A COVID-19 vaccine: In conversation with Adrian Hill

02 June 2020

This was recorded on 20 May 2020

About Adrian Hill

Professor Adrian Hill is Professor of Human Genetics at the University of Oxford, Consultant Physician and Fellow of Magdalen College, Oxford. His own vaccine research programme has developed one of the most promising potential vaccines for malaria, which is currently being trialled on infants in sub-Saharan Africa.

The vaccinologist is also the director of the Jenner Institute, which focuses on designing and developing vaccines for infectious diseases prevalent in developing countries, such as HIV/AIDS, malaria and tuberculosis.

Hill is currently working on the production of a vaccine to counter the Covid-19 pandemic. He discusses the complications in the development of a vaccine and where we are in process of potentially coming to a solution.

headshot of Adrian Hill

Here is the transcript of the podcast

Gerald Moser: Good morning, good afternoon and welcome to today's call. My name is Gerald Moser and I’m the Chief Market Strategist at Barclays Private Bank.I’m very pleased and honoured to be joined today on the call with Professor Adrian Hill, good morning Professor.

Adrian Hill: Good morning

Gerald Moser: So, Professor, you are Professor of Human Genetics at the University of Oxford, consultant physician and fellow of Magdalen College Oxford. But you are also the Director of the Jennifer Institute, which focuses on designing and developing vaccines for infectious diseases in developing countries such as HIV, malaria, tuberculosis, or Ebola.
Your institute is also currently working to produce a COVID-19 vaccine and it seems to be one of the most promising potential solutions to stop this pandemic. For this reason, it’s a great opportunity for us to speak with you this morning.

Finding a vaccine is, of course, something that everyone would like to see, but it is also a key topic for financial markets, we saw it for the last two days when some initial results of a test for another vaccine in the US looked promising and that led markets higher, but of course there’s always rumours and disappointment and yesterday it faded a bit.

Could you please tell us what are the different steps in the vaccine development process and also where you and your team are in that process currently?

Adrian Hill: Yes, and thank you. So very broadly vaccine development can be divided into pre-clinical and clinical phases. Pre-clinical for us started in the middle of January just a day after the sequence was disclosed from China on this new coronavirus that had never been seen before.

We had the advantage of having a vaccine technology, or a “vaccine platform” if you like that was designed to let us respond very quickly to a need to make a vaccine in months rather than years. 
And some of that is driven by our work in personalised cancer vaccines, but the same vaccine technology has been very useful in responding to COVID-19.

So the steps are to make the construct, we have a specialised facility to do that rapidly, to then test the new vaccine candidate in animal models usually starting in mice and then, ideally, moving on to non-human primates which are the best model for humans before you go into the clinic.

And then, of course, you have the larger investment in making the vaccine to a GMP [Good Manufacturing Practice] standard so that it can be brought into the clinic and that often produces delay because universities don’t generally have these facilities. We’re lucky to have one right outside our clinical centre. We stopped what we were doing there, started on COVID and we were able to manufacture that by the end of March.

We could have gone into the clinic right then, but we delayed a little bit to get safety data in animals and in non-human primates which came through by the mid-April and a few days later we did our first vaccination.

A little blurred

So that’s the pre-clinical phase. We’re now in the second part, which is clinical, where you go through, traditionally, phase 1, 2 and then 3, phase 3 being a large licensure trial often involving tens of thousands of subjects aiming for regulatory approval as soon as possible.

Well, in this remarkably rapid response, those phases have become a little bit blurred. We’re doing two trials, we’re calling them phase 1/2, where we’ve just finished vaccinating people and then we hope to start a phase 2/3 trial very soon.

In the first trial, we vaccinated more than a thousand individuals, half of them randomised to receive the COVID vaccine, half of them randomised to receive the relevant meningitis vaccine and we’re actually following those individuals for incidence of COVID infections so there’s an efficacy component even in that phase 1 trial. These are UK volunteers in Oxford and 4 other centres in the south of England, including some in London, and we are following those for cases of COVID as well as assessing vaccine safety.

We will then, based on the safety data in that phase 1 trial, move ahead to the phase 2/3 trail enrolling over 10,000 people at 18 centres Scotland, Wales and England trying to follow the pandemic to an extent because, although happily cases are reducing everywhere, we do need cases to test our vaccine. 
So we’re hoping we will have enough cases, 20 or 30 or so by maybe August / September time, faster if there’s a second wave, and get a result on whether this vaccine is actually working and how well it is working.

