Dr. Philip McMillan,  John McMillan

The first head-to-head study of three mRNA vaccines in people finds the genetic cargo and its fatty packaging persisting far longer than the public was led to expect, much of it in pieces.

For most of the pandemic, the reassurance was simple. The mRNA in a COVID-19 vaccine does its job and disappears. Public health guidance said cells break it down within a few days, and anyone who wondered aloud whether it might linger was accused of spreading misinformation.

That reassurance always sat awkwardly with the chemistry. Vaccine mRNA is not the molecule your cells make every minute of the day. Every uridine, one of the four letters of the RNA code, has been swapped for a modified version called N1-methylpseudouridine. The swap was deliberate, and clever: it lets the molecule slip past the body’s alarm systems and survive long enough to be read. A molecule engineered to resist breakdown that then vanishes in days? It never quite added up.

Now there are better numbers. In October 2026, researchers in Melbourne published the most detailed measurement yet of what happens to these vaccines once they reach the blood. The answer is nothing like a few days.

The Parcel And The Packaging

Think of an mRNA vaccine as a parcel in protective wrapping. The parcel is the mRNA, carrying the instructions for the spike protein. The wrapping is a lipid nanoparticle, a bubble of fats far smaller than a cell. Nearest the mRNA are the ionizable lipids, which hold the cargo together and help it escape into the cell: Moderna’s product uses one called SM-102, Pfizer’s uses ALC-0315. The outer layer is built from phospholipids and PEG lipids, whose chains of polyethylene glycol bristle from the surface like fine hair and keep the particles from clumping or being cleared too soon.

Delivering the parcel is only half the job. Once the instructions are inside a cell, something has to deal with the packaging, and with the parcels that never reached a door. Where does it all go, and how long does it take? For a product injected billions of times, these are basic questions. They are also questions nobody was required to answer in humans, because regulators have long treated pharmacokinetic studies, routine for ordinary drugs, as generally unnecessary for vaccines.

What The Melbourne Team Measured

Stephen Kent of the Doherty Institute and Yi Ju of the Olivia Newton-John Cancer Research Institute led the work, published in ACS Nano. Their team analysed 327 blood samples from 73 volunteers who had received one of three formulations of Moderna’s vaccine, Pfizer’s vaccine, or an investigational mRNA vaccine targeting the spike’s receptor-binding domain at one of three doses.

Droplet digital PCR, a technique that counts individual molecules, measured both total vaccine mRNA and the strands still intact. Mass spectrometry tracked the ionizable lipids. The three platforms behaved differently. Moderna’s mRNA cleared fastest, Pfizer’s sat in the middle, and the investigational vaccine lingered longest, with the lipids falling away in the same order. Intact mRNA disappeared twice as fast after Moderna as after Pfizer. In Moderna recipients, the intact mRNA and its lipid declined together. After Pfizer, the ALC-0315 packaging outlasted the mRNA it had carried.

And the timescale? In some recipients, vaccine mRNA was still measurable four weeks after the dose.

None of this came entirely out of nowhere. In 2022, a Romanian study found vaccine mRNA in blood two weeks after injection. In 2023, a Danish team picked up traces at up to 28 days. In 2024, the same Melbourne group followed 19 Moderna booster recipients for four weeks. The new paper adds scale, a direct comparison across platforms, and a way to tell whole strands from broken ones. Every one of these studies landed long after the reassurance had done its work.

“When they told us that there was no evidence that the spike protein and the mRNA continue to circulate, it was because the research had not been done,” said Dr. Philip McMillan.

The Fragment Puzzle

What shape is the lingering mRNA in? To find out, the researchers built a panel of ten assays spaced along the full length of the spike transcript, rather like checking whether every carriage of a ten-car train is still coupled. Much of what circulated in Moderna and Pfizer recipients was fragmented, and the strands were least intact toward the tail of the molecule.

There is a wrinkle. In Moderna recipients, the pattern of breakage mirrored the vaccine as it came out of the vial. Some of the fragmentation was there before the needle went in, so part of what turns up in blood may simply be parcels damaged before they were ever posted.

That still leaves the bigger puzzle. Unprotected RNA does not survive long in blood, where enzymes shred it within minutes. Whatever shows up weeks later is either still shielded inside its packaging or being released from somewhere: the injection site, the lymph nodes, or cells that took up particles and only partly processed them, faster in some people than in others. And if cells are the source, which ones? Immune cells, the lining of blood vessels, tissue in the gut? Nobody knows yet, and this study was not designed to find out.

An Immune System Already Primed

The second surprise involves that hairy PEG coating. Polyethylene glycol turns up in cosmetics, toothpaste, laxatives and a long list of medicines, so most immune systems have met it before. Before vaccination, anti-PEG IgG antibodies were already detectable in 83 percent of the Moderna and Pfizer groups and 73 percent of the investigational group. By day 28, Moderna’s vaccine had roughly doubled those levels.

The researchers expected those antibodies to speed up clearance of the particles. They did not. Instead, higher peak levels of vaccine mRNA and lipid went hand in hand with stronger antibody boosting.

What that means clinically is an open question. So is a more speculative one. Since 2021, embalmers have described unusual fibrous clots, and microclots have been reported in people with long-term illness after COVID-19. Could an immune response aimed at circulating vaccine material help explain how such clots build up, or why they sometimes seem to accelerate? Nothing in this paper tests that idea. The data simply makes the question harder to wave away, and nobody can answer it without looking.

No Evidence, Or No One Looked?

For years, the absence of data on vaccine mRNA in blood was presented as proof there was nothing to find. The work that finally measured it was funded by Australia’s Medical Research Future Fund, the National Health and Medical Research Council, the Australian Research Council and the Victorian Critical Vaccinees Collection, not by the companies that sold the doses. It appeared as a preprint in March and took seven months to reach a journal.

Emergency authorisation was a defensible call in a crisis. But it came with an obligation to watch for anything unusual and to keep asking how the product behaves in real bodies. The same logic applies to every other reassurance on offer. When officials say there is no link between the vaccines and cancer, sudden deaths or dementia, the first question is what was actually studied, and in whom.

The authors frame their findings as a practical tool: a way to benchmark platforms and build more stable vaccines that might work at lower doses with fewer side effects. That is a reasonable goal, and the gap between Moderna and the investigational vaccine shows how much design choices matter.

“The implications we don’t yet fully understand. But I cannot for the life of me think that this is good,” said McMillan.

For the original rollout, the horse has bolted. Almost everyone who intended to be vaccinated has been, and boosters mostly go to people who have already had several doses. The lesson is for what comes next, as mRNA moves into flu, cancer and beyond. Measure first. Do the research, autopsies included. Only then declare there is nothing to find.