Human challenge trials of covid-19 vaccines still have much to teach us

We have developed approved vaccines for covid-19, but this doesn’t make carrying out challenge trials less important—quite the opposite, say Nir Eyal and colleagues

Last year the UK government set out plans to support human challenge studies of SARS-CoV-2 vaccines, starting in 2021.1 Now that vaccines are being rolled out, some have argued that this renders those challenge trials unnecessary.2 However, while it changes the role of challenge trials, we’d argue that they are no less important and ethically justified now than they were earlier in the pandemic.

In human challenge studies, volunteers are typically randomized to receive the vaccine being investigated or a control (a proven vaccine, an experimental vaccine, or a placebo). In this case, participants would then be exposed to live SARS-CoV-2. Comparisons of rates of infection (based on quantitative polymerase chain reaction) and of infectiousness (based on viral titers in the nasal cavity) between vaccinated and control participants could reveal the degree to which the vaccine being investigated blocks infections compared with other vaccines, earlier infections, or to no intervention. For their own safety, participants would all be young and healthy—a population in whom the chance of adverse outcomes or hospitalization from covid-19 is small.3 Participants should also receive priority access to any necessary care during and after the trials.

While standard phase III field trials have already proven multiple vaccines to be safe and efficacious, and many countries are already rolling these vaccines out, it remains important to test whether these vaccines block infections (a crucial role not yet elucidated)4 5 and how long that protection lasts.6 The world also needs to test whether next generation vaccines competitively prevent infection, especially for vaccines that aren’t restricted by cold chains, the need for repeat doses, limited global vaccine production, or unaffordable pricing.7 Human challenge trials can offer insights in these areas. Challenge trials could:

  1.  Discern the efficacy of vaccines in the important role of preventing infection to the vaccinated person. Field trials and observations can also help to do this,5 but challenge trials would do so far more reliably and accurately.
  2.  Discern the efficacy of vaccines in the crucial role of preventing mucosal infection, a proxy of infectiousness to others.
  3.  Measure rapidly the comparative efficacy of different vaccines in these two roles.
  4.  Measure rapidly the comparative efficacy of different regimens for proven vaccines, e.g. half-dose, spaced out, etc. in these two roles.
  5. Measure rapidly the efficacy of vaccines against new viral variants in these two roles.
  6. Discern the correlates of vaccine protection (easier to discern in challenge than field trials).8 Known correlates of protection would serve to triage new vaccine candidates and to assess efficacy in different populations.
  7.  Quickly triage for unpromising next generation vaccine candidates, limiting the number of further large field trials needed.
  8.  Discern the duration and breadth6 of vaccine immunity (by challenging participants months after they have been vaccinated). 
  9.  Discern the quality and duration of post-natural infection immunity (by challenging people previously infected with covid-19).
  10.  Help test therapeutics.6

For some of these purposes, for example, discerning a vaccine’s impact on infectiousness, field trials would be inherently less efficient. But inefficiency is not the only problem. At least one arm of a field trial would usually receive an experimental vaccine or placebo instead of the proven vaccine, for at least a part of the trial.9-11 To show statistical equivalence or even superiority in blocking infections to a vaccine previously proven efficacious (but not primarily at blocking infections), phase III field trials would also need to be even larger than the trials now concluding. Denying the most promising existing protection (the proven vaccine) to thousands of participants not only raises questions about their fair treatment, it also complicates their recruitment. Its public health footprint on their community could be substantial.9 These complications may in some conditions be surmountable,11 but not easily.

A vaccine challenge trial, on the other hand, would recruit less than 1/100th the participants of a field trial—all of whom would be young and healthy. Consequently, the risk-benefit balance for these volunteers remains fair, as we shall go on to argue. These trials would also be sure to be rapid once begun and, unlike field trials, we would know that all participants have been exposed to the virus and exactly when.  

Whereas field trials face difficulties establishing equivalence or superiority of one vaccine over another, challenge studies could assess comparative effectiveness nimbly, and even between challenges at multiple sites: so long as the same challenge virus is used at the same dosage in different sites around the world, each could assess a different vaccine, with all vaccines compared to a single small placebo group—a scientific and an ethical merit.12 By contrast, comparing field trial results to one another is virtually impossible: different housing and social interaction patterns result in different natural exposure dosages in different global sites.

Let us address some common objections to challenge trials.

Non-generalizability to high risk populations?

