Again, and again in conversations with friends and family or on social media, I see that the fact that COVID-19 vaccines were developed in less than a year is mentioned as something that is reason for concern. It is often stated that this was too fast, and therefore the vaccines were not properly tested for efficacy, safety, or "long-term safety".
It is also often states that the vaccines are therefore still experimental. Here are a few examples:
It is also often stated that this development process has to take at least 8-10 years for this to be done safely. But that is simply not true, and in this post I will give 9 reasons why that is the case.
Even though there are several sources out there that have tried to debunk this myths, so far many have them have missed certain points or were, at least in my opinion, incomplete. There are many, many different factors that made the ultra-fast development possible, and I would like to present an explanation that is as complete as possible here. You will ask, why am I qualified for that? Let me just say that I have a degree in molecular biology and biochemistry, one of my subjects of choice was immunology. I also have a PhD in cancer biology and have worked in clinical trials and drug development for almost 10 years. Every day I work together with specialists in regulatory affairs, patient safety, biostatistics, biomarkers, pre-clinical studies, toxicology and other scientists and medical doctors. Together we design, start-up, conduct, analyse and communicate clinical studies conducted that are actually with new investigational drugs in patients.
So here is my take on how such a fast development and approval was possible and why it does not mean that any shortcuts were taken as far as the safety of these vaccines is concerned. I have identified 9 reasons.
Vaccine Platform technologies
Prior research on SARS and MERS
Many attempts to develop a vaccine at the same time
Parallel development
Prioritisation
Testing in the middle of an ongoing wave of the pandemic
SARS-CoV-2 is not like HIV
Long term safety studies are not usually necessary prior to the approval of drugs or vaccines
These vaccines are not as revolutionary as many people think
1. Vaccine Platform technologies
For decades scientists had been warning that another pandemic may be coming and that it cannot be predicted what kind of disease it may be. After 2003 even politicians and science funding bodies knew that it may be Coronavirus related to SARS and the west African Ebola virus epidemic brought such a danger even more to the consciousness not just of scientists or the general public. This brought a lot of research funding to projects that developed vaccine platforms. Platforms are technologies through which not just a specific vaccine, but the whole development from manufacturing to animal testing and human studies can be extremely streamlined and optimized in advance. Then, when a pandemic arises, all you need is the genome sequence of one of the virus's genes, and from that you can start manufacturing and testing batches in animals or humans within days or a few weeks. Two such promising technologies were in development for more than a decade prior to the start of the COVID-19 pandemic.
One is the mRNA technology on which the Moderna and the Pfizer/BioNTech vaccines are based, the other is the adenoviral vector technology that is used by the AstraZeneca, J&J and the Russian Sputnik vaccines. I will not go into detail on how these technologies work exactly because that has been described elsewhere, but the important point is that these are plug&play technologies. You just insert the viral sequence, and you can start. Almost all other issues such optimisation of the manufacturing process, finding clinical grade starting material, getting the process approved by the FDA, EMA and other authorities or even the storage conditions have already been solved. What also has been solved already is everything regarding safety and efficacy that is not related to the specific genomic sequence inserted to the platforms.
All the issues regarding that are related to the constituents (ingredients) such immunogenicity, vaccine reactions, allergic reactions or other animal or human toxicity had already been tested years ago, long before the pandemic, because vaccines based on these platforms were already in development long prior to 2020, had already been approved and found safe enough for human testing In clinical studies. The first use of mRNA vaccines in humans was already published in 2009 and was optimized in the years since. Adenoviral vector technology was also in development long before the COVID-19 pandemic, with a vaccine preventing Ebola already being approved by the FDA in 2019. Using these platforms, BioNTech for example, the German biotech that developed the vaccine together with Pfizer, started development of the vaccine in mid-January 2020. They were able to dose the first 12 clinical study participants by April 23, 2020. This is absolutely astonishing and remarkable. According to BioNTech, they can produce a batch of any new experimental vaccine within 1 week. Another benefit of the platforms is that long before the pandemic started, questions regarding optimal choice of nanolipids or chemical modification for mRNA vaccines or the choice backbone virus strains for adenoviral vaccines had already been answered. The same is the case for questions on formulation, storage, stability or the ideal route of administration. Further, the optimal dosing range (whether you need 1 μg, 100 μg or 10 mg) had already been solved as well so that trials could focus on few doses that were likely to be successful. It is true that these technologies had not been in extensively used in commercially available vaccines before, it is wrong to say that they were completely new.
