Bioethics, COVID-19, Health, Science, Top Stories

Human Challenge Trials—A Coronavirus Taboo

The idea is as simple as it is apparently repulsive: allow human challenge trials (HCTs) under which “low risk” and healthy young adult volunteers in double-blind studies would be given trial vaccines (or a placebo) and then intentionally exposed to the novel coronavirus. This would accelerate the assessment of the trial vaccine’s safety and efficacy and more generally expand our understanding of this virus in a controlled setting.

According to the World Health Organization (WHO), more than 70 vaccines are currently under development, five of which are already moving to clinical trials. Notwithstanding these Herculean efforts, the earliest we can realistically expect a readily available vaccine is in 12 to 18 months. This is in large part due to constraints imposed by traditional vaccine validation methods, which rely on large test groups and chance exposure to the virus by participants to assess the vaccine’s safety and efficacy. HCTs, using a relatively small low-risk group of volunteer participants, could potentially accelerate the release of a safe and effective vaccine by many months.

While conducting coronavirus HCTs raises various ethical questions, a compelling case can be made for their use based on the current data. By speeding up the release of a safe and effective vaccine, HCTs could save hundreds of thousands of lives or more, globally. Additionally, this could save trillions of dollars of unnecessary damage to the US and global economies, and spare millions (or even billions) of people untold despair and misery.

On March 15th, 2020, Dr. Anthony Fauci, the director of the US National Institute for Allergy and Infectious Diseases declared on national television that, in the fight against the novel coronavirus, “everything is on the table.” Until recently, however, absent scattered voices and academic papers, consideration of HCTs has been treated as a taboo subject by our public health officials and political leaders. To date, neither the Trump administration nor the US Food and Drug Administration (FDA) or Centers for Disease Control and Prevention (CDC) has provided any definitive public guidance or position on HCTs with regard to the coronavirus. Why is this potentially important tool such a taboo topic?

In my casual conversations with friends, family, and colleagues, the consistent reaction to the idea of HCTs has been first, an awkward silence, quickly followed by a rejection of the idea as either immoral, impractical, or both. Then—if I am able to continue the conversation long enough to walk them through the current data—a reluctant “maybe,” but usually with a lingering hint of distaste and suspicion, as if I had somehow manipulated them into an unpalatable and unwanted conclusion that violates their better judgment.

On the one hand, I find this instinctive revulsion at the idea of HCTs wholly understandable (and perhaps even laudable)—after all, even if the participants are all “volunteers,” HCTs sanction the deliberate exposure of otherwise healthy individuals to potentially dangerous (and in the case of this coronavirus, potentially deadly) experimentation. As one friend bluntly objected, “It all sounds a bit too Dr. Mengele for me.” One needn’t be a Nazi to perpetrate such atrocities. American 20th century history offers its own bioethical horrors, from the infamous 1932 Tuskegee Study to many lesser known tragedies and abuses of HCT and human experimentation in the name of science.

On the other hand, as individuals and as societies, we effectively conduct these types of human “experiments” on a daily basis (knowingly and unknowingly) by balancing risks, including the risk of death, against perceived benefits of various courses of action or inaction. When the US Environmental Protection Agency (EPA) sets air pollution and particulate standards, it explicitly calculates the number of expected annual deaths that will occur at various levels of air quality. The EPA then agrees a standard that accepts some level of expected air-pollution-related deaths based on this risk/benefit analysis.

HCTs are not a new idea; a WHO report in 2016 noted that “human challenge studies have been conducted over hundreds of years and have contributed vital scientific knowledge that has led to advances in the development of drugs and vaccines.” In past HCTs, doctors and scientists have infected volunteers with diseases that include influenzas, malaria, dengue, typhoid, and cholera. While rare today, even under established medical ethics guidelines and protocols, these human clinical experiments still can be deadly. In 2001, a healthy 24-year-old died as a result of her participation in a study at the Johns Hopkins Asthma and Allergy Center. So, caution and thorough consideration of any HCT are essential to ensure that the relevant risks/benefits are adequately understood. These must be assessed by the appropriate scientific and bioethicist review board and by the volunteer participants themselves, in keeping with the foremost Hippocratic principle of medicine: “First, do no harm.”

