The commercially available Pfizer-BioNTech mRNA COVID-19 vaccine is safe and was tested on more than 250 people, contrary to claim by statistician Christine Cotton

A 2020 clinical trial on more than 43,000 participants showed that the Pfizer-BioNTech mRNA COVID-19 vaccine had a 95% efficacy at preventing symptomatic forms of the disease^[1]. The positive result led to the vaccine authorization both in the U.S. and the EU in December 2020.

However, French statistician Christine Cotton, who shared disinformation on COVID-19 vaccines in the past, used that clinical trial’s protocol as a basis to call the amount of safety data that justified vaccine authorization into question. Cotton, who has more than 59,000 followers on Twitter, claimed that the population received a Pfizer vaccine that “was not the one tested into the clinical trial” and that we had “no results, or maybe up to 250 participants” for the vaccine that was actually distributed to the public.

Her claim was also shared by outlets like The Florida Standard, which relayed it in an article carrying the headline “Pfizer’s COVID Shot Was Only Tested on 250 People”.

However, if we take a closer look at the clinical trial protocol and documents from the European Medicines Agency (EMA), it becomes clear that this claim doesn’t accurately represent how the vaccine’s quality and safety were assessed.

In her video, Cotton alleged that Pfizer used two different manufacturing processes to produce its vaccine. The first, called Process 1, was used to produce the vaccine intended for the clinical trial that yielded a 95% efficacy against COVID-19. Process 2, on the other hand, produced the vaccine that was distributed to the population.  Cotton used statements on page 131 and 182 of the clinical trial protocol to claim that Pfizer only selected 250 participants to test the Process 2 product, implying that the vaccine which reached the general population wasn’t adequately tested. 

Thus we can see that Cotton’s reasoning rests on three points: first, that there are two different manufacturing processes. Second, that Process 2 was tested on 250 persons only. Third, that Processes 1 and 2 yield different products. But as we will explain below, Cotton’s claim ignores important information showing that the vaccines generated from both processes were ultimately of comparable quality.

Pfizer-BioNTech updated the manufacturing process to face increasing demand

The Committee for Medicinal Products for Human Use (CHMP) of the EMA conducted an initial assessment of the Pfizer-BioNTech vaccine candidate, before its authorization. In this assessment, the EMA provided many details on the manufacturing and on the quality assessment of the vaccine.

The EMA assessment report indeed indicates the use of two manufacturing processes. Process 1 was used for vaccine batches used in the clinical trials, and Process 2 was used for commercial batches (page 34 of the assessment report).

The large global demand for vaccines necessitated an scale-up of the manufacturing process, which motivated the switch from Process 1 to Process 2 (page 69 of the assessment report). However, both processes ultimately aimed to produce the same mRNA carrying the genetic information to produce the spike protein. The differences between the two processes mostly concerned the way the mRNA contained in the vaccine was synthesized and purified (page 32).

 “Process 2” vaccines was tested on more than 250 people

Contrary to what Cotton claimed, the clinical trial protocol doesn’t say that the product manufactured using process 2 was tested on a total of 250 participants. It says, ‘each lot of “Process 2”-manufactured BNT162b2 will be administered to approximately 250 participants’ (page 174). “BNT162b2” is the product name of the Pfizer-BioNTech mRNA vaccine.

The protocol reiterated the same information in other sections, for example: ‘BNT162b2 generated using the manufacturing process supporting an increased supply (“Process 2”) will be administered to approximately 250 participants 16 to 55 years of age, per lot, in the study’ (page 182).

While the total number of lots is unspecified, Cotton’s claim is misleading as she misconstrued the number of participants per lot of vaccine with the total number of participants involved in the clinical trial.

Vaccines resulting from “Process 1” and “Process 2” have comparable quality and properties

As we explained in our introduction, in order for Cotton’s claim to be valid, it would require Process 1 and Process 2 to yield different products in terms of quality and function. If both processes yielded comparable products with similar physical, chemical and biological attributes, then the clinical results from one product can be extrapolated to the other. This consequently means that the different processes don’t actually imply a difference in product nature or quality.