In normal development, when you do not use a vaccine platform, by the time the platform companies were already testing in animals and humans immunologically relevant doses, other approaches were still in the design and early manufacturing stages. They had to make sure the product was pure and clean enough, figuring out how to make large quantities in a consistent way and at high quality, without toxic contamination and that it is stable and can safely be injected into animals.
2. Prior research on SARS and MERS
While it is true that the COVID-19 pandemic caught most of the world by surprize, the same isn't true for scientist who had been working on the closely related Coronaviruses SARS and MERS. According to this article:
The SARS-CoV-2 genome share about 82% sequence identity with SARS-CoV and MERS-CoV and >90% sequence identity for essential enzymes and structural proteins. This high level of the sequence revealed a common pathogenesis mechanism, thus, therapeutic targeting. Structurally, SARS-CoV-2 contains four structural proteins, that include spike (S), envelope (E), membrane (M), and nucleocapsid (N) proteins. These proteins share high sequence similarity to the sequence of the corresponding protein of SARS-CoV, and MERS-CoV.
This means that as soon as scientists in china were able to sequence the genome of the coronavirus SARS-CoV-2 and shared the data internationally, the scientific community already knew quite a lot about the virus, even if the specific type was unknown until then. This means the scientific community knew immediately where in the genome to look for the structural and functional proteins that are found on the surface of the virus and that are responsible for helping the virus enter and infect its target cells. All this research had already been done after the discovery of SARS (SARS-CoV) in 2003 and MERS in 2012. I am not going to review the whole literature on the development of SARS and MERS vaccine candidates, but I will note that it had already been confirmed in studies before the advent of COVID-19 that the Spike protein, which is also present on the surface of those closely related viruses, is a good vaccine target and that vaccines targeting this protein can induce neutralising antibodies (neutralising means that not only does the antibody bind the virus, it also inactivates (neutralises) it by rendering it unable to infect cells.
All this prior research on the related viruses meant that instead of focussing on many different potential targets and therefore diluting the effort, the companies that had access to the most promising platforms were also able to focus all their effort on this one most promising target. For example, while BioNTech/Pfizer in the earliest stage went into clinical trials for 4 different vaccine candidates, all four were targeting the Spike protein.
3. Many attempts to develop a vaccine at the same time
One of the often cited arguments why we got the COVID-19 vaccines so quickly, is the idea that a lot of money was thrown at the problem. In my opinion however, this does not really explain much at all. We still have to look how the money was spent. Of course, each company developing a vaccine spent a lot of money. But that is not all. It also means that many different companies independently started developing vaccines immediately. normally if there a many companies already pursuing a target, this discourages other companies. This hesitation usually means that as humanity, we only bet on few vaccines to be in advanced development simultaneously.
But with the pandemic starting, the prize for the companies was just too large. Even if there were competitors who were already more advanced, others still kept investing without waiting for others to fail first. If this was a soccer game, we were playing for a team that wasn’t satisfied with going for the goal just a few times in a game. Instead, we shot many, many times which increased the likelihood of one of them going through.
4. Parallel development
This previous point, only explains part of the money aspect though. Another aspect is that each company also threw a lot of money at the development of their own vaccine. To truly understand this point though, we still have to look at how each company used all the money to actually speed up the development of their own candidate. One of the ways in which that was done, was to perform many of the steps simultaneously (in parallel) steps that are usually done one after the other. To see how this was possible, we have to understand what forces are under normal conditions driving companies to perform the steps sequentially. The main forces are 1. the risk to lose money and 2. the safety of the study subjects.
Drug development is costly. Successful drug candidates can consume hundreds of millions or billions of dollars before a company can sell a single dose of the drug. Further, for each drug that gets approved to be used in patients, there are many candidates that fail before they can even be tested for the first time in humans, for example because they don't work well enough in the lab or because they are too toxic when tested animals. And even of those that are actually tested in humans, most still fail because they are not tolerated well or because they do not work in the way that had been assumed. As a drug progresses from one stage to the next, the development gets more and more costly. Lab experiments are cheaper than clinical studies and first in human Phase 1 studies are cheaper than large randomized controlled trials.
