‘Against all odds’: The inside story of how scientists across three continents produced an Ebola vaccine

In the spring of 2014, as Ebola exploded across West Africa, a scientist named Gary Kobinger was following the news intently from Canada.

Kobinger was the head of the special pathogens unit at the National Microbiology Laboratory in Winnipeg. He and the team he led had a well-deserved reputation for their work on Ebola and other viral hemorrhagic fevers; Kobinger himself had led development of a promising Ebola therapy.

The Winnipeg lab also had been working for years on an Ebola vaccine, one that looked tremendously effective in animal models. The lab had even produced human-grade vaccine in the hopes of testing it in people. But as of April 2014, that still hadn’t happened. The vaccine had never been deployed in an outbreak. No major pharmaceutical company had expressed interest in developing it.

With Ebola appearing to spread rapidly in a country that had no experience trying to control it — Guinea — Kobinger contacted the World Health Organization to offer the vaccine.

The WHO declined the offer.

“They thought it was premature to advance it,” recalled Kobinger, who said he was told that Guinea lacked the infrastructure to approve use of an experimental vaccine. “That was true,” he added.

The reality was that, for years, scientists who studied Ebola, which belongs to a family of viruses called filoviruses, had poured their hearts into work to develop vaccines and drugs to combat these deadly scourges. And for years, they had seen promising work smash up against unscalable walls. There was no potential for drug makers to recoup development costs; and, with outbreaks only sporadic, there was little opportunity to subject experimental vaccines to rigorous tests.

But faced with the prospect of Ebola victims lying abandoned in the streets of African cities — and the world’s self-interested realization that the virus rampaging through West Africa wasn’t likely to stay there — the balance would eventually tip.

“That big outbreak was a game-changer and reminded people that this exotic virus could become a real threat to public health regionally as well as in a global perspective,” said Dr. Heinz Feldmann, Kobinger’s predecessor, who led the work to develop the vaccine.

By 2014 Feldmann had long since given up hope that the vaccine — known in the myriad studies he and others published on it as rVSV-ZEBOV — would ever get made. But through an unlikely series of twists and turns, some fortuitous and not-so-fortuitous, the vaccine has finally been developed by Merck, approved by regulatory agencies in the United States and Europe late last year, and used in the field to save lives in Africa. It is known as Ervebo.

It is a feat that built on the work of scientists in multiple countries on three continents who toiled in obscurity for years. And it ensured that when future outbreaks strike, health workers have a crucial new tool at their disposal.

“This vaccine … from the beginning to the end — it should have never happened. On so many levels … against all odds, it made it,” said Kobinger, now director of the Infectious Disease Research Center at Laval University in Quebec.

This is how it happened.

The story of the Ebola vaccine began, as scientific advances often do, with a good idea and a lucky break.

In the early 1990s, a Yale University scientist named John “Jack” Rose was trying to figure out a way to use a livestock virus called vesicular stomatitis virus, or VSV, as a vaccine delivery system. While it can infect people, VSV doesn’t sicken them. The immune system response to the virus is rapid and the levels of antibodies induced are surprisingly high.

Rose thought the virus could be an effective backbone for a vaccine — if it could be engineered to include genes of viral pathogens like influenza or HIV. The idea was that the harmless virus would teach the immune system to recognize harmful potential invaders.

But he and students in his lab had been trying for about six years to successfully manipulate VSV to add in the genes of other viruses. One very good student left his lab, he recalled, because she concluded the work was never going to pan out.

Then, in 1994, Rose heard that researchers in Germany had succeeded where he had struggled — with a rabies virus. Using their approach, he was able to recover modified VSV viruses in a few months.

“That opened up a whole new area of research on VSV for us and others,” Rose recalled.

To see if the system worked, his group added a protein from an influenza virus to VSV and injected it into mice. “The neutralizing antibody responses were fast and off the charts,” he said. “And of course the mice were completely protected after a single dose.”

Rose’s lab and others later used VSV as a backbone for experimental vaccines for bird flu, measles, SARS, Zika, and other pathogens. It always worked.

Without the high-security laboratories needed to handle the world’s most dangerous viruses, the researchers couldn’t work on Ebola. Rose, nonetheless, thought a VSV Ebola vaccine, in theory, would work as well.

Yale patented Rose’s VSV construct, and licensed it to Wyeth Pharmaceuticals.

