41588-025-02117-1

Modern humans descended from not one, but at least 2 ancestral populations that drifted apart and later reconnected, long before modern humans spread across the globe.

Our history is far richer and more complex than we imagined

Aylwyn Scally

Jose A Bernat Bacete via Getty Images

Plaster reconstructions of the skulls of human ancestors

Yes

Prioritise vaccine boosters for vulnerable immunocompromised patients, say scientists

cjb250 — Wed, 12 Feb 2025 19:00:46 +0000

̽��ֱ��findings, published today in Science Advances, suggest that such individuals will need regular vaccine boosters to protect them and reduce the risk of infections that could be severe and also lead to new ‘variants of concern’ emerging.

Almost 16 million people worldwide are estimated to have died from Covid-19 during 2020 and 2021, though nearly 20 million deaths are thought to have been prevented as a result of the rapid rollout of vaccines against SARS-CoV-2, the virus that caused the pandemic.

During the pandemic, researchers discovered that immunocompromised individuals had difficulty clearing the virus, even when vaccinated. These are people whose immune systems are not functioning correctly, either as a direct result of disease or because they are on medication to dampen down their immune systems, for example to prevent organ transplant rejection. This meant that their infections lasted longer, giving the virus more opportunities to mutate.

Research from early in the pandemic showed that chronic infections can give rise to variants of concern that can then cause new waves of infection in the wider population.

When an individual is vaccinated, their immune systems produce antibodies that recognise and launch an attack on the virus. Such a process is known as seroconversion. Additional ‘booster’ vaccinations increase seroconversion and hence the likelihood of clearing infection.

However, although most immunocompromised individuals will have received three or more doses of the Covid-19 vaccine, they still account for more than a fifth of hospitalisations, admissions to intensive care units, and overall deaths associated with the disease.

To see why this is the case, scientists at the Cambridge Institute of Therapeutic Immunology and Infectious Disease (CITIID) at the ̽��ֱ�� of Cambridge examined immunocompromised individuals who had been vaccinated against Covid-19. These patients, recruited from Cambridge ̽��ֱ�� Hospitals NHS Foundation Trust, were living with vasculitis, a group of disorders that cause inflammation of blood vessels. Data from this group was compared against individuals who were not immunocompromised.

Treatments for vasculitis rely on immunosuppressant medicines. These include drugs such as rituximab, which depletes the number of B-cells in the body – but B-cells are the immune cells responsible for producing antibodies. As such, these individuals are a severely at-risk population.

When the researchers analysed bloods samples from the vasculitis patients, they found that even though vaccination induced seroconversion, this in itself was not always sufficient to neutralise the virus. Every immunocompromised individual required at least three doses of the vaccine to protect them across a range of variants up to and include Omicron (the variant that appeared towards the end of 2021 and caused a new wave of infections). In some cases, even four vaccinations were not sufficient to adequately protect them.

Kimia Kamelian, a Gates Cambridge Scholar at CITIID and St Edmund's College, Cambridge, said: “We know that immunocompromised individuals are particularly vulnerable to diseases such as Covid-19 because their immune systems struggle to clear infections. Vaccinations offer some protection, but our study shows that only repeated vaccinations – often four or more – offer the necessary protection.”

Professor Ravi Gupta, also from CITIID and a Fellow at Homerton College, Cambridge, added: “This of course has implications for the individual, who is more likely to have prolonged infection and a much greater risk of severe infection, but it also gives the virus multiple opportunities to mutate.

“We know from our previous work that at least some of the variants of concern probably emerged during chronic infections. That’s why these individuals must be given priority for updated vaccines against new variants.”

̽��ֱ��research was funded by Wellcome, Gates Cambridge, Addenbrooke’s Charitable Trust and Vasculitis UK, with additional support by the National Institute for Health and Care Research Cambridge Biomedical Research Centre.

Reference
Kamelian, K et al. Humoral responses to SARS-CoV-2 vaccine in vasculitis-related immune suppression. Sci Adv; 12 Feb 2025; DOI: 10.1126/sciadv.adq3342

Vaccinations alone may not be enough to protect people with compromised immune systems from infection, even if the vaccine has generated the production of antibodies, new research from the ̽��ֱ�� of Cambridge has shown.