In order for a drug company to make the decision to invest in a large controlled Phase 3 trial, they will usually perform a smaller Phase 2 study (often also controlled) to confirm if there is an effect (this is called Proof of concept). Before investing in the Phase 2, the company will first want to check that the drug is tolerable and elicits the desirable effects in very few patients, this is a Phase 1 study. Because investors are usually risk averse and don't want to throw too much many at a drug that could ultimately fail, they will only start the costly preparations for a large trial, such as finding trial sites that can recruit patients, or submitting study documents to regulatory authorities and ethics review bodies to get approval when the results of the previous stages have been fully collected, reviewed and analysed. This means that the different phases are not just done one after the other, but that there can even be long time gaps between one step and the next.
All this didn't apply for the COVID-19 vaccines. as companies started investing heavily in the later steps while early steps were still ongoing. Things that prepared much earlier than is usual include for example the design and planning of the large Phase 3 studies, planning Ethic committee submissions, drafting informed consent forms and hiring the staff or planning subcontracting much earlier than is normally the case.
Of course, there are still ethical restrictions that limit how fast you can proceed. Remember, you are essentially doing experiments on humans. Even with the patients' informed consent, it would still be unethical to inoculate 20000 subjects in the first week, without having at least some results from a smaller cohort available. It would be unethical to expose a large number of patients to an unproven and potentially toxic drug, before very frequent and severe adverse reactions have been ruled out. You also have to be able to support why you think that many of them are likely to benefit from your drug. The money used just means that as soon as the results from the previous step are in, you can start including a lot more patients immediately after.
5. Prioritisation
COVID-19 was from the beginning a global concern, I am not going to go into the details (we have all been living on the same planet, I hope) but COVID-19 has massively affected us. Solving the crisis has been an absolute priority for every government that wasn’t led by populist lunatics or incompetent leadership. As a result, companies working on the vaccines have put all their focus on their vaccine program and prioritisation affected all aspects of what happens in pharmaceutical research and development, pushing the development of the vaccines forward quickly at every step. Where necessary, staff was reassigned from other projects and lab space/equipment etc. that was previously used for other drugs, was used for the vaccine program instead. I have no insider information, but I can well imagine lab animals bred and prepared to be at exactly the right age to be injected for testing any other drug, being repurposed at the last minute for the vaccines. In the middle of a global pandemic that threatens to kill millions and disrupt the global economy, all other experiments could wait.
But the development of the COVID-19 vaccines wasn’t just a priority for the companies. Regulatory authorities, who usually have several weeks or even months time to review applications for new clinical trials were pushing all other trials aside to focus on the vaccines as soon as they were submitted. Instead of lengthy review processes, with comments and questions going back and forth in a formal way official I can imagine many issues being solved in telefone conference with all the experts at the company and the authority at the same virtual table. Normally companies have to apply months in advance for such advice meetings, giving all participants time to prepare, while still keeping up with all their other work. Regulators are normally strict with enforcement of these rules, not just to have enough time for review, but also not to give the impression of favouritism towards certain companies.
But when officials at the FDA and other regulatory authorities got to see applications for a vaccine trial or the approval of a finished vaccine product on their desk, all other meetings and requests could wait, even if you meetings had to be cancelled at the last minute or other timelines for other drugs had to be delayed. The faster review times do not mean that the review wasn‘t done properly. Authorities even established so called rolling review, that means that the authorities could start reviewing the data packages long before the trial had completely finished, and were able to start evaluating the data quality, methods, adherence to protocol, and the results such as age distribution, COVID-19 infections and the occurrence of adverse events regularly, instead of waiting for the study to be completed. But companies and authorities weren’t the only institution and people who prioritised the vaccine development. Institutional review boards or ethics committees and trial sites prioritised them too. The clinics who are tasked with finding the volunteers and who get informed consent, administer the vaccine and the follow up with the participants for adverse reactions or COVID-19 infections were also very keen to take part. This also meant that a lot of such clinics could be identified and added to the trial faster. Due to the pandemic, activities for many other trials were reduced but vaccine trials would still be done and all aspects of doing the trial were prioritised in favour of the vaccine trials. Entering data into the study database? Scheduling appointments to examine participants or take blood? Legal review of the contract between sites and companies? Everything has the potential to be done just a little bit faster than is normally the case if everyone prioritises and maybe even does some overtime on evenings and weekends.
