Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Q&A

Developing COVID-19 vaccines

Glenda Gray is CEO of the South African Medical Research Council and the chair of the research committee on COVID-19 in South Africa. Julie Gerberding is Chief Patient Officer at Merck, former director of the US Centers for Disease Control and Prevention, and an associate adjunct professor of medicine (infectious diseases) at the University of California, San Francisco. Yoshi Kawaoka researches influenza and vaccine development and is a professor in the University of Wisconsin–Madison School of Veterinary Medicine. Rino Rappuoli is Chief Scientist and Head of External Research and Development at GlaxoSmithKline Vaccines. Hanneke Schuitemaker is VP Global Head Viral Vaccines Discovery and Translational Medicine at Johnson&Johnson. Lynda Stuart leads the vaccination team at the Bill & Melinda Gates Foundation.

Do you think that it is best to pursue several approaches, or do you think that we should focus on the one that will prevail?

GG: We need to explore several approaches simultaneously and with urgency. This is to mitigate the risk of scientific failure of any one of the potential approaches, or if there are multiple candidates that show efficacy, this approach improves our ability to scale up the production of multiple vaccines for global distribution. The global manufacturing and distribution of billions of vaccines will require Herculean effort from many players, including increasing pharmaceutical capacity to deliver vaccines. The more vaccines that are found to be efficacious, the better it is for the world. This is a devastating epidemic with no short-term remission in sight.

Glenda GraySouth African Medical Research Council

JG: We are racing against the disease, and this crisis will require the global scientific community to examine multiple solutions in parallel. At Merck, we believe a range of medicines and vaccines will be needed to end the pandemic, and we will continue to pursue multiple paths and collaborate with others. We have prioritized platforms that we believe have the most promise and where Merck has considerable expertise and experience to uniquely contribute to the global scientific response.

Julie GerberdingJulie Gerberding

YK: It is better to pursue several approaches, because you never know what works best in terms of both efficacy and side effects until vaccines are used in many humans. Also, because the production facilities differ depending on the vaccine platforms, it is better to use different vaccine-production facilities to produce greater quantities of vaccines so that we can cover as many people as possible.

Yoshi KawaokaYoshi Kawaoka

RR: I think the answer is easy. This pandemic is so serious that we should basically use every single approach that we can think of to solve it as quickly as possible. And therefore we should look at the new technologies and, at the same time, work with well-validated ones and consider consolidating technologies. Among the technologies, I think the three that are most popular are the synthetic vaccines with RNA, the viral vectors and the protein-based vaccines plus adjuvant. We know that the protein-based plus adjuvant is very likely going to work, but it's taking time to come because you have to take the synthetic gene, produce a protein, purify and then do trials, while with RNA you take the synthetic gene and you transcribe and you make your RNA and that's it. And so in one week you have a vaccine in the lab, and in two months Moderna puts the vaccine in the clinic. With viral vectors, it takes basically three weeks to make a vector, but then you have a vector ready to go. The challenge is that these two technologies, viral vectors and RNA, are not really mature. I mean, there were technologies that were probably going to get a license in five, ten years, and now we want that to happen in three months. And so we are taking a risk. But there may be an opportunity. If they work, they work.

Rino Rappuoli Rino Rappuoli

LS: The world is taking the right approach by pursuing a range of vaccine platforms at the same time. Since time is of the essence, as well as not knowing which vaccines might be right for different populations or geographies, a multi-pronged approach is warranted. This allows for failures of one or more of the approaches without overall program failure. It’s also unclear which vaccine, or more likely vaccines, will ‘prevail’, since the full efficacy and durability of candidates will only be known over time. It may be the case that certain vaccines produce a stronger immune response in older individuals or are safer for individuals with pre-existing conditions. Other vaccines, such as mRNA candidates, may require strict cold chains that are less feasible in lower-income settings.

Lynda StuartLynda Stuart

When it comes to the manufacturing of any vaccine, how do you think we should proceed with the mammoth task of manufacturing enough for the population worldwide? And how do you think it will be delivered to the people?

GG: We will require global cooperation and financing to support the scale-up of the manufacturing of vaccine candidates found to be effective. This will require co-ordination among international funders, governments, regulators and pharmaceutical companies.

JG: Global vaccination will require global collaboration at an unprecedented scale. No company or country can do this alone. The biopharmaceutical industry is working with many partners to plan for the allocation and distribution of vaccine doses as they become available. At Merck, we are investing at risk to speed and scale our manufacturing capabilities, already targeting hundreds of millions of doses. Vaccine delivery and uptake are even greater challenges than is production of the large number of required doses. The pandemic has disrupted existing childhood immunization programs in most countries. Many lack the experience and infrastructure to deliver vaccines to older children and adults, and these programs will need to be created at a time when people cannot safely convene for mass vaccinations. In addition, vaccine hesitancy is growing, and developing trust in SARS-CoV-2 vaccines is an urgent priority for all of us.

YK: The governments of many countries are supporting companies to increase capacity to produce a large number of doses of vaccines. This is the right thing to do. One of the issues with pandemic vaccines is prioritization. The expert panels for governments are now discussing how to prioritize vaccinations; i.e., who should be vaccinated first.

RR: The problem with manufacturing is that it normally it takes five years to make a manufacturing plant, from design, to building, to outfitting, to training workers. And then you need to validate that it works according to good manufacturing practice, and for it to pass inspection. Now, we are accelerating everything, and on top of this, we're going to need billions of doses. In which plants will we do this?

What has changed is that there is an unprecedented amount of money, so the money and the will are there to build or to find capacity to do these things. We'll only know in the fall whether we really have the billion doses people are talking about or whether we'll not have them.

In theory, we would like to have vaccines ready for billions of people and to be able to deliver them to all the poor and rich countries at the same time. But in practice, it is not going to happen like that. There are some countries that are already procuring vaccines, and they will get the first doses. Fortunately, in 2021, we're going to have enough vaccines for most of the countries, poor and rich, because there are plenty in development and we will have plenty of recombinant proteins plus adjuvants. But that means that all the technologies need to work. In 2020 we will not have many doses, but we would like to have few hundred million doses. It’s exciting because really we are really pushing the limits of technologies. Obviously we need to keep the balance between being fast and being safe.

HS: It will be a mammoth task to manufacture enough vaccines for the world, and it is fantastic news that several developers have now committed to produce large quantities. Of course, everyone is just ramping up now, so world demand will not instantly be met. But we and others are already preparing to start mass production, even before we know if our vaccine candidate will be efficacious. We are producing at risk in order to have a vaccine supply as quickly as possible should we get a positive signal from our clinical study. It is then the decision of regulatory authorities and local governments to approve the use of the vaccines and determine how they should be distributed. Mass vaccination campaigns will be quite an operation but not the most challenging problem to solve in the pandemic.

LS: The good news is this process is already underway. Governments, philanthropies and the private sector have begun working on, and making investments in, at-risk manufacturing capacity for some of the leading approaches. Of course, different vaccine platforms will require different types of manufacturing processes and facilities, and they are not necessarily interchangeable. Delivery of vaccines will be accomplished through existing infrastructure for immunization, and in lower- and middle-income countries, Gavi, the Vaccine Alliance will play a role, as they have over the past 20 years with other vaccine-preventable disease efforts.

What do you think should be the criteria for rolling out the vaccine? Will this be a ‘regular’ process? What will happen with respect to regulation if something doesn’t hold up at a larger scale?

GG: The vaccine should be safe and tolerable and have at least moderate efficacy. It is unclear whether we would achieve sterilizing immunity, but even a partially effective vaccine will have impact. We will require both an accelerated regulatory process as well as a global financing mechanisms to procure vaccines for poor countries. This will be an extraordinary process, and it should be flexible, innovative and equitable.

JG: The criteria for making a pandemic vaccine available ― either as an investigational product or as a licensed product ― must remain stringent and be based on efficacy, safety and good manufacturing practices. Speed can be improved by conducting some steps in the clinical development and regulatory processes in parallel, by coordinating the design and conduct of clinical trials, by harmonizing the regulatory approval process across jurisdictions, and by preparing for manufacturing in advance rather than waiting until the product is approved. While we work to develop an effective vaccine as quickly as we can, it is important to note that, as with all our medicines and vaccines, safety remains our priority. Post-approval assessments also will be necessary, just as they were for the vaccines used in the 2009 H1N1 influenza pandemic.

RR: Safety is always a balance between of the risk of the disease and the necessary safety of the vaccine. If you have a situation in which by the time we have vaccines, the virus has gone down like it is now in Europe, then obviously the risk you want to take with the vaccine is much less. The good thing is that the type of vaccines that we are developing I think are by definition pretty safe.

But in my experience, we need to be careful, because any time you bring something new into a very large population, something strange happens, and it may be completely unrelated to the vaccine, but it creates panic. In 1976 when there was a potential flu pandemic in July and we suddenly vaccinated a hundred million in the United States, coincidentally, there was an increase in Guillain-Barre syndrome, and that was a disaster for the vaccination but it probably had nothing to do with the vaccination.

HS: We currently assume that only vaccines with demonstrated safety and efficacy will be rolled out. If the efficacy of a vaccine candidate cannot be demonstrated due to low incidence of COVID-19 as a consequence of the measures societies now take, such as social distancing, face masks, hand washing, etc., other ways to demonstrate that the vaccine is likely protective will be applied. One can think of immunobridging from another vaccine that has demonstrated clinical efficacy and an immune correlate of protection, or immunobridging from an immune correlate of protection from preclinical challenge models. Regarding the actual rollout of vaccinations, we will collaborate closely with governments and other organizations. Ultimately, they are responsible for regulation and distribution. Even with advanced manufacturing, vaccine supplies may at first be limited, and it will require a good dialogue and mutual agreement to determine who should be vaccinated first. As a company, however, we remain committed to ensuring our vaccine is made accessible worldwide.

