‘Ethical Pressure’ a Factor
Ethicists Discuss Covid Vaccine Issues
The bioethicist and the auto mechanic share at least one thing in common. When you go to see one, you have a problem.
On Oct. 7, at the first of four Ethics Grand Rounds sessions held this academic year, a panel of three experts considered the question, “In vaccine trials for Covid-19, is there an obligation to offer the first vaccine shown to be effective to all participants?”
It is estimated that there are more than 200 covid vaccines in development worldwide, said Dr. Emily Erbelding, director of NIAID’s Division of Microbiology and Infectious Diseases. Five candidates are either planning or conducting phase 3 trials, 3 of which are underway in the United States already.
The stakes are high. As of Oct. 7, there were 7.4 million cases of Covid-19 in the U.S. and more than 210,000 deaths; 40,000 to 50,000 new cases were being diagnosed daily. (As of Nov. 5, new cases had topped 100,000 in a day for the first time and nearly 234,000 people in the U.S. had died of Covid-19.)
“The U.S. is leading the world in this set of grim statistics,” Erbelding noted.
Vaccines are undeniably effective, she added. Worldwide, they prevent some 3 million deaths each year, from a variety of diseases.
More than $10 billion funds Operation Warp Speed, a U.S. effort to develop covid vaccines and therapeutics, Erbelding reported. Independent of that effort, Pfizer has a phase 3 vaccine in trials; the product may eventually be purchased by OWS. And Merck has a vaccine in early stages of development.
The challenge for ethicists: What should investigators do if an efficacy signal is identified in one of the trials? Should they offer the vaccine to those in the placebo arm of that trial? Should they offer the vaccine to individuals participating in trials of other vaccine candidates?
The FDA has established an efficacy benchmark—a vaccine would have to demonstrate at least 50 percent protection vs. placebo. Assuming that a candidate achieves that standard, two concerns arise, said Erbelding: Breaking the blind might undermine longer-term data that a continued study might uncover. And, an interruption to declare success complicates other trials—should the people in the placebo arms of those trials be offered the vaccine that showed early success?
Addressing that tension was Dr. Joseph Millum of the Clinical Center’s department of bioethics and the Fogarty International Center.
Assuming the vaccine was not available outside of clinical trials, a dilemma arises, he said. Participants in the placebo arm are at greater risk of infection, “but there may be important scientific questions that can only be answered by continuing the study,” including what the duration of the vaccine’s efficacy might be.
Millum underscored the fundamental ethical challenge of all clinical research: It exposes participants to risks for the benefit of others, not necessarily for the benefit of the participants themselves.
A philosopher, Millum added other concerns: Is withholding the vaccine imposing a risk? If it is, a risk/benefit analysis is needed, as “social value can justify asking folks to take on this risk. Comparative judgments are always required in risk/benefit analysis.”
Does the added social value justify the added risk? It depends, Millum said. “What are [participants’] chances of infection? What underlying risk factors do they have? How confident are we in the estimate of efficacy?”
He concluded that risks could be justified if they were low and were necessary to gain socially valuable information.
Discussant Dr. Steven Goodman, associate dean of clinical and translational research and professor of epidemiology and population health and of medicine at Stanford University, said the terms of the argument reminded him of an old joke.
“How do you pronounce the capital of Kentucky—is it Louis-ville, or Louie-ville? Neither. It’s Frankfort.
“It’s a lame joke, but it is relevant to a serious question,” he said, suggesting that the way the question was framed led panelists down too limited an analytic path. He suggested an augmented, expanded analysis.
First off, a vaccine clinical trial is not therapeutic; it won’t cure a disease, but it may prevent one. And should herd immunity occur, “If everyone else is immune, you don’t need to be.” So some of our ethical intuitions derived from therapeutic trials need to be modified.
The key unit of analysis, Goodman argues, should be at the population level, and framed in terms of treating risk, not disease.
Unlike a therapeutic trial, people can take their own measures to reduce their risk. “People can avoid [Covid-19] risk by the usual techniques—distancing, mask-wearing, hand-washing,” he said. “Preventive options abound, in the absence of a vaccine…Self-protection is as, or more effective, than a vaccine.” So if they find out they were in the placebo arm, they don’t need the vaccine to protect them going forward.
And as to risk, Goodman noted that, in the U.S., there would be about a 1 percent chance of getting infected during the trial and 1 percent risk of dying from Covid-19 if you get infected. So an individual’s net chances of dying from Covid-19 in the trial are about 1 percent of 1 percent, or 1 in 10,000. “If the vaccine is 50 percent effective, then your risk of dying is reduced by 1 in 20,000.”
Not only is that absolute benefit very small, but “the folks in the placebo arm have also avoided the risk of the vaccine itself,” said Goodman. Therefore, there is little “ethical pressure” after the trial to offer the vaccine first to average-risk placebo group members over those at higher risk for Covid-19 death either outside or inside the trial.
All discussants acknowledged other important social factors influencing their opinions, including the possibility of “compounding injustice” to certain minority populations most heavily affected by Covid-19 if the trial population was not representative.
During a brief Q&A, someone asked if there is an obligation to tell everybody, in any trial, if there’s been an efficacy finding. Goodman said yes, and Erbelding added, “Unless you are living in a cave, you will find out anyway.”
Added Millum, “This underscores the importance of communicating with participants in any trial. We need to be clear, given the importance of their involvement.”
Said Goodman, “[A Covid-19 vaccine trial] is not like [a trial offering treatment for] cancer or serious cardiovascular disease, where there is much more ethical pressure” for treating those in the placebo arm.
The final question touched on public skepticism about vaccines in general. Would continuation of a trial, once efficacy had been shown, fuel even more distrust?
Answered Goodman, “Letting [trials] go to their planned end is the way to earn trust. It would be the death knell for vaccine trials to start cutting them off early for non-scientific reasons.”
The full discussion is archived at https://videocast.nih.gov/watch=38781.