Cotton claimed that Process 2 yielded a product of lower quality because “with process 2 they had a rate of mRNA integrity that was lower than with process 1”. She also implied that this would be detrimental to the population by immediately adding that “In real life, you got the process 2 jabs”.

However, the initial EMA assessment report of the vaccine already addressed that issue. It reported that while there was “a decrease in RNA integrity […] for the initial Process 2 batches compared to Process 1 batches”, Process 2 underwent later adjustments, giving “more consistent” and “reproducible” RNA integrity levels that were “more similar to levels achieved in Process 1 batches” (page 17).

The EMA did raise other concerns on how comparable Process 1 and 2 were (pages 18 and 19 of the report). This prompted the agency to ask for additional analysis to better characterize the quality and the physical and chemical attributes of the vaccine from both processes, in order to clearly establish that both were comparable.

Nevertheless, the EMA moved ahead with recommending conditional marketing authorization for the vaccine, as these aspects involved in the requested analysis didn’t alter the vaccine’s overall balance of risk and benefit (page 139).

The EMA grouped the required additional analysis under the name “specific obligation 1” (SO1 or SOB1). The agency then published an additional report in September 2022 where it announced that all the requirements from SO1 had been “fulfilled” (page 4).

The report explained: “Data on the quality SOBs have been submitted and assessed. The CHMP is of the view that all quality SOBs adopted with the initial marketing authorisation have been fulfilled” (page 13).

In summary, contrary to Cotton’s assertion that this is the “HUGEST scandal in the pharmaceutical industry history”, the data establishing that both processes generated products of similar composition and quality was available long before Cotton even made her claim. The EMA reports show that the work needed to demonstrate comparability was performed and that the vaccine from Process 2 met the necessary requirements. In short, Cotton’s claim, which carries the suggestion that the vaccine wasn’t adequately tested before reaching the public, is contradicted by the safety and comparability assessment of the EMA.

Conclusion

Available data contradicted the claim that the Pfizer-BioNTech COVID-19 vaccine that made it to the general population was inadequately tested. While it is accurate that the vaccine manufacturer used two different manufacturing processes, the process used for commercial batches wasn’t only tested on 250 people.

The EMA also closely monitored the vaccines resulting from Process 1 and Process 2 to ensure that both processes produced vaccines of comparable quality and biological activity. After an additional round of analysis, the EMA found both processes to generate comparable vaccines.

Finally, we now have a large amount of real-world data after two years of vaccinating the general population with Pfizer-BioNTech COVID-19 vaccines produced using Process 2. These data show that the Pfizer-BioNTech vaccine is safe and effective, rendering the claim moot^[2-4].

UPDATE (1 August 2023):

Following the publication of our review, Cotton reached out to us and tweeted to dispute the conclusions of this review. In her reply, she affirmed that the number of participants who tested the “Process 2” vaccine was actually 252—not 250. However, she didn’t provide the data and methodology leading to that estimate. Her claim thus remains unsupported. Furthermore, this doesn’t change the fact that her tweet from 9 July 2023 provided an incomplete account of the Pfizer-BioNTech clinical trial protocol, since she omitted important details as explained in our eleventh and twelfth paragraphs. We attempted to respond to Cotton via email, but the email bounced.

REFERENCES

• 1 – Polack et al. (2020) Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine. The New England Journal of Medicine.
• 2 – Liu et al. (2021) Effectiveness and safety of SARS-CoV-2 vaccine in real-world studies: a systematic review and meta-analysis. Infectious Diseases of Poverty.
• 3 – Barda et al. (2021) Safety of the BNT162b2 mRNA Covid-19 Vaccine in a Nationwide Setting. The New England Journal of Medicine.
• 4 – Shasha et al. (2022) Real-world safety data for the Pfizer BNT162b2 SARS-CoV-2 vaccine: historical cohort study. Clinical Microbiology and Infection.