We know that immunocompromised individuals are particularly vulnerable to diseases such as Covid-19 because their immune systems struggle to clear infections

Kimia Kamelian

NoSystem images

Vaccination of an senior male

Yes

Monoclonal antibodies offer hope for tackling antimicrobial resistance

cjb250 — Mon, 16 Sep 2024 10:31:11 +0000

A team lead by researchers at the ̽��ֱ�� of Cambridge has developed a monoclonal antibody drug, using a technique involving genetically engineered mice, that may help prevent infection from Acinetobacter baumannii, a bacteria associated with hospital-acquired infections, which is particularly common in Asia.

A. baumannii bacteria can cause life-threatening respiratory illness and sepsis in vulnerable individuals, particularly in newborn babies whose immune systems have not fully developed. It is usually spread through contaminated surfaces, medical equipment and via contact with others. In recent years infections with strains of this bacteria that are resistant to almost every antibiotic available have become common.

Professor Stephen Baker from the Cambridge Institute of Therapeutic Immunology and Infectious Disease at the ̽��ֱ�� of Cambridge said “A. baumannii is good at sticking to medical equipment, and if people are vulnerable or don't have a particularly well-developed immune system, they can succumb to this infection and get aggressive pneumonia requiring ventilation – and in many cases, the patients can acquire the infection from the ventilation itself.

“ ̽��ֱ��bacteria are naturally resistant to many antimicrobials, but as they’re now found in hospitals, they’ve acquired resistance to almost everything we can use. In some hospitals in Asia, where the infections are most common, there isn't a single antibiotic that will work against them. They’ve become impossible to treat.”

In a study published today in Nature Communications, the team produced monoclonal antibodies using transgenic mice – mice that have been genetically-engineered to have a human-like immune system, producing human antibodies instead of mouse antibodies. They went on to show that these monoclonal antibodies were able prevent infection with A. baumannii derived from clinical samples.

Monoclonal antibodies are a growing area of medicine, commonly used to treat conditions including cancer (for example, Herceptin for treating some breast cancers) and autoimmune disease (for example, Humira for treating rheumatoid arthritis, psoriasis, Crohn's disease, and ulcerative colitis).

Usually, monoclonal antibodies are developed from the antibodies of patients who have recovered from an infection, or they are designed to recognise and target a particular antigen. For example, monoclonal antibodies targeting the ‘spike protein’ of the SARS-CoV-2 coronavirus were explored as a way of treating COVID-19.

In the approach taken by the Cambridge team, however, transgenic mice were exposed to the outer membrane of A. baumannii bacteria, triggering an immune response. ̽��ֱ��researchers then isolated almost 300 different antibodies and tested which of these was the most effective at recognising live bacteria, identifying the single monoclonal antibody mAb1416 as the best.

Professor Baker said: “Using this method, we don't infect the mice with the live bacteria, but we instead immunise them using multiple different elements and let the mouse’s immune system work out which ones to develop antibodies against. Because these mice have ‘humanised’ immune systems, we wouldn’t then need to reengineer the antibodies to work in humans.”

̽��ֱ��team treated mice with mAb1416, and 24 hours later exposed them to A. baumannii isolated from a child with sepsis in an intensive care unit. They found that those mice treated with the drug saw a significant reduction in bacterial load in their lungs a further 24 hours later, compared to mice that were not treated.

All of the isolates used to produce and test the monoclonal antibodies were from patients in Ho Chi Minh City, Vietnam, but the isolate used to test mAb1416 was taken from a patient ten years later than the other isolates. This is important because it shows that mAb1416 was protective against A. baumannii bacteria that may have evolved over time.

Professor Baker said: “Using this technique, you can take any bacterial antigen or cocktail of antigens, rather than waiting for somebody that's recovered from a particular infection – who you assume has developed an appropriate antibody response – give it to the mice and extract the antibodies you think are the most important.”

More work is now needed to understand the mechanism by which mAb1416 protects against infection, as this could allow the team to develop an even more effective treatment. Any potential new drug will then need to be tested in safety trials in animals before being trialled in patients.

Professor Baker added: “We know that monoclonal antibodies are safe and that they work, and the technology exists to produce them – what we have done is identify how to hit bacteria with them. Apart from the cost effectiveness, there's no reason why this couldn’t become a medicine within a few years. Given the emergency presented by antimicrobial resistance, this could become a powerful new weapon to fight back.”

