A highly unusual clinical trial in Guinea has shown for the first time that an Ebola vaccine protects people from the deadly virus.

The study, published online today by The Lancet, shows that the injection offered contacts of Ebola cases 100% protection starting 10 days after they received a single shot of the vaccine, which is produced by Merck. Scientists say the vaccine could help to finally bring an end to the epidemic in West Africa, now more than 18 months old.

“This will go down in history as one of those hallmark public health efforts,” says Michael Osterholm, the director of the Center for Infectious Disease Research and Policy in Twin Cities, Minnesota, who wasn’t involved in the study. “We will teach about this in public health schools.”

“It’s a wonderful result and a fantastic illustration of how vaccines can be developed very quickly and can be used in an outbreak situation to control the disease,” says Adrian Hill, a vaccine researcher at the University of Oxford in the United Kingdom, also not involved in the work.

The vaccine, first developed by researchers at the Public Health Agency of Canada, consists of the Vesicular Stomatitis Virus (VSV), which causes disease in livestock but not people, with the Ebola surface protein stitched into it. It is one of two vaccines currently being tested in the Ebola-stricken countries; the other one is produced by GlaxoSmithKline (GSK). The study of the Merck vaccine was led by Ana Maria Henao-Restrepo of the World Health Organization (WHO) in Geneva, together with colleagues at the Norwegian Institute of Public Health in Oslo, the Guinean Ministry of Health, and others.

The decision to start the trial was taken in October, but it didn’t get off the ground until March. By then, Ebola cases had already begun to plummet, and they were scattered across a large area in Guinea. To show efficacy in a standard randomized controlled trial, the researchers would have had to enroll far more people than was feasible.

Instead, they opted for a design called ring vaccination, in which only contacts of new Ebola patients, as well as the contacts’ contacts, were vaccinated. The rings, or clusters, were randomized; in 48 of them, vaccination occurred as soon as possible after the detection of the Ebola case in their community. In the 42 other clusters, the vaccination teams came to give the shots three weeks later. The researchers then counted the number of new Ebola cases in each ring; because they weren’t sure how long it takes for the vaccine’s protection to kick in, they only included cases that occurred at least 10 days after vaccination in their primary analysis of the data. There were zero such cases among the 2014 people who were vaccinated right away, and 16 among the 2380 who got the shot 3 weeks later. That translates to 100% vaccine efficacy, at least in this study, the researchers write.

The idea of a ring vaccination design, never before used in a formal vaccine study, “was absolutely very creative,” says Osterholm, and it allowed the team to follow the epidemic wherever it went. “Had this been a standard, straightforward randomized controlled trial, we would never had this answer.”

“It surprised me how quickly you can intervene with a vaccination and have an effect,” says Jeremy Farrar, the head of the Wellcome Trust research charity, which co-funded the study. “It’s possible to do that sort of complex work in very, very complex environments—ethically, socially, culturally and scientifically. You can do it. That is a revelation for many people.”