The FDA's Green Light for a Vaccine Might Tank Ongoing Trials

The agency is close to authorizing Pfizer's Covid-19 shot. That raises questions about the fate of study volunteers and lost opportunities to collect their data.
pfizer COVID19 vaccine
Photograph: JUSTIN TALLIS/Getty Images

On Thursday night, after more than eight hours of public discussion, a panel of independent scientists reporting to the US Food and Drug Administration found in favor of BNT162b2—the official name for the Covid-19 vaccine made by Pfizer and BioNTech. On the question of whether its benefits outweigh its risks for individuals over the age of 16, the Vaccines and Related Biological Advisory Committee voted 17 in favor and 4 against, with one person abstaining. The recommendation paves the way for the FDA to issue what’s called an Emergency Use Authorization. It’s not quite an approval, but more of a temporary all-clear to fast-track promising treatments and vaccines in times of urgent need.

sanitation workers cleaning stairs
Here's all the WIRED coverage in one place, from how to keep your children entertained to how this outbreak is affecting the economy. 

It’s already obvious which way regulators are leaning. Earlier this week, FDA scientists released more than 100 pages of documents that provide the most detailed look yet at available data from the 38,000-person Phase 3 clinical trial, the last stage of drug testing. In them, FDA reviewers glowingly described Pfizer’s two-dose vaccine as “highly effective” in preventing symptoms of Covid-19, “with no specific safety concerns” that would stand in the way of an EUA. That decision could come as soon as later today.

Within 24 hours after that, trucks are expected to roll out of Pfizer’s plant in Kalamazoo carrying vaccines packed in special, ultra-cold containers bound for distribution sites around the US, Michigan Operation Warp Speed officials said this week. The hope is to start getting the first tranche of shots into the arms of health care workers shortly thereafter. That can only happen after the Centers for Disease Control’s Advisory Committee on Immunization Practices officially recommends the vaccine. Anticipating an imminent EUA decision, the CDC committee has scheduled emergency meetings for December 11 and 13. That Americans might be getting immunized against the coronavirus less than a year after it began circulating in the US is nothing short of remarkable. It’s also complicated.

That’s because the FDA has never used an emergency authorization to make an unapproved vaccine available to the general public. (In 2005, the agency issued one allowing an approved vaccine that protects against skin-contact anthrax to be used by the military for inhaled anthrax.) The unprecedented move will almost certainly alter the fate of Pfizer’s ongoing clinical trial. To get a full licensure, enabling the company to market its vaccine after the public health emergency ends, Pfizer will have to see the study through to its planned end. But that gets more difficult once its vaccine becomes available to non-study participants through an EUA. It’s not just ethically murky to deny a working vaccine to the placebo group—it might just be feasibly impossible. At Thursday’s meeting, a lawyer who serves on the FDA’s expert committee said that he has been receiving angry emails from vaccine study participants, complaining that they should have access to a shot as soon as an EUA for it is issued.

Still, it’s not like Pfizer’s vaccine is going to be widely available for a while. It could be next summer before the company can provide the US government with enough doses to vaccinate 100 million people. That should offer some additional incentive to study participants to stay in. And it means the bigger effect of the EUA decision will likely be for the nearly 60 vaccine candidates manufactured by other pharma companies that are currently being studied in humans. If a working vaccine is made available, fewer people may volunteer to test the competitors. That’s a problem because, as WIRED’s Adam Rogers has written, it’s going to take more than one or two or even three vaccines to exit this pandemic. Public health authorities want to avoid a situation in which the early success of effective but difficult-to-distribute vaccines slows the progress or even cripples the development of other shots that travel and store more easily, can be made more cheaply, or work better in certain subsets of the population.

“Any time there is any kind of preapproval access to a drug or vaccine, there’s always a concern that by providing that access we will reduce the population that’s willing to participate in clinical trials, delaying the research that is needed to fully understand how well it works,” says Patti Zettler, a former FDA associate chief counsel who now teaches regulatory health law at Ohio State University.