By his own estimate, Rose shared his VSV vector with at least 100 labs worldwide. One of them was located in a city in Germany with a rather auspicious name: Marburg.

It was there, in 1967, that laboratory workers and people related to them became sick with what was later named the Marburg virus. The source: primates imported for research purposes. (Nine years later, scientists would discover a related virus, Ebola.)

When a scientist named Hans-Dieter Klenk moved to the city in the 1980s to lead the Institute of Virology at the Philipps-University Marburg, there was no research being conducted there on Marburg or Ebola. Klenk decided that ought to change. He asked one of his students, Heinz Feldmann, if he wanted to continue to work on influenza, or move over to filoviruses like Marburg. “He did not think long,” Klenk said. “This is how it started.”

With Rose’s virus, Klenk’s team could study individual Ebola genes by putting them onto the VSV backbone. The beauty of the approach was that they could do the work at lower biocontainment levels than Ebola research is normally conducted, which made it safer, faster, and cheaper.

At first, the protein on the surface of the VSV virus — known as the glycoprotein or G protein — was swapped out and replaced with the Ebola glycoprotein. Later the group made a VSV virus with the G protein of the Marburg virus.

Klenk said that, even then, there was some discussion about whether the hybrid VSV virus could be made into Ebola or Marburg vaccines. But the group didn’t have high-containment labs in which to do animal studies, so it couldn’t test the theory.

Back across the Atlantic, though, Canada was building a new national microbiology laboratory — one that included biosafety level 4 facilities, the type needed to study Ebola. Feldmann was recruited to lead the special pathogens team there. And when he left Germany in 1999, he asked Klenk if he could take the VSV construct with him, so he could continue his work. Klenk agreed.

“This became ‘the Canadian vaccine’ — how it was known for many years. But certainly it has also roots in Marburg,” Klenk said.

As Feldmann recalls it, he wasn’t even thinking about using Rose’s VSV construct as a vaccine when he was in Marburg. “We had no vaccine program. We had no interest in vaccines,” he said. “We used it basically as a model system to study the glycoprotein.”

After he’d moved to the Canadian lab, though, Feldmann and Tom Geisbert, a friend and frequent collaborator, heard Dr. Gary Nabel, then head of the National Institutes of Health’s Vaccine Research Center, deliver a lecture on Ebola. He argued the glycoprotein was the cause of the profound damage Ebola does when it infects animals and people.

Feldmann and Geisbert, an Ebola expert who was then at the U.S. Army Medical Research Institute of Infectious Diseases, thought Nabel was wrong and that they could use the VSV construct to prove it.

In Winnipeg, Feldmann’s team infected mice with the VSV virus containing the Ebola glycoprotein. If Nabel’s theory was correct, exposure to the protein should have been toxic to the mice.

The rodents were unharmed.

As an afterthought, the group decided to expose the mice to Ebola to see what would happen. All the mice that had been infected with the VSV virus carrying the glycoprotein were fully protected from illness; the mice that had not been exposed to the VSV virus all died.

“I guess that was basically the start of the vaccine project, even though I don’t think we really jumped on it with a lot of priority right away,” said Feldmann.

In hindsight, that delayed response might seem striking. But at the time the Winnipeg group faced more urgent matters. In 2003, an alarming new disease — which came to be called severe acute respiratory syndrome, or SARS — burst out of China and spread to Hong Kong, Vietnam, Singapore, and Toronto. The special pathogens team at the Canadian lab joined the search to try to determine what was causing the new illness and how to contain it. Other work was put on hold.

With the Winnipeg team tied up, Geisbert agreed to replicate the mouse study in nonhuman primates, considered the best animal model for what happens when humans are infected with Ebola.

Like the mice before them, monkeys that had first been exposed to rVSV-ZEBOV survived what should have been a lethal Ebola challenge. A paper on the study was published in Nature Medicine in 2005 — and it “blew the doors off,” recalled Geisbert, who is now with the University of Texas Medical Branch in Galveston.

It was suddenly clear that the modified VSV vector, loaded with the Ebola glycoprotein, was not only safe, but that it could be used as the foundation of an effective vaccine.

Scientifically speaking, it was thrilling. But realistically, it was a nonstarter. Vaccines are estimated to cost in the neighborhood of $1 billion to develop. The pharmaceutical industry was not interested in making a product to protect against a disease that emerged only now and again in impoverished countries. At the time, Ebola had killed about 1,300 people over the nearly 30 years since it had been discovered.

