The 13 Pediatric Vaccine Trials That Used an Active Substance in Place of Saline

An Essay on how the childhood schedule was never actually tested against a placebo

This essay examines the clinical trial design of thirteen vaccine products on the US childhood schedule, drawing on package inserts, FDA licensing documents, trial protocols filed with regulators, and the FUTURE II consent forms obtained in Denmark. The material engages the vaccine industry’s own technical vocabulary: antigen, placebo, adverse event, control group, blinding, adjuvant. Where that vocabulary appears in direct quotation or in attribution to a document, regulator, or publication, it refers to the specific source. Where it appears in the author’s analytical voice, it refers to the industry’s categories as the industry defines them, and does not imply endorsement of the biological framework those categories presuppose. The question examined is the question of what the trials actually did, which is independent of the question of whether the diseases the vaccines are nominally designed to prevent are what the establishment says they are. The documents say what they say. That is what this essay reads.

---

A 2002 Danish consent form for Merck’s FUTURE II trial told the young women who signed it this: “One half of the participants receive the active vaccine while the other half receives a so-called placebo vaccine (i.e. a vaccine without an active substance).”¹ A September 2004 revision of the form used a word that was easier for a Danish eighteen-year-old to understand. Saltvand. Saline.²

The Merck Protocol document for the same trial, V501-015, specified something else. The control group received AAHS, Merck’s proprietary amorphous aluminum hydroxyphosphate sulfate, the aluminum adjuvant that provokes the inflammatory response Merck wanted to measure. The ingredient is named on page 23 of the amendment Merck filed with regulators.³

Both documents exist. Both were filed. One of them describes what the trial actually did. The other is what the Danish teenagers were told before they signed.

The gap between these two documents is the opening into a question that reaches across the entire CDC childhood vaccination schedule. In the pivotal clinical trials used to license the vaccines routinely given to American infants and children, the control groups did not receive saline, sugar water, or any inert compound whose adverse event profile could serve as a baseline. The compounds given to the control groups were bioactive. In most cases they were other vaccines. In some cases they were the same vaccine with one ingredient removed. In the Protocol 018 subset of the Gardasil trial, the control group received a four-chemical mixture that the FDA then described as a “true saline placebo.”⁴

The pattern that follows is not an argument from first principles. It is a reading of the package inserts, the FDA licensing documents, and the trial protocols. Where investigators have obtained the consent forms, those are read too. The documents say what they say. They are available for anyone who wants to check them.

This is one count of thirteen.

Essay PDF Download (Free Subscribers)

A Note on the New Essay PDFs

Most essays now come with a designed PDF, available to all subscribers for now. Built as a slim 6×9 monograph with a line-art cover, proper typography, and back-matter pages that compress the argument into four beats, pull the hardest numbers onto a receipts page, and distill practical takeaways into a quick-reference sheet. The format is new and still finding its shape. What works? What doesn’t? What would you add, cut, or redesign? All ideas welcome.

Audio Deep Dive (Paid Subscribers Only)

Support This Work

The essays are free and will stay free. What paid subscribers get is the library that surrounds them: the full digital book catalogue, audio deep dives, archived, and embedded in each essay, and the Questions for Your Doctor, Before You Consent, and Package Insert series. Platinum Supporters also receive the digital edition of every paperback on the day it’s published.

Whichever tier suits, thank you for being here. If you’d like to become a paid subscriber, upgrade a tier, or gift a subscription to someone who might value the work, the options are below.

Gardasil: the anchor case

The young woman in the 2002 Danish trial was Kesia Lyng. She was eighteen when she enrolled, recruited by a brochure that reached her mailbox in Copenhagen six months after her grandmother died of cervical cancer. The brochure read, in Danish, “FUTURE 2 er IKKE et bivirkningsstudie,” which translates as “the FUTURE 2 study is NOT a side effect study.” Half of participants, the brochure said, would receive the vaccine; the other half would receive saline.⁵ Kesia enrolled in Protocol V501-015 and received her first injection in September 2002.

The Protocol document described a different trial. V501-015 specified that the control group would receive AAHS, Merck’s proprietary aluminum adjuvant, administered in volumes matched to the vaccine dose and from syringes prepared to look identical to the vaccine syringes. The 2002 consent form that Kesia signed described her alternative as “a vaccine without an active substance.” The vaccine without an active substance was aluminum.

AAHS does not look like saline. Saline is clear. AAHS is cloudy. The purpose of using AAHS as the control substance, Merck told the FDA when asked, was blinding, so that neither the investigators nor the participants could tell who got what.⁶ The agency accepted the answer. The alternative, had Merck wished to pursue it, was a clear control solution and a protocol in which the injections were prepared by staff who did not otherwise interact with the participants. This procedure was used in Protocol 018.

Protocol 018 is the trial that the FDA, when approving Gardasil, described as “saline placebo-controlled.” It was, according to the agency’s clinical review, “of particular interest” because this group allegedly used a true placebo compared with the AAHS control groups.⁷ The description was wrong.

The Protocol 018 control group, 594 children aged 9 to 15, received not saline but what the trial documents call a “carrier solution.” The carrier solution contained polysorbate 80, sodium borate, L-histidine, and the genetically modified yeast used in the manufacture of what Merck calls the HPV antigens.⁸ Polysorbate 80 crosses the blood-brain barrier and has been associated with cardiac risk and reproductive effects.⁹ Sodium borate, also known as Borax, is banned by the FDA for use in food products, though not in vaccines.¹⁰ L-histidine is an amino acid; the yeast is Saccharomyces cerevisiae genetically modified to express the vaccine components. In 2007, Merck scientists published a paper describing the Protocol 018 control as containing “identical components to those in the vaccine with the exception of HPV L1 VLPs and the aluminum adjuvant.”¹¹ Not saline. The vaccine, with two components removed.