It’s a pattern seen over and over again during the pandemic. Though the FDA has used its authority to issue emergency authorizations during past health crises, like H1N1, Ebola, and Zika, Covid-19 supercharged the agency’s use of the preapproval mechanism. And under the Trump administration, each time it has been fraught. First there was the EUA for hydroxychloroquine, an antimalarial drug of dubious effectiveness endorsed by the president. The agency revoked that EUA in June. Then there was the one for remdesivir, which as of November the World Health Organization recommends against. And then there was one for convalescent plasma; the lack of evidence to show if it works has not stopped a run on the sticky, yellow liquid.

The risks this time are a bit different. The statutory criteria for issuing an EUA merely demands reasonable assurances of safety and the possibility of usefulness. But it’s “permissive,” says Zettler, which means the FDA isn’t required to issue an authorization for any product that meets that lower standard. And agency officials can require a higher standard if they want to, which is what they have done with the guidelines for developing Covid-19 vaccines. “The specific expectations that the FDA has for Covid-19 vaccines seem higher than the expectations it has had for Covid-19 drugs going through the EUA process,” says Zettler. Those expectations seem much closer to the standards the FDA would require for a full approval.

That should give Americans much more confidence in this particular emergency authorization. (If you’re still skeptical, consider that earlier this week the United kingdom began immunizing its citizens with the Pfizer shot, and on Wednesday, Canada’s drug regulator granted emergency use of the vaccine.)

The problem is that there are still a lot of important questions that the study hasn’t answered yet, because it hasn’t gone on long enough. Questions like: Does the vaccine work as well in older people as in young ones? What about in groups of people from different ethnic backgrounds? How long does its protective effect last—longer than the two months Pfizer researchers have observed their inoculated volunteers so far? And can it protect not just against severe symptoms of Covid-19, but also prevent people from catching and spreading the coronavirus?

Answering them with certainty will require study participants to stick it out for the next 18 to 24 months, unaware of whether they received the placebo or the active shot. Any unblinding could change people’s behavior, compromising the integrity of the trial. But once a vaccine is made available to the public, that becomes a much bigger request. And because study participants have the right to withdraw at any time, investigators and regulators have no control over people dropping out if they learn (or guess) that they’re in the placebo group. A mass exodus would jeopardize the study’s ability to answer these crucial questions. Losing people from the trial prematurely could also weaken the odds of detecting rare or delayed side effects. Science, and society, would suffer. That’s why the FDA has urged vaccine manufacturers to keep their studies running as planned for as long as possible.

According to the documents released this week, Pfizer researchers plan to tell any study participants who ask whether they received placebo or the vaccine. If that volunteer falls into a group eligible to receive the vaccine based on their state’s prioritization schedule, the company will provide the shot as part of the study. In other words, the volunteer won’t have to drop out to get the active shot. That means the company will be able to follow vaccinated participants for up to 18 months to monitor them for any side effects. As a further incentive, Pfizer is also proposing to provide vaccines to anyone in the placebo arm who completes 6 months of follow-up, regardless of where they fall in their state’s queue for priority in getting inoculated.

That’s not going to be good enough, according to Steven Goodman, an epidemiologist and Associate Dean of Clinical and Translational Research at the Stanford University School of Medicine. At Thursday’s meeting, Goodman addressed the ethical questions swirling around a potential EUA for the Pfizer vaccine. Echoing the case made last week by bioethicists from the National Institutes of Health in Science, he outlined how the continuation of placebo-controlled vaccine trials could be ethically justified, because there are still so many important things to learn from them about who the shot works best best and what its limitations are. He also pointed out that, even without the vaccine, participants have ways to lower their risks of catching Covid—masks, social distancing, the full Swiss Cheese suite. The issue with continuing the studies as planned, with a placebo group, isn’t the ethics, said Goodman, but rather feasibility.

