Jon Cohen: Issues continue to dog the testing of Ebola drugs and vaccines (16 Oct 2014)
At a U.S. congressional hearing today that examined the country’s public health response to Ebola, an official from the Food and Drug Administration (FDA) said it’s working to develop “a flexible and innovative protocol” to evaluate experimental treatments for the disease. The fact that no such common protocol already exists speaks to the complex practical and ethical issues that surround the use of untested drugs and vaccines in the midst of explosive spread of a virus that kills more than half the people it infects.
Given the epidemic’s unprecedented scale, a panel of bioethicists and infectious disease specialists convened by the World Health Organization (WHO) in August unanimously decided that it was ethical to use unproven treatments and preventions against this deadly disease. The panel also said there was a “moral obligation” to gather and share scientifically relevant data about whether these products were safe and effective. But it did not suggest how this should happen, and as the FDA official’s testimony indicated, new views are still emerging while others are being refined.
Over the past few months, subsequent WHO consultations and opinion pieces by prominent public health experts and ethicists have spelled out detailed visions of how to proceed with testing of experimental Ebola medicines. The issues, both practical and ethical, are starkly different for drugs and vaccines. Unproven drugs go to the sick, who are fighting for their lives and often have few options, whereas experimental vaccines are tested in healthy people—most will be first-line workers—in an effort to protect them from the deadly virus. “Ethical arguments are not the same for all levels of risk,” noted 17 prominent researchers and ethicists from 11 countries in an editorial about Ebola drug testing published online on 10 October in The Lancet.
The gold standard of clinical trial design for both drugs and vaccines is the randomized controlled trial (RCT), in which half the participants are randomly assigned to receive the experimental medicine, while the control group receives conventional care, sometimes including a placebo. Doctors Without Borders (MSF), which has treated more Ebola patients in West Africa than any other group, emphatically opposes RCTs in affected countries for either treatments or vaccines. MSF’s Annick Antierens, who oversees “investigational platforms” for experimental Ebola products, says “this cannot be defended ethically.”
The Lancet editorial, led by Piero Olliaro of WHO and the United Kingdom’s University of Oxford, sided with MSF with regard to treatments. In making their case, the authors question the meaning of “equipoise,” a fundamental ethical principle behind RCTs that says investigators should not know whether an intervention is better than what’s offered to the control group. “Equipoise is a useful principle, but it can break down when conventional care offers little benefit and mortality is extremely high,” the authors write. “This is precisely the problem with Ebola: current conventional care does not much affect clinical outcomes and mortality is as high as 70%. When conventional care means such a high probability of death, it is problematic to insist on randomising patients to it when the intervention arm holds out at least the possibility of benefit.”
The editorial rebuts an article published online on 11 September in The Journal of the American Medical Association by bioethicist Steven Joffe of the University of Pennsylvania (Penn) and an earlier one by bioethicists Annette Rid of King’s College London and Penn’s Ezekiel Emanuel that ran online on 21 August in The Lancet. Those authors contend that “compassionate use” of experimental Ebola medicines outside of RCTs, as has happened in a few patients with the antibody cocktail called ZMapp and a drug made by Tekmira, risks compromising the ability to gather scientific evidence and, as Joffe writes, “will not necessarily prevent more deaths than would administration of the drug in a well-designed clinical trial.”
Rid says Olliaro and co-authors wrongly assume that receiving an experimental treatment is necessarily better than receiving effective supportive care—which has not been available to many infected people in West Africa. She further contends that the editorial “underplays the possibility that experimental interventions can make people worse off” and “neglects the population-level concern that, even if the interventions don’t make individuals worse off, they may be ineffective and we would end up misallocating scarce resources.”