In 2015, Dr. Deirdre Little, an Australian gynecologist, wrote to the Therapeutic Goods Administration, Australia’s drug regulator, pointing out that the Gardasil package insert in Australia described the Protocol 018 placebo as saline. The TGA investigated and replied to Dr. Little in writing. The agency concluded that “the current information is a misrepresentation of the situation. That is, that the current reference to the placebo being ‘saline’ misrepresents the actual placebo in Protocol 018.”¹² The TGA informed Dr. Little that it had asked Merck to correct the package insert. As of 2018, when Mary Holland and co-authors wrote The HPV Vaccine on Trial, the Australian package insert still called the placebo saline.¹³

Protocol 018 had other peculiarities. The ratio of vaccine recipients to controls was 2:1, not the 1:1 ratio used in every other Gardasil protocol. The sample was small enough that the trial lacked statistical power to detect anything other than the most common and most immediate reactions. The age group was the ultimate target population, 9 to 15 year olds, who would soon be receiving the vaccine on the schedule. The dose received by the vaccine arm appears, from the trial documents, to have been half the final approved formulation.¹⁴

The FDA’s 2006 clinical review contains a table (Table 229) that shows, for the Protocol 018 carrier solution group, zero serious adverse events in the first fifteen days after injection. A separate table (Table 295) in the same review reports 43 serious adverse events in a group it labels “placebo.” The second table has pooled the Protocol 018 carrier solution data with the Protocol 013, 015, and 019 AAHS data. All 43 serious adverse events, by Merck’s own earlier accounting, occurred in the AAHS group. Unless the reader cross-references the two tables, the impression created is that serious adverse events in the “placebo” group occurred at a rate similar to the rate in the vaccine group.¹⁵

Kesia Lyng and another Danish FUTURE II participant, a young woman named Sesilje, were both unblinded in 2007. Kesia had received the vaccine. Sesilje had received what the consent form had called the placebo. Both women had become chronically ill after their first injection, with the same pattern of headaches, muscle pain, exhaustion, and insomnia. Both were still ill when they met, in 2016, through a Danish vaccine-injury support group. At that meeting, Sesilje learned for the first time that the “placebo” she had received was not saline. She had received three doses of Merck’s aluminum adjuvant.¹⁶ The trial the Danish consent form had framed as a comparison between an active vaccine and a vaccine without an active substance was in fact a comparison between two aluminum-containing compounds. The chronic illness patterns in the two arms, over the following fourteen years, were indistinguishable.

The Protocol 018 story is the Gardasil story in miniature. A regulator who called the control a “true saline placebo” when the control was a four-chemical mixture. A manufacturer who said in one document what was contradicted by another document the same manufacturer also filed. Danish girls who were told they might get saline and who, had they been randomized to the control arm of FUTURE II, would have received an aluminum adjuvant. A regulator in Australia who, when pressed, admitted the “saline” description was a misrepresentation but, seven years later, had not required its correction on the package insert.

The pattern Protocol 018 reveals, Protocol 015 confirms, and the Danish consent forms document, runs across the entire schedule.

The thirteen

The vaccines routinely administered to American children under five are licensed by the FDA on the basis of clinical trials in which the control groups received other vaccines, previously licensed versions of the same vaccine, the vaccine with one ingredient removed, aluminum-containing solutions, aluminum-and-mercury-containing solutions, or, in two documented cases, nothing at all. In none of the pivotal licensing trials for the vaccines on the under-two childhood schedule did the control group receive saline.

The description above is not an interpretation of the evidence. It is a summary of what the package inserts, the FDA licensing documents, and the trial protocols themselves say. Where the control substance has been described in those documents as a “placebo” or a “saline placebo,” the description has been contradicted by the Protocol or by the licensing document that describes what the trial actually used.

The thirteen trials, in order.

1. Pediarix

Pediarix is GSK’s five-in-one combination vaccine, administered to American infants at two, four, and six months. It combines components for protection against diphtheria, tetanus, pertussis, hepatitis B, and polio. The package insert describes safety testing conducted in 14 clinical trials involving 8,088 subjects.¹⁷

The largest trial, conducted in Germany, administered Pediarix together with a Hib vaccine to the trial group. The control group received Infanrix, GSK’s standalone DTaP vaccine, along with a hepatitis B vaccine, a Hib vaccine, and oral polio vaccine. Four injections given to the control group so that the three injections given to the trial group could be assessed for safety. In another trial explicitly described in the package insert, Pediarix was tested against a control group receiving Infanrix, hepatitis B vaccine, and inactivated polio vaccine. The remaining 12 trials, by the package insert’s own account, used “comparator vaccines” whose composition is not disclosed in the labeling document.¹⁸

The Pediarix package insert does not describe any trial in which the control group received saline. It does not describe any trial in which the control group received an inert substance of any kind. The adverse event rates reported for Pediarix are compared with adverse event rates in groups that received other vaccines, including the vaccine Pediarix was designed to replace. If the comparator vaccines themselves have adverse event profiles, which they do, the rates reported for Pediarix cannot be interpreted against a known baseline. The 8,088 subjects were studied in a trial design that systematically prevents the calculation of a true adverse event rate.

The FDA licensed Pediarix in 2002.

2. Kinrix

Kinrix is GSK’s four-in-one combination, identical to Pediarix minus the hepatitis B component. It is administered to American children at age four to six as the fifth DTaP dose and the fourth polio dose. In the vaccine’s largest clinical trial, the control group received Infanrix and IPOL, GSK’s standalone polio vaccine. All trial participants, in both the Kinrix arm and the control arm, were concomitantly administered a dose of the MMR vaccine.¹⁹

The design places the Kinrix question against a background of three other injections given to everyone. Any adverse event occurring in the Kinrix arm must then be distinguished from adverse events caused by the MMR dose given to the same child. The control-arm adverse events must be distinguished from the MMR dose and from the DTaP and polio doses they received. The comparison between the two arms is a comparison between four vaccine components and five vaccine components. The isolated adverse event signal for Kinrix is not calculable from this design, and the package insert does not report it.