̽��ֱ��research was funded by the Bill & Melinda Gates Foundation, the UK Medical Research Council Newton Fund, the Viet Nam Ministry of Science and Technology, and Wellcome.

Professor Baker is a fellow at Wolfson College, Cambridge.

Reference
Baker, S, Krishna, A & Higham, S. Exploiting human immune repertoire transgenic mice to identify protective monoclonal antibodies against an extensively antimicrobial resistant nosocomial bacterial pathogen. Nat Comms; 12 Sept 2024; DOI: 10.1038/s41467-024-52357-8

Monoclonal antibodies – treatments developed by cloning a cell that makes an antibody – could help provide an answer to the growing problem of antimicrobial resistance, say scientists.

We know that monoclonal antibodies are safe and that they work, and the technology exists to produce them – what we have done is identify how to hit bacteria with them

Stephen Baker

TopMicrobialStock (Getty Images)

A Petri dish with a culture of the Superbug Acinetobacter baumannii next to antibiotics

Yes

‘Diabetes distress’ increases risk of mental health problems among young people living with type 1 diabetes

cjb250 — Wed, 17 Jul 2024 09:00:45 +0000

̽��ֱ��findings highlight the urgent need for monitoring and support for the mental health of young people diagnosed with type 1 diabetes.

According to the charity JDRF, there are 8.7 million people living with type 1 diabetes around the world, including over 400,000 people in the UK. It is a chronic, life-threatening condition, usually diagnosed in childhood, that has a life-long impact.

Currently, people with type 1 diabetes rely on a routine of finger-prick blood tests and insulin injections or infusions, because their pancreas no longer produces insulin itself, although recent developments in artificial pancreas technology are helping transform this care.

Previous studies have shown potential links between childhood-onset type 1 diabetes and a number of mental health disorders in adulthood. However, it is not clear whether these links can be best explained by the impacts of living with the condition and its treatment, or whether underlying common biological mechanisms may be implicated, for example the impact of unstable blood sugar levels on the developing adolescent brain.

To help answer this question, a team of researchers turned to data from over 4,500 children with type 1 diabetes on a national register in the Czech Republic and from large-scale European DNA studies.Their findings are published today in Nature Mental Health.

From the national register data, the researchers found that children diagnosed with type 1 diabetes – compared to children without the condition – were over twice as likely to develop a mood disorder and more than 50% more likely to develop an anxiety disorder. They were also more than four times more likely to develop behavioural syndromes including eating and sleep disorders

Conversely, children with type 1 diabetes were at a much lower risk of developing psychotic disorders, such as schizophrenia – almost half the risk compared to their peers.

̽��ֱ��findings are consistent with the results from two other national register studies in Sweden and in Denmark, suggesting that the results would likely apply to other countries, too, including the UK.

̽��ֱ��team used a statistical technique known as Mendelian Randomisation to probe causal links between type 1 diabetes and these various psychiatric disorders, but found little evidence in support of a common underlying biological mechanism.

Tomáš Formánek, a PhD student at the ̽��ֱ�� of Cambridge and the National Institute of Mental Health, Klecany, Czech Republic, said: “Although we found a concerning increase in the risk of mental health problems among people living with type 1 diabetes, our study – and others before it – suggests this is unlikely to be the result of common biological mechanisms. This emphasises the importance of prevention and sustained attention to the mental health needs of children and young people with type 1 diabetes.”

̽��ֱ��researchers say that mental health problems in later life may be a result of children with type 1 diabetes being forced to make significant changes to their lives, with a constant focus on monitoring their food intake and a need to check blood sugar levels and administer insulin injections. This often leaves these children feeling excluded from social events and singled-out by peers, teachers and even family members.

Dr Benjamin Perry from the Department of Psychiatry, ̽��ֱ�� of Cambridge, said: “We know that people diagnosed with type 1 diabetes can experience ‘diabetes distress’. This can include extreme frustration with blood sugars and feelings of isolation and can lead to burnout, hopelessness, and a feeling of lack of control. It’s little wonder, then, that they are at risk of compounding mental health problems, spanning into their adult lives.”

Professor Peter Jones, also from the Department of Psychiatry, ̽��ֱ�� of Cambridge, added: “Our findings emphasise the urgent need to support children diagnosed with type 1 diabetes, look out for signs of mental health problems and offer timely, expert help. That way, it may be possible to help these children early, before these problems fully take root.”