So it’s been a bit of a race, not against other vaccine developers but against time and the pandemic diminishing in incidence, but we are probably the first group globally to go into phase 3 and to hopefully get a result, one way or the other, on whether our particular vaccine is working.

Gerald Moser: Thank you. You mentioned that universities usually don’t have the facilities to do the big manufacturing process, why is Oxford University participating in the global efforts to find a cure for Covid-19?

Vaccine development is a very diverse subject

Adrian Hill: This goes back to 2000 when it became very clear that there were - and there still are - a lot of academics at the University interested in translating their discoveries in microbiology, cell biology and animal models into something that could be a prophylactic tool or a vaccine, or even a therapeutic vaccine. 

The problem with vaccine development is that it’s a very diverse academic subject: it goes all the way from structural biology where you want to know, to atomic resolution the shape of your molecule that you're vaccinating with, to the health economics and market economics of a vaccine product that might be deployed globally.

To have that range of expertise is difficult, certainly for a new biotech, but in a very large research university like Oxford we’re lucky to have people with quite a range of expertise. So what we did in 2005 was bring together those capabilities into a new institute called after Edward Jenner, the founder of vaccination in Britain back in 1796, and this institute has a range of capabilities and core facilities that we were lucky to get funded and to take a vaccine candidate all the way from antigen to licensure.

Different focus

Now that’s difficult. In general we go into phase 1 and phase 2 trials and then partner with a company which takes it on to registration and market supply and so on. But our focus is different with most companies, what we saw was that it’s difficult even for a big Pharma to put investment into very tough diseases like tuberculosis, malaria, HIV and sustain, literally that over decades, so that you come up with a really good product.

Now in universities, we’re spending funders money, not our investors' money and we have a longer time frame. So that’s where we started and built our facilities like the GMP manufacturing facility, a new clinical centre, and biosafety level 3 animal facilities. Then what changed in 2014 was that the Ebola outbreak in West Africa came along and suddenly there was a time-sensitive need to make a vaccine against a virus that people hadn’t really paid much attention too.

Thanks to bio-defence funding, there were actually several candidate Ebola vaccines that had been manufactured but they’d never been given to humans, so things were moving very slowly. We were lucky to be involved so Oxford tested several of those vaccines and were involved with several of the companies, GSK, Janssen and Merck and others, trying to license the vaccine in record time, and what we came to appreciate was that these are very difficult disease targets to support for a company.

Outbreak pathogens

Ebola might never come back again. Some of the other pathogens, like Nipah virus, certainly don’t have a business case – they are too sporadic - and that provided an opportunity for an academic group like ours to focus in on that area, plan to build stockpiles of vaccine that then would be available very quickly should a new outbreak occur, and we’ve now invested heavily in that area.

We have 7 or 8 such candidates for these “outbreak pathogens” now in the clinic or about to go into the clinic. For these, led by Sarah Gilbert and others, we used the vaccine technology that we’re now deploying in clinical trials for Covid-19 and learned to accelerate the whole process.

So, we are unusual as a University to be able to compete with the likes of Janssen and Moderna and others, but we do have some advantages in that we are nimble and can move quickly and have a lot of relevant facilities.


Gerald Moser: It’s a very interesting positioning for the University. If I can just go back for a second to the process you described earlier on. In the financial market, we always look at the risk that you face.

Now when you develop the vaccine you mentioned that you are just about to go into phase 2 and 3, what is the most critical stage and the riskiest where there’s the highest likelihood of failure in a vaccine process. Is it in the pre-clinical, is it in phase 1? Is it when you go into the very last test phase 3? Where do you have the most failure?

Adrian Hill: So that’s easy. Far and away the biggest risk in the whole span of vaccine development is going from animals into humans and what you always find is that the magnitude of the immune response that you get in your animal model falls when you go into humans, which is very irritating and makes it harder to predict what you’ll when you go into the clinic, after investing heavily in doing that.

So the key measure that you should take with any vaccine technology is what it will do in a clinical trial, even the first clinical trial will often tell you that. And it’s remarkable, there are vaccine technologies out there that give you a strong immune response with a single dose and there are others that give an immune response that appears to be so weak that they’re not actually disclosed, even those several vaccine doses have been used.

So I would strongly advise any investor to look very carefully at that jump from animal to human and ask do you get good immune responses in humans, no matter what efficacy results are achieved in animal models.

So, we can often predict from the first clinical trial whether something has a chance of working or should really be dropped. If the immune responses are very weak you're going nowhere, if the immune response is strong it does not guarantee success, although there are a lot of other things you should judge carefully: for example, whether it’s a conserved antigen. The spike protein of SARS CoV-2 causing Covid-19 certainly is, so that’s encouraging.