The Kinrix package insert mentions no trial involving a placebo control group of any kind, inert or otherwise.²⁰

The FDA licensed Kinrix in 2008. The licensing document relies on the trial above and on GSK’s prior licensing of Infanrix and IPOL. Infanrix, as the next entry shows, was never tested against saline. IPOL, as the entry further down shows, was tested against the first-generation Prevnar pneumococcal vaccine. Neither of the two foundations on which Kinrix’s licensing rests involves a trial in which a child received an inert control substance.

Kinrix is on the pediatric schedule in all fifty states.

3. Infanrix

Infanrix, GSK’s standalone DTaP vaccine, is the backbone of the Pediarix and Kinrix combinations examined above and of a wider family of DTaP-containing combination vaccines manufactured by GSK and Sanofi Pasteur. The question of how Infanrix was itself tested is the question of what, precisely, the DTaP combination family is tested against.

The Infanrix package insert reports one clinical trial in which the control group received the predecessor DTP vaccine, the whole-cell pertussis formulation that Infanrix was designed to replace. The second trial reported in the insert had no control group.²¹

The DTP vaccine itself, which was replaced in the late 1990s after years of reporting on severe adverse events including seizures, encephalopathy, and death, was never tested in a modern randomized controlled trial against a placebo. It was developed in the 1930s and 1940s, at a time when the modern randomized controlled trial as a standard of pharmaceutical testing did not yet exist. The trials that led to DTP’s acceptance into the pediatric schedule contained no randomized control group whose adverse event profile could be compared with the trial group’s.²²

A search of ClinicalTrials.gov, the US government registry of clinical trials conducted since the year 2000, returns dozens of results for GSK’s Infanrix family. None of them specifies a randomized controlled trial with a saline placebo control group.²³

The DTaP backbone of the childhood schedule has not been tested against saline in any trial that is publicly documented. It is tested against DTP, which was never tested against saline, which was never tested at all. The combinations built on Infanrix (Pediarix, Kinrix, Pentacel, Quadracel) are tested against Infanrix. The architecture repeats downward. It does not terminate in a trial that would allow the adverse event rate of DTaP to be calculated against a known baseline.

4. Pentacel

Pentacel is Sanofi Pasteur’s five-in-one combination (DTaP, polio, Hib), the Sanofi analog of GSK’s Pediarix. It was tested for safety in four clinical trials during its licensing process. In three of the trials, the control group participants received an assortment of different vaccines in combinations that varied by study site and by trial.²⁴

The fourth trial appears, in the package insert, to have had a control group that received no vaccines. This would be the only trial in the Pentacel licensing dossier where the control group was not administered another bioactive compound. On examination of the clinical review document submitted to the FDA, the trial does not appear to have had a control group at all. The description in the package insert indicates a comparison group where the actual trial documentation indicates none.²⁵

The reader who checks the package insert and does not then check the FDA clinical review will understand Pentacel to have been tested, in one trial, against children who received no injection. The reader who checks the FDA clinical review will understand that this fourth trial has no control group. The package insert and the FDA clinical review, both filed by Sanofi Pasteur with the US government and both publicly available, describe the same trial differently.

Pentacel was licensed by the FDA in 2008 on the basis of these four trials. The absolute adverse event rate for Pentacel, against a known baseline, is not reported, because the trials were not designed to produce it.

5. Daptacel

Daptacel is Sanofi Pasteur’s standalone DTaP vaccine, the Sanofi analog of GSK’s Infanrix. It is the control substance in the Pentacel trial above and in the Quadracel trial (DTaP, polio). The question of how Daptacel was tested is the question of what the three-dose Daptacel foundation rests on.

Daptacel underwent four clinical trials during its licensing process. All four were randomized and controlled. In each of them, the control group received a different combination of DTaP or DTP vaccines, sometimes concurrently with other vaccines.²⁶

The design parallels the Kinrix design. The control group in every Daptacel trial was given a bioactive compound against which the Daptacel arm’s adverse event rate could be reported. The question of what Daptacel’s adverse event rate is, measured against a child who received no injection, is not asked by any of the four trials and cannot be answered from the trial record.

The Daptacel package insert does not mention a trial involving a saline placebo control group.²⁷

Daptacel was licensed by the FDA in 2002. Pentacel’s licensing in 2008 was based in part on Daptacel’s prior licensing. Quadracel’s licensing in 2015 was based in part on Daptacel’s prior licensing and in part on the licensing of IPOL. The combination-vaccine family manufactured by Sanofi Pasteur sits on Daptacel. Daptacel sits on a trial whose control groups received other vaccines. Those other vaccines sit on trials whose control groups received still other vaccines. The architecture matches the GSK architecture examined above. No trial in the entire family tested a DTaP-containing product against a baseline that could support the calculation of an absolute adverse event rate.

The Sanofi DTaP family and the GSK DTaP family together constitute the full set of pertussis-containing vaccines on the American pediatric schedule.

6. Hiberix

Hiberix is GSK’s monovalent Haemophilus influenzae type b vaccine, administered to American infants as part of the Hib series or as a booster in combination schedules. The package insert reports a single randomized controlled trial. In that trial, the vaccine was administered concomitantly with several other vaccines, including DTaP, polio, and hepatitis B. The two control groups received, respectively, a different Hib vaccine or a DTaP-polio-Hib combination vaccine, along with several other vaccines. Seven other trials mentioned in the leaflet were open-label and apparently had no control groups.²⁸

Two other monovalent Hib vaccines are approved for use in the US. ActHIB, manufactured by Sanofi Pasteur, was clinically tested for safety in three randomized controlled trials. In the first, the vaccine was administered simultaneously with the DTP vaccine, and the control group received the DTP vaccine alone. In the other two trials, the trial groups received a combination vaccine with ActHIB as one component, while the control groups received the same combination vaccine without the ActHIB component plus a separate ActHIB dose.²⁹

PedvaxHIB, Merck’s monovalent Hib vaccine, was tested in Native American infants. The package insert states that the control group received a placebo. All trial participants, both those in the vaccine arm and those in the control arm, were also given the DTP vaccine and the oral polio vaccine concomitantly.³⁰ The adverse event rate reported for PedvaxHIB is therefore not the adverse event rate of PedvaxHIB. It is the adverse event rate of PedvaxHIB plus DTP plus OPV, compared with the adverse event rate of placebo plus DTP plus OPV.

