A Dutch infectious disease researcher refuses to say how she got two students at her university to consent to experimental infection with delayed treatment.
The study obtained consent for exposure to a pathogen with a 12-week delay in curative treatment. Details that are withheld are relevant to human subject concerns in current COVID-19 vaccination research.
The article published in Annals of Internal Medicine sparked a brief, but intense expression of outrage on social media from infectious disease experts, clinical trialists, and ethicists.
Little sign of this controversy is preserved in the biomedical literature. What was said by experts on social media about the ethics of studying infectious diseases in this unusual study should be available in a systematic search of the literature. Especially with the urgency of determining the efficacy and safety of vaccines for COVID-19 with healthy research participants.
Sensitive to the controversy this article had elicited, the official Twitter account of the journal tweeted
Apparently, there were few on Twitter who took advantage of this offer who succeeded in getting a letter published.
What was said by experts on social media about the ethics of studying infectious diseases in this unusual study should be available in a systematic search of the literature. Especially with the urgency of determining the efficacy and safety of vaccines for COVID-19 with healthy research participants.
Our published letter to the editor has largely gone unnoticed because of an exceptionally stubborn journal paywall, combined with delays and glitches in the indexing of letters to the editor. I could not find this letter in recent searches of the journal website or through Google Scholar.
My best friend and colleague University of Pennsylvania Professor of Medical Ethics & Health Jon Merz, MBA, JD, Ph.D., had made polite, persistent, but ultimately futile attempts to get the details of what was told to the two students and what assurances were given in securing their informed consent.
Jon wrote emails to the first author, senior author, and Research Evaluation Board of the Leiden University Medical Centre, Leiden, The Netherlands.
After some exchange of niceties, the response seemed to be an emphatic “Echt, niet!”
The article’s senior author noticed that Jon was an ethicist and presumed that Jon might want to publish on this issue. She reserved the right to protect her team to be the first to publish what was done to ensure informed consent. The team was collaborating with ethicists at Leiden University and she suggested that they would be publishing details.
The senior author also noted that her research staff had not given serious consideration to self-experimentation involving deliberately infecting themselves.
She ended further correspondence with:
Given the ethical debate around controlled human infection studies, we believe that it is important to take the utmost care in informed consent procedures, guidance of our volunteers and the public opinion.
Subsequent publications from this group have not revealed what was told to these students, not even in advisement about how to manage risks to normal volunteers in COVID-19 vaccine research.
Actually, a strong case can be made for cohort-specific informed consent, which involves disclosing specific risks to prospective subjects with particular personal characteristics.
The Head of the REB replied that she preferred that Jon obtain information and consent forms from the senior author of the study. She did not respond to further inquiries.
The journal did not require us to sign over the copyright of the letter, so here it is in its entirety:
We read with interest the short report by Roestenberg and colleagues about the experimental infection of 2 graduate students with male cercariae of Schistosoma mansoni.1 The overarching hypothesis of this study is quite provocative. Results would be potentially clinically relevant, except data from only 2 of the originally proposed 17 subjects proposed in the trial registration were reported.2 Information in the trial registration indicates that recruitment is no longer occurring.
No reason is given for writing up the paper with such a small subsample of what was proposed. One is left wondering whether recruitment failed because of the risks and burdens of a study requiring exposure to a pathogen and a 12-week delay in curative treatment. No information is provided about recruitment strategies, target subject population, or information disclosed in the consent form. Was an incentive offered and was it consistent with the risks and burdens of the study? We requested the consent form from the senior author and the IRB, but our request was refused.
Despite recruiting only two subjects, there were six authors and a larger coHSI clinical trials team of 20. We wonder why no members of the team experimented on themselves. The Nuremberg Code provides some guidance for dangerous experiments that hold out no prospect of therapeutic benefit: “5. No experiment should be conducted where there is an a priori reason to believe that death or disabling injury could occur; except, perhaps, in those experiments where the experimental physicians also serve as subjects.”3 While death or long-term injury from this experiment was unlikely, the principle behind this provision of the Code has a persuasive rationale.
If an experiment is too dangerous or imposes too high a risk and burden, only those who are most informed and free of undue pressures and coercion should be experimental subjects — namely, the investigators. No researcher (or IRB member) should ask others to do anything that they themselves would not do. If recruitment failure was due to the inability to find enough unaligned people willing to undertake such an experiment, we find it ironic and telling that none of the investigators would step into the breach for the good of their science.
We hope the investigators will be open and transparent about their recruitment and consent methods, as others may face similar barriers in trying to replicate and extend their important work.