“Yes, it was exciting, but where would you go with that excitement?” Feldmann said when asked about the results. “You went to the bar next door and had a beer and continued working.”

“No one was interested in Ebola.”

In 2008, Feldmann left Winnipeg to become the head of the virology program at the NIH’s Rocky Mountain Laboratories in Hamilton, Mont.

In March 2009, a sudden crisis led to a critical decision.

A German researcher pricked her finger with a needle containing Ebola virus while doing a mouse experiment. The needle penetrated three layers of gloves; though the wound didn’t draw blood, her skin had been punctured.

The University Medical Center Hamburg, where she was taken, reached out to Ebola researchers in the U.S. and Canada to see if there was anything that could be done.

The experts on the call — a who’s who of Ebola researchers and field workers at the time — concluded that she should be offered the VSV vaccine. Some testing in animals had shown it had increased survival when given after exposure, even as late as 48 hours afterward — though whether that translated into a similar effect in people was unclear.

The Canadian government agreed to send the vaccine — which was not human-grade product, but material the lab had produced for animal studies. Roughly 48 hours after the accident, the woman, who was not publicly identified, was vaccinated.

The following day, she developed a fever. It’s not uncommon for a live virus vaccine, like rVSV-ZEBOV, to trigger a fever; it’s actually a sign that the immune system has activated. But fever could also have been the first symptom of an Ebola infection. With no way to know which scenario they were facing, the doctors monitoring the researcher transferred her into a specially erected bio-containment treatment unit.

The fever subsided; the woman did not develop overt symptoms of Ebola. It was impossible to know whether the vaccine had shut down a developing infection, or whether she had never been infected in the first place. Feldmann and others believed the latter was more likely.

But, importantly, there were no negative consequences to using the vaccine, a fact that would later give comfort to people who were struggling with whether to deploy rVSV-ZEBOV in a much larger emergency.

If two assets hadn’t come researchers’ way, it’s almost a certainty that Ervebo would have never come to fruition. One was money, the other was a rare talent.

The monetary asset was a $2 million grant awarded to the Winnipeg lab. The grant, though a drop in the bucket when it comes to scientific research, was hard-won. Feldmann and Steven Jones, who had done a lot of the animal testing in the lab, had repeatedly applied to U.S. government agencies for funding. Their applications were repeatedly rejected.

In fact, the whole special pathogens program was always in the crosshairs of government bean counters. Every year during the budget process, Frank Plummer, scientific director of the National Microbiology Laboratory from 2000 to 2014, would get pressed on why Canada needed to work on pathogens like Ebola. “I always had to defend it, and we always were scrambling for money,” he said.

The grant was doled out by a Canadian defense program that funded research into tools to combat bioterrorism. It was to be used to produce test lots of human-grade VSV vaccine for Ebola Zaire, the species of the virus that has been the most common cause of Ebola outbreaks.

The idea was to hire a German contract manufacturer, IDT Biologika, to produce the vaccine. First, though, the Winnipeg lab had to create the starter material from which it could do so. That work was laborious and tedious, and it fell to the second asset — Judie Alimonti, an unassuming immunologist and lab scientist dedicated to the cause.

Among other tasks, Alimonti had to develop tests to show that the materials being transferred to IDT did not contain any microorganisms that inadvertently contaminated the product. When IDT had produced vaccine, it shipped back vials to Alimonti who ran the tests to ensure the material was pathogen-free.

“Judie did that. … She spent two years on this alone, I think,” said Kobinger. “She put all her heart into it.”

Alimonti, who died of cancer in 2017, loved the project. Her former colleagues describe her as the unsung hero of the Ebola vaccine.

She was “a very meticulous, methodical scientist,” said Plummer, who oversaw the VSV vaccine project after Feldmann and then Jones left the Winnipeg lab.

Indeed, if IDT didn’t already have in hand the seed material to make more vaccine, the time needed to develop usable material would simply have been too great when a crisis struck. And the Winnipeg lab’s Ebola vaccine would have missed the chance to attract the big-league help needed to get the vaccine tested and eventually licensed.

“I think probably it would have never happened,” Kobinger said.

After securing a patent for the vaccine system — and obtaining permission from Wyeth Pharmaceuticals to use its platform to produce Ebola and other viral hemorrhagic fever vaccines — the Winnipeg lab talked to a variety of pharma companies, big and small, looking for a development partner. The only interest came from a tiny firm called BioProtection Systems Corp., a spinoff of NewLink Genetics, a biotech working on cancer vaccines.

