This article, about the recent summary judgment in Gardasil litigation, was written by Dorit Rubinstein Reiss, Professor of Law at the University of California Law San Francisco, who is a frequent contributor to this and many other websites, providing in-depth, and intellectually stimulating, articles about vaccines, medical issues, social policy, and the law.

Professor Reiss writes extensively about vaccination’s social and legal policies in law journals. Reiss is also on the Parent Advisory Board of Voices for Vaccines. This parent-led organization supports and advocates for on-time vaccination and the reduction of vaccine-preventable diseases. She is also a Vaccines Working Group on Ethics and Policy member.

On March 11, 2025, a federal district court judge granted Merck’s motion for summary judgment in the multi-district litigation against Gardasil, essentially ending the litigation. The judge found that plaintiffs’ warning defects claims could not be brought because Merck did not have new information on which to base warning doctors that HPV vaccines cause postural orthostatic tachycardia syndrome (POTS) or primary ovarian insufficiency (POI). In fact, the judge’s findings strongly suggest that such evidence as existed supported the absence of causation. However, the judge did not rule on causation directly.

The Gardasil litigation 

As described in the decision, the multi-district litigation in question here – In Re: Gardasil Products Liability Litigation – included over 200 cases brought by plaintiffs claiming that they suffered POTS and POI because of the Gardasil vaccine.

The judge, with the parties’ agreement, created an “Initial Bellwether Pool” of sixteen plaintiffs whose cases would be used to examine core issues that go across the case (p. 6). The litigation was handled by Judge Kenneth Davis Bell of the federal district court in the Western District of North Carolina, who served in several attorney positions for the United States federal government and then as a partner in several law firms before being appointed to the federal bench by President Donald Trump in 2018.

The plaintiffs initially brought claims of several product defects and fraud, but in March 2024 the court dismissed most of these claims. In another ruling in June 2024, the court also dismissed a constitutional challenge to the National Childhood Vaccine Injury Act, which was the basis for the dismissal of most of the claims. 

As a reminder, this is one of the lawsuits Mr. Robert F. Kennedy Jr., now secretary of Health and Human Services, was heavily involved in as a consultant (even while telling Congress manufacturers of childhood vaccines have “zero-liability”. Mr. Kennedy refused to completely divest his potential of earning money from this litigation, instead giving his earnings to his adult son, leaving it in the family.

The main claim left to the plaintiffs was Merck’s failure to warn the doctors of the risks of POTS and POI – what is referred to in legal terms as a warning defect claim. Although warning defect claims are claims under state law, the plaintiffs also had to contend with federal law – specifically, for this purpose, the FDA’s elaborate system of regulation of vaccines (and other products, but for our purposes, vaccines are what matters).

Most importantly, the FDA regulates in detail what manufacturers are allowed to put on the product label, and manufacturers can only add warnings without the FDA’s permission in certain circumstances. If the manufacturers cannot add warnings under federal law, federal law means they are not permitted to do so, and since federal law is supreme over state law, when both are valid, federal law “displaces, or preempts,” state law when they directly conflict.

So unless the manufacturers could legally add the warnings, there’s no case against them – even if the warning should have been given (as you will see, here there are no good grounds to say it should have been). The legal term for this is federal preemption. 

The plaintiffs suggested that Merck could add the warning by adding a “special labeling supplement” amendment, which Merck could do if it had “’newly acquired information’ that requires ‘add[ing] or strengthen[ing] a contraindication, warning, precaution, or adverse reaction for which the evidence of a causal association satisfies the standard for inclusion in the labeling under’” the relevant legal provision. (pp. 10-11 of the decision). The change must be supported by reasonable evidence, and it’s the plaintiffs’ job to give the evidence for that change. 

The court points out, correctly, that the discussion below is not meant to cast doubt on the plaintiffs’ very real suffering – and I want to echo it. I do not doubt that the plaintiffs sincerely suffered. The problem is that, as you will see, connecting their suffering to Gardasil, on the evidence they have, is problematic. Although the court here does not address causation, since the standard for adding to the label is whether there is enough reasonable evidence suggesting a causal link to justify adding the alleged syndrome to the label as an adverse event, the issues overlap. 

The court also reminded us that there are two reasons unreasonable language should not be added to labels. First, including risks without evidentiary basis can lead to people refusing beneficial treatments or, here, vaccines, for reasons that simply are not true. Second, including too many baseless warnings can lead to the real and important warning being drowned in the noise and missed. 

The court concluded – very convincingly – that plaintiffs failed to show that Merck had reasonable evidence to ask to add POTS or POI to the label. Why? 

Postural orthostatic tachycardia syndrome

POTS is a syndrome in which people suffer from “an excessive increase in heart rate (tachycardia) and other symptoms that worsen upon standing or sitting up, such as light-headedness, shortness of breath, chest pain, and palpitations” that persist for over six months. The condition can be highly debilitating. Incidence peaks at 14, but 50% of those affected are adults – and the vast majority of them are female.

