Aluminium in Vaccines Is Harmful

Aluminium in Vaccines Is Harmful
by Peter C. Gøtzsche at Brownstone Institute

It has been surprisingly difficult to get an answer to a simple and highly relevant question: Is aluminium in vaccines harmful? After having studied the best evidence we have, the randomised trials, in great detail, I conclude that the answer is yes. 

Like lead, aluminium is a highly neurotoxic metal. We will therefore expect vaccines containing aluminium adjuvants to cause neurological harms if the aluminium enters the nervous system in neurotoxic amounts. 

The aluminium in the adjuvant is important for eliciting a strong immune response in non-live vaccines and their efficacy is related to their toxicity at the injection site.^1-3 Immune-reactive cells engulf particles of aluminium adjuvant and distribute their load throughout the body, including to the brain, where they are killed, releasing their contents into the surrounding brain tissue where they can produce an inflammatory response.

The precise mechanism of action is not so important, but the data we have on the harms are, and they have been systematically distorted. 

False Information from the European Medicines Agency (EMA)

In October 2016, my research group complained to the European Ombudsman about the EMA’s mishandling of their investigation into the suspected serious neurological harms of the HPV vaccines.⁴ In his reply to the Ombudsman, EMA’s Executive Director Guido Rasi stated that the aluminium adjuvants are safe; that their use has been established for several decades; and that the substances are defined in the European Pharmacopoeia.^5,6 

Rasi gave the impression that the aluminium adjuvants in the HPV vaccines are similar to those used since 1926. However, the adjuvant in Gardasil, Merck’s vaccine, is amorphous aluminium hydroxyphosphate sulfate, ‎AlHO₉PS^-3 (AAHS), which has other properties than aluminium hydroxide, the substance Rasi mentioned. Moreover, its properties are not defined in the pharmacopoeia. AAHS has a confidential formula; its properties are variable from batch to batch and even within batches. The harms caused by the adjuvant are therefore likely to vary. When we investigated whether the safety of AAHS has ever been tested in comparison with an inert substance in humans, we were unable to find any evidence of this. 

Rasi mentioned that the assessment of the evidence for the safety of the adjuvants had been performed over many years by the EMA and other health authorities, such as the European Food Safety Authority, the FDA, and the WHO. 

However, none of his five references supported his claim about safety. Three links, to EMA, FDA, and WHO, were all dead. One worked two years later but contained nothing of relevance. A link to the European Food Safety Authority was about the safety of aluminium from dietary intake, which has nothing to do with aluminium adjuvants in vaccines. Very little oral aluminium is absorbed from the gut, and much of what is absorbed is eliminated by the kidneys. The last link was to a WHO report that was also unhelpful.⁵ It mentioned that the FDA had noted that the body burden of aluminium following injections of aluminium-containing vaccines never exceeds US regulatory safety thresholds based on orally ingested aluminium, which is irrelevant information. 

The Randomised Trials Document the Toxicity of Aluminium Adjuvants

As an expert witness for the Los Angeles law firm Wisner Baum, I have read 112,000 pages of confidential Merck study reports.⁷ If Merck’s aluminium adjuvant causes serious neurological harms, one would expect to see more harm with Gardasil 9 than with quadrivalent Gardasil because it contains five more HPV antigens and more than double as much adjuvant, corresponding to 500 µg vs 225 µg aluminium.

And this is what we see. Three trials have compared Gardasil 9 with Gardasil, but two of them were so small, only 1,095 patients in total, with only 3 serious adverse events, that they cannot shed any light on this issue. The third trial, however, was large, with a total of 14,215 females.⁷ 

Merck was not keen to reveal what they found. In the published trial report, in New England Journal of Medicine,⁸ there was no mention of serious harms. But on page 27, just before the last page, in a supplementary appendix on the web, which few people will ever find and read, it was revealed that there were more serious adverse events in females receiving Gardasil 9 than in those receiving Gardasil (3.3% vs 2.6%). There was no P-value, but I calculated P = 0.01 for this difference.

There was more that Merck did not publish in the NEJM, which I found in Merck’s confidential clinical study report. As expected, more patients on Gardasil 9 than on Gardasil experienced nervous system disorders. Again, there was no P-value, but I calculated P = 0.01.

