Health Policy
Squabbles Over Testing Methods Hamper Search for Ebola Vaccine
Researchers at odds over most effective way to trial treatments
Wall Street Journal
Thomas M. Burton
Updated April 9, 2015 4:42 p.m. ET

The Ebola virus outbreak in West Africa created a rare opportunity: New vaccines could be tested, and if they worked, serve as a firewall in future epidemics.

It now appears this chance is slipping away amid public health officials’ squabbles over the right way to test vaccines. As a consequence, there may never be a definitive answer about the vaccines’ effectiveness.

The study generally regarded as the most scientifically solid, which is run by the U.S. National Institutes of Health, began in Liberia but is struggling as new Ebola cases have subsided. The other studies, in Guinea and Sierra Leone, fall short of the scientific gold standard—a randomized, placebo-controlled study—partly because some medical officials have opposed giving a placebo to anyone at risk for the deadly disease. As a result, this outbreak could end without the vaccines’ being rigorously tested.

“I don’t see how it’s going to happen unless our trial gets expanded,” said Dr. H. Clifford Lane, deputy director of the NIH’s National Institute of Allergy and Infectious Diseases (NIAID), who is helping to lead the trial in Liberia. He said expanding the study to Guinea, or perhaps Sierra Leone, “is the right thing to do.”

Just this week, following inquiries by The Wall Street Journal, the World Health Organization said it and the government of Guinea will allow the NIH study to expand there. However, a senior Guinean health official said in an interview that no such decision has been made. Some other doctors in the Guinea trial oppose NIH study expansion into Guinea because it might harm their own research.

Randomized, placebo-controlled trials are overwhelmingly regarded as the best scientific way to evaluate medical products. People are randomly assigned to get the product or a placebo—in this case, a placebo vaccine. Ideally, such a study is “double-blind,” meaning that neither patients nor doctors know who gets the real thing.

At NIH, NIAID Director Anthony S. Fauci and Dr. Lane say a placebo-controlled study is both essential and ethical because otherwise no one would know if the vaccines actually are working. Early results of the NIH trial show it is safe, but testing hasn’t determined whether the experimental vaccines can actually protect against Ebola.

The NIH research in Liberia was set to enroll about 27,000 health-care workers. They were to be given a vaccine developed by NIH and GlaxoSmithKline PLC, another vaccine from the Public Health Agency of Canada, Merck & Co. and NewLink Genetics Corp., or a placebo.

But Ebola rates have plummeted since last year. The WHO reports that as of the week ended April 5, there were 21 new cases in Guinea, nine in Sierra Leone and none in Liberia.
In Guinea, the WHO is conducting a study along with Doctors Without Borders and other groups. They are using a design called “ring vaccination.” When a new Ebola case appears, people in contact with that person get vaccinated with the Merck/NewLink vaccine, establishing a “ring” around that first case. The next test group is formed when a person in, say, another village gets Ebola; people in that person’s ring get vaccinated 21 days later. The next ring is vaccinated immediately, the next 21 days later, and so on.

The researchers say the decision about giving an immediate vaccination or waiting 21 days—the end of the estimated incubation period for Ebola—produces a randomized study that could find a difference in the rate of susceptibility to Ebola from ring to ring. But Dr. Lane sees a big flaw: After 21 days, all of the test subjects have received a vaccine, making it difficult to tell differences between those who were inoculated and those who weren’t.

Dr. John-Arne Rottingen of the Norwegian Institute of Public Health, a leading figure in the Guinea study, said researchers in Guinea stayed away from using a placebo because they feared “that the outbreak would really continue into an epidemic.” He said the Guinean study nevertheless could produce definitive results about the vaccine by late summer, and argues against allowing NIH to expand its research into Guinea. He and others say people in Guinea might not be willing to be in a placebo-controlled study.

“I don’t think it’s feasible to do the NIH study at the same time as the ring vaccination study,” he said, partly because there may not be enough patients for both.

Study participants said they shied away from using a placebo because Guinean doctors didn’t want it. Neither did Doctors Without Borders.

“When there is a placebo, one person might get an effective vaccine, and another person might not get the vaccine,” said Annick Antierens, the international group’s deputy medical director. She said the humanitarian group “is reluctant to be engaged in a randomized, controlled trial in an Ebola context.” Still, Dr. Antierens and colleagues say their study is scientifically rigorous.

“The tragedy to me is that the protocol designs are not equivalent,” said the NIH’s Dr. Lane. The Guinea study, he said, “has the look of a drug distribution system in the guise of research. We don’t know that any of this stuff works, and they’re pumping it into people.”

Mandy Kader Konde, chairman of the Ebola Research Commission in Guinea, said some doctors in his country also have had concerns about giving some patients a placebo. Dr. Konde said it is “under discussion” to possibly allow the NIH to expand its research into Guinea but that, “We have to see a research protocol” first.

In Sierra Leone, the U.S. Centers for Disease Control and Prevention is helping run a study in which about 6,000 health-care workers get vaccinated with the Merck/NewLink vaccine, either right away or up to six months later. But all participants will know if they were vaccinated or not.

The main problems with this method are that vaccinated people could engage in riskier behavior, or that their bosses might order them into riskier situations, biasing the results. Also, those who weren’t might be more careful.

“We are aware that there are biases, and possible differences in behavior,” said Dr. Anne Schuchat, the CDC official heading its Sierra Leone effort.

“There is no doubt that the quickest, most definitive way to determine whether an Ebola vaccine is truly effective” and not harmful, said the NIH’s Dr. Fauci, “is to perform a double-blind, randomized, placebo-controlled trial.”