The interest had little to do with Ebola, or even infectious disease vaccine platforms, which is what BioProtection Systems would be licensing. The company was looking for assets to add to its portfolio to generate capital investment, Jones recalled. “It was a business decision for them that it would enhance their portfolio and make it easier to get funding to do the other work they were interested in,” he said.

The deal turned out to be a steal.

NewLink agreed to pay the Canadian government — which officially held the patent — about $156,000 for each product it developed. (An amendment a couple of years later would increase that amount to roughly $360,000.) The Canadian government would also get royalties from some sales, though in truth those royalties were never expected to deliver much.

The company, which would later be absorbed into a drug maker called Lumos Pharma, never pushed development of the Ebola vaccine. For all intents and purposes, it would have remained no more than a scientific idea gathering dust on a shelf.

And then came the West African Ebola crisis.

The outbreak, which probably started in late 2013, smoldered in the way Ebola outbreaks do. At first, the assumption is the people who fall ill have contracted malaria or some other disease. Eventually, health workers become ill. Finally, a diagnosis of Ebola is made.

The WHO reported a “rapidly evolving” Ebola outbreak in southeastern Guinea on March 23, 2014. By that point, there were already 49 cases and 29 deaths, making it larger than about half of all previous known Ebola outbreaks. The following day, the tally grew: 86 cases and 59 deaths.

Before the week was out, cases were reported in Guinea’s capital, the first time Ebola had taken root in an urban setting. By the end of March, one of Guinea’s neighbors, Liberia, was investigating possible cases.

Back in Canada, Kobinger’s offer to the WHO had been rebuffed. He heard that, a few weeks later, GSK, which was developing its own Ebola vaccine, also offered vaccine to the WHO. It, too, was turned down.

Still, Kobinger saw a bright side: “The seed of the [vaccine] being available was planted.”

He mentioned the offer to Dr. Armand Sprecher, an Ebola expert with Doctors Without Borders who he knew as a strong supporter of the VSV vaccine. As Ebola spread from Guinea to Liberia and Sierra Leone, the group, known by the acronym of its French name, MSF, had been emphatically warning the WHO and others that conditions on the ground were rapidly deteriorating. Prompted by Sprecher, MSF started pushing for use of the VSV vaccine.

On Aug. 8, 2014, the WHO declared the outbreak a global health emergency. A couple of days later, the Canadian government announced it would donate its vaccine to the agency.

It was a pivotal moment, but it also created a conundrum. Was the vaccine safe to use? What was an appropriate dose? And how could human trials be conducted in the midst of an epidemic?

It is widely considered to be unethical to use untested drugs or vaccines in Africa, where clinical safeguards are sometimes lacking and where memories linger of scandals like Pfizer’s use of a meningitis drug that resulted in the death of 11 children in 1996. Yet given the scale of the expanding crisis, experts were now scouring the medical literature, looking for any existing medicines that could be repurposed to fight Ebola, or experimental vaccines or drugs, regardless of where they were in the developmental pipeline.

The WHO convened a meeting to determine the best path forward. It concluded there was an “ethical imperative” to try experimental vaccines and therapies, given the extraordinary threat Ebola presented. But it was also decided that in order to use the donated Canadian vaccine, clinical trials first had to assess its safety and establish the appropriate dose. It was apparent to everyone that NewLink didn’t have the expertise or bandwidth to take on this work.

Marie-Paule Kieny, who then headed the division of the WHO tasked with trying to spur development of drugs and vaccines for diseases like Ebola, noted the company had never conducted a clinical trial. “So when we were saying, ‘We should do a clinical trial in Africa,’ they were completely lost,” Kieny recalled.

Researchers at the NIH and Walter Reed Army Institute of Research (WRAIR) started planning a Phase 1 trial to determine the appropriate dose of the vaccine. Others started working on Phase 1 trials that would be conducted in Switzerland, Germany, Gabon, and Kenya.

The WHO and others — including players in the U.S. government — began casting about to find a more experienced pharmaceutical company to partner with, or to acquire the vaccine from NewLink.

The list of potential white knights wasn’t long. Sanofi Pasteur wasn’t interested. Novartis had sold its vaccines division to GSK earlier that year. GSK was racing to test its own experimental Ebola vaccine. Johnson & Johnson’s vaccines division, Janssen, was also working on an Ebola vaccine, but it wasn’t as far along as the GSK or the NewLink vaccines. But Merck had experience producing vaccines in the types of cells the VSV vaccine was made in.

