Carol Levine and Jeremy Sugarman
Five of the six British men who suffered devastating illnesses after receiving an experimental compound - TGN1412 - in the first phase of human testing are now reported to be out of hospital, and the sixth is apparently out of intensive care. Although it is unclear whether they will suffer long-term health effects from this catastrophic experience, for now the news is mostly good.
The bad news is that this trial apparently went forward on a "business-as-usual" approach, and something like it could happen again. The basic problem is neither that the subjects were paid, nor that the research was not conducted at an academic medical institution. Paying research subjects for their time and inconvenience is accepted practice, although very high payments raise serious questions about "undue inducements" to participate. Furthermore, the three research subject deaths that caused so much concern in the United States in recent years occurred in trials conducted at well-respected, large academic centers - the University of Rochester, University of Pennsylvania, and Johns Hopkins University.
The basic problem in the TGN1412 study, as we see it, is that those involved ignored critical aspects of the research that should have warranted what we have called "special scrutiny." With our colleagues in the Consortium to Examine Clinical Research Ethics (CECRE), in a 2004 article in the Annals of Internal Medicine, we pointed out that traditional methods to protect research participants - such as informed consent and special protections for "vulnerable groups" - are necessary but have never been sufficient to protect participants' welfare.
In proposing a focused, special kind of scrutiny for research that raises serious moral issues, particularly in the new globalized, privatized research environment, we challenged institutional review boards, investigators, and sponsors to recognize and respond to the moral complexities of a protocol before it is implemented, not after it is criticized in the scientific or lay press - nor, we might have added, after subjects have already been harmed.
These are our criteria for special scrutiny:
Criterion 1. "The research involves initial experiences of translating new scientific advances to studies in humans, especially when the intervention is novel, irreversible, or both."
Criterion 2. "Without potential for offsetting direct medical benefit, there is a known or credible risk for significant harm (death or serious disability are the clearest examples) to humans as a consequence of the experimental intervention."
Criterion 3. "The protocol raises ethical questions about research design or implementation for which there is no consensus or there are conflicting or ambiguous guidelines."
Even based solely on information made publicly available from newspaper and journal reports, the TGN1412 trial appears to have met all three criteria for special scrutiny. Had it been subjected to that process, it might have never gone forward as it was designed and implemented. Some reports assert that it was initially rejected in Germany.
Here's what is publicly known: The TGN1412 study was the first test in humans of a new compound developed by a German company called TeGenero and manufactured by another German company, Boehringer Ingelheim. According to TeGenero's Web site, its "most advanced product candidate is TGN1412, a fully humanized CD28-SuperMAB®." This monoclonal antibody is designed to mobilize one part of the immune system specifically to attack another part. "TGN1412 is a promising novel approach," the company says, "to addressing the medical need in chronic autoimmune/inflammatory diseases, which require long-term therapy without severe side effects." In other words, if this compound works, it could have a huge market in treating devastating diseases such as rheumatoid arthritis and leukemia.
TeGenero did not conduct the clinical trial itself. That was a job for Parexel International, a Massachusetts-based major contract research organization (CRO), one of the intermediaries that now manage a great deal of clinical research in this country and abroad. Their responsibilities included obtaining ethical review of the protocol, which in this case was conducted by the Brent Medical Ethics Committee, a North London group that reviews protocols for National Health Service-sponsored trials in the Borough of Brent. It is authorized to review both Phase I (safety and toxicity trials) and Phase II trials (dosage and preliminary efficacy trials).
For routine protocols, which make up the bulk of clinical research, this kind of international, private-public collaboration is not particularly unusual and perhaps not worrisome. But someone at some point has to recognize that a particular protocol is not routine - indeed is highly unusual - and call for a more stringent and focused review.
For example, in this case the animal evidence required very careful review, since, according to one report, monkeys developed swelling of the immune tissue in their necks. TeGenero asserts that "TGN1412 had been tested extensively in laboratories and on rabbits and monkeys, with no adverse effects and no drug-related deaths." But in another statement it acknowledged that one animal (it is not clear whether it was a rabbit or monkey) "sadly had to be put down during the trial after suffering diarrhea caused by an unrelated bacterial infection."
Since the trial subjects were healthy - which is standard in phase I toxicity trials - there were no offsetting direct benefits to them; therefore, particular attention should have been paid to the potential risks of a compound designed to effect a major alteration of the human immune system. There was no reason - other than efficiency - to give the drug to all six subjects at approximately the same time, and in fact this approach is very unusual in first-in-human studies. By all available accounts, the first person to get the drug started having major symptoms (pain and swelling) almost immediately; yet there was apparently no delay in giving it to the others.
In its investigation the British Medicines and Healthcare Products Regulatory Agency (MHRA) found that there had been no deviation from the protocol or contamination of the drug, so the sequence of administration must have been in the protocol itself - a serious design flaw. Further, it remains unclear why two subjects were given placebo (the lucky ones, as it turned out), since the trial was not meant to determine efficacy. Nor could placebos be justified on the grounds that the subjects would likely have a confounding underlying condition that would make assessing safety difficult in those who received the experimental compound.
Finally, according to two ethicists who reviewed the 13-page consent form, the language was complex. The consent form did not make clear that this was a "first-in-humans" trial, and minimized the risks of altering the immune system. Reportedly the form stated that "No significant side effects had been seen in the animal studies," suggesting, perhaps, that there had been some side effects but that the company did not consider them "significant." There was also no information about who participants could contact at the ethics review committee should they have any questions during the trial.
It is easy to point to the profit motive as the source of the trouble, and this may have influenced the unusual study design. However, it is critical to acknowledge the important ways in which the research landscape has changed in the past few decades. There are now so many organizations and individuals involved that the connection between the participant and the researchers may be remote. Among the details missing from the accounts so far is the name of the principal investigator - the person who is ultimately responsible for the conduct of the trial. The only statements have come from corporate officials. No sponsor, CRO, or even ethics committee can take the place of an individual researcher who values participant safety and well-being above product development.
Even when the principal investigator is conscientious, however, others still have important responsibilities, which sometimes include asking hard questions that may delay or even stop a trial. So be it: the consequences for individual subjects and the research enterprise as a whole are momentous. Special scrutiny, as we have envisioned it, suggests the circumstances when heightened attention is warranted, which is not the case for all research. In fact, in assuming a "one size fits all" approach, the current oversight system sends inappropriate messages regarding its critical role. Clearly there is room for enhancing efficiency in the ethical review of some research, but first-in-human studies, such as that involving TGN1412, are not the place to start.