To that end, he offered a potential compromise. In the immediate future, manufacturers like Pfizer could pivot to a trial design called a “double crossover.” Instead of telling participants which study arm they’re enrolled in, investigators could offer everyone another round of shots when they become eligible for the vaccine—based on both national and local priority schedules. This time, anyone in the placebo group would get the vaccine, and those in the vaccine group would get placebo. Everyone would get vaccinated, but none of the subjects would know who had originally been in the placebo group. This would protect some of the study’s blinding, while eliminating the incentive for participants to drop out. While that design might mean it would take longer to get answers to questions about durability and long-term safety, it would at least preserve the ability to answer them at all.

Down the road, as more vaccines become available via EUA, Goodman proposed that trials will have to evolve toward a head-to-head model, comparing vaccines to one another without a true placebo arm, similar to the types of trials that yielded the first-ever Ebola treatment last year.

Executives from Pfizer didn’t commit to make any changes to its proposed plan during Thursday’s meeting. William Gruber, the company’s senior vice president of vaccine clinical research and development, said it would be one of the things Pfizer would be hashing out with both the FDA and CDC in coming days and weeks. The company estimates that healthcare workers, who will be first in line for a vaccine made available through an EUA, make up 20 percent of its Phase 3 study cohort. Gruber said that losing that group wouldn’t derail the study entirely. Where it could get dicey is if CDC officials quickly expand the number of groups they are recommending to be immunized, he said. “We want to be conscientious of providing vaccine to people who qualify.” Gruber suggested that would be easiest under Pfizer’s current plan. He voiced skepticism that it would be possible to pull off a double crossover. “The logistics are not trivial,” he said.

The FDA has legal authority to set contingencies on the distribution of a vaccine under an EUA, says Zettler. That could mean limiting it to people from certain age groups or imposing requirements on how frequently and for how long those people are monitored for side effects. “The law gives FDA a lot of flexibility to shape how the product is used under an EUA,” she says. But the statute doesn’t explicitly state whether or not the agency can impose provisions on how a vaccine manufacturer like Pfizer manages its ongoing trial of that product as a condition of the emergency authorization. “Maybe?” guesses Zettler. “I don’t know that there’s a clear answer to that.”

Read all of our coronavirus coverage here.

It’s something they’ll have to figure out soon. The Vaccines and Related Biological Advisory Committee will convene again on December 17 to consider the emergency use of a similar genetic vaccine developed by Moderna, working with the National Institutes of Health. AstraZeneca is also expected to be submitting data soon for its vaccine, developed with the University of Oxford, which uses a different technology. A committee meeting to evaluate that data has not yet been scheduled.

The companies that aren’t Pfizer or Moderna have already started to struggle with enrollment, says Peter Bach, director of the Center for Health Policy and Outcomes and the Drug Pricing Lab at Memorial Sloan Kettering Cancer Center. He worries that’s because of overly optimistic projections by some officials within the federal government about how quickly most Americans will get access to Covid-19 vaccines. “It’s going to be eons before it’s widely available, next summer at the earliest. So hopefully we can get people to understand that at least for now the range of choices is going to be nothing or a 50-50 shot at an effective vaccine,” he says.

For everyone who’s not at the front of the vaccine line, or who can’t or doesn’t want to participate in a trial, Bach says, the key message of the coming EUA is entirely counterintuitive: Get excited to wait. “I’m not a psychologist, but I think the endlessness and the unpredictability of when we’d have a light at the end of the tunnel has made it harder to buckle down,” he says.

But now that the FDA is about to green light the first shot, there’s a pretty good sense of how much longer it will be—around six months. And scientists have a pretty good idea of how to prevent deaths between now and then. That’s masks. And distancing. And avoiding crowded indoor spaces. And getting your flu shot. And basically just hunkering the hell down. During a week when daily Covid-19 deaths in the US topped 3,000 for the first time, the stakes have never been higher.

“We’re at the beginning of the end,” says Bach. “Barring production problems and unforeseen safety issues, a year from now we’re going to be in a really different place. Wouldn’t it be nice to have our friends and relatives who we might otherwise lose around for that?”


More From WIRED on Covid-19