At the height of the Ebola crisis in West Africa, an experimental drug produced in tobacco plants appeared to help save the lives of a few people who were infected, including some American aid workers. But in the midst of the outbreak, scientists were not sure if the treatment truly worked.
Findings published Wednesday of the first randomized, controlled trial of the treatment, ZMapp, don’t provide a definitive answer either, but they could provide scientists with crucial data to consider in the case of future outbreaks.
While ZMapp seemed to help more people survive an Ebola infection, it narrowly missed the statistical threshold that researchers set to prove the drug’s effectiveness, according to the new study, published in the New England Journal of Medicine. Plus, the researchers said they were left with inconclusive results because of the small number of patients enrolled.
As a result of those factors, health officials are recommending that anyone infected with Ebola receive ZMapp, said Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases and one of the paper’s authors.
“Given that I know it’s safe, and given that I know it’s just missing the statistical cutoff point of being significant, we feel that on an emergency basis, we can distribute it,” Fauci said.
Other researchers questioned whether the design of the trial was problematic. Among the concerns: the inclusion of very sick people in the trial, some of whom died before they had received the second of three infusions of ZMapp. It’s possible that those people were so sick and had sustained such extensive tissue damage by the time they were given the therapy that no drug could have saved them.
“You can’t reverse that, it’s too late,” said Gary Kobinger, the director of the Infectious Disease Research Center at Laval University in Quebec City, who helped develop ZMapp. “There’s real damage in there.”
Importantly, the trial did establish that ZMapp is safe, with little risk of dangerous side effects.
ZMapp consists of three antibodies, proteins the immune system uses to recognize and attack viruses. Demand for the drug grew during the Ebola crisis, which ultimately killed more than 11,000 people, after it showed major successes in studies done in monkeys and appeared to help some, though not all, people who received the few regimens that were available.
The problem was that by the time production had ramped up to meet the demands of a clinical trial, the crisis was starting to wane. It’s a dilemma faced by other Ebola treatments and vaccines: Clinical trials need a certain number of sick people on whom to test the therapy for the results to be valid.
With the ZMapp trial, the researchers had originally wanted to enroll 200 patients, 100 of whom would received ZMapp in addition to the standard of care, and 100 of whom would only receive the standard of care. Instead, they enrolled just 72 people total (71 of whom were evaluated in the end). That, the authors wrote, “affected our ability to reach definitive conclusions.”
The researchers enrolled patients from March 2015 to November 2015 in Liberia, Sierra Leone, and Guinea, and included an American health worker who was evacuated from Sierra Leone after getting sick. In the trial, 13 of the 35 patients who received the standard of care in their country died within 28 days, while eight of the 36 patients who also received ZMapp died.
Before the trial, the researchers established that the probability that incorporating ZMapp was better than standard of care alone had to reach 97.5 percent to show ZMapp worked. But based on the analysis of these patients, the researchers found the probability was 91.2 percent.
Fauci said he thought that if the researchers had been able to enroll the full 200 patients, the results could have shown a statistically significant result. He also said that the study showed it was possible to run a randomized, controlled trial in the middle of an international health emergency, which is important to make sure any new drugs do not cause more problems than they address.
Given concerns that some patients in the clinical trial were simply too sick to respond to ZMapp, Kobinger said he would have treated the severely ill but not included data about them in trial; their deaths might be not preventable at all.
In the paper, the researchers wrote that the patients’ viral load at the beginning of the trial “was strongly predictive” of both dying and how long someone needed to be in the hospital.
With the results of the trial in, health officials are discussing how to proceed with additional tests for ZMapp should more cases of Ebola arise, said Gary Disbrow, who helped lead Ebola response at the US government’s Biomedical Advanced Research and Development Authority.
Disbrow said the authority, other US agencies, and international partners continue to support the development of Ebola treatments, vaccines, and diagnostics so they are available in the case of future outbreaks.
“It’s important for us to finish what we started … and to make sure we get the data, the safety and efficacy that’s necessary to support the approval and licensure of therapeutics and vaccines,” Disbrow said.
Meanwhile, scientists are also at work on other treatments. Erica Ollmann Saphire, a structural biologist at Scripps Research Institute, said she and other researchers are investigating whether other antibodies could be more effective in fighting Ebola; she said she thinks scientists could improve on the two of the three antibodies in the ZMapp cocktail.
A more effective antibody cocktail could induce a stronger immune response, require a smaller dose, or fight the other species of Ebola as well. (The one responsible for the West African crisis is formally called Zaire ebolavirus.)
“There’s reason for optimism, but I think we can do better,” Saphire said about the trial. “I think science can do better.”