By its very nature, COVID-19 presents a unique opportunity to use HCTs to accelerate the development of a safe and effective vaccine. COVID-19 has many bad attributes, but one potentially useful attribute is that the infection fatality rate (IFR) for the relatively young (20 to 39 years old) and healthy population segment is comparatively low. This characteristic allows for the identification of a group of healthy young volunteers to be part of one or more HCTs. In addition to potentially significantly shortening the time required to develop and validate a safe and effective vaccine, use of HCT volunteers could also quickly expand our basic understanding of the virus, and potentially answer important questions such as: Does the amount of initial viral exposure affect the severity of the illness? Can asymptomatic carriers transmit the disease? Are individuals with COVID-19 antibodies protected from reinfection? Do certain medications provide prophylactic protection against infection? These citizen volunteers, along with all the frontline and other essential personnel, could make a historic and critical contribution to vanquishing the threat of COVID-19.

The evidence in support of adding HCTs to our coronavirus toolkit has grown in recent weeks. New data gleaned from preliminary antibody tests indicates that the spread of the virus may have been significantly more widespread than initially believed. And it looks increasingly certain that the risk of hospitalization and death from COVID-19 is relatively low (though by no means nonexistent) for healthy young adults aged between 20 and 39. At the end of April 2020, Governor Cuomo announced that antibody testing of 3,000 New York State citizens (updated as of April 27th, 2020) indicated that 14.9 percent of the general population in New York State and 24.7 percent of the NYC residents had already contracted COVID-19. New York State has a total population of roughly 20 million. Of these, approximately 5.3 million are between the ages of 20 and 39. If 14.9 percent of this group has developed COVID-19 antibodies, it would imply that 789,700 New Yorkers in the 20 to 39 age bracket have already been infected.

As of April 24th, 2020, there had been 304 reported COVID-19 deaths in this age cohort. Of these 304 deaths, 194 had one or more identified underlying health conditions (hypertension, diabetes, obesity, renal failure, cancer, etc.), leaving 114 deaths in the 20–39 age range that were not associated with an underlying comorbidity. If the antibody sampling percentages are accurate, that would mean that the morbidity/IFR for this 20 to 39 age and health cohort is roughly 0.014 percent or 1.4 deaths for every 10,000 infections. In other words, the survival rate in this cohort is 99.986 percent. Similarly, a study of excess deaths in Italy reports an IFR of 0.017 percent in those aged between 20 and 41 years old (without regard to the presence of any comorbidities). In an HCT, we would therefore expect a similar (or lower) risk of death amongst the volunteers in the same age and health cohort who actually contracted COVID-19.

While any coronavirus deaths are tragic, 1.4 deaths per 10,000 is remarkably low. For comparison, the fatality rate in the US for a kidney donor is roughly three deaths per 10,000 (more than double the estimated risk of a HCT coronavirus volunteer)—a fatality rate that both society, from an ethical cost/benefit standpoint, and each individual kidney donor, from a personal safety standpoint, willingly accept. And coronavirus HCTs are arguably supported by even more compelling justifications in terms of the total potential number of lives saved and other societal benefits. Similarly, the 0.014 percent fatality risk for coronavirus volunteers even compares favorably with the job-related fatality risk faced each year in many professions. According to US Department of Labor statistics, fishermen, loggers, air pilots, roofers, trash haulers, steel and construction workers, truck drivers, ranchers, farmers and other agricultural workers, lawn and groundskeepers, and many other professions all have a higher risk of death from their jobs each year than would a HCT coronavirus volunteer.

If we expected, say, 1,000 HCT coronavirus volunteers to contract the virus in HCTs, we would, statistically, not expect to have even one volunteer (out of the 1,000) die as result of the trial because the expected IFR would be 0.14 deaths per 1,000 (an average of one death per 7,000+ infected HCT volunteers). Obviously, to the extent scientists and doctors can design the relevant studies to have fewer than 1,000 people actually contract the virus, this risk of having even one death would become even more remote. In the absence of a vaccine, a significant percentage of HCT coronavirus volunteers (particularly those living in large, dense urban environments) would likely contract the virus over time in any case, but with potentially worse outcomes—they would become infected without the high level of supervision, monitoring, and potentially without the prompt and preferential medical treatment that would presumably be made available to them as HCT volunteers (including preferential access to any promising novel treatment therapies). And, to the extent that the severity of the reaction to the virus is dependent on the initial exposure dosage, the HCTs could, by limiting the exposure levels, limit the potential of a volunteer to become severely affected until the vaccine’s efficacy can be confirmed.