̽��ֱ��research was supported by the National Institute for Health and Care Research Applied Research Collaboration East of England at Cambridgeshire and Peterborough NHS Foundation Trust and the Ministry of Health, Czech Republic, with additional funding from Wellcome and the UKRI Medical Research Council.

Reference
Formánek, T et al. Childhood-Onset Type 1 Diabetes and Subsequent Adult Psychiatric Disorders: A Nationwide Cohort and Genome-wide Mendelian Randomization Study. Nature Mental Health; 17 July 2024; DOI: 10.1038/s44220-024-00280-8

Children diagnosed with type 1 diabetes are at significantly higher risk of a number of mental health issues, including mood and anxiety disorders, a study from a team in the UK and the Czech Republic has found.

We know that people diagnosed with type 1 diabetes can experience ‘diabetes distress’. It’s little wonder, then, that they are at risk of compounding mental health problems, spanning into their adult lives

Benjamin Perry

kyotokushige

Teenager With Type 1 Diabetes Takes at Home Test

Yes

Journeys of discovery: Steve Jackson and a life-saving cancer drug

lw355 — Mon, 15 Jul 2024 07:00:08 +0000

What excites Steve Jackson is understanding how biology works and why it sometimes goes wrong. But what galvanises him is knowing there are people alive today as a result of his discovery of how to create a cancer drug.

Artificial intelligence outperforms clinical tests at predicting progress of Alzheimer’s disease

cjb250 — Fri, 12 Jul 2024 22:30:52 +0000

̽��ֱ��team say this new approach could reduce the need for invasive and costly diagnostic tests while improving treatment outcomes early when interventions such as lifestyle changes or new medicines may have a chance to work best.

Dementia poses a significant global healthcare challenge, affecting over 55 million people worldwide at an estimated annual cost of $820 billion. ̽��ֱ��number of cases is expected to almost treble over the next 50 years.

̽��ֱ��main cause of dementia is Alzheimer’s disease, which accounts for 60-80% of cases. Early detection is crucial as this is when treatments are likely to be most effective, yet early dementia diagnosis and prognosis may not be accurate without the use of invasive or expensive tests such as positron emission tomography (PET) scans or lumbar puncture, which are not available in all memory clinics. As a result, up to a third of patients may be misdiagnosed and others diagnosed too late for treatment to be effective.

A team led by scientists from the Department of Psychology at the ̽��ֱ�� of Cambridge has developed a machine learning model able to predict whether and how fast an individual with mild memory and thinking problems will progress to developing Alzheimer’s disease. In research published today in eClinical Medicine, they show that it is more accurate than current clinical diagnostic tools.

To build their model, the researchers used routinely-collected, non-invasive, and low-cost patient data – cognitive tests and structural MRI scans showing grey matter atrophy – from over 400 individuals who were part of a research cohort in the USA.

They then tested the model using real-world patient data from a further 600 participants from the US cohort and – importantly – longitudinal data from 900 people from memory clinics in the UK and Singapore.

̽��ֱ��algorithm was able to distinguish between people with stable mild cognitive impairment and those who progressed to Alzheimer’s disease within a three-year period. It was able to correctly identify individuals who went on to develop Alzheimer’s in 82% of cases and correctly identify those who didn’t in 81% of cases from cognitive tests and an MRI scan alone.

̽��ֱ��algorithm was around three times more accurate at predicting the progression to Alzheimer’s than the current standard of care; that is, standard clinical markers (such as grey matter atrophy or cognitive scores) or clinical diagnosis. This shows that the model could significantly reduce misdiagnosis.

̽��ֱ��model also allowed the researchers to stratify people with Alzheimer’s disease using data from each person’s first visit at the memory clinic into three groups: those whose symptoms would remain stable (around 50% of participants), those who would progress to Alzheimer’s slowly (around 35%) and those who would progress more rapidly (the remaining 15%). These predictions were validated when looking at follow-up data over 6 years. This is important as it could help identify those people at an early enough stage that they may benefit from new treatments, while also identifying those people who need close monitoring as their condition is likely to deteriorate rapidly.

Importantly, those 50% of people who have symptoms such as memory loss but remain stable, would be better directed to a different clinical pathway as their symptoms may be due to other causes rather than dementia, such as anxiety or depression.

