NO VACCINE IN HISTORY HAS BEEN SO EAGERLY ANTICIPATED as the one that will prevent COVID-19. Without one, it’s hard to imagine life returning to normal, and as the economic shadow of the lockdowns begins to lengthen, there’s a pressure to get the job done quickly.

But speed is not a hallmark of vaccine development. Usually it takes a decade or more to develop one, though scientists say they might, with moonshot effort, shorten that time to 18 months. For an exhausted public, even that is too long to wait. “There’s a well-worn pathway to how vaccines are tested for safety and efficacy,” says Libby Hohmann, an infectious disease physician at Massachusetts General Hospital, “and this situation just blows that up.”

Hohmann is also the director of the hospital’s institutional review board (IRB), a body that exists at every U.S. research institution and protects the rights and welfare of its human subjects. The IRBs will be the waystations that look out for those who receive these vaccines—most of which will not work and some of which may be harmful in themselves.

Ethical questions around vaccine development have already begun to surface. How safe is “safe enough” before a vaccine can be released to millions of people? Should laboratories deliberately infect volunteers to get quick results? For all these questions, IRBs have the unenviable, impossible task of weighing the urge for speed against its human cost.

Q: How is the push for a COVID-19 vaccine different than other vaccine efforts you’ve worked on?
A: Typically we have a long road for developing vaccines, and we have a very clear pathway for how we test them. We study how the vaccine itself operates in healthy individuals. We look at immunogenicity—how much immune response a vaccine provokes—and we see how safe it is for the subjects. We aim to do this well before we put them at any risk of infection. Then we do phase 3 studies in settings where the disease might be present, in a hard-hit community for example, and we can see if the product prevents the disease in practice. That’s the normal way of doing things.

This situation just blows that up, because everybody wants the testing process to be completed in two months. That’s a very tough thing to do, because it calls for a number of compromises, and each of those comes with a potential cost.

We also don’t just know enough about this virus. Is it going to mutate? And how will COVID-19 move through the population? What if we look at Italy in nine months and there’s 90% serological immunity there, a solid immunity, which means that people have become protected through natural exposure and they’re never going to get this sucker again? Maybe then we don’t need a vaccine at all. All of those unknowns need to factor into how we approach these vaccine trials and what we’re willing to consider.

Q: Your institutional review board and others like it will weigh in on the testing protocols. We’d like to ask about placebos, and the ethical baggage those carry. There have been a few passionate articles in recent weeks in JAMA and The New England Journal of Medicine begging doctors to wait for placebo-controlled trials before trying new treatments. It must be difficult to put a patient on a placebo arm of a trial when something potentially lifesaving is out there?
A: When you don’t know if a new treatment works—and by know, I mean have solid, peer-reviewed evidence—there’s a lot of downside to just rolling it out across the population. I can give you a lot of examples. Take bone marrow treatments for breast cancer. These were routinely happening in the 1990s , and people were petitioning the government and their insurers to cover them. But when the clinical studies finally took place, they were shown not to prolong survival. In fact they killed people. This was because people weren’t waiting for the right trials to be done.

Of course it can be difficult to talk about the placebo arm with patients, and wait for well-planned studies, if a treatment seems to offer hope. A study I’m involved with right now is placebo-controlled, one-to-one. That means a person has a 50% chance of getting the drug and a 50% chance of getting the placebo. I tell patients, “I don’t know if this drug works. It looks really good in a test tube. You only have a 50% chance of getting it — but you have a zero percent chance if you don’t sign up here. And this will help us most rapidly determine whether the drug works.”

Q: Is there ever a scenario in which you would want to do a two-to-one or three-to-one treatment-to-placebo ratio?
A: Yes. In this study, I would be much happier if I could tell people that this is a two-to-one, that they have a two-to-one chance of getting the active drug. And those kinds of trial designs are becoming more common.

Q: There is a real hunger for studies that can quickly yield results. Some have floated the idea of doing “human challenge” trials. Researchers give people an experimental vaccine and then intentionally infect volunteers to see if it works. If that proposal came before your IRB, what would you say?
A: I think that’s a rotten idea and I’ll tell you why. I used to do challenge studies back when I worked on Salmonella and typhoid fever. Our young volunteers were infected with bacteria that could potentially be cured with antibiotics. These trials were very medically challenging, but they were useful, and, yes, there is a place for them.

What I’ve seen in the hospital here with COVID-19 would not make me feel good about a human challenge model. The proposals floated so far have talked about using people in low-risk groups, young folks who don’t have any preexisting medical conditions. I have patients that I’m taking care of who are in their thirties with no medical problems whatsoever. They are dying. We can’t predict who is going to crash and burn. People who are otherwise healthy and uninfected would die from that trial.

I consider myself a pretty gutsy investigator—but this virus makes me really scared, and the people who are writing that nonsense have not been seeing what I see. I feel very strongly about that.

Q: As of Friday May 1, there is an emergency use authorization for remdesivir, which seems to help in severe cases. Does that affect the equation at all?
A: Yes, I was an investigator in that study. It MIGHT make the challenge idea a more viable possibility. We need to more fully assess the several studies that are underway, but this could be an effective agent, especially if used early in disease in carefully monitored volunteers.

Q: What are the other alternatives?
A: There are absolutely other ways to go about testing these vaccines. There are immunogenicity studies which will be required in any case, and we can roll these out among health care workers who are exposed on a daily basis. They would volunteer and might give us lots of information about how effective the vaccines are.

I think you could also do things like the Ebola ring vaccinations. During the 2018 outbreak they had a plan that, when a hotspot of Ebola erupted, researchers would identify a confirmed case and then vaccinate their immediate contacts, their household contacts, their community contacts. They made a ring around that index case, and you could eventually see if the vaccine protected them. The way that COVID-19 is currently tearing across cities this might not be feasible right now, but when testing and contact tracing have advanced it might be a fit.

If I can come back to the challenge model for a minute, I don’t think anybody should make that kind of proposal until they’ve stood in the ICU and done this front-line work. Because it’s sobering and heartbreaking. We’ve had families in here where both parents are in intensive care. We have people who have come through the disease but who can’t go back home because they’re in a communal living situation. All manner of chaos is breaking out when someone comes down with this infection.

Q: Let’s talk about other potential trial designs and possible shortcuts. The Moderna vaccine candidate is being moved into human trials without first being tested in animals, which is extremely unusual. Is that a safety concern?
A: Some institutional review boards considered that. There are some vaccine platforms that we are familiar with—and I’m not certain of that particular one—that, in this public health disaster, could be acceptable to move directly into humans. We know a lot about some vaccines and immunology and adjuvants already. I could see that, and I could volunteer in such a thing, depending on some of the details.

Q: Should everyone get the same chance to be involved in a trial? I’m thinking particularly of people who have pushed back against what you do—the anti-vaxxers, the people who spread disinformation and defy public safety orders. Should they be allowed to receive a vaccine or treatment before someone who is on the side of science?
A: That’s a tough one. First off, I don’t know that researchers would have any way of ascertaining whether a person did those things. But even if we could… you know, we treat them all the same—prisoners, criminals, anyone. We try to offer what we have to everybody. It would be against our oath to say that such a person should go to the back of the line.

But how you pick people to be in a trial—that is a perennially awful question, and it has been a disturbing factor in this pandemic response. If I have 10 patients who qualify and I can only enroll one today, how do I pick that person? It’s heartbreaking to be getting emails all day long saying, “Hi Libby, here’s a great candidate.” I’m sorry, but I can’t take any more today. There are so many people who are sick, and we would like to save them all. They will not all be saved.