Trials Unlikely to Show if COVID-19 Vaccine Prevents Severe Cases
Trials Unlikely to Show if COVID-19 Vaccine Prevents Severe Cases

Trials Unlikely to Show if COVID-19 Vaccine Prevents Severe Cases

A vaccine expert explains that such answers likely won’t come until a vaccine has already been approved and administered to millions of people.

Max Kozlov
Max Kozlov
Oct 30, 2020


Nearly 50 vaccines to protect against COVID-19 are speeding through clinical trials at an unprecedented pace. These studies are designed to test the vaccines’ safety and efficacy, but a review published in The Lancet earlier this week (October 27) outlines the challenges in determining whether a vaccine candidate really is efficacious. 

The authors of the review note that there are many definitions of efficacious—reducing the likelihood of developing severe symptoms is one, for instance, reducing the number of deaths is another. Presently, the ongoing clinical trials are mainly designed to determine if recipients have a reduced risk of a coronavirus infection, but knowing whether a vaccine can prevent people from developing severe COVID-19 or dying is a long way off. Vaccine developers have not produced evidence of such a benefit yet, rather, a number of them have reported evidence of an immune response in recipients, data that do not necessarily mean the vaccine will prevent infection, the authors write. 

The Scientist spoke with University of Oxford infectious disease specialist and coauthor Susanne Hodgson, who’s working on the UK clinical trials of an AstraZeneca COVID-19 vaccine, about the timeline for generating evidence of vaccines’ effect on morbidity and mortality rates.

The Scientist: All eyes are currently on the vaccines undergoing clinical trials. What kind of evidence does a clinical trial show and not show?

Susanne Hodgson: COVID vaccines could work in a number of ways, for example, it could prevent you getting infected in the first place, it could reduce the likelihood of you developing severe disease, reduce the likelihood of you transmitting infection to other people, or do a combination of all three of those. And the hope is in Phase 3 clinical trials that we would start to get some data to tell us if the candidate COVID vaccine is actually able to do any of those three.

At the moment, unfortunately, because transmission rates are quite high, vaccine developers seem to be quite confident that we should get a measure of whether a COVID vaccine can prevent infection. However, because death and severe disease from COVID are a relatively rare consequence of infection, it’s unlikely that our clinical trials that are ongoing will be statistically powered to provide data to answer, ‘Do these vaccines prevent severe disease?’ 

Susanne Hodgson, University of Oxford

TS: Why is that an important distinction? 

SH: One of the reasons we wrote the article is because people often ask me, ‘Will this vaccine work?’ That seems such a simple question, and it’s so complex. There are so many caveats to answer: What do we mean when we say a vaccine ‘works?’ Does it mean it prevents diseases and infection, or does it mean to prevent transmission? And I think the most important vaccine that we need is one that prevents severe disease and death because that is what causes the most burden in our hospitals and of course loss of life. That’s what we’re all aiming for. 

That’s not to say that [the clinical trials] are not going to give us very important information—the safety data and an indication about whether they prevent infection, which of course are very important. But what’s most likely is that a vaccine will be deployed, given to a very large population on the basis of the clinical trial data, and only then, when a large number of people have had the vaccine, will we actually know if the vaccine prevents severe disease and or death. 

TS: Will these vaccines need to be monitored after approval, and why?

SH: There were already very clear protocols in place and infrastructure to monitor vaccines which are deployed and licensed in population. A COVID vaccine would be no different to existing vaccines and in terms of its monitoring post licensure. There would be ongoing, active monitoring of safety and side effects in COVID vaccines, so that if there were relatively rare side effects seen with the vaccine that only happened in one in many thousands of people we would still be able to get that information. 

See “COVID-19 Vaccine Researchers Mindful of Immune Enhancement

Post-licensure authorities will be very keen to look for any evidence of vaccine-associated enhancement of COVID. This is individuals who’ve received a vaccine and then develop an immune response against COVID, who, when they then encounter COVID in the community, actually have a worse disease because of the vaccine-induced immune response. The reason this is a concern is some animal studies of vaccines that have been developed for other coronaviruses such as MERS and SARS have demonstrated this phenomenon. Fortunately, in the animal studies of COVID vaccines to date, this phenomenon hasn’t been seen, and there’s no suggestion yet to date that I’m aware of that this has been seen in humans in relation to COVID vaccines. But of course, that’s a reason why there will be ongoing intense monitoring of safety post-deployment of a COVID vaccine.

TS: What makes it so challenging to determine if a vaccine is efficacious?

SH: First, when you just decide whether a vaccine works enough, you need to decide what the definition of a vaccine working is. As I mentioned before, a vaccine could reduce infection rates, transmission rates, or disease, and so for each of those outcomes you need to be very careful and understand how you will measure those outcomes. A lot of those outcomes rely on laboratory assays, for example, the real-time qPCR assay, which tests for the presence of antigen and confirms COVID testing. As I’m sure a lot of people are familiar, this test is imperfect. . . . Other people have suggested perhaps using antibody levels as a way of telling if somebody has become infected or not, but as we’ve seen in a lot of the papers that are coming out, not everybody develops antibody levels, and if you do they don’t stay elevated for a long time. And, of course, the many vaccines are aiming to induce antibodies, so you need to be able to distinguish vaccine-induced antibodies from those which are induced by infections specifically. 

Clinical trials that are assessing efficacy take volunteers, typically will give half a COVID vaccine and half a placebo vaccine or a comparator, vaccinate them all, and then follow them up in the community, and you’re reliant on individuals in the trial meeting COVID in the community, becoming infected, and then you compare the infection rates in your test vaccine group compared to your placebo group. If your vaccine is working, you would hope that statistically more people get COVID in your comparator group compared to your vaccine group. Of course, that relies on ongoing transmission in the community, which is at odds with all the public health interventions that we’re trying to adhere, so paradoxically the better we are at controlling COVID transmission in the community, the longer it takes for us to get an efficacy read-out in terms of infection from these clinical trials. 

TS: Given your background in vaccines, what gives you hope about the vaccine development process, looking forward to next year?

SH: It’s the speed at which people are moving. People [are] also publishing extremely quickly their data and sharing that, which is really encouraging. I think all ethical bodies, regulatory bodies, and, internationally, people are very supportive of the vaccine development program for COVID. 

Having worked on malaria vaccines for more than a decade now, I am optimistic with regards for a COVID vaccine because of all the things I’ve just mentioned. But I’m also a bit sanguine because we’ve seen so many malaria candidates that look promising in pre-clinical studies and in Phase 1 studies—you generate immune responses—but when it actually comes to the challenge study, which we use in malaria studies, it’s disappointing and you have to go right back again. I’m hopeful and of course the large number of candidates you have the more likely you are to get some that will work, but I think we need to just be realistic that this isn’t necessarily a matter of just, “One more month, one more month,” that seems to be presented in the media.

I just hope that the public should have faith in the vaccine development community—they’re a formidable professional group who have very stringent regulations that they’re adhering to and they’re prioritizing safety. This is not about headlines and just getting a vaccine to meet a headline or a political deadline. This is an international group working very hard and doing what they’ve been doing for many years, for many other pathogens, to get us a safe, efficacious vaccine as soon as we can. That can’t be rushed, unfortunately—people are moving as quick as they can. This is unprecedented pace of development, but it will only be ready when it’s safe and it actually works.

S.H. Hodgson et al., “What defines an efficacious COVID-19 vaccine? A review of the challenges assessing the clinical efficacy of vaccines against SARS-CoV-2,” The Lancet, doi:10.1016/S1473-3099(20)30773-8, 2020.

Editor’s note: The interview was edited for brevity.