Easy Targets

By and large the outbreak pathogen vaccines that are being made have a better chance of success than the old complex targets like malaria and HIV because they haven’t been worked on, they might be easy and nobody has invested in them. Contrast that with the tens of billions that have gone into HIV vaccine development where after 30 years we still aren’t close to a vaccine.

So these might be easy targets, people just have to take them on and do them and be able to respond quickly with a vaccine technology that can go into humans in a couple of months and that’s the focus.

Gerald Moser: You’ve just mentioned in your answer immunity, I mean there’s a lot of talk and confusion at least for me when I read everything about the antibody test what does it mean, does that then mean that you are immune, there’s a lot of discussion on whether you can get it again, the Covid-19 once you’ve had it already and we even heard it from the Chief Medical Officer that without natural immunity for a prolonged period a vaccine is not really a solution.

Could you please shed a bit of light on that story about antibodies natural immunity being built and what it means for a vaccine?

Adrian Hill: Yes, well, that’s a great question and very topical as you say. Probably the background to remember is that there are actually two arms to the immune response, antibodies you’ve already mentioned, and that is how most vaccines work, but over the last 10 or 15 years there has been a lot of attention to the other arm of the immune system producing cellular immunity and that’s driven partly by having targets for cellular immunity like cancer.

Neutralising antibodies

And there vaccines are beginning to work and there you will definitely need a potent T-cell or cellular immune response to get cancer vaccines to work.So ideally everybody would agree you’d have a vaccine technology that gives you those good antibodies and good T-cell of cellular immunity.

Then within the antibody area, you would like to have antibodies that are able to not just find the pathogen and recognise them but mop them up and get rid of them, and stop them entering cells. And in the virology field, there’s a special name for those antibodies, neutralising antibodies, so a big focus in COVID is on whether people can get neutralising antibodies after vaccination.

Now the confusion I’d like to highlight is the assumption that if with a natural infection you get modest not very durable immunity you will find it difficult to make a vaccine that will give you long-term immunity. It sounds rational but it actually doesn’t follow, so just let me explain that.

With natural immunity, you get a huge variation between individuals in the magnitude in the immune response that’s generated by the infection and if you think about the range of clinical presentations from being totally well and asymptomatic which happens in a small percentage of people, to ending up in ITU on a ventilator with huge viral loads in the respiratory tract, your talking of tens, hundreds or thousand-fold differences in the amount of virus in those different people.

So it’s not surprising the level of immunity with a very mild infection is lower than in somebody who has spent ten days being ventilated. So you need to look carefully at what patients you are talking about when asking is there a durable immunity and not surprisingly the more virus you’ve had, the stronger the immune response you get.

Different technologies

But the key point is that just because this is a relatively non-immunogenic infection, and some people come away with very weak antibodies after recovery, doesn’t imply that a vaccine is going to find that hard to do. We’re using two completely different technologies with modern vaccines compared to the whole virus infection.

So there are vaccine technologies, for example in malaria, that will produce a hundred times better immune responses that you would get from a lifetime of exposure of to the parasite in West Africa. So that’s good news actually. It should not be impossible to produce a durable Covid-19 vaccine, I’m pretty sure it will be.

I’m actually pretty sure that several vaccines will work of the large number being rushed into the clinic, And then the key question of durability, I’m afraid, just takes time to work out because we’re going to have to follow individuals and see how long their immune responses last.

Some people are interested in setting up a so-called “challenge model” to try and find that out more quickly and see if a year after vaccination people are still protected, but that’s all for the future.

Safety and efficacy

Gerald Moser: It’s quite interesting and I want to go back to some of the responses you have just given. What you said also in the first question, you said that you delayed moving from phase 1 into phase 2 and 3 because you wanted to have good, safety data.

I mean from what I read, safety and efficiency are the two key main points for a vaccine. You said from your side you’re being very careful with the safety, of course that’s paramount for a vaccine, but then what about the risk like influenza which you say changes every year, which is why there’s always a new vaccine or a different type of vaccine [correct me if I’m wrong] for the flu when it comes every year.

What is the risk that Covid-19 mutates and that we need to adjust the vaccine or that the efficacy of the vaccine is a bit like the efficacy vaccine in around 50/60 persons?

Adrian Hill: Yes, so there’s been a lot of discussion on this. It’s important to remember that ‘flu is unusual, of the 28 vaccines that are licensed for human use, ‘flu is really the only one that we make as you said every year or at least adjust the strain of the vaccine to match the circulating strain in the northern hemisphere, or 6 months later adjust to what’s down in the southern hemisphere.

