The deadly Ebola outbreak in West Africa two years ago, the worst in recorded history, led to a vaccine that seemingly would stop the next Ebola epidemic in its tracks.
In December, doctors from the World Health Organization, Doctors Without Borders and other groups reported in the British medical journal The Lancet that a vaccine from Merck & Co. and NewLink Genetics Corp. tested during the outbreak proved to be 100% effective at preventing people from contracting the hemorrhagic fever once the vaccine's protection kicked in.
Now there are questions about that claim. A panel of scientists from the esteemed U.S. National Academy of Medicine has challenged the methodology of that 4,160-patient trial in the African nation of Guinea. They conclude the vaccine "most likely provides some protection to recipients" but that protection "could in reality be quite low."
"There is greater uncertainty about how effective this vaccine is," said one of the National Academy authors, Kathryn Edwards, chief of pediatrics and a vaccine specialist at Vanderbilt University.
Merck and NewLink didn't immediately comment on the finding.
The finding is of great consequence. In the event of another Ebola epidemic, a vaccine with modest usefulness may still prevent deaths but not provide a total firewall. The last outbreak involved 28,616 Ebola cases, and led to more than 11,300 deaths.
The 287-page National Academy report tells the comprehensive story of vaccines and drugs tried in West Africa. The drugs mostly failed, and the Merck vaccine study remains the most important work so far.
The National Academy is a private nonprofit organization that serves as an advisory group to the U.S. government. The Ebola study was sponsored by the National Institutes of Health, the Food and Drug Administration and the U.S. Department of Health and Human Services.
Those entities were parties to a fierce scientific debate that erupted during the Ebola epidemic. Clinicians from groups like Doctors Without Borders and Britain's University of Oxford argued it was unethical to conduct a normal clinical study with patients randomly assigned to a vaccine or a placebo. That is the gold standard in science, but one that would exclude half of patients from potentially preventive treatment. They contended it was immoral to exclude villagers and health-care workers on the front lines from the vaccine.
Doctors at the NIH, FDA and HHS countered that no one knew if the vaccines and drugs worked or were harmful, so a randomized study was essential.
On this question, the National Academy report was definitive: "This committee found...that the randomized controlled trial was an ethical and appropriate design to use, even in the context of the Ebola epidemic."
But the Guinea vaccination trial was a compromise. Instead of randomizing people, it randomized groups of people who had had contact with an infected person. Researchers identified a ring of human contacts in the area where Ebola broke out. In the first ring, villagers were to be vaccinated right away. The next ring of contacts, or "cluster," received a delayed vaccination after 21 days, the outside time period for someone to show Ebola symptoms. That randomization was then expanded to more areas throughout Guinea.
Estimating that a vaccine would take 10 days to be effective, researchers counted Ebola cases that erupted 10 days or more after the first case. In this analysis, there were no new Ebola cases in the immediate-vaccination clusters, and 16 in delayed clusters; hence, the 100% statement.
One of the Guinea vaccine-study researchers, Ira Longini of the University of Florida, said in an interview, "Not a single vaccinated person became ill after being vaccinated."
However, people did get ill in the first nine days after vaccination. During that period, 20 of 3,232 participants got Ebola in nine of 51 immediate-vaccination rings, versus 21 of 3,096 in seven of 47 clusters randomized to receive delayed vaccination.
So there is agreement the vaccine works. But the National Academy team concluded that varying ways of analyzing the data lead to differing estimates of how effective the vaccine is.
One confounding issue is that there are indications that exposure to the illness varied from town to town. "The small proportion of clusters in which Ebola cases were reported raises a concern about the comparability of risk across clusters," the panel wrote.
Also, about one-third of the people assigned to immediate vaccination didn't get vaccinated (some villagers declined) -- a phenomenon that generally raises red flags in research. A concern is that health-conscious people who got vaccinated had different characteristics from the others, damaging the effort toward an impartial study.
"The rules of science and research don't change during an outbreak," said H. Clifford Lane, deputy director of the NIH's National Institute of Allergy and Infectious Diseases. "The widely publicized number of 100% was not based on the most rigorous analysis of the data."
Write to Thomas M. Burton at firstname.lastname@example.org and Michelle Hackman at Michelle.Hackman@wsj.com
(END) Dow Jones Newswires
April 24, 2017 18:36 ET (22:36 GMT)