Professor Sarah Gilbert on the science behind one of the world’s most advanced COVID-19 vaccine trials, and what still needs to be done to prove it is safe and effective
As professor of vaccinology at the Jenner Institute and Oxford’s Nuffield Department of Clinical Medicine, Sarah Gilbert has been making and testing vaccines for over twenty years, initially focused on immunising against malaria and influenza, before working on emerging pathogens. She is now leading the project to develop and test a vaccine for COVID-19 through the Oxford Vaccine Centre.
Her team was one of the first to report preliminary findings from human trials of a potential COVID-19 vaccine, generating enormous amounts of publicity and interest from around the world. Initial results indicated no early safety concerns and found the vaccine induced a strong immune response, in both T-cells and antibodies, in participants.
The vaccine, known as ChAdOx1 nCoV-19, is based on the established adenovirus vaccine vector ChAdOx1, which has been engineered to express the characteristic SARS-CoV-2 spike protein when it infects cells. Gilbert and her colleagues have been besieged by those wanting to know when the vaccine will be ready, but more expansive longer-term studies are required to understand if the vaccine actually prevents COVID-19 infections and is safe to use in the general population.
There are now many different types of vaccine and hundreds of different COVID-19 vaccine projects underway around the world. Why did you choose this particular approach, in terms of this vector and the DNA sequence for the spike protein?
I have worked in vaccine development for more than 20 years, and spent a long time making vaccines against malaria. For malaria we wanted vaccines that would induce protective T-cell responses, not just antibody responses, and that’s when we started to use DNA and viral-vectored vaccines. You can use ‘replication competent’ viruses or ‘replication deficient’ viruses. Obviously the replication deficient ones are going to be much safer because even if somebody is completely immunocompromised and they can't make any immune response to control the vaccine virus, it doesn’t replicate.
We use an adenovirus which infects our cells, expresses the antigen added to it, then cannot replicate. They're safe but also very good at inducing an immune response to the gene that is added, because you're actually getting a viral infection. Even though it's very short lived, that infection means that the cells of the immune system recognise that they need to respond. They're looking for what's different, and what's new is the spike protein of the coronavirus, being produced in large quantities. As the adenovirus can't replicate, it's lost from the body pretty quickly, but you're left with that immune response and the memory of the immune response.
You mentioned the T-cell element of immunity. When most people think about gaining immunity they think of antibodies – can you tell me a little bit how about how T cells can remember a pathogen and help learn to fight it again in the future?
The function of antibodies is to bind to the viral pathogen and stop it from infecting cells in the body. If you've got a neutralising antibody response to the coronavirus in your respiratory tract, it can bind to the virus and stop it from getting inside cells. But if the pathogen does get inside your cells, the antibodies can't reach it anymore. Cytotoxic T-cells can identify cells that have been infected with the virus and destroy them, stopping the virus spreading that way too. Another sort of cell, the CD4 helper T-cell, works with the B-cells that produce the antibodies.
If you've been vaccinated or encountered the pathogen before then you get the cytotoxic T-cell expansion very rapidly, and the B-cells with the help of the CD4 helper T-cells will then expand and produce more antibodies, and again that happens very quickly.
There's been a lot of interest in the initial results from your phase I/II trials recently which showed that the vaccine generated a robust immune response. Those subjects will be followed for an entire year; what else needs to happen before the vaccine is approved or not approved for use?
That was the first study in humans, which is, as always, in healthy adults aged 18 to 55. We're reporting the reactogenicity of the vaccine, i.e. the adverse event profile, and the antibody and the T-cell responses. It's unusual to publish papers so early – as you say we would normally collect data for a year after vaccination but in these times, people need to know quickly, so we published with only a month follow-up.
Once the participants have been vaccinated, they continue doing what they normally do – if they’re exposed to the coronavirus they may become infected. If anybody who's in the trial has symptoms they contact us straight away, we arrange for testing and we record who has had a positive result for COVID-19. Eventually, when enough infections have accrued, the statisticians will unblind the study and see which group those people were in – the group with the COVID-19 vaccine or the meningitis vaccine [the control group] – and that will tell us the vaccine efficacy.
The timing of this is what everybody wants to know. It's impossible to say because to determine the vaccine efficacy we must get these cases. Obviously transmission in the UK is pretty low at the moment, although the past couple of days* signal that it's picking up a bit, so we'll see where we get to.
The next thing that happens is we need to expand the age range. In the next groups of people that were vaccinated they're between 55 and 69 years or 70 years and older. We know that vaccines don't usually work so well in older people, so that's why we're looking at that separately, having shown that it was safe in the younger people.