That is unusual because there are circulating strains of flu that are really quite divergent so much so that if you vaccinate with the wrong strain you get no protection against the other strain.

Now a huge amount of work has been done on SARS-CoV-2 genetics. We know that it is divergent, there are mutations, that’s not surprising as it’s an RNA sequence that will mutate. But the scale of divergence is totally different to what you see in influenza viruses. And if you think about it that’s not too surprising because this coronavirus has only been in humans diversifying for the last 4 or 5 months.

So there are accumulated differences but they are very minor and in our vaccine we use more than a thousand amino acids, so you might see one or two variations within that, but it’s minimal it’s not relevant to vaccine-induced protection this year or next year. In the long term, the virus might diversify a lot more to get to the level where ‘flu viruses are, or more famously the super-high level where HIV viruses are at, but we’re nowhere near that at the moment.

Also people have studied vaccinated animals, taken the antibodies that are induced by their vaccine and asked does it work on all strains of Covid-19 and the answer is yes. So I think that viral diversity is probably one of the least of our problems at the moment or in the next couple of years.

So that is a part of the answer to the question “will we need a vaccine every year for Covid-19, just like as we do to flu”, but there are two reasons to keep topping up our flu immunity every year, one is to get the strain right for what’s circulating that year and the other reason is that the flu vaccines don’t work terribly well.

They work pretty well for just 3-4 months, which is why it’s important to have them in say November, but 6 months later their immunity has dropped really quite a lot.

So is that going to be an issue with COVID, that we will have just short term protection? Well, that comes back to what we were talking about in the last question on the durability, and I’m not sure.

My guess is that we’ll get 2 or 3 years protection from a good COVID vaccine. I don’t think you’ll get lifelong immunity, I suspect that it might need to be topped up and you might even have a seasonal COVID vaccine every year, along with the flu vaccine just for practicality, but we do really need to see how durable the current batch of vaccines coming along are and how well they protect.

Gerald Moser: Yes and another key topic when you think about building enough immunity worldwide is the amount of vaccines that would be needed, we’re talking about 70% of the population it would be nearly 5 billion vaccine doses if there’s only a need for one dose.

How difficult in your experience would it be to manufacture at that scale and how long it would take to get to that level of vaccination if we found a vaccine?

Ambition is there

Adrian Hill: Again we’re working it out, as we’ve really done nothing like this in living memory in terms of new vaccine manufacture. What you’ve just said highlights that a single dose vaccine will be hugely beneficial compared to one that takes 2 or 3 doses, not just for cost but also for logistics and for manufacturing supply. We were faced with this very early on, some months ago. Luckily we have a very manufacturable vaccine technology and the manufacturing yields are very good.

So what we decided to do was take a distributed manufacturing model, talk to contract manufacturing organisations in the UK and continental Europe and those are good, but they’re not really large scale so we ended up talking to much larger manufactures in India and China who were extremely keen to take this on.

We haven’t yet scaled up to two thousand litres bioreactors or ten thousand and as you scale up productivity can fall, but I will say this. The ambition is absolutely there, we may well have the first vaccine to show an efficacy result in humans, and if that were in August or September they’d be huge pressure to manufacture globally at a huge scale and we are gearing up to be in a position to do that.

And importantly a lot of manufacturing has already started at risk because you really have to. You can’t make a manufacturing process work in a few weeks.

So that I think is a greater challenge than taking a vaccine into the clinic quickly and making it work, we’ve done rapid vaccine trails before, but the world has never produced billions of doses of vaccine in months or in less than a year before.

So you know when people ask us are we competing with the big pharmas of this world, maybe, but we’d be really really happy if at least three vaccines worked by quarter four of this year and were being manufactured. Because to make a billion doses is a big ask, to make 4 or 5 billion doses for even a very big company, well, it’s just never been done. So hopefully the work will be spread out across several successful vaccines.

Gerald Moser: Well Professor Hill, thank you very much for taking the time to speak with us this morning, it has been quite eye-opening in the sense of where we are, the challenges that you are still facing you and your team in the quest for a vaccine and what we should be expecting over the next few months and couple of years in terms of development on the vaccine side.

So thank you very much for your time. That concludes the call, thank you everyone for dialling in. I hope you found it as useful as I did myself in learning and maybe deciphering a bit in what we hear in the news, what you read in the press with one of the world-class experts in the area. So I wish you all a very good day and an end to the week.

Other Podcasts