Three monovalent Hib products on the US schedule. None has been tested against a true inert control in a randomized controlled trial that isolates the Hib vaccine from concomitantly administered compounds. The combination products that include the Hib component (Pentacel, Vaxelis) rely on these licensings as their point of departure.

7. Prevnar

The original Prevnar, Pfizer’s pneumococcal conjugate vaccine, was tested in 1998 in a trial whose control group received a meningococcal vaccine that had not yet been licensed and was described in the trial documents as “experimental.”³¹

In 1998, saline and sterile water were both available as inert control substances, and no other licensed pediatric vaccine required co-administration with the pneumococcal vaccine. There was no ethical argument for giving the control group of the Prevnar trial any bioactive compound, let alone an unlicensed experimental vaccine. The decision to use an experimental meningococcal vaccine as the control substance for Prevnar was a decision taken by the manufacturer and accepted by the FDA.

The current Prevnar formulation on the US schedule, Prevnar-13, which covers thirteen pneumococcal serotypes rather than the original seven, was licensed on the basis of trials in which the control group received the original Prevnar.³² Prevnar-13’s adverse event rate is therefore reported against the adverse event rate of the vaccine that was tested in 1998 against an experimental meningococcal vaccine. Each successor vaccine’s safety profile is reported against the safety profile of the preceding vaccine, which was tested against the vaccine that came before it.

In the history of pediatric pneumococcal vaccination in the United States, no pneumococcal vaccine licensed for pediatric use has been tested against a saline control in a randomized controlled trial.

8. IPOL

IPOL is Sanofi Pasteur’s inactivated polio vaccine, the standalone polio vaccine on the US pediatric schedule and the polio component incorporated into Pediarix, Kinrix, Pentacel, and Quadracel. The package insert mentions no pre-licensure randomized controlled trials for the vaccine at all.³³

A 2018 Freedom of Information Act release from the FDA revealed that IPOL underwent two clinical trials, neither of which meets the current standard for a Phase 3 randomized controlled trial. The first trial, conducted in 1980 to 1983, enrolled 371 subjects in the IPOL arm and a comparable number in the control arm. The control group received the oral polio vaccine manufactured by Lederle, which was the only polio vaccine licensed in the United States at the time. All participants, in both the IPOL arm and the control arm, concomitantly received the DPT vaccine. The second trial, conducted in Buffalo, New York, in the late 1980s, enrolled 114 children who received three doses of either IPOL, oral polio vaccine, or a combination of both. Most of the children in that trial also received the DPT vaccine. The second trial was not controlled, not randomized, and not blinded.³⁴

The 1954 Salk Phase III field trial, enrolling nearly two million children, is sometimes cited as evidence that pediatric polio vaccines in the United States were, at least once, tested against a true placebo. The claim should be examined against the actual trial documents by any reader relying on this historical reference. The 1954 Salk trial is not among the trials examined in this essay; the trials examined here are the pre-licensure trials for the vaccines currently on the schedule.

The polio vaccine, the vaccine whose public reputation is most strongly associated with mid-twentieth-century clinical testing, is on the current pediatric schedule in a formulation whose licensing documents disclose two small trials, one of which was not a controlled trial and one of which used the oral polio vaccine as the comparator.

9. Engerix-B

Engerix-B is GSK’s hepatitis B vaccine, administered to American newborns in the hospital at birth and in two further doses in the first six months. The package insert describes the pivotal safety data in a single sentence. “The incidence of local and systemic reactions was comparable to those of plasma derived hepatitis B vaccines.”³⁵

That is the complete safety description. The vaccine was tested against the prior-generation hepatitis B vaccine, which was a different formulation of what the manufacturer calls the hepatitis B antigen, manufactured from blood plasma rather than from recombinant yeast. The adverse event rate of the plasma-derived predecessor was not itself established against a known baseline. The predecessor, Heptavax, was the first hepatitis B vaccine licensed in the United States, in 1981, and was no longer in use after recombinant alternatives became available in the mid-1980s.

A second trial reported in the Engerix-B package insert enrolled children aged 11 to 15. In this trial, the trial group and the control group both received Engerix-B, with different dosing regimens. Neither arm of the trial contained a child who received no injection.³⁶

Recombivax-HB, Merck’s hepatitis B vaccine, is the other recombinant hepatitis B vaccine licensed for use in the United States. The package insert for Recombivax-HB does not describe any randomized controlled trial of the vaccine’s safety in infants.³⁷

Twinrix, GSK’s combination hepatitis A and hepatitis B vaccine, was tested in clinical trials in which the control group received GSK’s standalone hepatitis A and hepatitis B vaccines (Havrix and Engerix-B), administered separately.³⁸

Three hepatitis B products on the US childhood schedule. None has been tested against saline. The one with the most explicit description of its safety testing is tested against another hepatitis B vaccine of a different formulation. The other two are tested against vaccine-containing comparators or on data that does not include a randomized controlled trial at all.

10. Havrix

Havrix is GSK’s hepatitis A vaccine, administered to American children at twelve to eighteen months. In a Thailand trial enrolling over 40,000 participants, the Havrix trial group was compared with a control group that received Engerix-B, GSK’s hepatitis B vaccine.³⁹

The design is instructive. The participants in both arms received a vaccine. The question of whether Havrix produced more or fewer adverse events than Engerix-B could, in principle, be answered. The question of whether Havrix produces adverse events in a child who would not have had them otherwise could not. The trial produces a comparison between two vaccines and establishes nothing about either vaccine’s absolute adverse event rate.