According to the decision, the condition affects 0.2-1% of people in the United States, though later in the case plaintiffs alleged the actual prevalence is 0.06-0.07% (so 6-7 in 10,000 are expected to have it). This summary is from the court decision but also reflects the NIH’s summary. NIH explains that “[s]cientists still do not know exactly what causes POTS.” 

The plaintiff’s evidence connecting it to Gardasil needed, to meet the requirements, to be “newly acquired information”, but the plaintiffs also addressed the clinical trial evidence, so I will address that too. To be newly acquired evidence, strictly speaking, the evidence for a link between Gardasil and POTS needs to be between 2006 and the plaintiffs’ initial date of 2011. The plaintiffs, however, changed their mind and tried to increase the date to an undetermined date although the court told them they could not do that, it nonetheless addressed evidence up to 2015, and then later. 

So what did the evidence show? 

Between 2006-2011, the plaintiffs’ expert witness identified one (yes, 1) published case report of POTS after the Gardasil vaccine. As the court pointed out, although the woman in the case study developed POTS two weeks after the vaccine, that temporal association alone does not show causation – and one case, when tens of millions of women received the HPV vaccine, is really, really not good evidence here.

The expert also found additional cases between 2014 and 2017 that covered eight (yes, 8) women – out, again, of tens of millions of doses. Even if the rate of POTS in the population naturally was the plaintiffs’ 0.06% – that should be 6-7 out of a 1000. Out of 10 million, that would exceed 9 (total) by an order of magnitude. These cases don’t even reach the start of the baseline rate of POTS according to the plaintiffs’ own estimate. 

The plaintiffs also submitted Vaccine Adverse Reporting System (VAERS) reports as evidence. In the 2006-2011 time frame, they found 10 VAERS reports mentioning both POTS and Gardasil. The court reminds us of the caveats related to using VAERS:

According to the CDC, “VAERS data alone cannot determine if the vaccine caused the reported adverse event.” This is because VAERS “accepts all reports of adverse events following vaccination without judging whether the vaccine caused the adverse event. Some VAERS reports might represent true vaccine reactions or side effects; others might be coincidental adverse events not related to vaccination.” CDC, About the Vaccine Adverse Event Reporting System (VAERS).

It then goes on and explain: 

Even if all of these unverified reports were actually POTS cases, these reports would be far below the background rate for POTS given the tens of millions of individuals who had received Gardasil.

Well, yes. Out of tens of millions of doses, ten unverified reports are not much. 

Even if we put the published case reports throughout with the unverified reports and assume they all are real cases of POTS after receiving the Gardasil vaccine, we get to 19 such cases in total. That low number cannot be used to show causation – it can easily be coincidence.

What about studies? During the 2006-2011 period, the plaintiffs bring two studies. One is a 2009 report of VAERS reports, and that report found “51 reports of autoimmune disorders.” It does not, however, help plaintiffs. Why? I’ll let the court tell us:

…the study does not mention POTS. Id. The report concludes that most of the reported adverse events “did not meet the FDA definition of serious” and “the safety profile described by these data for frequent [adverse events] is consistent with prelicensure data, with the exception of syncope and [venous thromboembolic events].” Id. Thus, not only does Slade not support Plaintiffs’ position; it supports Merck’s.

The second article cited is an “‘editorial’ in a scientific journal” by Harper – maybe Dr. Diane Harper, whose early writing on HPV vaccines anti-vaccine activists liked (though her later writing was supportive of HPV vaccines). In it, the author suggested that “doctors should more extensively discuss the possibility of adverse events of all kinds with patients because cervical cancer is most often not fatal.” The editorial does not focus on POTS, and is not a study, and hence could not support causation. 

The court concludes: 

In summary, Bellwether Plaintiffs’ alleged “newly acquired information” of any connection between POTS and Gardasil before January 2011 is one published case of POTS and less than 20 unverified reports of POTS. It is an understatement to conclude that such evidence does not rise to the level that qualified scientists could find a “causal association” between POTS and Gardasil.

The court concluded this is not reasonable evidence on which Merck could have added POTS to the label, even if it wanted to. 

Concerning the clinical trial data, plaintiffs tried to use the fact that the insert included clinical trial reports of autoimmune diseases to support their claim. Putting aside the fact that autoimmune diseases general are not POTS, the court pointed out that the FDA did not ignore the question of autoimmune diseases generally, referencing a “post-marketing study required by FDA evaluating the risk of new-onset autoimmune conditions following Gardasil vaccination and concluding that there was no evidence of an increased risk of autoimmune conditions associated with vaccination).”

If 2015 is used as the end date, the court points out that “[b]y June 2015, there were still only 83 cases of POTS reported worldwide out of well more than 100 million administered doses of the vaccine.” Further, in 2015, the European Medicines Agency, responding to concerns raised by the Denmark Health Authority, conducted a review of the evidence on whether HPV vaccines caused POTS – and found that the evidence did not support such a causal link. (The court calls out the plaintiffs for ignoring the study). The court pointed out that later large studies did not find an association to POTS either. 

Even the one supportive study the plaintiffs brought – by Chandler from 2017 – based on unverified reports of Europe’s parallel to VAERS, which was not specific to POTS – “concluded that “a causal association between HPV vaccination and [possible POTS adverse events] remained uncertain [but] “the study authors believed that more extensive analysis of spontaneous reports can better identify relevant cases for thorough signal evaluation.”

