By Rob Verkerk PhD, founder, executive & scientific director, ANH-Intl
It won’t have escaped you that the airwaves are currently booming with news from an interim report of the Phase 3 trial on the Pfizer-BioNTech mRNA-based vaccine, snappily named BNT162b2. It’s one of 48 vaccines under clinical evaluation that aims to protect against Covid-19. The primary source of the news, headlined by a “90% effective” claim, isn’t a peer reviewed journal article. Nor is it the World Health Organization (WHO). Rather, it’s a media release issued on Monday by Pfizer, the commercial partner linked to the young German biotech company BioNTech, that developed the vaccine.
Pfizer was conspicuous, given its position as one of the largest drug companies in the world, in excluding itself from the US government Operation Warp Speed. The downside for Pfizer was that it didn’t benefit from the US government (= taxpayer) funding support that the likes of Moderna, Johnson & Johnson and Astra-Zeneca have been privy to. But don’t feel too bad, BioNTech received funding from the German government. The plus side for Pfizer was that it didn’t need to be dictated to by others, and it didn’t need to data share or have its data analysed by a shared, Operation Warp Speed data monitoring committee. Remember this as you read on.
So, what do the headlines really mean? Here’s the first problem. We only have a press release to go on. At ANH, we’re always keen to get to primary data sources so we had to look further. The press release refers to an interim report on the Phase 3 trial by an “external, independent Data Monitoring Committee (DMC)”. An article by the Kaiser Family Foundation suggests that Pfizer’s DMC is anything but independent. The anonymity of its 5 members also makes it anything but transparent.
We weren’t going to give up looking for the interim report given widespread calls, such as that in a September editorial in one of the world’s most influential journals, Nature, which argued “COVID vaccine confidence requires radical transparency”.
We’d seen the protocol, but if better-than-expected interim primary outcomes were going to be cited, we had to assume Pfizer or BioNTech would make the data behind the results available. The first place we went hunting for the interim report was the WHO “Draft landscape of COVID-19 candidate vaccines” page, that incidentally "disclaims any and all liability or responsibility whatsoever for any death, disability, injury, suffering, loss, damage or other prejudice of any kind that may arise from or in connection with the procurement, distribution or use of any product included in any of these landscape documents".
Alas, while the WHO page was updated yesterday, the report isn’t there. What you will find there is the now outdated trial design for the vaccine as a link to the NIH ClinicalTrials.gov portal. But no interim study results. A Google search using the terms ‘“Data Monitoring Committee” BNT162b2 BioNTech Pfizer’ and similar didn’t make it magically appear either, just lots of references to the “90% effectiveness” claim across a plethora of media channels. Suffice to say, I think we can confirm that there are no supporting data for the claim.
What we’ve been told
Back to the press release. The key pieces of information that fall out of it are as follows:
The claim in its full glory: “more than 90% effective in preventing Covid-19 in participants without evidence of prior SARS-CoV-2.”
A link to an updated study design (protocol) which presumably replaces the one listed on ClinicalTrials.gov.
43,538 participants enrolled to-date (89% of which have received the second dose)
42% of participants have “ethnically diverse backgrounds” (which begs the question what ethnicity is the non-diverse 58%; surely not Caucasian? But probably)
94 people out of the 43,538 participants (i.e. just 0.2%) have contracted Covid-19 so far, these being split between the vaccinated and placebo groups, the split not being reported
“No serious safety concerns have been observed” – which does not mean moderate or severe adverse reactions have not been observed. Two months of safety data will be available at the time Pfizer applies for Emergency Use Authorization with the US Food and Drug Administration (FDA)
The clearest statements of all appear in the disclosure notice and commentary on forward-looking statements at the end of the release that appear to be directed at the stockmarket. They effectively abrogate Pfizer and BioNTech of any responsibility for performance of the final product, especially in relation to effectiveness, safety or supply. The lawyers are clearly being attentive to their clients given the world’s eyes are on them.
What can we make of this?
In terms of the key statement on effectiveness, let’s say that just 10% of the infected group of 94 were infected. If you round that up to 10 people, a 90% claim that would make it 9 out of 10 were able to prevent infection post-vaccination. Not a lot of people potentially – which is why Pfizer and BioNTech should have been clear about numbers because 60 is a lot different to 10. But, remember that the press release states “over 90%”? Dialling in the inner Sherlock Holmes, just 11 of the infected group also being vaccinated, with 10 of these exhibiting immunogenicity as per predetermined trial endpoints would do it (9 divided by 10 gives 90.9%).
