Received Wisdom vs Settled Science: The EV trash bin

Science is supposed to be self-correcting. The textbooks, the funding committees, the peer reviewers, and the senior professors will, eventually, follow the evidence. That's the theory. The reality is messier, more human. Throughout history, some of the most important biomedical advances have come not from confirming what everyone believed, but from someone stubborn enough, or simply naive enough, to question it.

What Does "Settled" Actually Mean?

At the heart of the problem is a phrase that does enormous damage: settled science. It sounds reassuring. It implies that evidence has been weighed, alternatives have been tested, and the question is closed. Sometimes that's genuinely true. The germ theory of disease is settled. The structure of DNA is settled. Vaccines prevent the diseases they are designed to prevent. Mutation is commonplace, and speciation is driven by mutation. These conclusions rest on multiple independent lines of evidence, have withstood every serious attempt at falsification, and make predictions that have been confirmed across entirely separate fields of enquiry. Challenging them requires extraordinary counter-evidence, and none has emerged.

But "settled" is also a political word in the context of science, and it can be applied to ideas that haven't earned it. The distinction that matters is between conclusions that have been exhaustively tested and those that have simply gone unchallenged long enough to feel permanent. Medicine, in particular, has a long history of the second kind masquerading as the first, with consequences that range from embarrassing to fatal.

The EV waste disposal hypothesis

Extracellular vesicles (EVs), including the subclass known as exosomes, tell a remarkably similar story, and in this case the history is unusually precise. Exosomes were first described in two papers published within a week of each other in 1983, one by Harding, Heuser and Stahl in the Journal of Cell Biology, and one by Pan and Johnstone in Cell, both observing that maturing red blood cells jettisoned unwanted surface receptors by packaging them into tiny vesicles and expelling them. The interpretation was straightforward: this was a disposal mechanism. The framing was cemented by Rose Johnstone, one of the original discoverers, in a 1991 paper in the Journal of Cell Physiology proposing that exosome release was "a major route for externalization of obsolete membrane proteins." The word "obsolete" said everything about how the field viewed these particles. For another five years, that framing largely held, and the vesicles attracted little serious research investment as a result.

The reappraisal came in 1996, 13 years after the first observation of EVs, when Raposo and colleagues published a landmark paper in the Journal of Experimental Medicine demonstrating that exosomes released by B lymphocytes carried MHC class II molecules on their surface and could directly stimulate antigen-specific T cell responses. Cellular rubbish, it turned out, did not present antigens to the immune system in a precisely regulated, functionally specific manner. Something else was going on. That paper broke the waste disposal consensus and opened the field. By 1998, Zitvogel and colleagues had shown that dendritic cell-derived exosomes could eradicate established tumours in mice, effectively proposing them as a cell-free cancer vaccine platform. From that point, the field accelerated dramatically. What had been dismissed as obsolete packaging is now understood as a sophisticated and highly regulated intercellular communication system, carrying proteins, lipids, RNA, and signalling molecules between cells and tissues, and implicated in immune modulation, tissue repair, cancer metastasis, and neurodegeneration. The cells had not changed. Only the assumptions of the researchers looking at them had.

In all of these cases, the conclusion was reached not through rigorous negative evidence but through an absence of looking. We called it junk because we couldn't see what it did. We said the brain couldn't grow because we hadn't looked carefully enough. We called exosomes rubbish because we hadn't asked what they were carrying. That is not settled science. That is the absence of curiosity dressed up as certainty.

The Man Who Drank Bacteria

The most dramatic example is Barry Marshall, an Australian physician who in 1984 proposed that peptic ulcers were caused not by stress and excess stomach acid, as every gastroenterologist knew with absolute certainty, but by a bacterium called Helicobacter pylori. The received wisdom was so entrenched, and the scepticism from the medical establishment so complete, that Marshall resorted to drinking a solution containing the bacteria himself, successfully giving himself gastritis, and then treating it with antibiotics. It took another decade and a Nobel Prize in 2005 before the medical world fully accepted what he'd shown with a glass of bacterial culture and sheer bloody-mindedness.

The ulcer consensus was never truly settled science. It had never been rigorously tested against infectious alternatives. It felt settled because it was old, widely taught, and commercially convenient for the antacid industry. That is not the same thing.

The Doctor Who Knew Too Much

A century before Marshall, Ignaz Semmelweis identified in the 1840s that doctors who moved directly from performing autopsies to delivering babies were killing mothers through contaminated hands. His data was clear. His solution, handwashing with chlorinated lime, was simple. The medical establishment rejected him, in part because germ theory hadn't yet provided a satisfactory explanation for why it worked, and in part because accepting his conclusion meant accepting that doctors themselves were vectors of death. He was eventually committed to a mental institution and died there. Germ theory, arriving shortly after, vindicated him entirely.

Semmelweis had the data. What he lacked was a mechanism that the establishment was prepared to accept. Science, in practice, sometimes demands not just evidence but a framework congenial to those holding institutional power.