In other words, even the plaintiffs’ best study did not actually show an association, but those authors would prefer to go with individual case studies (which generally cannot show causation) over the data that scientists usually use to examine causation. 

With this evidence, the court concluded that plaintiffs did not show that Merck had reasonable evidence to add a potential link to POTS to the label. In fact, although the court did not determine causation (since it determined Merck could not have changed the label, it did not reach the issue), the court’s review strongly suggests that the evidence it saw did not support causation.

Primary ovarian insufficiency

Our friend Orac, and the Skeptical Raptor, have previously written about the claim that Gardasil causes POI and the lack of evidence for it.

The issue here is “loss of ovarian activity before the age of 40” – in the case of plaintiffs, loss of it at a very young age, in most cases in the teens. Most women with POI will suffer from infertility. The court reminds us that “POI impacts approximately 1% of women under age 40, 0.1% of women under age 30, and 0.01% of women under age 20. Doc. No. 197-3 at 56-57. For up to 90% of diagnosed POI cases, the cause is currently unknown. About 4% of POI cases are estimated to have an autoimmune etiology”. Note that this tells us that without Gardasil, still, unfortunately, 1: 10,000 women would experience it. Give the vaccine to 10,000,000, and you will likely have cases in vaccinated women because of the base rate alone. 

The evidence here includes Merck’s animal studies in rats – but unfortunately for plaintiffs, those studies “revealed no evidence of impaired female fertility or harm to the fetus due to GARDASIL” – supporting, as the court pointed out, Merck’s position, not the plaintiffs.

In addition, in the trial, the plaintiffs’ experts point out that under new medical conditions, there were some reported cases – unfortunately, as the court points out since there were ““two ovarian failure cases … in the qHPV vaccine group and four in the placebo group”, the evidence goes the other way here, too. In terms of published case reports, we have one in 2012 and three more in 2013 – so four in total. The plaintiffs asked for an end date of 2013, and there were no others before, but even after, the plaintiffs submitted only one additional study with two more. So we have six case reports out of tens of millions of doses.

In terms of reports – VAERS had, between 2006-2013, 12 reports of POI – out of tens of millions of vaccinated women. Merck reported another 11. So adding case reports and treating all these unverified reports as true, we have 29 cases together of POI after the vaccine in tens of millions of women – well below the base rate due to chance alone. The court concluded: “considering the background rate of POI in the population, these few cases of reported POI following tens of millions of Gardasil vaccinations are not sufficient to generate a safety concern.”

There were no large studies up to 2013, but a large study published in 2018 covering “199,078 female patients from 2006 to 2014” found very few cases and concluded there was no significant risk. A 2021 study with “996,300 Danish-born girls and women aged 11 to 34 years, followed from 2007 to 2016” also found no significant increase in risk in the vaccinated. The same was true of another large study from 2022. 

As the court points out, “not only were all these scientific studies published well after November 2013, they conclude that Gardasil does not appear to increase the risk of POI.” (pp. 32-33). 

The court adds a criticism of the plaintiff’s expert’s approach to the study that is worth reproducing since it reflects a common approach on the anti-vaccine side:

Dr. Amato argues that the studies have several limitations, including that they are “underpowered” and “the small number of included studies.” In light of Dr. Amato’s willingness to rely on just a few cases of POI to support his opinions, these critiques, are, at best, incongruent.

I do not know the plaintiffs’ expert and do not suggest he is anti-vaccine, he may be simply falling prey to a very common and very human confirmation bias, but the approach does reflect something we see often from the anti-vaccine side: holding studies supporting vaccines to an impossible standard, while being willing to accept evidence that does not really show causation at all in support of claims of vaccines risks. 

Conclusion

Based on this evidence, the court convincingly concluded Merck did not have a basis to change the label without FDA approval, and therefore, it was legally forbidden from adding the warning plaintiffs claim it should have added – and the plaintiffs’ case failed. I hope readers realize why, although the court did not address causation, I think its analysis strongly implies plaintiffs did not have evidence to show that more likely than not – more than 50% likely – Gardasil caused their POTS and POI. 

The court opened the decision with this strong language that I think is worth quoting: 

As of 2013, when Plaintiffs contend that the latest of the warnings should have been given, more than a hundred million doses of Gardasil had been administered worldwide. However, by that time, there had been only one published, verified case of POTS and four published case reports of POI. …To the extent there were any broader studies or analyses before the relevant dates, they did not establish causal associations. Simply put, no scientist could reasonably conclude there is a causal association between POTS and POI and Gardasil based on this paucity of evidence.

Just so.  

What next? Although this decision was to the bellwether cases, it will likely be applied across the cases, and they will also be dismissed. Plaintiffs can try and appeal, but it would be hard to challenge this decision. At least one law firm seemed to suggest that plaintiffs acknowledge that and will refocus on state cases. This likely is the end of the multi-district litigation, though some plaintiffs may decide to appeal, and although there are still a handful of cases in the state courts – this decision may help Merck there, too.