In addition, hopefully many of the trial vaccines administered will turn out to be safe and efficacious, so a large number of the 1,000 HCT volunteers would not actually become infected, leaving only a small percentage of volunteers (mostly those given a placebo) who would actually contract COVID-19 and be at risk. It may also be possible to further circumscribe the applicant pool by accepting only 20 to 30-year-olds and/or by screening for hidden comorbidities or genetic vulnerabilities in order to further reduce the risk to volunteers. Finally, by their very nature, HCTs offer the opportunity to test vaccines on smaller sample sizes. Non-HCTs require many more people to be inoculated, so HCTs would provide additional protection if it later turned out that the vaccine administered was for some reason dangerous or unsafe. Given the current data indicating that it is statistically unlikely that even one coronavirus volunteer would die as a result of the HCT, and the potential of HCTs to reduce the time required to create a safe and effective vaccine, thereby saving thousands of lives and trillions in economic carnage, HCTs should be given close consideration and should be part of an open discussion and debate among public health officials, political leaders, the scientific community, and the general public.

If the Trump administration and the public health officials at the FDA and CDC were to conclude that coronavirus HCTs are a worthwhile endeavor in the battle against the coronavirus, an important question remains: Would sufficient volunteers actually come forward to participate in such trials? We will not know the answer to that question for certain until we ask, but I suspect that properly positioned, many individuals would willingly volunteer, just as many volunteered to help out (or enlisted in the military) after 9/11 or volunteer every day to donate a kidney or engage in other acts of altruism. In fact, we have positive indications from a non-profit (1 Day Sooner) that, as of May 1st, 2020, nearly 9,000 individuals from 52 countries had already expressed a willingness to take part in coronavirus HCTs.

The chance to play an important role in conquering this disease (even at some level of personal risk) would, I suspect, be attractive to many. An opportunity for service, purpose, and even self-sacrifice, especially at a time when most of us are being told that the best—and frankly only—thing we can do is to withdraw and self-isolate at home, might resonate with the public. This willingness of individuals to volunteer for HCTs also comports with the results of numerous sociological and management studies confirming that humans seek (and highly value) having purpose and meaning in our lives—often more than financial and other material rewards. An appeal to individuals whose primary motive is to help the world defeat this virus is likely to be sufficient to secure a sufficient number of qualifying volunteers, without requiring any large financial incentives, besides costs and expenses.

If we can overcome the taboos surrounding HCTs, they can become a game changer in combating the coronavirus and limiting its ruinous effects on countless lives. A recent paper by a Harvard epidemiologist, a Rutgers bioethicist, and a London School of Hygiene & Tropical Medicine epidemiologist sets forth proposed protocols for how a coronavirus vaccine HCT might be conducted. In the US, meanwhile, some lawmakers are beginning to look at the issue. Recently, 35 bipartisan members of Congress signed a letter to the FDA and US Department of Health and Human Service in support of coronavirus HCTs. Time, however, is of the essence.


William Wiebe is a lawyer in San Francisco, who also farms grapes in Napa Valley and designs and develops property in Northern California and Bali. 


  1. Good idea. So long as they’re volunteers, what’s the issue really?

  2. using a relatively small low-risk group of volunteer participants

    This will produce results of a self-selected group that is already low risk. How will this address issues in children, elderly, sick, pregnant, etc.?

  3. Because this is a novel virus in humans, scientist believe that no individuals (young or old, healthy or sick) have natural immunity to the virus and that all of us are at risk of contracting Covid-19.

    As such, the key point here is not that the healthy young people who would participate in these trials are immune from the Coronavirus, but instead is only that they are less likely to have a sever or fatal reaction to the virus.

    Thus, if the trial vaccine is effect in preventing this group of test subjects from ever becoming infected, it should have the same protective effect on the rest of population with little or no difference based on age or other factors…, which I believe is typically the case for most vaccines.

  4. There is, of course, no good evidence for the supposition that all age groups are equally susceptible to infection by COVID-19. A lot of data even suggest otherwise.

  5. That actually doesn’t follow. The subject is, as usual, more complicated. The question is not whether you are susceptible to infection, when you are looking at who is stricken by the disease, the question is whether or not your immune system is strong enough to fight it off and keep you asymptomatic, so that you can pass it on, but not suffer the ill-effects yourself. Young people can certainly do that, but the question is are they more resistant to the virus actually getting into their systems and starting something, which is a different question.