Junk DNA and the Brain That Couldn't Grow

More recently, the assumption that the adult human brain cannot generate new neurons, established as textbook fact for most of the twentieth century, has been substantially overturned. The dogma was so complete that early researchers who found evidence of neurogenesis in adult animals struggled to get published. Similarly, the vast majority of human DNA that doesn't code for proteins was confidently labelled "junk" for decades, an assumption that turned out to reflect the limits of our analytical tools more than the actual biology. The ENCODE project and subsequent research revealed extensive regulatory activity throughout regions once dismissed as evolutionary debris.

The Fat We Were Told to Fear

The dietary fat story is perhaps the most consequential modern example. Ancel Keys' hypothesis linking saturated fat to heart disease became nutritional orthodoxy from the 1960s onwards, shaping public health policy, food reformulation, and medical advice for generations. The picture that has emerged since is far more complicated, with sugar's role actively minimised, partly, it now appears, through deliberate lobbying by the sugar industry. The certainty with which fat was condemned was never justified by the evidence available, and the population-wide dietary experiment it produced has had consequences we are still working through.

So What Is Wrong Right Now?

History is comfortable. It's easy to nod along to Semmelweis and Marshall because we already know the ending. The harder and more interesting question is: which of today's biomedical certainties are tomorrow's embarrassments?

Four candidates perhaps deserve more scrutiny than they currently receive.

The amyloid hypothesis of Alzheimer's disease has dominated neuroscience research for three decades, directing billions of dollars of research funding toward the assumption that amyloid plaques are the primary cause of the disease. However, clinical trial after clinical trial has failed. A major fraud scandal in 2022 revealed that key supporting data in a landmark 2006 Nature paper had been manipulated, and the paper was ultimately retracted in 2024. The hypothesis may still prove correct, partly correct, or it may represent one of the most expensive misdirections in the history of medical research. It has the hallmarks of received wisdom rather than settled science: strong early adoption, enormous institutional investment, and a troubling pattern of anomalies explained away rather than investigated.

The serotonin theory of depression, the idea that depression is caused by a chemical imbalance and specifically by low serotonin, has underpinned antidepressant prescribing for thirty years and been absorbed into popular culture as a settled fact. A substantial 2022 umbrella review by Moncrieff et al., published in Molecular Psychiatry, found no consistent evidence that people with depression have lower serotonin levels. The review caused considerable controversy, and the debate continues. The drugs may work, but the explanation for why has likely been wrong all along. A treatment working does not validate the theory behind it, a logical error that medicine has made repeatedly.

Viruses as drivers of chronic disease may be dramatically underappreciated. A landmark 2022 study by Bjornevik et al., published in Science, analysed data from over ten million US military personnel and found that the risk of multiple sclerosis increased 32-fold after infection with Epstein-Barr virus, providing compelling evidence that EBV is the leading cause of MS. Herpes viruses are under serious investigation as potential contributors to Alzheimer's. The assumption that autoimmune and neurodegenerative diseases arise from purely genetic or lifestyle factors, with no infectious trigger, may look very different in twenty years. We have been here before: the idea that a virus could cause cancer was once considered implausible, until it wasn't.

Finally, the somatic mutation theory of cancer, the organising framework of modern oncology, holds that cancer is fundamentally a disease of accumulated DNA mutations. A competing metabolic theory, rooted in observations made by Otto Warburg in the 1920s and subsequently marginalised, proposes that cancer is primarily a disease of mitochondrial dysfunction and disordered energy metabolism. The mutation theory has delivered genuine advances, but its dominance may have slowed exploration of metabolic approaches to treatment that are now quietly attracting renewed interest. Whether it will prove to be the amyloid hypothesis of oncology remains to be seen.

The Pattern Behind the Pattern

What do these stories have in common? In every case, the consensus hardened before the evidence warranted it. Careers, funding, textbooks, and institutional prestige became invested in a particular view. Challengers were dismissed not primarily on the merits of their evidence but on the implausibility of their challenge to the established order. And the mechanisms by which science is supposed to self-correct, peer review, replication, open debate, were slowed or captured by the very consensus they were meant to test.

The phrase "settled science" is not the problem when it is applied carefully, to conclusions that have genuinely earned it. The problem is its casual extension to ideas that are merely old, widely taught, or commercially entrenched. Genuine settled science welcomes challenge because challenge has repeatedly failed to dent it. Received wisdom tends to discourage challenge entirely. Learning to tell the difference is one of the most important and undervalued skills in medicine.

None of this means that consensus is usually wrong. It isn't. The contrarian is wrong far more often than the establishment. But certainty should be earned continuously, not inherited. The most important question in any field isn't "what do we know?" but "what are we most confident about that we haven't seriously tried to disprove?"

Marshall didn't just discover a bacterium. He demonstrated that a glass of unpleasant liquid, consumed in defiance of expert opinion, could be worth more than decades of comfortable agreement.

The unsettling implication is that right now, in every corner of biomedicine, there are things everyone knows that are simply wrong.

IMAGE The complexity of EVs CREDIT Rai et al / Cell Guidance Systems

Learn more about powerful technologies that are enabling research: