Inside the global scramble to build and deliver COVID vaccines

New book from Pandemic Center's Seth Berkley recounts how scientific breakthroughs, supply chain bottlenecks and political battles shaped the pandemic response.

Excerpted from Fair Doses: An Insider's Story of the Pandemic and the Global Fight for Vaccine Equity by Seth Berkley '78, M.D. '81, senior adviser to the Pandemic Center, adjunct professor of the practice of epidemiology. He served as CEO of Gavi, the Vaccine Alliance from 2011 to 2023:


"There are many different types of vaccine, and some new varieties were developed for COVID-19, but the thing that most of them have in common is that they are biologics—drugs made from living organisms or containing complex organic molecules derived from living organisms. So, as a starting point we’re going to need living cells to grow and harvest a secure and controlled supply of countless millions of the infectious viruses (sometimes bacteria) that cause the disease we want to vaccinate against. (Let’s park the debate about whether viruses are truly alive for now and assume they are.)

These infectious agents usually need a cell host to develop. So already we’ve doubled the complexity and are now working with two living systems. The chickenpox vaccine, for example, is made from real chickenpox viruses grown in well-characterized lines of human cells maintained in a lab. For flu, we usually grow the influenza viruses inside incubating chicken eggs. 

As we’ve discussed, some vaccines are made of whole viruses, which are weakened or killed to make sure that, while they can trigger the immune system, they don’t cause the full-blown disease. Other vaccines use just a piece of the grown virus (or bacteria), such as a protein or a strand of DNA—whatever is enough to trigger the immune response.

This active piece of the vaccine, called the antigen, isn’t enough on its own. To get the sensitive antigen into billions of people’s arms as a routine vaccination, it must be packaged and transported, usually in vials of liquid that assures its stability. That’s more complicated than the inert ingredients used in pills to keep them solid and stable. 

Before it goes into the vial, the vaccine must be formulated to keep the antigen stable and to assure that the antigen is presented appropriately to the immune system by the addition of various other ingredients. A vaccine might need preservatives, which prevent it from becoming contaminated once the vial has been opened, if the vial will be used to vaccinate more than one person. Stabilizers—typically sugars, amino acids, or proteins—are included to prevent chemical reactions within the vial and to keep the vaccine components from sticking to the vaccine vial. Surfactants keep all the ingredients in the vaccine blended, preventing any damaging settling and clumping of liquid elements. And some sterile water is often added to dilute the vaccine to the correct concentration. (We want to give each person just enough to trigger immunity, but no more.)

Some vaccines also contain adjuvants, which improve the immune response to the vaccine, sometimes by keeping the vaccine at the injection site for a little longer or by stimulating local immune cells. Typical adjuvants include tiny amounts of aluminum salts (aluminum phosphate, for instance, or potassium aluminum sulphate). Some are more exotic. The adjuvant QS-21, much prized for its powerful immunological action, and a component of several important commercial adjuvants, could until recently only be scraped from a freshly harvested Chilean soapbark tree.

As an aside, it’s often these additives, rather than the active immune-triggering antigen itself, that spark controversy about the safety of vaccines. For decades, vaccine manufacturers used a preservative called thimerosal, which contained a form of soluble mercury. Although there were no data suggesting thimerosal was unsafe, some manufacturers started to phase out its use in 1999 as a precaution, although it is still used in some multidose vials in developing countries with continued evidence of its safety.

“ It’s often these additives, rather than the active immune-triggering antigen itself, that spark controversy about the safety of vaccines. ”

Seth Berkley '78, M.D. '81 senior adviser to the Pandemic Center, adjunct professor of the practice of epidemiology

So, we have our vaccine formulated. Is that the end of the manufacturing process? Not quite. We now move to what’s called fill-and-finish, which means getting the formulated vaccines into the containers, usually vials, but sometimes prefilled syringes or blow- fill-seal mechanisms for vaccines packaged in plastic. Some vaccines are lyophilized (or freeze-dried) and made into a powder that helps with long-term stability. Fill-and-finish rarely gets much attention, probably because it seems routine and low-tech compared with the science that comes before. But fill-and-finish is often the key to the entire enterprise, especially if you want to produce large numbers of vaccine doses quickly. 

That’s because, while it sounds simple, most small vaccine manufacturers simply don’t do it. They don’t have the facilities, the staff, or the desire. So, they tend to outsource fill-and-finish jobs to a limited number of specialist contract companies, which are usually booked up for months, and so rarely sit idle waiting for a global pandemic.

