The frustration, at times approaching despair, has been evident for years when talk among filovirus scientists turned to Ebola vaccines. In high-containment biosafety level 4 (BSL-4) laboratories, several vaccines have saved animals from what should have been lethal doses of the virus. Mice, guinea pigs and nonhuman primates—the best stand-ins for man—have been protected. A few of these experimental vaccines even progressed to the tantalizing point where they were injected into humans, in first-in-man trials in the U.S. But there the research always stalled.

All that is changing. Ebola vaccines are finally getting their shot.

Starting as early as January clinical trials designed to test two leading experimental Ebola vaccines will start to enroll volunteers in Liberia. Trials in Sierra Leone and Guinea will follow. With the rate of new infections dropping in Liberia, time is of the essence if the goal of proving whether the vaccines protect people is to be met.

No one involved in the Ebola response wants to see more cases. But the reality of vaccine research is that you can only find out if these experimental preparations work in settings where the targeted pathogen is spreading. If infection rates drop too low, the planned Liberian trial—slated to enroll 27,000 people—would have to be expanded, adding to the cost, complexity and the time it will take to get to the answers.

Organizers are still hoping to avoid that, says Charles Link, Jr., CEO of NewLink Genetics, an Iowa-based biotech that is developing one of the vaccines. “I don’t think anybody wants to go in that direction. I think they want to chase the spots where the infection is just more rampant.’’

The work to get the clinical trials up and running has been lightning swift by vaccine testing standards but agonizingly slow when measured against Ebola's rampage through the afflicted countries—more than 14,000 cases and more than 5,000 deaths. Jeremy Farrar, director of British charity foundation Wellcome Trust, which is funding the work, hopes the regulatory dotting of the i’s and crossing of t’s has not squandered an opportunity to get a proved Ebola vaccine—or several. “What a tragedy if again we missed the opportunity to assess vaccines which have been on the shelf. We have not even had to develop them from scratch,” he says.

It is hard to make sense of it today, knowing what we know now about how large and catastrophic an Ebola outbreak can become. But when the first cases in the current epidemic started moving this virus around eastern Guinea 11 months ago, there were valid reasons why no experimental Ebola vaccine had made it out of the regulatory pipeline.

In the absence of a large outbreak there was no way to test that the promise seen in labs would translate into protected people. The largest one before this was 34 times smaller; it sickened 425 people in Uganda and killed more than half of them in 2000–01. To test efficacy meant using unproved vaccines in an outbreak, an idea which until now was dismissed as ethically iffy. It was also almost impossible in logistic terms. Outbreaks flare up unexpectedly and are generally over in the few months it would take to get a clinical trial off the ground.

Under these circumstances no major drug companies would step forward to do the expensive development work needed to gain a license from a regulatory body such as the U.S. Food and Drug Administration. The demand for Ebola vaccines is not so much a market as a humanitarian project. Ebola vaccines are only needed sporadically by clusters of poor people, Ebola researchers and outbreak responders. Successful vaccines are either high volume (childhood or flu vaccines) or high cost (experimental cancer vaccines). Ebola vaccines can be neither.

So Ebola outbreaks continued to occur, killing 30, 50, sometimes even 150 people. And researchers who take their lives in their hands every time they enter a BSL-4 laboratory still had only moonsuits, helmets, boots and gloves to protect them from the killer they study.

But as case numbers in Guinea, Liberia and Sierra Leone exploded in late summer, agencies guiding the response started to conclude that an effective vaccine might be the only way to halt the epidemic. A constellation of global health heavy hitters have come together to propel products that were far from ready to the red carpet. The organizations include: the World Health Organization, National Institutes of Health, FDA, Wellcome Trust, Bill & Melinda Gates Foundation and World Bank, along with Big Pharma and small biotechs. 

It is a singular response to an enormous task: trying to test and scale up production of vaccines in a few months, a job that normally takes five to 10 years to achieve. The need for creative solutions has inspired executives of rival companies to openly share data and discuss using their products in combination. The hope: if neither of the frontrunners succeeds, maybe using company X’s vaccine as the prime and company Y’s as the boost might work. Discussions that "never happen" are happening now. “There is a willingness to consider trials that include combinations of the vaccines, to see how best they might be able to be used together, for example, in prime–boost strategies. This is very unusual. But it is going on,” says Ripley Ballou, vice president of biologics for U.K.-based GSK (aka GlaxoSmithKline), which owns the vaccine that is furthest along in this modified pipeline.

NewLink is talking with GSK about just such a strategy, if it is needed, Link acknowledges. "In my lifetime it's unprecedented to see this level of collaboration. It's the way, if the world was perfect, that maybe science would always work."

GSK’s vaccine, called cAd3-EBOV, was originally developed by scientists at the National Institute of Allergy and Infectious Diseases; GSK acquired the rights to it when it bought Swiss vaccine developer Okairos in 2013. It is an inactivated (killed) vaccine that uses a genetically modified chimp adenovirus to present an Ebola gene to the immune system.

The NewLink vaccine, rVSV-ZEBOV, was designed by scientists at the Public Health Agency of Canada. It is composed of a live virus (modified vesicular stomatitis) coupled with the primary protein found on the Ebola virus’s surface. Vesicular stomatitis viruses—VSV for short—sicken livestock but are harmless to people. It is a replicative vaccine; the virus in it generates a low-grade infection that provokes the immune system to pump out antibodies against Ebola. But the vaccine cannot trigger the disease itself.