LS: The immunization schedule and mass campaigns will need to be determined by the WHO and national regulatory and public health agencies. They will need to see more data before programs can be developed, as they will be determined by the characteristics of the vaccines that prove safe and effective in clinical trials. Nothing about it will be ‘regular’ in the sense that we haven’t seen mass immunization campaigns such as this for a generation at a global scale. Public health and regulatory agencies will need to carefully coordinate, watching the safety and efficacy signals, and making changes accordingly.

What do you think is the best approach if immunity wanes ― that is, if immunity is not durable?

GG: There will be a need to ensure that there is post-licensing surveillance to further validate effectiveness and safety. As for HIV, we need to find other biomedical interventions to support the control of this pandemic: we need antivirals, monoclonal antibodies, therapeutics and vaccines. We need to shut transmission down, treat the ill, and prevent deaths.

JG: It is too early to speculate on the durability of the response to any SARS-CoV-2 vaccine. If initial population vaccine coverage is high and occurs relatively fast, transmission rates should quickly fall, making new infections rare. In that case, the durability of protection becomes less important. It is also possible that waning immunity could still provide some protection or lessen the symptoms of the disease. Alternatively, people could be given a booster inoculation before immunity declines.

YK: Durability of immunity conferred by vaccines is an important aspect. However, even if immunity wanes, it will still provide some protection. If an antigenically different virus emerges, we will just have to change vaccine strains.

RR: If we want to really eliminate this virus and control it forever, we'll need to vaccinate a lot of people worldwide. To do that we also need durable vaccines. At this point we have no idea what's the protective level of immunity, and we don't know how long it's going last. So that's something we need to learn from the clinical trials. It’s possible that we may have vaccines that are good in the short term, but they may not be good in the long term, or vaccines that are a good for a fast first intervention, but then we need other vaccines to boost the immunity and to make that durable.

That's why it's important to have multiple platforms and multiple vaccines, because maybe some vaccines would be fast, but not durable, or vaccines that are not fast may be durable. And so we need to mix and match and see what's going to happen.

HS: If immunity wanes, which can happen over time after vaccination, it could be recommended to get a booster vaccination (another shot to strengthen the immune memory that was primed by the first vaccination). But it remains to be seen if this is needed. Because of the incubation time between SARS-CoV-2 infection and the development of symptoms, it is possible that the kinetics of the anamnestic response to incoming virus, so the recall of vaccine-elicited immune memory, is such that it can still help prevent development of COVID-19 disease.

LS: We are already seeing that most of the vaccines are likely going to require two doses. It would also not be unusual, or particularly concerning, if we did see waning immunity. It may be the case that booster doses are needed on a regular schedule ― yearly or some other cadence that brings antibody levels back to the level necessary to prevent infection or progression of disease. All these scenarios will play out as researchers have more patients in trials and the earliest patients have more follow up.

What sort of process is needed to monitor vaccine efficacy in the future ― for example, if a new mutant of the virus appears, should we go back to the drawing board?

GG: Hopefully we can address antigenic drift like we do seasonal flu vaccines or find a vaccine approach that has cross protection.

JG: Coronaviruses do evolve, but so far major changes affecting immune response have not been reported. Virus isolates from all over the world are being tracked and genetically fingerprinted. If changes that could affect vaccine efficacy are detected, their impact on the efficacy of the various vaccines would need to be assessed.

HS: We will need to monitor circulating SARS-CoV-2 for new mutations in the spike protein. If we see new variants emerge, we could test serum from vaccinated preclinical models and trial participants for the ability to neutralize these newly emerging virus variants. Only if we see escape variants, viruses with mutations that are no longer being neutralized, would we need new vaccine designs. Fortunately this virus does not seem to change too much, so let’s hope the vaccines currently in development will do the job, and if they do, at least for a significant period of time.

LS: We have systems in place to monitor vaccine safety and efficacy, through reporting and post-approval surveillance studies. These will be particularly important in the case of COVID-19 vaccines, which will likely be used at scale more quickly than any vaccine in history. The clinical trials will give a good sense of safety and efficacy, but surveillance will be important to see how it plays out in millions of people rather than the tens of thousands involved in trials. We should have the platforms developed that will allow us to both monitor the virus and continue to study whether the strains included in the vaccines need to be modified. This might be quicker with an mRNA vaccine than with other approaches, but the protein used is conserved across a range of coronaviruses.


Nature Careers


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing


Quick links