Many commentators have argued that challenge trials could not help us discover vaccines that serve older adults and people with comorbidities, who for their own protection are excluded from challenge trials.8 13 However, challenge trials can and probably should always be followed up with immune-bridging (and safety) studies in large high risk populations14 to determine whether the vaccine being investigated induces these same correlates of vaccine protection in older adults and people with comorbidities—if it does not, then the vaccine doesn’t work in them.8 Immune bridging based on correlates of vaccine protection is the investigational strategy currently planned for efficacy testing in populations that were excluded from the field trials, e.g. children. The main difference is that correlates of protection discerned in challenge trials would be both easier and quicker to obtain.8

No read on product safety?

Challenge trials are small, and generate almost no data on vaccine safety.13 15 However, as mentioned, so long as they are followed up by a large safety study,14 this is no problem.

Undermining trust in vaccines?

Some have speculated that challenge trials would destroy public trust in SARS-CoV-2 vaccines,13 16 17 but there is no basis for this speculation. The only empirical data we could find on the public’s opinion of challenge trials are from a deliberative workshop in the UK,18, and a survey of 5920 participants in the UK and seven other countries.19 Both suggest public enthusiasm for challenge trials.

Impossibility of informed consent?

Some authors have claimed that “The uncertainty of information about risks to [challenge] participants from both the infection and the vaccine makes adequate disclosure next to impossible in the informed consent process.”13 Yet when the risks are uncertain, adequate trial disclosure remains possible—it simply includes acknowledgement of that uncertainty.20 If uncertainty of information about the vaccine precluded adequate disclosure,13 no trial of that same vaccine could ever adequately disclose risks. Indeed, high quality informed consent is more realistic with a small cohort of dedicated participants than with tens of thousands of people.

Unfair risk-benefit balance for study participants?

But do challenge trials ask study participants to take on a fair level of risk? The British plan reportedly includes administration of therapeutics,1 16 and more efficacious therapeutics may be discovered by the time they take place, but so far, there is no proven rescue therapy for covid-19, just therapeutics of modest efficacy.8 13

Any plan for a challenge trial should ensure access to lifelong care as necessary. But the main assurance of safety comes from participant selection, which according to the recent estimates of one group of researchers would keep risk of death around 0.000031 per participant.3 That’s 10 times below the risk of perioperative mortality following live kidney donation,21 a widely accepted practice. It is true that even young and healthy adults who contract SARS-CoV-2 have a chance of morbidity, which will either resolve after months or, possibly, remain as long term disability or even a risk factor for long term mortality.13 As high as that disturbing chance of morbidity is, it is just as high for many tens of millions of people worldwide, and only increases the need for fast vaccine development for humanity. We can afford lifelong priority care for a small number of participants, not to millions of people. 

There were great uncertainties about the risks of SARS-CoV-2 vaccine field trials when they began—for example, whether they might enhance disease severity—but that didn’t mean they weren’t ethical to carry out at that time.22 In the face of mass global death, morbidity, and impoverishment due to covid-19, the ineliminable risks of carefully done challenge trials are similarly well justified. It would be overbearing to prevent altruistic adult volunteers from helping humanity find answers to so many actionable questions based on risks smaller than ones already tolerated in consensual practices with far smaller global health benefits.

Unfair risk-benefit balance for the surrounding community?

A final warning about challenge trials focuses on their risk to the community around the trial site—from secondary transmission8 13 or from undue competition with local pandemic response.13 23 However, now that vaccines are coming to sites with high community spread, the impact on surrounding communities favours challenge trials. Infection is far more likely to come from the assignment of thousands of field trial participants to control than from the viral exposure of dozens of participants in challenge trials who remain isolated until confirmed uninfected. Unlike field trials, challenge trials do not have to happen where viral spread and surge in demand for care are high. 

Multiple vaccines for SARS-CoV-2 have proven highly efficacious in preventing disease. Global health requires testing both these and further vaccines in the critical additional role of blocking infections, and in defining the correlates and duration of vaccine protection. Large randomized field trials comparing vaccines to controls have now become much harder to perform. Yet human challenge studies, such as the ones that the UK is planning, can and should play an important role in the necessary investigations we still have to carry out.

Nir Eyal is a population-level bioethicist at Rutgers University. He is the lead author of the first peer reviewed article to have called for challenge trials for testing SARS-CoV-2 vaccines. 

Arthur Caplan is a leading bioethicist based at the NYU Grossman School of Medicine, New York City. 

Stanley Plotkin played a key role in developing the rubella vaccine and authored a seminal textbook on vaccines, Plotkin’s Vaccines, now in its 7th edition. 

Conflicts of Interest: We have read and understood BMJ policy on declaration of interests and have the following interests to declare:

NE and AC serve on the advisory board for 1DaySooner, a group of intended volunteers for human challenge trials—an unpaid position. SP consults many companies manufacturing vaccines to fight covid-19. AC is an unpaid consultant for Moderna.