Both experimental vaccines have pros and cons. The GSK vaccine is further advanced in testing; there should be sufficient phase I safety and dosing data to start a phase II/III trial in health care and burial team workers as well as other Ebola front liners in Liberia in January. The NewLink vaccine trails a little; its phase I trials started later. But the vaccine is easier to make than the GSK product and many more doses are already available; just how many depends on what the phase I studies show is needed to generate good levels of antibodies.

There are concerns the GSK vaccine might not be able to protect with a single dose. A two-dose delivery regimen—especially one that uses different vaccines for priming and boosting—would be phenomenally difficult given the state of the health care infrastructure in the affected countries. It is expected the NewLink vaccine will require only one shot but there are concerns it will induce mild but nonetheless confusing side effects in recipients. Those side effects are likely to be low-grade fever, chills, muscle aches, headache—in other words, precisely the same cluster of symptoms that foretell Ebola’s arrival.

On the one hand, developing these side effects within 12 to 24 hours after vaccination will tell recipients and researchers the vaccine is working, says Heinz Feldmann, a filovirus virologist who led the work on the VSV product. (Feldmann is now chief scientist for BSL-2–4 virology labs at the NIH’s Rocky Mountain Laboratories in Hamilton, Mont.) But in a world that uses those symptoms to detect Ebola infections, this will make sorting the sick from the well in the outbreak zone more challenging.

NewLink CEO Link says nothing is easy about the Ebola vaccines project: "The complexities are off the chart."

The soon-to-start phase II/III trial in Liberia will contain three arms: Some recipients will receive the GSK vaccine; some the NewLink vaccine; and the remainder, it appears, will receive a placebo, perhaps a flu shot or hepatitis B vaccine. A number of prominent scientists argued placebo-controlled trials in this situation are unethical. “None of us would consent to be randomized in such circumstances,” they wrote last month in The Lancet. But the FDA has pushed for placebo-controlled trials. "We need to learn what helps and what hurts at the soonest possible time and in the most definitive way. It's going to be important for generations to come and we have to get this right," says Luciana Borio, the agency’s lead on the Ebola response.

Farrar had been hoping for another approach, but is resigned at this point. “I’m not absolutely comfortable with it. But with a vaccine, which you are giving to healthy people when you don’t know its safety profile and you don’t know its efficacy, I actually can accept either a cluster randomized or step-wedge design or a placebo-controlled design.”

A step-wedge trial may follow in Sierra Leone. That trial design uses the fact that it is impossible to vaccinate everyone at once to create a control group; you compare the rate of new infections in areas that have already received the vaccine with those in places where rollout has not yet taken place. The benefit: everyone gets vaccine; the disadvantage: it may take longer to determine if a vaccine works.

Guinea, too, will see some type of trial although it is likely to be something less ambitious. The country’s infrastructure is in worse shape than those of its neighbors, making operating clinical trials an even more difficult task there. Marie-Paule Kieny, the WHO assistant director general for Health Systems and Innovation who is leading the drive to develop Ebola vaccines and drugs, says the Guinean trial may test whether using two different vaccines in a prime–boost strategy would work. Meanwhile the Gates Foundation may fund a trial to see whether ring vaccination—vaccinating around a known case to try to prevent onward transmission—would be effective.

There are a series of others vaccines at different stages of development; some are thought to be as promising as the GSK and NewLink products—or more so. And a few are close to being ready to start phase I trials. But those behind the GSK and NewLink vaccines face a tough reality: In this race fourth or fifth place is not likely to count. The future market for Ebola vaccines will be limited. WHO or GAVI or The Vaccine Alliance will likely stockpile product for use in future outbreaks. And some affluent countries will surely buy supplies as a shield against bioterrorism—that’s it. So unless one of the frontrunners fails, those at the back of the pack may fall away. "The ones that come after the two first, they have a place only if the first two fail," Kieny says.

Ballou insists it is always a good idea to have more than one supplier, in case a manufacturer runs into trouble with a production run. But he cannot envisage that the world could support multiple Ebola vaccines. Still, some of the vaccines may have properties that make them look better than the competition. The NewLink vaccine, for instance, may work as both a preexposure vaccine—given, like a childhood vaccine, to prevent infection—and as a postexposure treatment, in the way that rabies vaccine is generally used. If so, labs around the world would stock it for when scientists are accidentally exposed to the virus—Ebola’s equivalent of an EpiPen. A vaccine aimed at nonhuman primates might also find a place on the market. Great apes are susceptible to Ebola and infected primates can trigger human outbreaks. The butchering of infected bushmeat is thought to be how some Ebola outbreaks start, so reducing infections among the apes would protect them—and us.

Teams of experts have been on the ground in west Africa for weeks trying to work through the logistics of conducting these trials. As the decline in cases in Liberia shows, it won’t be easy. Both products require storage at –80 degrees Celsius (–112 degrees Fahrenheit), a temperature tricky to maintain in a world where constant access to electricity and refrigeration is not a given.

Despite the fact that the outbreak is not growing at projected rates, Kieny insists that good news cannot get in the way of testing the vaccines. "The vaccine development work must go ahead, because if we want to have a vaccine for the next outbreak, which will happen, there needs to be some testing…. With a start in January, there should still be a way to have some data on efficacy. What may not happen is a very large-scale deployment of vaccine. Population wide."