For the injections, pain was by far the most common adverse event. A table in NEJM showed that 4.3% vs 2.6% had severe pain (P = 6 · 10^-8) and 36.8% vs 26.4% had moderate or severe pain (P = 10^-40). There were also more cases of severe swelling, 3.8% vs 1.5% (P = 9 · 10^-18) and of moderate or severe swelling, 6.8% vs 3.6% (P = 2 · 10^-18). Yet again, there were no P-values, but I calculated them.

There was nothing in the NEJM article about systemic adverse experiences. Merck concluded in its internal study report that most patients experienced such events, “most of which were of mild or moderate intensity.” This is very misleading. As mild events are easily tolerated according to Merck’s own definition, Merck should have focused on systemic adverse experiences of moderate or severe intensity. A table showed that 11.7% vs 10.8% of the patients had severe systemic adverse experiences (P = 0.08) and that 39.3% vs 37.1% had moderate or severe systemic adverse experiences (P = 0.007, the number needed to harm was only 45; my calculations).

It is not likely that it is the five additional antigens are responsible for the increased toxicity of Gardasil 9. It is far more likely that it is the larger dose of the aluminium adjuvant that is responsible for the harm. 

I also did a dose-response study of the trials where I compared the maximum contrast, the vaccine versus placebo, with an intermediate contrast, the vaccine versus the adjuvant, and with the minimum contrast, Gardasil 9 versus Gardasil.⁷ There was a clear dose-response relationship for all adverse events (P < 0.00001), and there was little difference between groups two and three. This means that the aluminium adjuvant is similarly harmful as the adjuvant plus the vaccine.

It was also illuminating to read Merck’s confidential animal studies.⁷ Merck admitted that its adjuvant causes harm but argued that, since the harms were similar to those caused by a high-dose vaccine, this meant that they had “minimal toxicological significance.” This conclusion is false. 

Even worse, since the HPV vaccines and their adjuvants have similar harm profiles, the manufacturers and regulators concluded that the vaccines are safe. This is like saying that cigarettes and cigars must be safe because they have similar harm profiles.

Human and animal studies have also shown harms for the other adjuvant, aluminium hydroxide, which is used in Cervarix, the HPV vaccine from GlaxoSmithKline. In a large randomised trial in humans, influenza vaccines caused 34% more adverse events when they contained adjuvant than when they did not, risk ratio 1.34 (95% confidence interval 1.23 to 1.45, P < 0.0001) and also more severe adverse events, risk ratio 2.71 (1.65 to 4.44, P < 0.0001) (my calculations),⁹ even though these adverse events were recorded up to only three days after the vaccination. 

Merck, GlaxoSmithKline, and the EMA called the toxic aluminium adjuvant placebo and girls recruited to Merck’s trials were told that half of them would get placebo.⁷ This is fraud, as fraud is defined as a deliberate intent to deceive. According to Merck’s own definition, an aluminium adjuvant is not a placebo: “A placebo is made to look exactly like a real drug but is made of an inactive substance, such as a starch or sugar.”¹⁰  

Our Systematic Review of the HPV Vaccines 

My research group did a systematic review of the HPV vaccines based entirely on clinical study reports we had obtained from the EMA because they are vastly more reliable than what drug companies publish in medical journals.⁷ 

Against all odds, as the control groups, apart from two small studies, had active comparators, we found that the HPV vaccines increased serious nervous system disorders significantly: 72 vs 46 patients, risk ratio 1.49 (P = 0.04).¹¹ We called it an exploratory analysis, but it was the most important one because the suspected harms to the autonomic nervous system were what caused the EMA to assess vaccine safety in 2015.

Two important neurological syndromes are the postural orthostatic tachycardia syndrome (POTS), in which a change from lying to standing causes an abnormally large increase in heart rate that may be accompanied by light-headedness, trouble thinking, blurred vision, and weakness, and Complex Regional Pain Syndrome (CRPS). They are rare syndromes that are difficult to identify, and we knew – which EMA trial inspectors had confirmed – that the companies had deliberately concealed what they found.⁷ This was also clear from the fact that no cases of POTS or CRPS were mentioned in the clinical study reports. 

I documented in my expert report to the law firm that Merck had committed scientific misconduct in many ways, which included refusing to register POTS cases investigators tried to report to Merck during the clinical trials.⁷ 

To assess if there were signs and symptoms consistent with POTS or CRPS in the data, we did another exploratory analysis where we asked a blinded physician with clinical expertise in POTS and CRPS to assess the MedDRA preferred terms (code terms the companies use to categorise and report adverse events). We found that the HPV vaccines significantly increased serious harms definitely associated with POTS (P = 0.006) or CRPS (P = 0.01).