Vaccine manufacturing is tough enough for vaccines that have been produced before. It’s still a big ask to get each of the steps right every time. But because COVID-19 was a new disease (although we had experience with other coronaviruses, which was lucky, as some of the vaccine science had already been worked out) and the vaccines we were going to need had not been developed before, much less manufactured on the kind of scale we were going to require, we knew that manufacturers were going to have to invest serious money to build the production lines, work out the vaccine formulation, and commit to huge, new fill-and-finish contracts. And to make enough vaccine quickly enough to fight the pandemic, they would need to make these investments at risk— before they had a working, safe vaccine approved and ready to go.

Within a few months of the COVID-19 outbreak, there were dozens of vaccine candidates being developed, which used all the different established mechanisms for developing vaccines.

But many were taking newer, and in some cases completely novel, approaches. These vaccines didn’t rely on using whole viruses to stimulate the immune system; instead, they tried to do the same job using just a small piece. This is seen as safer than using whole viruses, but it makes the manufacturing job more complicated. Many companies, for example, thought they could make COVID-19 vaccines based around viral proteins, such as those the virus uses to infect cells (so-called spike proteins). The original, even most optimistic estimates guessed that it would be 18 months to two years before vaccines would be available—and in small numbers; but because the science of a stabilized prefusion spike as an antigen had been proven for other coronaviruses, which was itself an advance in vaccinology, many different groups used this technology, leading to a much higher success rate than in standard vaccine development. Such subunit vaccines tend to not produce as strong an immune response as whole viruses, so more effort is needed to find the right adjuvant additives, and thus the vaccine formulation.

The most complicated vaccines developed against COVID-19 were mRNA vaccines, which introduced new complexities related to formulation and delivery, but were also a novel class of vaccines. These take the vaccine a step further away from the actual virus by injecting people with a strand of genetic material that allows their own cells to make a piece of the spike protein, which then triggers the immune response.

“ The most complicated vaccines developed against COVID-19 were mRNA vaccines. ”

Seth Berkley '78, M.D. '81 senior adviser to the Pandemic Center, adjunct professor of the practice of epidemiology

The manufacturing of mRNA vaccines offered some serious advantages over conventional production techniques. These vaccines were the first not to rely on living cells—no more eggs or cell lines!—because the mRNA strands could be synthesized from reagents and enzymes in a lab, and this would speed things up significantly. Also, they could be made from the published genetic sequences, so no biological material needed to be transported. But they also looked—at least initially—to be a real nightmare to formulate and had challenges in stability.

It wasn’t enough to inject them into muscle tissue, as we do with conventional vaccines, and then rely on circulating immune cells to find them and trigger the protective response. Because these new vaccines needed to borrow the cell machinery of the person being vaccinated (to produce the viral protein), the companies needed a way to get the mRNA strands into enough of the individual muscle cells, after the vaccine was injected. And human cells have evolved to be fussy about what they allow inside.

To get around that, and into the cells, the mRNA strands needed to be coated in a fatty skin. Called lipid nanoparticles, their creation is a laborious skill that depends on special chemicals and numerous steps. And even once the lipid nanoparticles were available, companies needed to invest in specialized equipment that could reliably and precisely pump and squirt miniscule volumes of liquid around to get the mRNA inside or cleverly have the particle form around the mRNA itself. Although mRNA vaccines were new and had been hard to develop, they are incredibly fast to produce and ended up being the most commonly used COVID-19 vaccines in high-income countries—and, as we will see, by COVAX.

The process of manufacturing vaccines can be temperamental, unstable, and unpredictable. But that’s nothing compared with how people can behave. Faced with new discoveries—and amplified by the stress of a pandemic—they sometimes retreat into fear of the unknown or into the latest conspiracy theories. And even when people understand how important, and how safe, vaccines are, getting doses to the people who need them requires mass organization to produce the vaccine, create the systems to deliver them, develop messages to counter the inevitable misinformation, monitor side effects, find ways to reach the most marginalized and those at highest risk, and share the know-how so that vaccines can be made at scale across the world."

Reprinted from Fair Doses: An Insider's Story of the Pandemic and the Global Fight for Vaccine Equity by Seth Berkley, MD, courtesy of University of California Press. Copyright 2025.