In three other Havrix clinical studies, the trial groups received Havrix concurrently with another vaccine and the control groups received MMR, varicella, or other licensed pediatric vaccines in various combinations.⁴⁰

Vaqta, Merck’s hepatitis A vaccine, is the other hepatitis A product on the US childhood schedule. According to FDA licensing documents, Vaqta was tested in two clinical trials. The first, known as the “Monroe” study, had no control group. The second, in addition to having no control group, administered Vaqta along with two other vaccines.⁴¹

The Vaqta package insert tells a different story about the Monroe study. The package insert states that the Monroe trial had a control group that received a placebo. Examination of the trial documents reveals that the compound given to the “placebo” group contained the vaccine’s aluminum adjuvant and the preservative thimerosal, a mercury-based compound removed from most US pediatric vaccines in the early 2000s.⁴² The Monroe study “placebo” was aluminum plus mercury.

Two hepatitis A products on the US childhood schedule. Neither has been tested against an inert control in a randomized controlled trial. One of them was tested against an aluminum and mercury solution the manufacturer called a placebo.

11. MMR II and ProQuad

Merck’s MMR II is the measles, mumps, and rubella vaccine administered to American children at twelve to fifteen months and again at four to six years. The MMR II package insert does not mention any safety trials. A 2018 Freedom of Information Act release from the FDA revealed that MMR II was tested in the mid-1970s in eight small clinical trials involving approximately 850 children. The control groups in all of these trials received either the predecessor MMR vaccine, a measles-rubella combination vaccine, or a single-dose rubella vaccine. Some of the trials were randomized. None were blinded.⁴³

The predecessor MMR vaccine, licensed in 1971, was itself tested in a small number of trials in which the vaccine was given to approximately a thousand infants and children. The control groups totaled about one-tenth that number, were mostly siblings of the vaccinated children (which violates the randomization principle), and received no injection at all.⁴⁴

Merck’s ProQuad, the four-in-one measles, mumps, rubella, and varicella combination, was tested for safety in several randomized clinical trials, most of which were not blinded (”open label”). The largest randomized controlled trial compared ProQuad with a control group that received the older-generation MMR II vaccine and the Varivax vaccine administered at the same time. Another study, conducted in children aged four to six years, divided the subjects into three groups. The first group received ProQuad and a “placebo.” The second received MMR II and a “placebo.” The third received MMR II and Varivax. In the open-label studies, participants in both trial and control groups received additional vaccines.⁴⁵

The three licensed MMR-containing products on the US pediatric schedule (MMR II, ProQuad, and the historical MMR) have been licensed on the basis of trials in which the control groups received other vaccines, received the predecessor vaccine, or received no injection at all. None of them has been tested against a saline control in a randomized controlled trial.

12. Varivax

Varivax, Merck’s single-dose varicella vaccine, has a safety section in its package insert that describes “a double-blind, placebo-controlled study” of 914 healthy children and adolescents. According to the insert, “only two mild symptoms, pain and redness at the injection site, occurred at a significantly greater rate in vaccine recipients than in placebo recipients.”⁴⁶

The reader who stops at the package insert will understand this as a trial in which a vaccine was tested against saline and found to produce only two mild local reactions at a greater rate than saline.

The reader who proceeds to the published paper describing the Varivax trial will find a different description. The compound given to the “placebo” group in the Varivax trial was the test vaccine with what the manufacturer identifies as the attenuated-virus component removed. The adjuvant and the other excipients were present. The attenuated varicella component was not.⁴⁷

The reader who understands that the “placebo” was the vaccine with one ingredient removed has a different interpretation of the trial result. Pain and redness at the injection site were the only symptoms that occurred at a greater rate in the vaccine group than in the “placebo” group. If the “placebo” group received a compound containing the same adjuvant as the vaccine, then the trial result shows that the attenuated-varicella component itself produces pain and redness at the injection site, and that no other adverse event in the trial was more common in the vaccine arm than in the arm receiving the vaccine with that one ingredient removed.

A separate controlled Varivax study compared two different formulations of the vaccine. According to the package insert, the safety profiles of the two formulations were comparable.⁴⁸

Varivax has not been tested in a randomized controlled trial against an inert control. The varicella component of the MMRV combination (ProQuad) has not been separately tested in that way either.

13. RotaTeq and Rotarix

The rotavirus vaccines occupy a particular position in the childhood schedule because the circumstances of their licensing eliminated the standard arguments against saline-controlled testing. The first rotavirus vaccine, RotaShield, was pulled from the market in 1999 after it was linked to intussusception, a telescoping of the intestine that can be fatal in infants.⁴⁹ The second generation of rotavirus vaccines, Merck’s RotaTeq and GSK’s Rotarix, were developed in a market where no comparator vaccine was available. The oral polio vaccine, which could in principle have been used as a comparator for the oral rotavirus vaccines, had been withdrawn in the United States.

The ethical arguments against inert controls, which require an available alternative vaccine to be given to the control arm, did not apply. Sugar water, administered orally, would have provided a baseline. Saline, administered orally, would have provided a baseline. The manufacturers chose otherwise.

The Rotarix pivotal trial, enrolling approximately 63,000 infants, used as its control substance a compound described in the GSK FDA licensing document as the vaccine without its antigenic component.⁵⁰ The RotaTeq pivotal trial used a control substance whose composition was deleted from the FDA licensing document and marked as a trade secret. In a separate RotaTeq clinical trial, the control group received the vaccine-sans-antigen, the same compound used in the Rotarix trial.⁵¹

The adverse event rates reported in both trials were similar between the trial arm and the control arm. In the Rotarix “placebo” group, approximately one in thirty infants experienced a severe medical event, 16 infants suffered intussusception, and 43 died. In the RotaTeq “placebo” group, one in forty subjects experienced a serious adverse event, 15 suffered intussusception, and 20 infants died.⁵²

The Rotarix package insert states, on the specific question of intussusception: “No increased risk of intussusception was observed in this clinical trial following administration of ROTARIX when compared with placebo.”⁵³ The sentence is true. The sentence conveys the opposite of what the words suggest. The “placebo” was the vaccine with the active component removed. The adverse event rates in the two arms were similar because the two compounds were similar. The 43 deaths in the Rotarix “placebo” group were deaths in a trial arm whose composition is a trade secret in some documents and is described as the vaccine-sans-antigen in others.

