Covering: Biology - Neuroscience - MedicineCovering: Biology - Neuroscience - Medicine
All explained papers

Lancet, 2026 Aug 22 · 6 min read

Risk of transmission of amyloid beta pathology via transfused blood products

Every year, millions of people receive blood transfusions, one of the most common medical procedures in the world. A paper published in the Lancet raises a question that has been building in the field for years: can a transfusion carry a key molecular feature of Alzheimer disease from donor to recipient?

The claim needs stating carefully. The paper examines whether amyloid beta, a protein that clumps in the brains of people with Alzheimer disease, can pass from a donor's blood into a recipient's bloodstream and, critically, into the recipient's brain. It does not claim that a transfusion gives someone Alzheimer disease. It asks whether a transfusion passes on a molecular pathology that might eventually contribute to disease in some recipients.

Amyloid beta is a small protein fragment made naturally as brain cells go about their normal work. In a healthy brain it is cleared away continuously, like rubbish collected every day. In Alzheimer disease that collection fails or is overwhelmed, and amyloid beta piles up between neurons as plaques. Plaques are a hallmark of the disease and show up on brain scans decades before symptoms. Whether they cause the disease, speed it up, or mostly follow from other processes is still debated, but they sit at the centre of the biology and of every major drug target in the field.

Amyloid beta is also in the blood, at very low levels. The blood tests developed for early detection of Alzheimer disease work because a fraction of the brain's amyloid beta leaks into the bloodstream. Donors in the early, silent phase of the disease, years before any memory trouble, can carry measurable amyloid beta in their blood without knowing it. That is the group this paper is about.

The hypothesis is that a recipient could get blood from someone in that silent stage, and that in some fraction of cases the amyloid beta could reach the recipient's brain and start the clumping that, over time, feeds plaque formation. The idea behind this is seeding. A misfolded protein can act as a template, a bad mould that bends normal copies of the same protein into its wrong shape. This is how prion diseases spread, and lab and animal studies show amyloid beta can behave in a similar way.

Prion diseases are the model. In Creutzfeldt-Jakob disease, a protein called prion protein misfolds and then pushes normal prion protein to misfold too, in a chain reaction. Researchers have long asked whether other clumping proteins in brain disease, amyloid beta and tau in Alzheimer disease and alpha-synuclein in Parkinson disease, can act this way under the right conditions. This paper looks at that question in a real clinical setting.

What does it show? This account is based on the published abstract; the full text is behind a paywall. The paper examines the risk of transmitting amyloid beta pathology through transfused blood products and reports evidence consistent with a transmission risk in at least some circumstances. The exact design cannot be fully described from the abstract. It could be an epidemiological study following people who received blood from donors later confirmed to have silent Alzheimer disease, a biomarker study, a registry analysis, or some mix.

That uncertainty matters a great deal. A population study showing more Alzheimer disease in people who received blood from donors who later developed it is a very different kind of evidence from a lab study showing that amyloid-laden blood seeded clumping in an animal. Publication in the Lancet, one of medicine's strictest journals, means reviewers found the methods and conclusions credible. Still, the finding needs independent replication before it changes clinical practice.

Today, blood donation screening targets infectious agents: viruses like HIV, hepatitis B and C, and others. It does not screen for silent Alzheimer disease. Until recently, no blood test was sensitive enough to detect early amyloid pathology at scale. That has changed. Plasma amyloid beta ratio tests are now accurate enough for clinical diagnosis, and whether blood banks should use them is the policy question this paper raises directly.

In the United States, the Food and Drug Administration cleared a plasma amyloid beta test for clinical use in 2024. It measures the ratio of two forms of the protein, A-beta 42 and A-beta 40, and that ratio shifts years before memory symptoms appear. Major reference laboratories now offer it. Its purpose is to help doctors decide whether a patient with memory concerns needs a confirmatory PET scan. It was not built for blood banks, but the technology behind it is what makes the screening conversation possible at all.

So the technical barrier is much lower than it was five years ago, as the tests' accuracy has improved dramatically in the last few years. The remaining gap is between two jobs. In a clinic, the patient already has symptoms, and a false positive may mean an unnecessary PET scan. In a blood bank, the aim is to find rare silent donors across a whole population, where the performance bar is different and a false positive means turning away a healthy donor, which has its own cost to the blood supply.

Screening is possible in principle but costly in practice. The world's blood supply involves hundreds of millions of donations a year, and testing every unit would add cost and complexity to a system already running on tight margins. On the other side, if the risk is real and significant, the cost of not screening falls on recipients who develop Alzheimer disease years or decades later, with no way to trace the cause. The balance depends on how large the risk actually is, which needs the large, forward-looking data this paper is presumably starting to supply.

There is precedent. The blood supply has been updated many times as new risks appeared. West Nile virus screening was added after the 2002 outbreak in the United States, and Zika screening during the 2016 epidemic. Each time, it took years to show transmission, build a reliable test, and roll it out. The Alzheimer case is harder: the pathology is not an infectious agent in the usual sense, and any effect would take decades to appear.

That long delay is the core research problem. If someone received amyloid-containing blood in their forties and developed Alzheimer disease in their seventies, proving cause would be very hard, because the background risk at that age is already substantial. Separating a transfusion's contribution from that baseline takes very large groups of people, long follow-up, and precise biomarker tracking that has only recently become possible. This paper is likely a step in building that evidence, not the final word.

It is important to keep the level of concern right. Anyone who needs a transfusion should get it. In a life-threatening situation, the immediate benefit vastly outweighs a theoretical long-term risk that is unproven in human populations and, if real, would affect a small fraction of recipients after a very long delay. The finding matters for blood safety policy and for how scientists think about how Alzheimer disease might spread. It is not a reason to avoid a needed transfusion.

Screening would raise its own hard questions. A donor deferred for hepatitis has a clear, active infection. A donor with a high amyloid ratio may simply have a genetic tilt or a biological variant that never becomes disease. Telling that person they carry an early marker for Alzheimer disease carries psychological weight, and possibly consequences for insurance and employment depending on where they live. Blood banks would need guidance on what to tell deferred donors, worked out with neurologists, ethicists, and regulators, not just test makers.

The paper adds a new dimension to a field focused mostly on genes, lifestyle, and drug targets: the possibility of an iatrogenic route, meaning disease caused by medical care itself. Medicine has been here before. In the 1980s, growth hormone taken from the pituitary glands of cadavers was found to have passed Creutzfeldt-Jakob disease to hundreds of patients, and it was replaced with synthetic, lab-made hormone. Whether transfusions pose a comparable risk for Alzheimer disease is now a live scientific question, and this paper is part of answering it.

Scale is what makes it significant. Alzheimer disease affects tens of millions of people worldwide, and transfusions reach hundreds of millions each year. Even a small rise in risk, if real, would add up to many cases across a population. That is a strong case for investigating rigorously and weighing screening options while the direct evidence is still being gathered.

The paper is PMID 42624157, published in the Lancet on August 22, 2026; this account is based on the abstract, as the full text is behind a paywall.