New onset diseases definitely associated with POTS were also increased (P = 0.03).

Almost all Observational Studies Are Seriously Misleading

Serious harms of vaccines and other drugs are often overlooked in observational studies. Of the many biases in such studies, the most important one is the healthy vaccinee bias, which no amount of statistical adjustment can make up for.^7,12 

During my deposition in Los Angeles, Merck’s lawyer repeatedly referred to flawed studies as evidence that Gardasil does not cause serious harms, an argument I rejected.⁷ 

When the best evidence we have, the randomised trials, have clearly shown that a drug or a substance is harmful, there are always a myriad of observational studies claiming that there is no harm. I call this the UFO trick: If you use a fuzzy photo to “prove” you have seen a UFO when a photo taken with a strong lens has clearly shown that the object is an airplane,¹³ you are a cheat. 

A notorious example is psychiatry, which is full of UFO tricks and can only survive as a medical specialty because its practitioners lie systematically about the wonders their drugs can achieve.¹⁴ Randomised trials have shown that antidepressants increase suicides and that antipsychotics increase mortality, but leading psychiatrists and their organisations say the opposite, referring to flawed observational studies. I consider this a crime against humanity because it is deadly.¹⁵

In July, a large observational study of aluminium-containing vaccines was published, which got a lot of media attention, with some headlines declaring that the debate had been settled. It was a Danish study and its authors concluded that they did not find evidence supporting an increased risk for autoimmune, atopic or allergic, or neurodevelopmental disorders associated with early childhood exposure to aluminium-adsorbed vaccines.¹⁶

However, the study is seriously flawed, which the 22 comments uploaded with the article document.¹⁶ Yaakov Ophir noted that 25 of 34 adjusted hazard ratios indicated a lower risk for adverse outcomes with higher aluminium exposure, and 13 of these unexpected inverse associations were even statistically significant, including those for food allergy, autism spectrum disorder, and ADHD.

Ophir wrote that assuming aluminium is not a miracle compound that reduces the risk of numerous unrelated conditions, this pervasive pattern suggests a strong systematic bias in the data, inadequately addressed despite adjustment for various confounders: “The most plausible explanation is the healthy vaccinee bias, according to which families with better health status or stronger adherence to preventive care are more likely to follow vaccine schedules.”

Sarcastically, Catherine Sarkisian asked if we should be recommending more aluminium to children.

The Danish researchers avoided presenting the data for their unvaccinated group, which they lumped with a group with low exposure to vaccines. This is highly inappropriate when one wants to do a dose-response analysis of exposure to aluminium. Christof Kuhbandner calculated the missing data and found that, in unadjusted analyses, there were marked reductions in disease risk among unvaccinated children with statistically significant results for several allergic outcomes and autism: “It is noteworthy that these results – given their partially high statistical significance – were neither reported in Andersson et al.’s study nor addressed in response to critical commentary.”

Christine Stabell Benn and colleagues, who seemed to have peer-reviewed the study, pointed out errors and issues with the authors’ data sources. 

In their replies, which Yaakov Ophir demonstrated were wholly unconvincing, the authors beat about the bush and tried to explain away the most critical issues, which essentially killed their study. 

The Distorting Role of the Media and Medical Journals 

In a press conference on 22 September, US President Donald Trump declared: “We want no mercury in the vaccine. We want no aluminum in the vaccine.”¹⁷ His Secretary of Health, Robert F. Kennedy, Jr., had already declared that mercury should be removed from vaccines, as equally good vaccines were available without mercury. So why allow a toxic metal in vaccines? 

Two months earlier, Kennedy had criticised the Danish study for much the same reasons as the commentators did.¹⁸ However, the media constantly distort and derail the debate by citing conflicted experts for saying that there is no documentation that mercury and aluminium in vaccines are harmful. This is like putting the cart before the horse. It should have been demonstrated in randomised trials that adding toxic metals to vaccines is safe, before they were approved by drug regulators, but this was never done. 

It is a steep, uphill battle to improve the safety of vaccines in such a hostile environment where the media are critical towards any such attempts. The taboo is so pronounced that some people got fired for questioning the safety of the Covid-19 vaccines, or just for raising relevant questions, e.g. about the wisdom of vaccinating children or of recommending multiple boosters, even for those who had already been infected and had acquired a much better immunity than a vaccine can provide.  