Thirteen vaccine products on the childhood schedule. The thirteenth, like the first twelve, was tested against a bioactive comparator.

The Infanrix inheritance

The thirteen entries above repeat a pattern, and the pattern has a structure. The structure is architectural, in the sense that each vaccine’s licensing rests on the licensing of another vaccine, and the inheritance of licensing trial design propagates downward through the schedule.

Pediarix was tested against Infanrix plus other vaccines. Kinrix was tested against Infanrix plus other vaccines. Pentacel was tested against Daptacel plus other vaccines. Quadracel was tested against Daptacel plus other vaccines. The combination vaccines rest on the standalone DTaP vaccines.

Infanrix was tested against DTP. Daptacel was tested, in each of its four trials, against a different combination of DTaP or DTP vaccines. The standalone DTaP vaccines rest on their predecessors or on other licensed vaccines.

DTP, the whole-cell pertussis predecessor, was developed in the 1930s and 1940s. It was never tested in a modern randomized controlled trial. The licensing documentation for DTP, insofar as it exists in modern form, consists of accumulated clinical experience and the post-market surveillance that was conducted after the vaccine was already on the schedule.

The pneumococcal line has the same architecture. Prevnar-13 was tested against the original Prevnar. The original Prevnar was tested against an experimental meningococcal vaccine. The experimental meningococcal vaccine was never licensed.

The polio line has the same architecture. IPOL was tested, in the one trial of IPOL that was controlled, against the oral polio vaccine made by Lederle, which was the only polio vaccine then on the US market.

The hepatitis A line has the same architecture. Havrix was tested against Engerix-B. Engerix-B was tested against Heptavax, the first-generation plasma-derived hepatitis B vaccine. Heptavax was the first US-licensed hepatitis B product and was itself not tested against an inert control.

The architecture terminates, in each line, in a trial that did not involve a control group receiving an inert substance. The termination points vary. Infanrix and DTP terminate in trials from the 1930s and 1940s whose designs predate the modern randomized controlled trial. Prevnar terminates in a trial against an experimental compound. Varivax terminates in a trial against the vaccine with one component removed.

What the architecture does not do, at any point along its downward trajectory, is terminate in a trial whose control substance was saline. The thirteen vaccines on the schedule were licensed on a foundation that nowhere in its entire structure contains such a trial. The foundation is the subject of this essay.

The ring of Helsinki

The Declaration of Helsinki, formulated by the World Medical Association in 1964 and revised multiple times since, is the current international statement of ethical principles for medical research involving human subjects. Its provisions include the following. Physicians may not be involved in a research study involving human subjects unless they are confident that the risks have been adequately assessed and can be satisfactorily managed. A potential research subject who is incapable of giving informed consent must not be included in a research study that has no likelihood of benefit for them unless the research entails only minimal risk and minimal burden.⁵⁴

Infants cannot give informed consent, and the question of what the “placebo” actually contains cannot be put by them.

In the Rotarix and RotaTeq pivotal trials, infants in the control arms received the test vaccine with the antigenic component removed. The compound provided no potential benefit; the removed component is the one that, according to the manufacturers, prevents the condition the vaccine is marketed against. The compound carried significant potential for harm; the demonstrated adverse event rates in the control arms (16 intussusceptions and 43 deaths in Rotarix; 15 intussusceptions and 20 deaths in RotaTeq) establish the harm as not theoretical.⁵⁵

The Nuremberg Code, drafted in 1947 in the aftermath of the Nazi medical experiments, states that an experiment “should be so conducted as to avoid all unnecessary physical and mental suffering and injury.”⁵⁶

The suffering was not necessary. Sugar water, administered orally, was available in the late 1990s when the second-generation rotavirus trials were being designed, as was oral saline. The use of the vaccine-sans-antigen as the control substance in trials of infants constitutes, by the ordinary application of the Declaration of Helsinki and the Nuremberg Code, a research protocol that could not be ethically justified. The protocols were approved anyway. The vaccines are on the schedule.

The documents say what they say

A 2002 Danish consent form for Merck’s FUTURE II trial told the Danish teenagers who signed it that they would receive either the vaccine or “a vaccine without an active substance.” A September 2004 revision of the same consent form used a clearer word. Saltvand.⁵⁷

The actual Protocol V501-015 specified AAHS, Merck’s proprietary aluminum adjuvant, administered to the control group in volumes matched to the vaccine dose.

Both documents were filed. Both were signed. One of them described what the trial actually did. The other described what the participants were told.

In 2015, when an Australian gynecologist pressed the Australian regulator on the “saline” description of a different Gardasil protocol, the regulator wrote back that the description was a misrepresentation of the situation.⁵⁸ The regulator asked the manufacturer to correct the package insert. The package insert was not corrected within the subsequent years that are examined in the public record.

The misrepresentation is on thirteen package inserts. Not one.

The package inserts, the FDA licensing documents, and the Pentacel clinical review that contradicts the Pentacel package insert on the question of whether the fourth trial had a control group at all are on the FDA website. The 2007 Merck paper describing the Protocol 018 control as the vaccine minus two components is in the peer-reviewed literature. The 2015 TGA letter to Dr. Deirdre Little is in her correspondence and has been reproduced in print. The Danish FUTURE II consent forms are in the files of the Danish Cancer Society.

A document that reads differently from the trial it describes is still a document. The protocols exist. They say what they say. The children on the schedule were injected on the basis of trials whose control substances were not what the labeling documents said they were.

The documents are open. The childhood schedule is on the public record. The regulator in Australia has already said, in writing, what the matter is.

How to Explain It to a Six-Year-Old

Imagine a candy company is making a new kind of candy. The company wants to find out if the new candy makes kids feel sick. So the company says it will do a test. The company picks a big group of kids. Half the kids will eat the new candy, and the other half will be the comparison group. If the kids who eat the new candy get sick about the same amount as the kids in the comparison group, the company will say the new candy is safe. That is the whole plan.