And finally I want to mention the volunteers who participated. Often, recruitment of participants is a very slow part of performing trials. Patients who fit the inclusion criteria still have to agree to take part. Clinic staff have to take time from their busy schedule to inform participants, document their consent to participate and perform all other study related activities. But for the COVID-19 vaccines, trial sites were able to find willing participants with very little effort. And that is because the development of the vaccines was a priority for the whole society. Volunteers were hoping for the 50% chance they were not in the control group and that they may even benefit from the vaccines. Identifying and recruiting about 40000 participants, normally a huge task that on its own can normally take many months or even years, could be accomplished in few weeks.
6. Testing in the middle of an ongoing wave of the pandemic
It is easy to underestimate how important the aspect of statistics is here. In many types of clinical trials, what you are drying to show is that your intervention or drug reduces the likelihood of certain unwanted outcomes or events in the future. For example, while a vaccine may reduce the likelihood of catching and suffering from an infectious disease, a statin may reduce the likelihood of a heart attack and a cancer drug may reduce the chance of the tumour growing or coming back.
In order to analyse these trials statistically, you then compare the number of these bad outcomes observed in your study participants over time and compare that to your control group, often tested with a placebo, that you are observing at the same time. These trials are often blinded. Even the scientists at the pharmaceutical company don't know if more events occurred in one group or the other until the results are unblinded. But that creates a conundrum for the team who has to analyse the data. When should they unblind to take a look? After a fixed time? What if not enough people in the control group get infected because of measures such as masks, social distancing or lockdowns? How do you know, that any difference you see will already be large enough to have a statistically significant result? The trick is to suppose a particular effect size for your drug and then calculate the number of events you need in see in total, in both the placebo and the test group to get a significant result. Then, as soon as you get to that number of events, you do your analysis and take an unblinded look at the data.
Waiting for the events to occur, is one of the reasons why some trials take years and years to complete. Imagine you are testing a cholesterol lowering drug in patients who already had a heart attack, to see if your drug can lower the number of second heart attacks these patients get. Now, assume that your drug was to reduce the number of heart attacks by a quarter. If your patients initially had a risk of 4% for having a heart attack in the next 3 years, you would need to watch ~16000 patients, 8000 with placebo, 8000 treated with your drug for 3 years to get a statistically significant result. That is because even with these numbers, you would still only have a total of 560 events (320 vs 240)in the groups. In a cancer trial for example where you know that almost all patients will likely progress within 6 months, and you hope that your drug can change that by 30%, you may get the 60 events you need within those 6 months. A similar effect happened with the vaccines.
Now say that your vaccine should be at least 50% effective to get approved by a regulator. You actually only need a very small number of infections in your groups to get a statistically significant result. However, for a potentially dangerous infectious disease it would be highly unethical to go around and expose your trial participants to the disease on purpose (at least in most cases). Therefore you have to wait for people to get infected by chance. And remember, people are actually doing everything they can to avoid that, including wearing masks, social distancing, reduced mobility, meeting outdoors instead of indoors etc. Some of your participants may have been in countries with strict lockdowns.
This put the people planning the studies into some difficulties. How do you make sure you get enough events in a reasonable amount of time, if the weekly incidence is just a handful of cases per 100000 inhabitants? Their answer was to enrol a staggering ~40000 participants in total, and to do that in countries of having a chance of getting a strong outbreaks. At the end, both Pfizer and Moderna got their required number of infections in just a few months, exactly because of the ongoing global pandemic. With the highly efficacious vaccines, very convincing statistical significance was achieved with just around 100 cases, almost all of which occurred in the control groups.