The authors’ reply
We thank the authors for their interest in our research and sharing their ethical deliberations. Recruitment for this trial has been completed as proposed (n=17) and final trial visits for the last group of volunteers are currently taking place as planned. We aim to provide a full report of the trial results as soon as possible and anticipate final data collection in April 2019. Volunteers were financially compensated for their time and travel expenses according to accepted Dutch standards for medical research with healthy volunteers.
We have performed a thorough review of our informed consent procedures, volunteer expectations, risk assessments and motivation. The results of this study will also be made publicly available. Given the importance of the clinical findings of the first participants in this study to the medical community, we have been specifically requested to publish the clinical details of these initial cases as soon as possible and not await the end of the trial. We apologize for any concerns we might have caused with regards to failed recruitment or early termination and will keep you updated on trial publications.
Note that this reply seems to promise that the results of the larger study will be published. It did not say whether these publications would include the details of the protection of human subjects, including this study’s procedure for ensuring informed consent.
It is very odd to publish results obtained from the first two research participants when a larger study is in progress. This article could be seen as premature and inappropriate because any effects may prove false positives with the accumulation of more data. The authors want to attract attention to their preliminary results, but risk the likelihood of promising results getting more attention than the not-so-promising, but more valid results they will have later.
Responsible researchers do not publish preliminary results from an ongoing study without a very good reason. That is Best Research Practices 101.
Particularly since this group is taking the extraordinary step of publishing results obtained from the first two of a planned larger study, the medical and scientific community have the right to know how consent was obtained and if there were any participants who declined participation once they were fully informed.
If the results of this study are so important that they must be shared at this preliminary stage — after collecting data from only two subjects — then so are the procedures by which the results were produced, including informed consent.
There are unaddressed ethical concerns about this study. What were prospective participants told about possible adverse events, including the risk of death? How much were participants compensated in recognition of this risk? Are there special issues in these participants being students at the research team’s university and their perhaps feeling pressured to give consent?
Our suggestion that the investigators consider doing this research experimentally expose themselves — not someone else — to the pathogen has strong precedent.
A systematic review identified 465 documented instances of medical self-experimentation, including by researchers who went on to receive the Nobel Prize.
Walter Reed’s Yellow Fever Commission in Cuba in 1900 allowed mosquitoes suspected of carrying the deadly disease to bite Commission members, but Reed excused himself from having to do so. One member of the commission developed yellow fever and suffered lifelong chronic ill health. Another member died from the disease.
A classic paper in the New England Journal of Medicinethat is regularly assigned reading in medical ethics classes discussesAuto-Experimentation — An Unappreciated Tradition in Medical Science.
The paper notes
Obermeier, in his efforts to find a cure for cholera that then was epidemic in Berlin, allegedly injected blood from a patient with cholera into himself, and died.
Swiss chemist Albert Hofmann accidentally poisoned himself with LSD and then began purposely auto-experimenting with lower dosages to discover its psychedelic effects.
The NEJM article poses an important ethical question for the Leiden research team:
Investigators cannot serve as subjects in all scientific experiments. But the widespread practice of auto-experimentation raises a fundamental question about the philosophy of research: Is it ethical to subject another person to an experiment if the researcher did not do the experiment on himself first?
What are Katayama syndrome and schistosomiasis?
According to Lancet Infectious Disease, Katayama syndrome
Is an early clinical manifestation of schistosomiasis that occurs several weeks post-infection with Schistosoma spp (trematode) worms…Katayama syndrome appears between 14–84 days after non-immune individuals are exposed to first schistosome infection or heavy reinfection. Disease onset appears to be related to migrating schistosomula and egg deposition with individuals typically presenting with nocturnal fever, cough, myalgia, headache, and abdominal tenderness.
According to the World Health Organization (WHO), schistosomiasis:
Is an acute and chronic parasitic disease caused by blood flukes (trematode worms) of the genus Schistosoma. Estimates show that at least 229 million people required preventive treatment in 2018. Preventive treatment, which should be repeated over a number of years, will reduce and prevent morbidity….
Transmission occurs when people suffering from schistosomiasis contaminate freshwater sources with their excreta containing parasite eggs, which hatch in water…
Chronic schistosomiasis may affect people’s ability to work and in some cases can result in death. The number of deaths due to schistosomiasis is difficult to estimate because of hidden pathologies such as liver and kidney failure, bladder cancer and ectopic pregnancies due to female genital schistosomiasis.
Are psychology graduate and medical students members of vulnerable populations?
