By Jonathan Pugh, Dominic Wilkinson and Julian Savulescu.
A group of citizen scientists has launched a non-profit, non-commercial organisation named ‘RaDVaC’, which aims to rapidly develop, produce, and self-administer an intranasally delivered COVID-19 vaccine. As an open source project, a white paper detailing RaDVaC’s vaccine rationale, design, materials, protocols, and testing is freely available online. This information can be used by others to manufacture and self-administer their own vaccines, using commercially available materials and equipment.
Self-experimentation in science is not new; indeed, the initial development of some vaccines depended on self-experimentation. Historically, self-experimentation has led to valuable discoveries. Barry Marshall famously shared the Nobel Prize in 2005 for his work on the role of the bacterium Helicobacter pylori, and its role in gastritis –this research involved a self-experiment in 1984 that involved Marshall drinking a prepared mixture containing the bacteria, causing him to develop acute gastritis. This research, which shocked his colleagues at the time, eventually led to a fundamental change in the understanding of gastric ulcers, and they are now routinely treated with antibiotics. Today, self-experimentation is having something of a renaissance in the so-called bio-hacking community. But is self-experimentation to develop and test vaccinations ethical in the present pandemic? In this post we outline two arguments that might be invoked to defend such self-experimentation, and suggest that they are each subject to significant limitations.
The Liberal Argument
Self-experimentation in science can be supported by liberal arguments concerning the value of individual freedom. John Stuart Mill’s ‘Harm Principle’ states that “The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others”. If self-experimentation only poses risks of harm to the individual herself, then by the lights of the harm principle, it would not be justifiable to prohibit such experimentation. To do so would be unduly paternalistic, regardless of whether the expected harms of self-experimentation outweigh its expected benefits.
This liberal argument in favour of self-experimentation relies on the premise that self-experimentation does not pose a risk of harm to others. However, in the case of vaccine self-experimentation in a global pandemic, it is not clear that this condition is met. First, such self-experimentation may provide ‘false reassurance’. This will occur if individuals self-administer an ineffective vaccine that they incorrectly believe will prevent them from transmitting the virus to others. Such false reassurance may mean that these individuals engage in riskier transmission behaviours, or fail to take more effective vaccines, than they would have if they had not self-administered an ineffective vaccine.
Second, widespread confidence in vaccination is crucial to ensuring sufficient uptake. The failure of unapproved open source vaccines could potentially undermine confidence in vaccines that have been subject to regulatory approval, especially if they led to unexpected serious adverse events. Third, health needs generated by any such adverse events would impose additional burdens on health services that are already stretched to capacity in the pandemic. The risks of self-administered vaccination are enhanced amongst untrained members of the general public.
Cumulatively, it might be argued that these risks of harm to others are enough to prevent a Millian justification of vaccination self-experimentation. Yet, even if this were not so, it should be noted that the liberal argument in favour of self-experimentation is not decisive in other forms of medical research. A central tenet of research ethics is that clinical trials may only expose participants to ‘reasonable risks’. Of course, there is considerable debate about where the threshold of ‘reasonable risk’ is located, and whether the imposition of reasonable risk thresholds to voluntary research participation is justifiable. The important point here though is that there are limits to the degree of risk that we allow research participants to willingly be exposed to in research. Indeed, this is part of the reason why some have objected to the use of challenge trials in the pandemic. If self-experimentation on vaccines is to abide by conventional tenets of research ethics, then the fact that it is undertaken voluntarily and does not violate the Harm Principle is not sufficient to establish its moral permissibility. Amongst other things, such experimentation would also have to pose only a level of risk to the experimenter that is ‘reasonable’; we return to this issue below.
The liberal argument champions individual autonomy over paternalistic interference. However, a final challenge for the liberal argument is that it is not clear that all participants will make autonomous decisions to take on the risks of self-experimentation. The RaDVaC White Paper is quite clearly prefaced with informed consent information stating that individuals using the information in the paper must agree that they are consenting adults who take responsibility for using the information. Standard clinical trials in human subjects involve far more rigorous methods of obtaining informed consent, and even these methods do not ensure that participants will make an autonomous decision to participate in research. Of course, some self-experimenters can clearly give highly-informed consent when they are intimately involved in developing the rationale and design of the experiment; but this is not true of all those who may use this open source information.