One thing to be done in the future that we haven't started on yet is testing the vaccine in children, because it may be necessary to vaccinate children in future to prevent transmission. We won't start that until we’ve got a lot of safety data from the adults. We will also need to have some other separate cohorts in what are called ‘special populations’, including pregnant women and HIV positive people.
The vaccine is needed around the world and it's important that we get the safety and immunogenicity and the efficacy data from other countries as well. So there are trials going on in Brazil and in South Africa at the moment too.
To get to this stage in under six months is an amazing achievement. In what ways have you managed to condense the time-frame for vaccine development and trial approval?
Part of it goes back to planning from years ago. We've been preparing for ‘disease X’, and I've been working on emerging pathogen vaccines for a while, including other coronaviruses like MERS. We had already thought through what we're going to have to do in terms of the antigen sequence, the manufacturing. Some things were already mapped out.
We've published the results of quite a number of clinical studies using ChAdOx1 vectored vaccines. So we know how to manufacture it, we know what the safety profile looks like, we know what the immune response looks like, we know if we put a new antigen into this vector what's going to happen with it. We are not starting from scratch – there's a whole load of safety data that underlies all of this.
What has really helped is the regulatory and ethical bodies that have to give approvals for the clinical trials. Normally it takes months – the regulatory dossier has to be absolutely complete before it gets reviewed. We worked with regulators to get them as much information as we could really early, accepting that it was missing key test results from the manufacturing. As soon as we got the last testing results from the manufacturing, there was a very small amount of extra information to review and they can do that very quickly.
With the ethical committee, again, normally that is quite a long process and there's only a certain number of meetings, perhaps one a month, and if they're full on that day then you have to wait until the next month. That changed in response to COVID-19, and we had ethical committee meetings by phone for the first time. Both the regulators and the ethical committee were really helpful in speeding up that process without missing out a single step. The whole process happened as before, but we broke it down into sections, reviewing as much in advance as possible.
The more you use a platform technology the easier it becomes. The first use of any new technology is clearly going to take a long time to review all the data, but repeated use of the same technology becomes quicker because everybody's happy with what happened previously. So, it speeds up and you're just reviewing the parts of it that are new.
Finally, for this vaccine, we were given special permission to publicise the trial and screen volunteers before we had all the approvals in place, something you are not normally able to do.
Personally, how have you found working amid so much interest and pressure to find a vaccine – do you feel the weight of the world on your shoulders? Are you working longer hours and feeling more stressed by it, or does this feel like a similar drug development process to ones you've worked on in the past?
It's all the same things that I've done in the past, it's just happening faster with a much larger team. I was working very, very long hours from February to April, but it’s better now the trials are well underway. As the project lead, I started things moving early on and got the funding, and now those who are running the trials are very busy.
There is not much point in thinking about the weight of the world. We do what we do, we know how to do this, we know it's worth doing. If this doesn't work, I don't think anything will work. Nobody can say that this vaccine is definitely going to be effective, but I think we've got at least as good a chance as anybody else, and better than some of them. The type of vaccine that we've made does everything that we want to see. Other people will be testing their vaccines, and many of those will work as well. And that's good because we're going to need more than one vaccine for 7 billion people.
If the vaccine is approved and mass-produced, how much do you worry about the effects of anti-vaccination messaging? Are you already preparing and strategising to try and counter that kind of movement and its messages?
That’s something that a part of the Oxford Vaccine Group has worked on for quite a while. They have a website called the Vaccine Knowledge Project with a lot of information for people about what vaccines are, how they work, what's in them, what they do. I was just talking today to the head of the manufacturing facility and she said she has had messages on Twitter saying ‘you don't know what's in these vaccines’, and people talking about microchips and other nonsense. She said, ‘I do know what's in it, because I made it’.
The completely mad and ‘out there’ ideas are the ones that we can't really do a lot about. If you’re worried about microchips and being tracked then you shouldn’t have a mobile phone or be online! But we do have a lot of information that people can look at, so they can understand what a vaccine is and why we need it and how it works, and the dangers of not being vaccinated.
We don't have a specific anti anti-vaxxer strategy for this vaccine. We understand that people are cautious about a new vaccine being introduced, and we are using our established channels to address the main concerns. We have a lot of information available, and a well thought out approach to responding, not to the extremists, but to those who are hesitant.
Have you or your colleagues ever felt like your security was threatened by the fact that you're making this vaccine that is at the centre of some quite extreme conspiracies?
No, there was the story about Russian hackers but that hasn't impacted me directly. Obviously we are careful about IT security all the time anyway, and we were all familiar with scams trying to get money. As for the rest of it, what can you say, people have been setting fire to 5G mobile phone masts, what is that all about? You can't really understand it or predict it and so we just get on with doing what we do.
Interview by Tom Ireland
*This interview was conducted at the end of July.