    Actually this has some uses. One of the things that would be helpful is, if you are testing safety, to get a good idea of what the safety is. It is something that you do for all medications, so it’s an excellent idea to do it for this one.

    There is also a way to get some idea of its efficacy, because you should be able to get plasma and determine antibody titer to covid-19. It’s crude, but as long as you can measure a change and see that it does produce antibodies against it, you have some idea of whether or not it is working. Long-term efficacy studies, however, are also important, because you do need some more important and precise data on how well the antibodies raised are doing.

    In other words, one of the points of antibodies is to recognize the virus and mount a specific response without having to do too much damage to yourself with the rest of the immune response. So the question of whether or not the antibodies that you can detect in the plasma are actually doing the job they need to do may be a little tricky. You would need a longer efficacy study where they would need to be exposed to the virus. The way it’s going around, that might just include a day trip to New York and a sign saying hug me. Or you could go around law enforcement in a big city and you do the same, I have a friend who has lost multiple fellow officers to this, some of them in their 30s.

    One of the things that would also be useful is getting plasma from these people, if the antibody titer is high enough, because apparently injecting that into people functions much the way mother’s milk does, in that you are providing antibodies to something that the system cannot necessarily detect on its own. So my friend, when he finally gets his clearance, since he is still tired and weak but much better, and still tests positive, is going to see if he can donate plasma. Once he is virus free, his antibodies may help people who have severe cases of covid-19.

    So if you vaccinate people, and the vaccine is shown to be safe, there are still some useful things you can get out of these people besides their safety.

    Edit: I realized that what I said might be confusing, because the article already says expose them to the coronavirus after they’re vaccinated. Thing is, I don’t think it’s necessarily helpful to expose them to a lab strain, necessarily. In other words, you want an area where there are multiple mutant strains wandering around, because that is actually the ultimate test. In other words, if it can protect against the single strain or a couple of strains that you’re testing against them, that’s good, but what you want is something that can provide resistance against the virus as it changes, which might be well served by a trip to New York City.

  6. Good question, and it would depend on the data and the number of people in the trial. One of the things that can be very frustrating for people in a clinical trial is when they see something that makes a difference, and then they still have to wait for another clinical trial phase and then for the FDA to approve it before they can get the treatment and start taking it again.

    Usually what you do in these cases is a small study for safety, not a huge number of patients, but what you’re looking for is huge reactions, and to check your dosages. In other words, you want to know what the safe dosage is to give someone. You also want to know if it causes a massive amount of reaction. So, for example, if people start swelling up, then it’s possible that you have put in something that’s causing an allergic reaction or what not. Usually you want to monitor people for a while and it can take some months to collect that data and then process it. I would expect that to be significantly expedited because you should be able to get a lot more people in the study and have the FDA and the CDC actively involved in what you were doing, which means they are going to be much quicker to approve it because hopefully they are looking at the data at the same time as you are.

    Normally it can take years, but I’m hoping that another 6 months might be enough to get one out. If we can get enough people in the trial, we could probably cut it down to two or three months, simply because we have enough people to get good data and don’t need to do what a lot of studies have to do, which is have a steady trickle of people starting and ending the study over a long period of time in order to make up the numbers.

    So as far as giving you a number, I couldn’t. I don’t know enough about what the researchers in question are doing specifically, nor how the CDC and FDA are going to be working with them, in a trial like this, though I would hope to get massive participation. The nice part is that at the very least, it would be a small economic Boon, because you are paid an honorarium for your efforts. In this case, I would hope for a slightly larger honorarium, but let me tell you that when I was a grad student, those fifty bucks came in handy.

  7. Lots still to learn about the virus, but according to the W.H.O. Children and young people ”are just as likely to become infected as any other age group”. Maybe further research will prove otherwise but this is what W.H.O. is saying on their website today…

    “Research indicates that children and adolescents are just as likely to become infected as any other age group and can spread the disease.

    Evidence to date suggests that children and young adults are less likely to get severe disease, but severe cases can still happen in these age groups.”

