select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
A study published in npj Dementia looks at apolipoprotein E genetic variants of Alzheimer’s disease.
Prof Anneke Lucassen, Professor of Genomic Medicine, Nuffield Department of Medicine, University of Oxford, said:
“Whilst the authors are themselves more nuanced in the press release, the headlines around this paper are misleading: The claim that ‘most cases’ are linked to a single gene is simply an artefact of the authors’ choice to use the rare, protective ε2/ ε2 genotype as their baseline, effectively labelling approximately 95% of the population (who have ε3 or ε4 alleles) as being ‘at genetic risk’.
“The headline also conflates the very high risk conferred by ε4/ ε4 homozygosity (where a person has 2 copies of ε4) with the more moderate risk moderate risk of ε4 in combination with 3 or 2. The headline suggests a genetic determinism that simply doesn’t exist for most individuals.
“In reality, unless you carry two copies of the ε4 variant (which is rare, only ~2% of people have this combination), your risk is heavily influenced by lifestyle. The study essentially blames the gene for cases that might actually be prevented by better heart health and lifestyle. This is a bit like saying 99% of sunburns are attributable to having fair skin. While statistically true that darker skin protects against sunburn, it doesn’t mean having fair skin is a “disease state” that causes the burn—exposure (lifestyle) is still the driver which could be avoided.
“More detailed responses to some of points in press release:
It implies that if we could magically edit everyone’s DNA to be the rare ε2/ ε2 type, 90% of Alzheimer’s would vanish. This confuses susceptibility with causality. Just because a gene makes you susceptible doesn’t mean the disease “would not occur” without it; it might just require different environmental triggers.
This comment tries to rebrand the ε3 allele carried by 60% population as a “risk allele.” With the rare protective ε2 allele as neutral. Framing the lack of protection as risk is misleading. It’s a bit like saying: Not wearing a bulletproof vest is a risk factor for gunshot wounds. This is statistically true, but we don’t normally label wearing normal clothes as high-risk behaviour.
“Again statistically true for this paper but a consequence of labelling ε3 as a risk factor (when previous papers only labelled ε4 as one). If one expands the definition of at risk then the attributable risk numbers naturally skyrocket.
“The role of the environment needs much greater recognition in the reporting around this paper: If a person gets Alzheimer’s because they had the gene AND for example, a poor diet, removing either one might have prevented that specific case. So just because this paper finds a 90% Population Attributable Fraction (PAF) for APOE does not mean lifestyle doesn’t matter. It just means the gene is a necessary “partner” in the crime for many cases, but then so are most genes!”
Dr Najaf Amin, Associate Professor, Molecular Epidemiology, Oxford Population Health, said:
“It has long been established that the APOE ε4 allele, present in approximately 11–13% of individuals, confers an increased risk of Alzheimer’s disease, whereas ε2, the rarest allele, is associated with a protective effect. The ε3 allele is the most common, occurring in 70–80% of individuals of European ancestry, and is therefore typically treated as the reference (neutral) group in genetic risk estimation.
“In contrast, the study by Williams et al. estimates dementia risk using the rare protective ε2 group as the reference, rather than the neutral ε3 group. This choice necessarily inflates the apparent risks associated with the ε3 and ε4 groups, leading to estimates that are both imprecise and difficult to interpret clinically, as acknowledged by the authors themselves.
“Our own genome-wide and proteome-wide association analyses of Alzheimer’s disease, stratified by APOE ε2/ε3/ε4 status, demonstrate that the ε3 group is highly heterogeneous, with variants in other genes—such as TMEM106B, GRN, and MAPT—playing a more prominent role in disease risk (see: (https://www.medrxiv.org/content/10.1101/2025.05.07.25327065v1.full.pdf, https://www.medrxiv.org/content/10.64898/2025.12.08.25341836v1.full.pdf)). This observation is expected, given that ε3 represents the majority of the population, and does not imply that the ε3 allele itself confers an elevated risk of dementia.
“We do, however, agree with the authors that APOE should be prioritised in dementia research, rather than treated as a nuisance covariate. It is striking that nearly three decades after the initial discovery of the association between APOE and Alzheimer’s disease, the underlying biological mechanisms remain poorly understood.”