The company started fielding approaches from the WHO, the Biomedical Advanced Research and Development Authority (BARDA), and others asking if Merck would step up. Merck had already been debating what it could do to help with the outbreak, and the vaccine seemed like a good fit.

“We already understood how to scale production of vectors in that system, and we knew how to manage the whole scaling process. We had an enormous amount of know-how that was kind of complementary to the work that had already been done,” said Dr. Julie Gerberding, Merck’s executive vice president and chief patient officer for strategic communications, global public policy, and population health.

Gerberding said there was “moral clarity” at Merck that this was something the company should do.

In the anxious autumn of 2014, when Ebola was ravaging West Africa, it seemed like rumored negotiations between NewLink and Merck were taking forever. But in terms of a pharmaceutical deal, negotiations happened at warp speed.

“[From] Merck deciding to get involved from the initial sort of exploratory discussions at the beginning of October to licensing the vaccine in mid-November, that is unprecedented in terms of internal decision-making within the company because people recognized how urgent it was,” said Dr. Mark Feinberg, the company’s chief science officer at the time.

Merck agreed to pay NewLink $50 million for the license. The deal was announced Nov. 24, 2014.

That same month, around the time the researchers at NIH and WRAIR arrived at an agreement about the appropriate dose for the vaccine, Kobinger made a heart-stopping discovery. It was about the vaccine being used in the trials, donated by the Canadian government.

From its earliest iteration, when Feldmann and his team saw that it saved mice from Ebola, the vaccine had been made with the glycoprotein from the Ebola Zaire strain known as Mayinga. But somewhere along the line, a key feature in the vaccine had been changed.

The human-grade vaccine made by the German contractor contained the glycoprotein from a different Ebola Zaire strain.

No one had told Kobinger the change had been made; he discovered that Alimonti, who had done all the work to prepare the materials for IDT Biologika, had used the different glycoprotein on instructions from one of her supervisors. No one outside the lab realized that the vaccine being tested in people was not identical to the one that had been studied so thoroughly in animals.

Feldmann believed the switch would have no impact on whether the vaccine was effective. “Quite frankly, from a scientific prospective, it doesn’t matter,” agreed Kobinger. “From a regulatory perspective, it matters a lot.”

Kobinger quickly fired off an email to the Food and Drug Administration to inform the agency of his discovery.

He feared fireworks. That’s not what he got. “We never heard back,” he said, noting the fact that Phase 1 human trials were already underway may have helped.

Still, Kobinger urgently shipped off some doses to Montana so Feldmann could test the vaccine in primates to ensure the change had not affected the vaccine’s efficacy. It had not.

As researchers who conducted the Phase 1 and Phase 2 trials crunched their data, others were planning pivotal Phase 3s. The earlier trials were to determine if the vaccine was safe to administer; the Phase 3s would tell the world if it actually worked.

The NIH had reached an agreement with Liberia to test both the GSK vaccine and the VSV vaccine that Merck had acquired. Scientists from the Centers for Disease Control and Prevention were to test the vaccines in Sierra Leone.

The government in Guinea was also keen to host a trial. But the country’s health infrastructure was weaker than those of its neighbors, making it a more challenging place in which to conduct a study. When no other group stepped forward, the WHO announced it would conduct a trial there, with help from MSF.

The plan was to use an approach known as ring vaccination. People who had had direct contact with anyone infected with Ebola were to be vaccinated, as were their contacts. The goal would be to both protect people in the virus’s path and to block it from spreading.

In place of a placebo control, the rings were randomly assigned to either immediate vaccination, or vaccination after a 21-day delay. If there were more cases among the people in the rings that were vaccinated after the delay, the vaccine was working.

The approach was distinct from the one used in classical trials, in which participants are randomly selected to get either an intervention or a placebo, with neither the researchers nor the participants aware of which was administered.

To proponents, ring vaccination, a type of adaptive trial design, was the most feasible approach. Not everyone would agree.

Guinean health professionals made up the bulk of the team that conducted the ring vaccination trial.

Dr. Abdourahmane Diallo, a public health physician who works for Guinea’s health ministry, was one of those who answered the WHO’s appeal for help. He recalled that his colleagues were excited at the prospect of taking part in the study. “The only thing in our mind was that we wanted to assess if the vaccine worked or not because we wanted to contribute if possible to find a solution,” Diallo told STAT via email.