The design of real-world studies to test whether Ebola vaccines work has similarly triggered impassioned discussions. Many researchers at a WHO consultation held 29 and 30 September came to the meeting thinking that traditional RCTs were off the table: Sentiment seemed to be leaning toward a strategy known as stepped-wedge that would give all participants the Ebola vaccine at different points in time and then look to see whether people who received it later were more vulnerable to infection. But at the meeting, Ripley Ballou from GlaxoSmithKline (GSK), which has the Ebola vaccine furthest in development, won wide support for his argument that the fastest, most ethical way to assess whether the product works is with an RCT that uses an “active control”—such as a proven vaccine against an unrelated disease—rather than a dummy placebo shot.
In the wake of that meeting, some have questioned whether an active control is more ethical than a placebo. Peter Smith, an epidemiologist at the London School of Hygiene & Tropical Medicine who attended that meeting as well as a 2013 WHO consultation on the use of placebos in vaccine trials, contends the issue is practical, not ethical: An active control may persuade more participants to join the study. “If it’s easier to do the trial if you use an active control rather than a placebo, then fine, do the trial that way,” Smith says. “But to believe one is more ethical than the other is not the issue.” He stresses that the main benefit of joining a vaccine trial, especially in resource-strained countries like these, is that people who do develop the disease “are generally looked after better than people not in trials.”
Arthur Caplan, a bioethicist at New York University in New York City who co-authored the latest Lancet editorial opposing RCTs for Ebola drugs, supports their use for Ebola vaccines. But he worries that an active control may cause problems. Immune responses triggered by the vaccine in the active control arm, he notes, “could complicate interpreting the results” in people who received the Ebola vaccine. It also raises ethical dilemmas if an active control arm uses a vaccine that the country cannot afford to use routinely.
Bioethicist Nir Eyal of the Harvard School of Public Health in Boston, who also supports RCTs for Ebola vaccines, says it’s “an enormous privilege” to be offered a chance to participate in a trial that gives people a 50% chance of receiving a vaccination with some promise of success. “Of course nobody wants the placebo,” Eyal says. “But the point of medical trials is not to provide the intervention that’s medically best for the research subject. It’s to establish something that’s important—and this point is crucial—for a far larger population and to prevent human catastrophe.”
Even MSF does not rule out other trial designs that offer the Ebola vaccine to all participants. Aside from the stepped-wedge scheme, researchers could stage less rigorous “observational” trials, which are typically used after vaccines come to market to see how well they work in entire populations. “After licensure, there are plenty of observational studies that give us very, very useful and meaningful results that we believe,” says epidemiologist Arthur Reingold of the University of California, Berkeley.
In this scheme, vaccine is distributed without an organized study and investigators would look at Ebola rates in a cohort of vaccinated people—say, the health care workers at one hospital—or assess vaccination rates in people who develop Ebola. “The key issue when you start calculating vaccine effectiveness is whether people who get the vaccine and people who don’t get the vaccine are relatively comparable,” Reingold says.
Yet another hot-button issue is who should be eligible to receive experimental drugs or treatments. Scarce treatments like ZMapp have preferentially gone to health care workers because, as the August WHO consultation emphasized, they put their lives at risk for others and they are needed to control the epidemic. Rid and Emanuel question whether this makes sense, noting that health care workers have special ties to the medical community and relatively higher levels of income. “Their priority might therefore be viewed as further privileging of the already well-off, especially by contrast with those who provide care without being trained as health professionals.”
On the vaccine front, GSK may have up to 20,000 doses ready for efficacy tests in January, but that still means there likely will be far more interested participants than product. The latest WHO consultation says front-line workers should go first—not just doctors and nurses—a group that includes anyone who helps care for patients or even buries those who die.
For practical reasons, meeting participant Michael Selgelid of Monash University, Clayton, in Australia says it makes the most sense to offer the vaccine to the “traditionally conceived” notion of a health care worker. “They are best able to give proper informed consent, and it’s crucial in this scenario that we have really good informed consent,” he says. At the end of the day, Selgelid says, regulators like FDA likely will heavily influence trial design as they are the ones who will ultimately decide whether these products can come to market. “Just how flexible the regulators are going to be is a question for them rather than me,” he says.