Prof Masud Husain, Professor of Neurology, University of Oxford, said:
“This is a really important study. It reveals how significant the different variants of the APOE gene are in affecting risk of developing Alzheimer’s disease. But it also raises the question of whether knowing your APOE genotype would be useful. Currently, this is not available routinely in the NHS. This is largely because it is unclear what someone could do if they found that their risk of developing dementia is high. We definitely need clinical trials that focus on these higher risk people to establish whether new treatments can make a difference to them.”
Dr Richard Oakley, Associate Director of Research and Innovation at Alzheimer’s Society, said:
“This large-scale study offers a clearer picture of how the APOE gene influences the risk of developing Alzheimer’s, suggesting its impact may be even greater than we previously understood.
“While these findings offer a better understanding of the role of genetics, it is important to remember that having a high-risk form of the gene is not a certain diagnosis. Alzheimer’s remains a complex condition influenced by a mix of people’s backgrounds, genetics, and lifestyle. As we continue to further our understanding of risks and causes, we must not lose sight of the risk factors that remain within our control.
“There are steps people can take today to reduce their risk of dementia, such as exercising, not smoking and managing long-term health conditions like blood pressure and diabetes. We know a holistic approach to good health is the best way to lower your risk of dementia.
“There is no single approach to treating dementia. We must continue to fund diverse research to help us find effective preventative measures and treatments that work for everyone, regardless of their genetics.”
Prof Timothy Frayling, Professor of Human Genetics, University of Geneva, said:
“While the figures presented seem reasonable and in keeping with previous research, saying that “Potentially more than 90% of Alzheimer’s disease cases would not occur without the contribution of a single gene (APOE)” is a little like saying that “Potentially more than 90% of road traffic deaths would not occur without the contribution of cars”. People should not worry if they have the risk versions of the gene, because 99.4% of us do!”
Prof Gill Livingston, Professor of Psychiatry of Older people, UCL, said:
“This is good quality research. We have known for a long time that people with one or two copies of the APOEε4 gene are at higher risk for Alzheimer’s disease (AD) and that this is the gene most linked to older onset AD and most people with AD are older. We have also known that the APOEε2 which is relatively rare. Most people in these studies carried the ε3 or ε4 variant -over 99% in each sample. Therefore, it would be astonishing if most people with AD or any other illness did not have one of these genes. They are overrepresented in people with AD. As over 99% of the population here have one of these genes; to suggest potential genetic manipulation of this group would mean genetic manipulation of virtually all humans and we do not know the outcome. While the risk with any of the ε3 and ε4 variants is higher, these genes did not make AD inevitable and over 98% of people without dementia also had one of these genes. We know from other work that even people with the ε4 variant which is a higher risk have a reduced risk if they tackle other modifiable risk factors, including leading a cognitively, physically and socially active life; treating visual and hearing loss, not smoking and drinking in moderation and keeping blood pressure, sugar and cholesterol under control. The authors do not account for these in the analyses.”
Prof Atticus Hainsworth, Professor of Cerebrovascular Disease, from the School of Health & Medical Sciences at City St George’s, University of London, said:
“There is a danger that this report leads to unease in the general public. Carrying the APOE4 gene is a risk factor, it is not a determinant of dementia. As we tell medical students, there are 80-year-olds with two copies of APOE4 who are cognitively healthy.”
Prof Tara Spires-Jones, Director of the Centre for Discovery Brain Sciences at the University of Edinburgh, Division Lead in the UK Dementia Research Institute, and Past President of the British Neuroscience Association said:
“The causes of Alzheimer’s disease are complex with a combination of ageing, genes, and lifestyle all significantly contributing to disease risk. This study looked at a well-known genetic risk factor, the Apolipoprotein E (APOE) gene. For over 30 years, it has been known that the E4 variant of this gene substantially increases risk of Alzheimer’s disease, vascular dementia, and dementia with Lewy bodies. The E2 variant of the gene is associated with substantially lowered risk of dementia. The E3 variant, which is most common, has been considered “neutral” in terms of Alzheimer’s risk. This new study compared the risk of both E4 and E3 to the very rare situation of people who inherit two of the protective E2 genes (one from each parent). Their data indicate that between 71%-92% of Alzheimer’s cases could be due to inheriting either APOE3 or APOE4. The data are interesting but not surprising. The study is well-conducted and looks at data from 3 large cohort studies but is limited by the relatively small number of people with two copies of APOE2 and by the populations being studied being largely of European descent as there are known differences in APOE associations with dementia in people with different ancestries. There is currently research ongoing into how variants of APOE make the brain vulnerable to dementia and there are trials in development targeting the gene.”