  8. Well, this is entirely possible. One of the problems a virus has to overcome is getting into cells. This is one of the weird things about things like swine flu and avian flu, because they would bind to proteins on intestinal cells in pigs and birds, and these proteins, which were on the outside of the cell, would then be transported in by the various cellular processes that they were involved in, and the virus could then escape and go on a rampage. Now when they got into humans, the way they could get into cells by binding to proteins that were very similar to the ones in the intestines of pigs and birds, which were actually in human lungs. So what was an intestinal flu in them, and would be transported and transmitted by poop, was a lung flu in us. The fact that the symptoms of the flu in them versus in us were very different is not due to the virus at all. It is due to how our immune system was reacting and responding to the thing, and the different ways that we had to try to get rid of the virus based on where it was showing up.

    Anyway, I was making a point here, I know I was. Oh yes, right. So you have a virus which needs to bind to a protein, in this case it is able to bind to a protein in your lungs. What happens if you have a mutated version of that protein that lacks the necessary binding site? Maybe the virus can bind to something else oh, but it’s not as good, usually. So that virus is now going to have a much harder time getting into your cells. That would make it harder for it to replicate and spread through your body at all.

    So that is a way to win the battle before it starts.

    There are also some other ways. One of the things that a virus has to do is infiltrate a cell, hijack the cellular Machinery to produce more of itself, and then keep doing so until the cell bursts and lets out a huge blast of virus to go do it again. The more of the virus multiplies, the harder it is going to be for your body to contain and deal with it. So suppose that your body’s NK (natural killer) cells have an easier time spotting the infected cells, possibly because there are more of them due to a healthier immune system. They will be able to put a full-court press on the infection, and will have more immune cells able to spot the virus and figure out how to resist it. They can keep the virus contained, infecting relatively few cells, and possibly even beat it, while the adaptive immune system is working on how to make specific antibodies to it to target it for specific attack.

    Unfortunately, these people may be asymptomatic, but they are still able to spread the virus because it is slowly multiplying in their lungs, and they can spread it by coughing, sneezing, loud speech, etc.

    This is also true of seasonal flu, btw. It is why I have been recommending vaccination, as you may have noticed. The adaptive immune system is capable of vastly quicker response once you are vaccinated, and so it can cut out the infectious phase much more quickly.

  9. Healthy individuals routinely donate a kidney to help a sick patient. It is not risk free for the donor and the donors are usually viewed as heroic. I don’t see a problem with volunteering to be a vaccine guinea pig to benefit your fellow citizens.

  10. I see it more as a dress rehearsal for eliminating democracy in any meaningful sense, a la dear leader Justin’s recent gun ban. You will learn to respect their authoritah.
  11. Great post. You explain a lot very clearly. A side note that people might find interesting is that there are some people who are totally immune to HIV. In most cases, the HI virus will attack to a protein receptor on a white blood cell. The specific receptor is CCR5, but some people don’t have this receptor or it is so small that the HI virus just can’t attach to it. You also have people without that mutation who have partial immunity to the virus in that they only contract it after multiple exposures.

  12. I know, I just didn’t want to go into it in here. People have complained about walls of text, for some reason.

    Edit: Aaaaand I just found out that my testes have taste receptors. God I love biology.

  13. There’s some cheerful ideas Trevor. Naturally the world’s militaries will already be exploring this, but as you say one looks forward to the day when every incel can experiment with new viruses in his parent’s basement.

    I also noticed that Justin did this by proclamation. But not many seem to have noticed. Ditto all the corona – the authorities now have the right to control you in any way they choose at any time by proclamation. Very like Canadians not to notice or care.

    Good to know!

  14. These are very strange things to say about @Tj2mag , a biologist with first-hand virology experience/knowledge who painstakingly explains the complexities for the larger audience here at QC. As a teacher, he has a gift, and he’s shown tremendous patience in trying to explain the complexities to those (like you) with little to no expert experience/knowledge.

    Your comments quoted reflect your priorities rather than anything real about @Tj2mag ’s contributions. imo.

  15. Thanks for this article. I’m sure that HCTs could make sense and could make sense here. However I’d like to have seen this considered in the much wider context of more adaptive responses to the problems we face on COVID19 generally. Regarding testing for a vaccine, surely a lot of these trials could be done on medicos treating the virus with the control group infection rate being compared with the treatment group. But more generally, there should be lots of experimentation going on – in reducing infection and treating it – with some serious attention being given to how to optimise others capacity to access the resulting ‘knowledge commons’ via online platforms and so on.

Continue the discussion in Quillette Circle

42 more replies


Comments have moved to our forum