There were hints that it was working, he remembered. Others agreed. Neighborhoods where transmission had been intractable stopped producing cases after vaccination occurred. “But that’s not proof, of course,” said Kieny of the WHO. “That’s just a feeling.’’

In June, however, the trial’s data and safety monitoring board concluded there were not likely to be enough additional cases to change the outcome of the study. The vaccine had worked.

From 10 days after vaccination — the time needed for the immune system to respond to the vaccine — there were no cases among people who had been vaccinated in the early rings, but there were cases among the delayed vaccination rings.

The data and safety monitoring board recommended that health workers vaccinate anyone who had come in contact with people infected with Ebola as quickly as they could be found, rather than delaying some vaccinations.

On July 31, 2015, less than a year after the Canadian government donated the vaccine, the findings of the trial were published by the journal The Lancet. In less than 12 months, 12 clinical trials running the gamut from a “first in man” dosing study to a Phase 3 efficacy trial had been conducted. “That has never happened,” said Feinberg, who is now CEO of the International AIDS Vaccine Initiative.

In an editorial, The Lancet called the trial “a remarkable scientific and logistical achievement.”

“That such a trial was even possible is a testament not only to the skill of the research teams but also to the commitment of communities to defeating an epidemic that has devastated their nation,” the journal’s editors wrote. “Before this work, no clinical trial on this scale had ever been performed in the country.”

The Guinea vaccine trial — the trial that almost hadn’t happened — was the only one to reach a conclusion. The trials in Sierra Leone and Liberia ended without having enrolled enough patients to do so.

Despite the success, the study produced a backlash that was almost instantaneous in some quarters.

While everyone wanted an effective Ebola vaccine, there was heated debate over whether adaptive design studies were sufficient to prove that the Merck vaccine met that threshold. The detractors were vocal.

“It was ugly, frankly. It was ugly,” Kieny said. “I thought everybody would be happy to say: ‘This is great.’ But actually, this is when the bashing started. ‘This is not a study.’ … ‘Only a randomized controlled trial.’ The campaign against these results was flabbergasting.”

Both the findings and the approach were critiqued — and to this day are challenged by some experts. In the spring of 2017, the National Academy of Sciences issued a report on conducting research during disease outbreaks that called into question the way the trial was conducted and its findings.

“We concur that, taken together, the results suggest that the vaccine most likely provides some protection to recipients — possibly ‘substantial protection,’ as stated in the preliminary report,” the authors wrote. “However, we remain uncertain about the magnitude of its efficacy, which could in reality be quite low or even zero, as the confidence limits around the unbiased estimate include zero.”

In the trial, the vaccine had been found to be 100% effective. But the number of people enrolled was limited, and no vaccine works every single time. Still, the results were strong enough to convince Merck to push forward with the vaccine.

It did so with support from BARDA, which began funding rVSV-ZEBOV during the West African outbreak. The agency’s director, Rick Bright, estimated that it has spent about $175 million supporting production of vaccine and validation of Merck’s production facility for the vaccine in Germany.

When Ebola broke out in Equateur province in the Democratic Republic of the Congo in the spring of 2018, the country agreed to use the vaccine under a “compassionate use” protocol — similar to the protocol used in a clinical trial when there is no approved therapy. Vaccination began again in the country in the current outbreak, this time eight days after it was declared. Since then more than 260,000 people have been vaccinated.

“I’m really proud of that,” Rose said of the role played by the vaccine. “We worked night and day … trying to get VSV to work and finally got it to work.”

On Nov. 11, 2019, Ervebo was approved by the European Commission, the first time it had been licensed by any regulatory agency. On Dec. 21, the FDA approved the vaccine in the United States.

Beth-Ann Coller, who has been the project lead at Merck — another unsung hero of the vaccine, said Kobinger — choked up a little describing her reaction to the approval of the vaccine. “We are thrilled and we are proud,” she said.

Kieny waxed a bit philosophical about the unlikely success of rVSV-ZEBOV.

“You know, when things go really wrong, quite often it’s a succession of little issues in which none by themselves could have derailed the train. And sometimes for something good to happen it’s the same,” she said. “It’s just bringing together a number of discrete actions and discrete facts, which each alone would not have made it. But everything together makes it a success.”