7. SARS-CoV-2 is not like HIV in its biological characteristics
Another reason humanity was able to develop the vaccines quickly, has nothing to do with our technology or the science, but was entirely rooted in good luck. The virus has relatively simple lifecycle: first infection of the cell, then production of viral proteins and replication of the genome, followed by assembly of new virus particles and their release outside the infected cell. There is no insertion of the SARS-CoV-2 genome into our own DNA to hide from immune cells as some viruses do and there is also not the very high mutation rate, that generates so many new variants even within a patient that makes it hard to develop a vaccine, as is the case in HIV. Also, the virus doesn't infect and damage the cells of the immune system directly (as others do). Taken together, all these factors make the biology of SARS-CoV-2 much more similar to all the viruses that we can already vaccinate against (e.g. the flu), and much less like the viruses that we have not yet succeeded in creating vaccines against, like HIV.
8. Long term safety studies are not usually necessary prior to the approval of drugs or vaccines
This may sound surprising or even controversial at first but please keep reading. In order to really understand what this means, we have to first understand what we actually meant with long term safety. There are lot of misconceptions about safety and long term safety, but the definitions are important to understand this point. When we say a drug is safe that does not mean that it has absolutely no side effects. It doesn’t even necessarily mean there are no very severe side effects, or even death. We also do not mean that it doesn’t have side effects in most or all patients. Take for example chemotherapy. Many types of patients have very strong adverse reactions, from vomiting, nausea, anemia, damage to the liver or immunosuppression. All patients will experience at least one of these events, but mostly several in the hours, days and weeks following the treatment. Sometimes the side effects can even lead to the patient's death and others may even remain with a patient for years, long after the chemotherapy was stopped. An example for that is neuropathy. And yet, those drugs are being used for the treatment of many types and stages of cancer, despite the adverse events. And the reason for that is, that the alternative (not treating the cancer effectively) is considered to be worse. For a chance to beat the certain or very likely death from cancer, or even to live a few months longer, patients willingly accept these side effects. Despite all these events, these drugs are considered safe and effective for their approved uses. The same safety profile would not be acceptable for a drug intended to treat a much less serious disease, say a skin rash, but the same principle is still valid, and an occasional mild allergic reaction, or mild burning sensation may be acceptable. But even a rare side effect of moderate strength may be considered too much for a drug like that. Safety is therefore relative, and always to be considered together with the benefit of the drug.
But what is long term safety? Let me first explain what we do not mean. We do not mean a toxicity, that occurs within hours days or a few weeks of administration, but that is so rare that it cannot be discoverd even in a large trial, but that is only discovered once the drug is on the market and has been administered to a very large number of people.
According to one definition I found, ”A ‘long-term’ study is one that gathers data on research subjects for 5 years or more”. Importantly, these studies are almost never required prior to the approval of the drug to be used in patients and when they are when they are required, tit is because the subjects continue to be exposed to the drug a long time. That is because the drug or its effects have to continue to be present to do the slow cumulative damage that causes those long term adverse reactions.
But what does this mean for the long term effects of vaccines? The vaccines themselves, with that I mean the chemicals, proteins, RNA, lipids or adenoviral virions that are injected, are lost from the body within days or a few weeks and cannot continue to do more damage. Even when a vaccine is administered more than once, the administered total dose over the months or years is absolutely negligible compared the doses of drugs taken day after day for After the second, or third dose, each of which is usually tiny as compared to most other drugs, a vaccine may not be administered again for years and therefore cannot continue to do damage.
Update 7 Dec 2021: a guidance document by the FDA confirms that the regulator does not think that Long-Term Safety studies are necessary for RNA based adenoviral vector based therapies. The document covers gene therapies and other therapies applied through gene delivery systems, even if they do not permanently alter or edit the genome of the human cells.
The yellow No in the right column indicates that Long term safety studies are not necessarily required. This is further explained in Figure 1 of the same document.
As the RNA and adenoviral vectors do not alter the DNA of the cells permanently, the first question is answered no, the second is also a no because the RNA is injected into the body. The third one is answered in the pharmacokinetic studies performed in animals and is also a no. End of Update 7 Dec 2021.