Prof Chris Fox, Professor of Clinical Psychiatry, University of Exeter, said:
“This is an interesting finding, but this research is conducted in a specific population – we need more research in more diverse groups in terms of ethnicity and different levels of deprivation. While we know APOE genes cause dementia risk for some people, the vast majority of cases are not caused by genetics. Instead, it’s the interplay between lifestyle factors and different diseases which is important. For that reason, we generally don’t recommend getting a genetic test for dementia unless you have multiple family members affected, as it can cause unnecessary worry.”
Dr Sheona Scales, Director of Research at Alzheimer’s Research UK, said:
“A person’s risk of developing Alzheimer’s disease is shaped by a complex mix of factors, including age, lifestyle and genetics.
“Findings from this study provides important insight into the link between variants of the APOE gene and Alzheimer’s disease risk. Evidence around APOE3 contributing to Alzheimer’s and dementia risk is significant, as this has been largely thought of as having a ‘neutral’ effect on risk. However, it’s important to note that not everyone with these APOE gene variations will go on to develop dementia, as other risk factors also have an influence.
“This research raises lots of interesting questions, such as how APOE3 and APOE4 could drive Alzheimer’s risk, their effects in people of non-European ancestry and whether targeting these variants could be a promising avenue for treatment and prevention of Alzheimer’s.
“Because of the complexity of Alzheimer’s risk, APOE testing is not available on the NHS for people worried about their risk of developing Alzheimer’s in the future. If you are worried about your dementia risk, you should speak to your GP about this.
“Alzheimer’s Research UK is delighted to be funding a continuation of this research that will be studying the APOE gene in more detail. This work will help to gain a better understanding of how genetics alongside lifestyle and ethnicity affect our dementia risk. Knowing more about the biological mechanisms that underpin dementia will be essential to helping reach a cure for this devastating condition.”
‘The proportion of Alzheimer’s disease attributable to apolipoprotein E’ by Dylan M. Williams et al. was published in npj Dementia at 10:00 UK time on Friday 9th January.
DOI: https://doi.org/10.1038/s44400-025-00045-9
Declared interests
Prof Anneke Lucassen: My salary comes from university of Oxford and the NHS.
Dr Najaf Amin: No declarations of interest
Prof Masud Husain: I don’t have any conflict of interest.
Prof Timothy Frayling: I know some of the authors and have collaborated briefly with Emma Anderson on an unrelated project.
Prof Gill Livingston: I am in the same department as the last two authors but did not take part in the research. I also lead Lancet commissions on dementia prevention, intervention and care which considers modifiable risk factors and discusses genes but not as currently modifiable
Prof Atticus Hainsworth: AHH has received honoraria from Eli-Lilly and from NIA. He serves as a consultant for AriBio Co. Ltd and chairs the DementiasPlatformUK Vascular Experimental Medicine group.
Prof Tara Spires-Jones: I have no conflicts with this study but have received payments for consulting, scientific talks, or collaborative research over the past 10 years from AbbVie, Sanofi, Merck, Scottish Brain Sciences, Jay Therapeutics, Cognition Therapeutics, Ono, and Eisai, and direct a company Spires-Jones Neuroscience, Ltd to act as a consultant. I am also Charity trustee for the British Neuroscience Association and the Guarantors of Brain and serve as scientific advisor to several charities and non-profit institutions.
Prof Chris Fox: No conflicts of interest.
Dr Sheona Scales:/ARUK: We are currently funding the lead author, Dylan Williams, in a Senior Research Fellowship in which he is continuing research into APOE. However, we did not fund the work done in this paper as this was before his Senior Research Fellowship started.
For all other experts, no reply to our request for DOIs was received.