But for this interim endpoint the protocol lists primary and secondary endpoints for effectiveness – principally the prevention of Covid-19 symptoms in those showing evidence of infection (by nucleic acid amplification tests [NAAT]), as well as secondary ones – the development of various serological (antibody) results such as neutralizing antibody titers, S1-binding IgG and/or RBD-binding IgG levels, N-binding antibody. But which ones and by how much? And for how long (too soon for that of course)?
What else would we need to know to understand better the BioNTech/Pfizer interim result? Key information that hasn’t yet been put into the public domain and could have accompanied the interim release includes:
The number of people in the vaccinated or placebo groups who were exposed to SARS-CoV-2
The demography (age, gender, ethnicity, etc) of those who were infected and how many of these represent the most vulnerable groups i.e. the elderly or those with comorbidities
On what particular outcome parameters was the effectiveness determined? Was it, for example, based on lack of Covid-19 disease symptoms combined with elevated antibody responses, and if so which ones?
How serious was the manifestation of Covid-19 disease in the equivalent vaccinated and placebo groups (i.e. similar ages, gender, ethnicity and underlying disease pattern)?
What is the composition of the placebo that is being delivered to 50% of the 44,000-strong study population? Does it include the lipid nanoparticle minus the mRNA that encodes for the full-length spike protein of SARS-CoV-2?
What was the response of the cell-mediated (T cell) side of the adaptive immune system (which are not included in the endpoints according to the trial protocol)? The October publication of the Phase 1 results in the New England Journal of Medicine indicated it was the sister vaccine, BNT162b1, that produced a strong T-cell response but this was dropped as the vaccine induced some serious adverse reactions
What was the nature, severity and extent of adverse events to BNT162b2 reported until now in the Phase 3 trial?
How long will immunogenicity against SARS-CoV-2 persist in different people?
There are also many unknowns that may well remain unknowns. Top of my list would be these two:
The possibility that some of the NAAT-confirmed cases involve infection with other human coronaviruses and it is these non-SARS-CoV-2 viruses that have triggered the measured immunogenicity and the NAAT results are either false positives or the result of SARS-CoV-2 viral fragments
The presence of non-replicable viral fragments of (‘dead’) SARS-CoV2 have triggered immunogenic reactions so infection could not anyway have occurred
"The lack of data is very concerning....All we have right now is a headline by Pfizer" Dr Peter Doshi, associate editor, BMJ
What shall we make of it?
I’ve taken you through my thoughts triggered by the press release. You can see it’s led to a number of dead ends and it’s raised more questions than it’s answered. But let me distil my views down to the following:
In my view, it’s misleading, far too premature and disingenuous for Pfizer to be telling the world that the trial has demonstrated 90% effectiveness so far because most people will assume that that means, regardless of age, health status, ethnicity or underlying conditions, if they get vaccinated with an emergency approved vaccine that has undergone just 2 months of safety evaluation, they will have only a 10% chance of getting seriously ill if they become infected - and it will be safe. There are insufficient data to support either outcome.
It's difficult to judge the effectiveness claim against what your chances might be if you remain unvaccinated. But if you're healthy and under 75-years-old, the chances of serious disease are very low, probably much less than 10% following infection. But accurate estimates still cannot be made because we continue to be blind to the number of people who have been infected, and therefore, also to the true rate of serious disease and mortality among those infected. What we are also aware of are big variations in estimates of the number who are likely asymptomatic (infected but without symptoms).
Nothing can be said about the risk of harms from vaccination with two doses of BNT162b2 such as triggering autoimmune conditions as the trials need to run their course and most of these kinds of problems are generally not picked up until years after the product is first marketed. And that’s assuming a normal 6-year development program. Two months of post-vaccination adverse event reporting just doesn't cut it if you want a proper handle on safety.
If this vaccine fails to be effective a few months after its second doses has been administered, is there going to be a justification made for its mass roll-out, given the huge economic cost to society, the risk of harms, and the fact that healthy people seem to tolerate SARS-CoV-2 more than adequately? Remember, it was the alternate BioNTech vaccine, BNT162b1, that was found to enhance the T-cell response more, but had an unacceptable safety profile so was dropped
Disturbingly, this announcement, and the lack of data associated with the press release, demonstrates that Pfizer and BioNTech have an incapacity for real and meaningful transparency.
This leads to one central question: Can the public truly afford to trust vaccine companies who deliberately withhold information and data and have preyed on the public’s desperation to escape lockdowns, while, at the same time, reaping the rewards from the stock market that has responded to a premature and unsupported announcement?
Call me a conspiracy theorist, if you like, for asking this question.