Regulations vary by country, but in the U.S., federal rules require that institutions overseeing research conducted with human subjects take special precautions to ensure that consent is obtained is fully informed, voluntary, and without coercion. The concern is heightened when individuals are in situations in which they can easily be manipulated, or they may be a convenient and readily available study population.
U.S. regulations especially cite children, economically deprived persons, and prisoners as vulnerable populations. Students and employees of an institution are not explicitly mentioned.
Yet, researchers can expect that in the review of applications to experiment with students, consideration will be given to how consent is obtained without coercion and how students and employees can decline participation or withdraw from research without fear of reprisals.
Special issues with Leiden University psychology graduate students
The Dutch academic environment is highly competitive. Career advancement is dependent on the demonstration of research productivity, as quantified by the number of authorships on peer-reviewed publications.
Dutch Ph.D. theses consist of a collection of peer-reviewed publications with interspersed commentary. Co-authorships on research conducted by graduate students figure heavily in the hiring, promotion, and tenure of junior and senior faculty.
The publications listed on the curriculum vitae of many department and laboratory heads in the Netherlands consist mostly and sometimes almost entirely of authors stemming from Ph.D. student research.
Dutch psychology students face particular challenges in being able to recruit patients for Ph.D. research from medical hospitals and clinics staffed and controlled by other disciplines. Most of us who have done research in those setting find that authorships can be expected for anyone willing to authorize access to medical patients. These academic administrators are responsible for securing publications for their own colleagues, staff, and Ph.D. students and can be expected to look after them.
I have had to acquire a new data set at another Dutch medical center for a completed manuscript reporting results based on secondary psychometric analyses of self-report measures.
The head of the clinic had given permission for use to some data from a study completed long ago in the clinic, before her time. But she did not respond to repeated emails offering authorship for a minimal of effort sufficient to comply with journal guidelines for determining authorships.
We asked at the last minute before the manuscript was uploaded for her to provide permission to name her in an acknowledgment. The clinic head insisted that she be able to recruit her two new postdocs to read and make comments on the paper. The three of them should then become authors or else the manuscript could not be submitted. We got a new data set elsewhere to reanalyze. As customary, we again offered an opportunity to the source to become an author, but we received a note thanking us but declining.
Decision-making about authorships in Dutch medical settings often involves some compromise between international ethical standards for awarding authorships and the need to maintain collegial relationships with those in control.
In recent years, Leiden University has had a number of scandals about research integrity, gift authorships, and denial of credit for student and early career persons’ contribution to research in the Schools of Education and Psychology.
This past year, the Netherlands Board on Research Integrity (LOWI) concluded that a former professor of psychology at Leiden University was guilty of breaching several rules of scientific integrity. This conclusion was based on the findings of the Academic Integrity Committee (CWI) of Leiden University:
The CWI identified four violations of scientific integrity: (1) blood sample testing without the legally required permission of the Medical Ethics Review Committee, (2) irregularities with publications, (3) repeatedly selectively omitting research results without reporting or explaining this, and (4) submitting grant applications with incorrect (incomplete and manipulated) research data.
Transparency in research ethics during the pandemic
Achieving control over the world COVID-19 pandemic depends on unprecedented international cooperation and the sharing of data obtained from diverse settings and the ethics and informed consent with which these data were obtained.
These were always Best Research Practices and now, more than ever is the time to enforce them.
Academic institutions are expected to share data as well as basic details of their methodology to allow their research to be replicated. Especially in the midst of a pandemic, researchers should be forthcoming and disclose fully the details of how they informed health volunteers about the risks of unprotected exposures to pathogens in their research and how truly voluntary consent was obtained.
We did not anticipate that pandemic would soon be upon us when our letter was published in Annals of Internal Medicine. But we saw no reason why these authors from LUMC should not answer a reasonable question about how they protected human subjects.
Dear Professor Meta Roestenberg of Leiden University Medical Centre,
Could you please inform the scientific community about how you persuaded two students to put their health at risk in your study by getting exposed to pathogens?
- 1.Langenberg MCC, Hoogerwerf MA, Janse JJ, van Lieshout L, Corstjens PLAM, et al. Katayama Syndrome without Schistosoma mansoni eggs. Ann Intern Med 2019; Jan. 8. doi:10.7326/L18–0438.
- 2. Roestenberg M. Single-sex controlled human Schistosomiasis infection: safety and dose-finding. https://clinicaltrials.gov/ct2/show/NCT02755324 (accessed January 20, 2019).
- 3. Nuremberg Code. United States v. Karl Brandt et al, 6 Fed Rules Decisions 305 (1949).