UK clinical trial disaster highlights how much the industry still doesn’t know

Six young men became deathly ill

The details from a phase I clinical trial in London, England, sound more like a clinical trial going wrong after unapproved deviations from the protocol than the worst case scenario after a trial has been conducted, apparently, according to industry standards.

Six young and healthy male volunteers developed serious symptoms, including inflammation, vomiting, severe pain, swollen heads, and unconsciousness almost immediately after being injected with TGN1412, a humanized agonistic anti-CD28 monoclonal antibody, created by TeGenero AG of Wurzburg, Germany.

The men were administered the agent in the clinical trial unit at Northwick Park Hospital in London. The phase I clinical trial was conducted by Parexel International of Boston.

After the men displayed symptoms of a serious adverse event (SAE), they were transferred to the hospital’s critical care unit. A week after the incident, four of the six men had responded well to treatment, and they were making a steady recovery, but two of the men remained in critical condition and were continuing to receive organ support, according to Northwick Park Hospital’s media update. (See early timeline and facts of case.)

"This event is a real tragedy in England, and we certainly hope the six people who were harmed make full recovery," says Greg Koski, MD, a senior scientist with the Institute for Health Policy, Massachusetts General Hospital, Partners HealthCare System in Boston. Koski also is an associate professor of anesthesia at Harvard Medical School, and he is the former director of OHRP.

"This is much like a plane crash," Koski says. "Even though we have a system in place for trying to minimize risks and avoid these situations and even if you have an effective air transportation system, it doesn’t mean you’ll never have a plane crash."

"It seems rather bizarre in this day and age of testing to have that many immediate and severe reactions," says Marc P. Cardinalli, JD, an assistant general counsel and administrative code officer at the University of Nevada, Las Vegas.

As news of the disaster crossed the Atlantic, questions emerged about the way the study product was investigated in animals, as well as why the clinical trial team decided to test it simultaneously in the six volunteers.

"Six people were apparently given a compound concurrently in a phase I setting, and that really raised eyebrows for a number of people because I think we all assume that in a phase I trial that is intended to look fully at safety that the drug would be given to one individual at a time," Koski says. "It should screen for SAEs sort of sequentially rather than simply doing a shotgun approach."

While parallel processing would improve efficiency, it obviously opens the door for this kind of disaster, Koski adds.

"My suspicion is that every IRB or ethics committee more or less assumes the sequential model will be followed, but that point is not addressed in the regulations," Koski says. "There’s nothing that says you can’t do six at once."

The clinical trial’s concurrent design was surprising, says Anthony T. Dren, PhD, a consulting professor at the Duke University School of Nursing in Durham, NC.

"When I was in the clinical trial industry, we would take nine volunteers and start off with three, putting two on the study drug and one on a placebo, and they would get the drug first," Dren explains. "Then a week later the other six people, with four on the study drug and two on a placebo, would get their first dose, and the original three people would get a second dose."

Still, it’s not unusual for a clinical trial to put all participants on the study drug at the same time, notes George "Trey" Turner, III, RPh, MA, RAC, an assistant clinical professor in the Clinical Research Management Program at Duke University School of Nursing.

"We’ve done phase I studies with cohorts of eight," Turner says. "To do that you need to have early animal studies that had very low toxicity, and from the information I’ve seen about this particular study, it looks like the information the drug company had did not demonstrate any toxicities."

Given these facts, the study design would not be improper, Turner says. "But clearly that probably is going to change and probably more so in Europe than here."

Media reports during the week after the incident said that German officials had already said that phase I trials could not begin all volunteers on the study compound concurrently.

TeGenero in a question and answer news release said that two monkeys in pre-clinical tests experienced a transient increase in the size of lymph nodes, which was possibly due to the way in which TGN1412 works on the immune system. Company officials called this a mode of action of the drug rather than a drug-related side effect, and they noted that this is entirely different from the symptoms seen in the volunteers.