Senior author Professor Zoe Kourtzi from the Department of Psychology at the ̽��ֱ�� of Cambridge said: “We’ve created a tool which, despite using only data from cognitive tests and MRI scans, is much more sensitive than current approaches at predicting whether someone will progress from mild symptoms to Alzheimer’s – and if so, whether this progress will be fast or slow.

“This has the potential to significantly improve patient wellbeing, showing us which people need closest care, while removing the anxiety for those patients we predict will remain stable. At a time of intense pressure on healthcare resources, this will also help remove the need for unnecessary invasive and costly diagnostic tests.”

While the researchers tested the algorithm on data from a research cohort, it was validated using independent data that included almost 900 individuals who attended memory clinics in the UK and Singapore. In the UK, patients were recruited through the Quantiative MRI in NHS Memory Clinics Study (QMIN-MC) led by study co-author Dr Timothy Rittman at Cambridge ̽��ֱ�� Hospitals NHS Trust and Cambridgeshire and Peterborough NHS Foundation Trusts (CPFT).

̽��ֱ��researchers say this shows it should be applicable in a real-world patient, clinical setting.

Dr Ben Underwood, Honorary Consultant Psychiatrist at CPFT and assistant professor at the Department of Psychiatry, ̽��ֱ�� of Cambridge, said: “Memory problems are common as we get older. In clinic I see how uncertainty about whether these might be the first signs of dementia can cause a lot of worry for people and their families, as well as being frustrating for doctors who would much prefer to give definitive answers. ̽��ֱ��fact that we might be able to reduce this uncertainty with information we already have is exciting and is likely to become even more important as new treatments emerge.”

Professor Kourtzi said: “AI models are only as good as the data they are trained on. To make sure ours has the potential to be adopted in a healthcare setting, we trained and tested it on routinely-collected data not just from research cohorts, but from patients in actual memory clinics. This shows it will be generalisable to a real-world setting.”

̽��ֱ��team now hope to extend their model to other forms of dementia, such as vascular dementia and frontotemporal dementia, and using different types of data, such as markers from blood tests.

Professor Kourtzi added: “If we’re going to tackle the growing health challenge presented by dementia, we will need better tools for identifying and intervening at the earliest possible stage. Our vision is to scale up our AI tool to help clinicians assign the right person at the right time to the right diagnostic and treatment pathway. Our tool can help match the right patients to clinical trials, accelerating new drug discovery for disease modifying treatments.”

This work was in collaboration with a cross-disciplinary team including Professor Peter Tino at the ̽��ֱ�� of Birmingham and Professor Christopher Chen at the National ̽��ֱ�� of Singapore. It was funded by Wellcome, the Royal Society, Alzheimer’s Research UK, the Alzheimer’s Drug Discovery Foundation Diagnostics Accelerator, the Alan Turing Institute, and the National Institute for Health and Care Research Cambridge Biomedical Research Centre.

Reference
Lee, LY & Vaghari, D et al. Robust and interpretable AI-guided marker for early dementia prediction in real-world clinical settings. eClinMed; 12 July 2024; DOI: 10.1016/j.eclinm.2024.102725

Cambridge scientists have developed an artificially-intelligent tool capable of predicting in four cases out of five whether people with early signs of dementia will remain stable or develop Alzheimer’s disease.

We’ve created a tool which is much more sensitive than current approaches at predicting whether someone will progress from mild symptoms to Alzheimer’s

Zoe Kourtzi

Yuichiro Chino (Getty Images)

Brain on molecular structure, circuitry, and programming code background

Yes

AI able to identify drug-resistant typhoid-like infection from microscopy images in matter of hours

cjb250 — Mon, 08 Jul 2024 07:00:00 +0000

Antimicrobial resistance is an increasing global health issue that means many infections are becoming difficult to treat, with fewer treatment options available. It even raises the spectre of some infections becoming untreatable in the near future.

One of the challenges facing healthcare workers is the ability to distinguish rapidly between organisms that can be treated with first-line drugs and those that are resistant to treatment. Conventional testing can take several days, requiring bacteria to be cultured, tested against various antimicrobial treatments, and analysed by a laboratory technician or by machine. This delay often results in patients being treated with an inappropriate drug, which can lead to more serious outcomes and, potentially, further drive drug resistance.