Here is what the comparison group gets to eat. The comparison group does not get a glass of plain water. The comparison group gets a different candy. Sometimes it is an old candy the company already sells. Sometimes it is almost the same as the new candy, except one of the ingredients has been taken out. One time the comparison group got a candy nobody had ever eaten before, because the company said it was more interesting to compare the new candy to that one. The comparison group never, ever gets a plain glass of water. The company has been testing new candies for a long time, and the comparison group has always been given another candy.

If both groups of kids feel sick, the company says the new candy is safe. The company says, look, the comparison kids feel sick too, and we never gave them our new candy, so the new candy did not do it. But the comparison kids were eating candy. They were eating the kind of candy that already made kids feel sick, which is why the company made the new candy in the first place. Nobody has ever tested a candy made by this company against a glass of plain water, which is the only test that would tell you if candy itself makes kids feel sick. And in the test the company runs, the kids all keep feeling sick, and the company keeps saying every new candy is safe. The kids are not drinking any water in this story. The kids are only eating candy. There is a lot of candy, and the kids are small.

---

In Print

The Unbekoming library is available in paperback, printed to order through Lulu and shipped worldwide. The shelf begins with the paradigm question underneath everything else — No Virus, the isolation problem, the collapse of virology’s foundational claims, and a disease-by-disease reappraisal — and moves through the suppressed compounds mainstream medicine set aside: The DMSO Book, Chlorine Dioxide: The Forbidden Remedy, The Iodine Book, and The Hydrogen Peroxide Book. Two more recover what’s still on the kitchen shelf: Baking Soda and The Castor Oil Book. Two more recover the minerals modern soil, water, and processing quietly stripped from the diet: The Magnesium Handbook and The Boron Book. Sitting alongside these is No Contagion, co-authored with Jamie Andrews — the case against germ theory itself, catalogued through 258 failed contagion experiments.

The critique books cover what medicine, dentistry, psychiatry, and veterinary practice have become. The Unvaccinated treats the completely unvaccinated as a comparison group across twenty chapters and five appendices. Medicalized Motherhood follows a woman through 123 documented interventions from teenage pill to postpartum discharge. Drilling for Profit treats cavities, gum disease, and crooked teeth as the dietary problem they are. What Your Vet Can’t Tell You applies the same critique to pets. Escape from Psychiatry documents the fabrication of the DSM and the specific damage of every major psychiatric drug class. The Vitamin K Injection covers what happens in the first hours of a newborn’s life.

The full shelf is at lulu.com/spotlight/unbekoming. A physical book reaches the person a Substack post never will — the skeptical relative, the friend who won’t click a link but might open a book, the visitor whose eye lands on a coffee table. Buy one to keep, and one to give away.

References

1. Mary Holland, Kim Mack Rosenberg, and Eileen Iorio, The HPV Vaccine on Trial: Seeking Justice for a Generation Betrayed (Skyhorse, 2018), Chapter 5, “The Denial of Informed Consent,” reproducing the 2002 Danish informed consent form for the FUTURE II trial. Source cited: Danish Cancer Society.

2. Holland, Rosenberg, and Iorio, The HPV Vaccine on Trial, Chapter 5, discussing the September 2004 revision of the Danish FUTURE II consent form using the word “saltvand.”

3. Merck Protocol V501-015 (FUTURE II), Protocol/Amendment No. 015-00, at 23, available at https://www.scribd.com/document/367386168/V501-015-00-PRO-VD. Discussed in Holland, Rosenberg, and Iorio, The HPV Vaccine on Trial, Chapters 2, 5, and 8.

4. FDA 2006 Gardasil Clinical Review, cited in Holland, Rosenberg, and Iorio, The HPV Vaccine on Trial, Chapter 8. The FDA described the Protocol 018 control arm as “saline placebo-controlled” and “of particular interest” for that reason.

5. Holland, Rosenberg, and Iorio, The HPV Vaccine on Trial, Chapter 2, “Injured in the Trials: Testimony from Denmark,” discussing the FUTURE 2 recruitment brochure sent to Danish eighteen-to-twenty-three-year-olds.

6. Letter from the FDA to the authors of The HPV Vaccine on Trial, cited in Chapter 8, explaining that AAHS was used as the control substance because it “looked similar” to the vaccine.

7. FDA 2006 Gardasil Clinical Review, cited in Holland, Rosenberg, and Iorio, Chapter 8.

8. Composition of the Protocol 018 carrier solution: Merck Protocol V501-018 and the 2007 Merck paper (see note 11), cited in Holland, Rosenberg, and Iorio, Chapter 8.

9. Published research on polysorbate 80, cited in Holland, Rosenberg, and Iorio, Chapter 8.

10. FDA regulatory restriction on sodium borate in food products, cited in Holland, Rosenberg, and Iorio, Chapter 8.

11. Merck scientists et al., 2007 paper describing the Protocol 018 control arm, cited in Holland, Rosenberg, and Iorio, Chapter 8. The exact citation of this paper should be verified against the original publication before republication of this essay.

12. Letter from the Therapeutic Goods Administration (Australia) to Dr. Deirdre Little, 2015, quoted in Holland, Rosenberg, and Iorio, Chapter 7, note 13. The TGA’s conclusion that “the current reference to the placebo being ‘saline’ misrepresents the actual placebo in Protocol 018” is reproduced in that chapter from Dr. Little’s correspondence.

13. Holland, Rosenberg, and Iorio, The HPV Vaccine on Trial, Chapter 7, noting that the Australian Gardasil package insert still described the Protocol 018 placebo as saline as of July 2018, three years after the TGA’s letter asked Merck to correct it. The current state of the Australian package insert in 2026 should be checked by any reader relying on this claim.