The Beneficence-Based Argument
Accordingly, there are limits to the liberal argument in favour of self-experimentation. However, advocates of self-experimentation need not just appeal to this liberal argument and the value of freedom. Advocates of self-experimentation might also argue that such experimentation will benefit them. Accordingly, a further beneficence-based argument in favour of RadVaC’s open source approach to vaccine development and testing is that it gives individuals the opportunity to use an intervention that may reduce their risk of disease, before other approved interventions become available. RaDVaC published the first version of their white paper in July 2020, months before any other vaccine was approved. It was able to do so in part because the development and testing of the vaccine bypassed the various kinds of regulatory approval that must be obtained prior to initiating a traditional clinical trial.
Moreover, RadVaC claim to have developed an intranasal vaccine, in contrast to the intravenous vaccines that have so far been approved for use against COVID-19. Few other intranasal vaccines are currently under development for this purpose. Of course, the fact that the instructions for making the vaccine are freely available online means that access to the RaDVaC vaccine will also not be subject to the same limitations as those that are subject to patents in commercial vaccine production. Those who object to the use of patents in the context of vaccines may thus find something to admire in the open source RaDVaC model. Finally, it should be noted that the RaDVaC vaccine is intended to function as a supplement to other approved vaccines as they become available, not a replacement.
Of course, this beneficence-based argument in favour of the open source approach assumes that the potential benefits of the experimental intervention outweigh its potential harms. In defence of this assumption, RaDVaC point out that intranasal vaccines of the sort that the organisation has developed have an excellent safety record in animal studies and clinical trial. However, although no serious adverse effects have yet been observed amongst the self-administrating participants, the absence of sufficient data means that safety cannot be guaranteed. The same is true of efficacy. Although the published White paper details methods for testing and the rationale for the RaDVaC vaccine design, it does not include any efficacy data. It also does not offer detailed plans for systematically collecting such data. Accordingly, this moral argument in favour of the open source approach must be tempered by the significant uncertainty surrounding the safety and efficacy of the vaccine. As such, the beneficence argument may ultimately collapse into a permutation of the liberal argument; perhaps we should allow people the freedom to make their own choices about what risks are reasonable for them to take in this kind of research, given the limited information that is available.
Going Forward – Regulating Citizen Science
Regulators have taken an increasing interest in self-experimentation in the context of bio-hacking given the potential risks it may poses to others. Similarly, as Christi Guerrini and colleagues have argued, there are moral reasons to consider regulating self-experimentation for open source vaccines.
Of course, individual self-experimentation is extremely difficult to regulate, and many self-experimenters labour under the illusion that they are entirely exempt from regulation. However, as Guerrini et al point out, this belief is legally and factually incorrect in the US; the FDA is authorized to regulate any intermediate vaccine components, such as reagents, that cross straight lines. These authors suggest a number of ways in which regulatory authorities could support citizen science whilst subjecting self-experimentation and open source science to an appropriate degree of governance. These recommendations include encouraging citizen scientist organisations to seek review by an independent Institutional Review Board (IRB).
The success of such regulatory efforts will depend on developing a relationship of trust between citizen scientists and regulators. In order to gain the trust of citizen scientists it is crucial that regulators clearly communicate why the important moral justifications underlying the approval processes that slow down scientific research are necessary. However, it is equally important that the approval processes are themselves clearly morally justified for the research in question. If IRB approval is to be championed for citizen science, IRBs must also attend to the important differences between the aims of citizen science and more traditional forms of clinical research. One key question here is whether the reasonable risk thresholds invoked in the ethics of standard clinical research should also be invoked to determine the permissibility of self-experimentation when we approach the question of regulation.
Acknowledgements: This is an output of the UKRI Pandemic Ethics Accelerator project.
Authors: Jonathan Pugh, Dominic Wilkinson and Julian Savulescu.
Affiliations: Oxford Uehiro Centre for Practical Ethics, University of Oxford
Competing interests: None declared
Social media accounts of post authors: @JonnyPugh3 @NeonatalEthics @juliansavulescu