In research published in Nature Communications, a team led by researchers in Professor Stephen Baker’s Lab at the ̽��ֱ�� of Cambridge developed a machine-learning tool capable of identifying from microscopy images Salmonella typhimurium bacteria that are resistant to the first-line antibiotic ciprofloxacin – even without testing the bacteria against the drug.

S. Typhimurium causes gastrointestinal illness and typhoid-like illness in severe cases, whose symptoms include fever, fatigue, headache, nausea, abdominal pain, and constipation or diarrhoea. In severe cases, it can be life threatening. While infections can be treated with antibiotics, the bacteria are becoming increasingly resistant to a number of antibiotics, making treatment more complicated.

̽��ֱ��team used high-resolution microscopy to examine S. Typhimurium isolates exposed to increasing concentrations of ciprofloxacin and identified the five most important imaging features for distinguishing between resistant and susceptible isolates.

They then trained and tested machine-learning algorithm to recognise these features using imaging data from 16 samples.

̽��ֱ��algorithm was able to correctly predict in each case whether or not bacteria were susceptible or resistant to ciprofloxacin without the need for the bacteria to be exposed to the drug. This was the case for isolates cultured for just six hours, compared to the usual 24 hours to culture a sample in the presence of antibiotic.

Dr Tuan-Anh Tran, who worked on this research while a PhD student at the ̽��ֱ�� of Oxford and is now based at the ̽��ֱ�� of Cambridge, said: “S. Typhimurium bacteria that are resistant to ciprofloxacin have several notable differences to those still susceptible to the antibiotic. While an expert human operator might be able to identify some of these, on their own they wouldn't be enough to confidently distinguish resistant and susceptible bacteria.

“ ̽��ֱ��beauty of the machine learning model is that it can identify resistant bacteria based on a few subtle features on microscopy images that human eyes cannot detect.”

In order for a sample to be analysed using this approach, it would still be necessary to isolate the bacteria from a sample – for example a blood, urine or stool sample. However, because the bacteria do not need to be tested against ciprofloxacin, this means the whole process could be reduced from several days to a matter of hours.

While there are limitations to how practical and cost effective this particular approach would be, the team says it demonstrates in principle how powerful artificial intelligence could be in helping the fight against antimicrobial resistance.

Dr Sushmita Sridhar, who initiated this project while a PhD student in the Department of Medicine at the ̽��ֱ�� of Cambridge and is now a postdoc at the ̽��ֱ�� of New Mexico and Harvard School of Public Health, said: “Given that this approach uses single cell resolution imaging, it isn’t yet a solution that could be readily deployed everywhere. But it shows real promise that by capturing just a few parameters about the shape and structure of the bacteria, it can give us enough information to predict drug resistance with relative ease.”

̽��ֱ��team now aims to work on larger collections of bacteria to create a more robust experimental set that could speed up the identification process even more and allow them to identify resistance to ciprofloxacin and other antibiotics in a number of different species of bacteria.

Sridhar added: “What would be really important, particularly for a clinical context, would be to be able to take a complex sample – for example blood or urine or sputum – and identify susceptibility and resistance directly from that. That's a much more complicated problem and one that really hasn't been solved at all, even in clinical diagnostics in a hospital. If we could find a way of doing this, we could reduce the time taken to identify drug resistance and at a much lower cost. That could be truly transformative.”

̽��ֱ��research was funded by Wellcome.

Reference
Tran, TA & Sridhar, S et al. Combining machine learning with high-content imaging to infer ciprofloxacin susceptibility in isolates of Salmonella Typhimurium. Nat Comms; 13 June 2024; DOI: 10.1038/s41467-024-49433-4

Artificial intelligence (AI) could be used to identify drug resistant infections, significantly reducing the time taken for a correct diagnosis, Cambridge researchers have shown. ̽��ֱ��team showed that an algorithm could be trained to identify drug-resistant bacteria correctly from microscopy images alone.

̽��ֱ��beauty of the machine learning model is that it can identify resistant bacteria based on a few subtle features on microscopy images that human eyes cannot detect

Tuan-Anh Tran

Rocky Mountain Laboratories, NIAID, NIH

Colour-enhanced scanning electron micrograph showing Salmonella Typhimurium (red) invading cultured human cells

Yes

Licence type:

Public Domain