[1] Callaway E. Dozens to be deliberately infected with coronavirus in UK ‘human challenge’ trials. Nature 2020 doi: 10.1038/d41586-020-02821-4 

[2] Cornwall, W. New challenges emerge for planned human challenge trials. Science 2020 370(6521): 1150

[3] Manheim D, Wiecek W, Schmit V, et al. Exploring Risks of Human Challenge Trials for COVID-19. Under consideration. Preprint published at

[4] Mehrotra DV, Janes HE, Fleming TR, et al. Clinical Endpoints for Evaluating Efficacy in COVID-19 Vaccine Trials. Ann Intern Med 2020 doi:10.7326/M20-6169 

[5] Mandavilli A. Here’s Why Vaccinated People Still Need to Wear a Mask. New York Times 2020

[6] Douglas AD, Hill AVS. Immunological considerations for SARS-CoV-2 human challenge studies. Nat Rev Immunol 2020 doi:10.1038/s41577-020-00472-0 

[7] Duke Global Health Innovation Center. Mapping COVID-19 Vaccines Pre-Purchases Across the Globe 2020. Available from: 

[8] Deming ME, Michael NL, Robb M, et al. Accelerating Development of SARS-CoV-2 Vaccines — The Role for Controlled Human Infection Models. New England Journal of Medicine 2020 doi:10.1056/NEJMp2020076 

[9] Cohen J. Early approval of a COVID-19 vaccine could stymie the hunt for better ones. Science 2020

[10] Monrad JT. Ethical considerations for epidemic vaccine trials. J Med Ethics 2020:1-5. doi:10.1136/medethics-2020-1062351 

[11] Rid A, Lipsitch M, Miller FG. The Ethics of Continuing Placebo in SARS-CoV-2 Vaccine Trials. JAMA 2020 doi:10.1001/jama.2020.25053 

[12] Langenberg MCC, Dekkers OM, Roestenberg M. Are placebo controls necessary in controlled human infection trials for vaccines? Lancet Infect Dis 2020;20(4):e69-e74. doi:10.1016/s1473-3099(20)30020-7 

[13]  Kahn JP, Henry LM, Mastroianni AC, et al. Opinion: For now, it’s unethical to use human challenge studies for SARS-CoV-2 vaccine development. Proceedings of the National Academy of Sciences 2020 doi:10.1073/pnas.2021189117 

[14] Eyal N, Lipsitch M, Smith PG. Human challenge studies to accelerate coronavirus vaccine licensure. Journal of Infectious Diseases 2020;221:1752–56. doi:10.1093/infdis/jiaa152 

[15] Spinola SM, Zimet GD, Ott MA, et al. Human Challenge Studies are Unlikely to Accelerate Coronavirus Vaccine Licensure due to Ethical and Practical issues. Journal of Infectious Diseases 2020 doi:10.1093/infdis/jiaa457/5897421

[16] Kirby T. COVID-19 human challenge studies in the UK. Lancet Respir Med 2020 doi:10.1016/S2213-2600(20)30518-X 

[17] Dawson L, Earl J, Livezey J. SARS-CoV-2 Human Challenge Trials: Too Risky, Too Soon. Journal of Infectious Diseases 2020;222(3):514–16. doi:10.1093/infdis/jiaa314 

[18] Gbesemete D, Barker M, Lawrence WT, et al. Exploring the acceptability of controlled human infection with SARCoV2—a public consultation. BMC medicine 2020;18(1):1-8.

[19] Broockman D, Kalla J, Guerrero A, et al. Broad Cross-National Public Support for Accelerated COVID-19 Vaccine Trial Designs. Vaccine 2020 doi:10.31219/ 

[20] Steel R, Buchak L, Eyal N. Why continuing uncertainties are no reason to postpone challenge trials for coronavirus vaccines. Journal of Medical Ethics 2020 doi:10.1136/medethics-2020-106501

[21] Lentine KL, Lam NN, Segev DL. Risks of Living Kidney Donation: Current State of Knowledge on Outcomes Important to Donors. Evidence-Based Nephrology 2019;14:597-608. doi:10.2215/CJN.11220918

[22] Peeples L. Avoiding pitfalls in the pursuit of a COVID-19 vaccine. Proc Natl Acad Sci USA 2020

[23] Shah SK, Miller FG, Darton TC, et al. Ethics of controlled human infection to study COVID-19. Science 2020;368(6493):832-34. doi:10.1126/science.abc1076 (2020)