The immune response however remains. The question arises whether it is possible for the immune response to remain, and continue to slowly do damage to the body, even if no damage is visible immediately. Admittedly, such an effect is theoretically possible under a philosophical examination. But it practice we can still dismiss such an effect as it has never been observed before for any vaccine, infectious disease or anything else that stimulates immune responses. Symptoms mediated by the immune response have so far always been noted within days or a few weeks of vaccination. And most importantly, if such effects were even remotely possible, they would also be caused by the disease. The immune system is able to recognise foreign bits and is able to mount very specific antibody and T cell responses within less than 2 weeks. Sometime after a couple of weeks, the strongest and most specific immune response is active, and after that slowly becomes less. If the immune response is for some reason toxic, for example because the immune system was tricked into attacking part of our own body, that is because of a similarity between the viral antigens and certain proteins expressed in our own cells. Because the body has mechanisms to reign in such reactions, these reactions start to wane after a few weeks and the damage stops. Most importantly, because these reactions occur because of a similarity between the vaccine and own body, if the vaccine contains exactly the same antigen as the virus, then the virus will also have the tendency to cause the same autoimmune diseases. Some people will ask, but what about the adverse events discovered after the vaccines started to be widely used, such as the thrombosis coupled with thrombocytopenia for the Astra-Zeneca vaccine or the myocarditis for the mRNA vaccines, don’t they prove that the vaccines were approved too fast? Again the answer is no. These events are not long term safety events that arise following years of exposure, but short term events, occurring about 11-12 days after the administration, but in just a tiny minority of patients such as 1 in 100000 or less. Extremely rare events like these are never discovered in clinical trials prior to the approval of any drug, but always identified in post marketing surveillance, as happened with these vaccines within a relatively short time of just weeks. If we required drug manufacturers to identify such events for all vaccines prior to approval, the trials would require millions of subjects treated with vaccine or placebo and would make development of any vaccine so expensive that it would be impossible in practice. Even without the identification of such events in advance of approval, regulatory agencies grant approval based on the overall risk and benefit profile, knowing that even if such very rare serious events are discovered, the benefit would still outweigh the risks.
9. These new vaccines are not as revolutionary as many people think.
A lot of hype was made about the fact that mRNA vaccines or viral vector vaccines were relatively new technologies and had never, or only once been used in an approved vaccine before. Scientists, journalists and commentators explained that these vaccines are different, because instead of injecting a viral constituent directly, we were instead delivering the viral genome as RNA or DNA into the human cells, which then used this as a template to build the viral proteins directly.
Especially because of scepticism against genetically modified organisms or historical failures of gene therapies, comparisons were between the vaccines and genetic modification or vaccine therapy. Even of the critics who were not unnecessarily concerned that the vaccines could alter their own genes (a completely unfounded fear), some were still worried that this new technology constitutes a completely new principle for vaccines that warrants increased vigilance regarding the safety profile. But how new is the principle really? Adenoviruses, as used in the vector based vaccines cannot integrate their DNA into the human genome and are very common human viruses that are responsible for significant proportion of common cold occurring endemically throughout the year. They also insert their genome into our cells temporarily and use that to create new viral proteins. This happens all the time, without this principle constituting any danger. mRNA similarly is nothing new to our cells. mRNA exists naturally in every cell of our body. While the mRNA in the vaccines has been artificially modified, the modified nucleosides (the parts that make up RNA) also occur naturally in our body. Further, the lipid nanoparticles that are used to package it in the vaccine are also not a new principle. Lipid nanoparticles have been used before to deliver drugs into cells of the body or even in cosmetics.
Finally, even the supposedly new principle of viral genes being delivered into human cells for the purpose of vaccination has been used many times before. Where all the journalists and scientist describing the mRNA and viral vector vaccines then wrong? Yes, and no. While the technologies are new, the principle is at play when live vaccines are used. Live attenuated vaccines are actually complete viral particles that are fully able to enter human cells, release their genes, as a normal virus would do, and even use those genes to create all the viral proteins to make new viral particles. Even the first vaccine ever developed, the smallpox vaccine discovered by Edward Jenner in the late 18th century belonged in this category. Other examples include the oral polio vaccine, that was used to vaccinate many billions of people worldwide, and without them, the eradication of polio would not have been possible in the way it was done.
As the new vaccines do exactly the same thing as other vaccines have been doing for centuries and still are (just replacing the attenuated viruses proteins with other virus proteins or nanolipids), there was no reason for the approval of these vaccines to be delayed, just because they used the new platform technologies.
I hope I was able to explain and list all the reasons why these vaccines could be developed and approved safely in such a short time. If you have any comments or questions, if you think something is unclear or if I missed another reason, please leave a comment below.