To put it mildly, the medical journals are also unhelpful. In September 2016, Karsten Juhl Jørgensen and I submitted a paper to BMJ about EMA’s mishandling of their investigation into the suspected serious neurological harms of the HPV vaccines. This started a most bizarre and absurd Odyssey for us that lasted three years.⁷ BMJ involved their lawyers and the messages we received from the editors were conflicting. We tried to do the impossible and rewrote our paper four times but to no avail. BMJ killed our paper but did not have the guts to tell me. We then submitted it to BMJ Evidence-based Medicine where it was accepted after additional peer review and published.⁴ This was 4.5 years after we submitted it to the BMJ. 

What a tragedy this was for freedom of speech in science. In contrast, the most recent article I published in the Brownstone Journal came out two days after I submitted it.¹⁹ 

Conclusions 

Aluminium adjuvants are toxic and can, in rare cases, cause serious neurological harms such as POTS and CRPS. Aluminium should be avoided in vaccines. 

References

1. Awate S, Babiuk LA, Mutwiri G. Mechanisms of action of adjuvants. Front Immunol 2013;4:114.
2. Demasi M. A chat with ‘Mr Aluminium.’ Substack 2025;Sept 30.
3. Shardlow E, Mold M, Exley C. The interaction of aluminium-based adjuvants with THP-1 macrophages in vitro: Implications for cellular survival and systemic translocation. J Inorg Biochem 2020;203:110915. 
4. Gøtzsche PC, Jørgensen KJ. EMA’s mishandling of an investigation into suspected serious neurological harms of HPV vaccines. BMJ Evid Based Med 2022;27:7-10.
5. Gøtzsche PC, Jørgensen KJ, Jefferson T. Our Comment on the decision by the European Ombudsman about our complaint over maladministration at the European Medicines Agency related to safety of the HPV vaccines. Deadlymedicines.dk 2017;Nov 2. 
6. Gøtzsche PC. Vaccines: truth, lies, and controversy. New York: Skyhorse; 2021.
7. Gøtzsche PC. How Merck and drug regulators hid serious harms of the HPV vaccines. New York: Skyhorse; 2025.
8. Joura EA, Giuliano AR, Iversen O-E, et al. A 9-valent HPV vaccine against infection and intraepithelial neoplasia in women. N Engl J Med 2015;372:711–23.
9. Liang XF, Wang HQ, Wang JZ, et al. Safety and immunogenicity of 2009 pandemic influenza A H1N1 vaccines in China: a multicentre, double–blind, randomised, placebo-controlled trial. Lancet 2010;375:56-66.
10. Lynch SS. Placebos. Merck 2022; Sept.
11. Jørgensen L, Gøtzsche PC, Jefferson T. Benefits and harms of the human papillomavirus (HPV) vaccines: systematic review with meta-analyses of trial data from clinical study reports. Syst Rev 2020;9:43.
12. Deeks JJ, Dinnes J, D’Amico R, et al. Evaluating non-randomised intervention studies. Health Technol Assess 2003;7:1–173.
13. Sagan C. The demon-haunted world: science as a candle in the dark. New York: Ballantine Books; 1996.
14. 14 Gøtzsche PC. The Only Medical Specialty That Survives on Lies. Brownstone Journal 2025; Sept 8.
15. 15 Gøtzsche PC. Is psychiatry a crime against humanity? Copenhagen: Institute for Scientific Freedom; 2024 (freely available).
16. 16 Andersson NW, Bech Svalgaard I, Hoffmann SS, Hviid A. Aluminum-Adsorbed Vaccines and Chronic Diseases in Childhood : A Nationwide Cohort Study. Ann Intern Med 2025; Jul 1517 President Trump makes an announcement on medical and scientific findings for America’s Children. YouTube 2025; Sept 22.  
17. 18 Kennedy RF, Jr. Flawed science, bought conclusions: The aluminum vaccine study the media won’t question. TS News 2025; Aug 1. 
18. 19 Gøtzsche PC. A Giant in Medicine: Tribute to Drummond Rennie. Brownstone Journal 2025; Oct 2.

Aluminium in Vaccines Is Harmful
by Peter C. Gøtzsche at Brownstone Institute – Daily Economics, Policy, Public Health, Society