14. Holland, Rosenberg, and Iorio, Chapter 8, on the Protocol 018 ratio, dosing, and age group anomalies.

15. FDA 2006 Gardasil Clinical Review, Tables 229 and 295, discussed in Holland, Rosenberg, and Iorio, Chapter 9.

16. Holland, Rosenberg, and Iorio, The HPV Vaccine on Trial, Chapter 2, “Injured in the Trials: Testimony from Denmark,” on the stories of Kesia Lyng and Sesilje, their chronic illness beginning after their first FUTURE II injections, and Sesilje’s realization in 2016 that the “placebo” she had received was Merck’s AAHS aluminum adjuvant, not saline.

17. Pediarix package insert, cited in Turtles All The Way Down: Vaccine Science and Myth (Anonymous, revised English edition), Chapter on the clinical trials of childhood vaccines.

18. Pediarix package insert, cited in Turtles All The Way Down.

19. Kinrix package insert, cited in Turtles All The Way Down.

20. Kinrix package insert, cited in Turtles All The Way Down.

21. Infanrix package insert, cited in Turtles All The Way Down.

22. Turtles All The Way Down, discussing the history of the whole-cell DTP vaccine trials in the 1930s and 1940s.

23. ClinicalTrials.gov search for the Infanrix family, discussed in Turtles All The Way Down.

24. Pentacel package insert, cited in Turtles All The Way Down.

25. Pentacel FDA Clinical Review, cited in Turtles All The Way Down.

26. Daptacel package insert, cited in Turtles All The Way Down.

27. Daptacel package insert, cited in Turtles All The Way Down.

28. Hiberix package insert, cited in Turtles All The Way Down.

29. ActHIB package insert, cited in Turtles All The Way Down.

30. PedvaxHIB package insert, cited in Turtles All The Way Down. The pre-licensure trial was conducted in Native American infants with concomitant DTP and oral polio vaccine administration to all arms.

31. Prevnar clinical trial documentation (1998), discussed in Turtles All The Way Down. The control group received an experimental, unlicensed meningococcal vaccine.

32. Prevnar-13 package insert and FDA licensing documents, cited in Turtles All The Way Down and in Edward Geehr, Unavoidably Unsafe: What Parents Should Know About the Vaccine Schedule (2024), Chapter on the pneumococcal vaccine.

33. IPOL package insert, cited in Turtles All The Way Down.

34. 2018 FDA Freedom of Information Act release concerning IPOL licensing trials, cited in Turtles All The Way Down. The 1980–1983 trial of 371 subjects per arm and the late-1980s Buffalo trial of 114 children are both described in that document.

35. Engerix-B package insert, cited in Turtles All The Way Down.

36. Engerix-B package insert, cited in Turtles All The Way Down.

37. Recombivax-HB package insert, cited in Turtles All The Way Down.

38. Twinrix package insert, cited in Turtles All The Way Down.

39. Havrix package insert, Thailand trial, cited in Turtles All The Way Down.

40. Havrix package insert, additional trials, cited in Turtles All The Way Down.

41. Vaqta FDA licensing documents, cited in Turtles All The Way Down.

42. Vaqta Monroe study documentation, cited in Turtles All The Way Down and Geehr, Unavoidably Unsafe, Chapter on the hepatitis A vaccine.

43. MMR II package insert and 2018 FDA FOIA release describing the mid-1970s licensing trials, cited in Turtles All The Way Down.

44. Historical MMR vaccine trials (1971 licensing), cited in Turtles All The Way Down.

45. ProQuad package insert and published trial documentation, cited in Turtles All The Way Down.

46. Varivax package insert, cited in Turtles All The Way Down.

47. Varivax clinical trial publication, cited in Turtles All The Way Down.

48. Varivax package insert, cited in Turtles All The Way Down.

49. Withdrawal of RotaShield, 1999: CDC and FDA records, cited in Turtles All The Way Down and Geehr, Unavoidably Unsafe, Chapter on the rotavirus vaccine.

50. GSK Rotarix FDA licensing document, cited in Turtles All The Way Down. The pivotal Rotarix trial is reported in Ruiz-Palacios G, Pérez-Schael I, Velázquez F, et al., Human Rotavirus Vaccine Study Group, “Safety and efficacy of an attenuated vaccine against severe rotavirus gastroenteritis,” New England Journal of Medicine 2006;354(1):11–22. The CDC’s Pink Book (14th edition) confirms the pivotal Phase III trials of both rotavirus vaccines enrolled “more than 60,000 infants each.”

51. Merck RotaTeq FDA licensing documents, cited in Turtles All The Way Down. The pivotal RotaTeq trial is reported in Vesikari T, Matson DO, Dennehy P, et al., “Safety and efficacy of a pentavalent human-bovine (WC3) reassortant rotavirus vaccine,” New England Journal of Medicine 2006;354(1):23–33. The composition of the RotaTeq pivotal trial control substance was deleted from the FDA licensing document and marked as a trade secret; a separate RotaTeq trial used the vaccine-sans-antigen compound used in the Rotarix trial.

52. Adverse event data for the Rotarix and RotaTeq control arms, as summarized in Turtles All The Way Down and drawn from the Ruiz-Palacios 2006 and Vesikari 2006 publications (see notes 50 and 51) and the respective FDA licensing documents. The specific counts (approximately one in thirty severe medical events, 16 intussusceptions, and 43 deaths in the Rotarix control arm; one in forty serious adverse events, 15 intussusceptions, and 20 deaths in the RotaTeq control arm) are drawn from Turtles’ reading of those primary sources and would benefit from direct verification against the published trial reports.

53. Rotarix package insert, intussusception language, cited in Turtles All The Way Down.

54. World Medical Association, Declaration of Helsinki, current revision. The specific provisions are discussed in Turtles All The Way Down in the context of the rotavirus trials. The authoritative text of the Declaration should be checked for the exact wording cited.

55. See notes 50, 51, 52.

56. Nuremberg Code, Article 4, as cited in Turtles All The Way Down. The authoritative text of the Code should be checked for the exact wording cited.

57. See notes 1 and 2.

58. See note 12.