TGN1412 also had been tested on rabbits, and there were no drug-related side effects and no drug-related deaths despite scientists administering doses up to 500 times the dose used in the phase I clinical trial, the TeGenero media release said.

Also, the informed consent included information about the small chance of temporary swelling as the drug worked on the immune system, according to TeGenero officials.

Until the investigations are complete and a cause for the SAEs is found, researchers can only speculate on what went wrong in the clinical trial. But the way the compound harmed all of the volunteers, and so quickly despite animal studies indicating low risk, shows that changes need to be made in pre-clinical research involving monoclonal antibody agents, Turner and Dren say.

"My guess is there’s probably something about producing monoclonal antibodies that is not a good science yet," Turner says. "It’s easy to have contamination and have a product that is different from what was done with the earlier animal studies."

It’s possible the pharmaceutical company shifted its manufacturing to a different location and may have produced a product in the new manufacturing process for use in the clinical study and that was different than the product produced earlier and used in the animal study, Turner says.

"The monoclonal antibodies are designed to interact with a certain receptor in the human system, and there may not be an equivalent receptor model in animals to test against," Turner says.

A potential solution would be to create transgenic animals to have some human responses, Turner says. "It will need better pre-clinical models," he says.

Still, it’s too early to determine what went wrong, and there will have to be a close examination of both the manufacturing and some of the research data used to bring this drug to the clinical trial phase, Dren says. "I hope everyone will release the results of the investigation so we can learn from this experience," Dren says.

It may be some time before the clinical trial industry experiences fallout from this incident, but it’s possible the repercussions will be minimal in the United States, where there has been close review of phase I studies for years, Turner says.

"The FDA has more say over phase I studies in this country than the MHRA does in the United Kingdom," Turner says. "There was a time in the UK where phase I studies using normal healthy volunteers could be conducted without submitting significant information to the regulatory agency."

In the United States, researchers have to submit full IND reports, background literature searches, and other details, he says.

Also, the history of phase I research has shown that these studies typically have low incidences of SAEs, Turner notes.

"Typically what you do is push the dose on the phase I study until you get some minor toxicities, and then at that point you simply stop the studies," Turner says.

The international media coverage of the incident highlights the way clinical trial research is set back each time a research accident occurs. For instance, one Agence France-Presse article referred to the six volunteers as human guinea pigs on three separate references. An MSNBC commentary by Arthur Caplan, PhD, questioned the wisdom of having a clinical trial industry dominated by for-profit clinical trial companies and called for Congress and other regulatory bodies to take a hard look at whether for-profit businesses and their sponsors are putting the safety and welfare of subjects first.

From a legal and regulatory perspective, it’s likely the incident will result in some changes, Cardinalli says. (See story on previous research problems and regulatory responses, p. 53.) "Given the litigious nature increasing in these sort of studies, I would not be surprised to see lobbying by the industry to limit liability," Cardinalli says. "But it’s not something I would support." Also, it’s possible that testing in animals will be made considerably more rigorous as a result of this event, Cardinalli predicts. "Further, I believe our own HHS needs to make an immediate determination if this is an aberration."

Koski, Dren, and Turner say they doubt the disaster will have a chilling effect on recruitment for phase I trials.

"Just because there’s a plane crash in Omaha doesn’t mean there won’t be people flying around the world," Koski says. "It does mean we engage in a robust process so volunteers really understand what they’re getting into with these phase I trials."

The incident was such a rare event that it probably won’t have the impact on the clinical trial industry that it appears to have at first glance, Dren says. "In fact there have been a variety of clinical study sites with adverse events that have occurred, and those have not had an impact in today’s world," Dren says. "There have been seizures, serious rashes, and a number of events in phase I studies, and they haven’t affected recruitment of volunteers at all."

One reason the impact on recruitment likely will be minimal is because in the United States a lot of the people who enroll in phase I studies are professional drug research subjects, Turner notes.

"A lot of them do this for a living, and they’ve gotten pretty savvy about what they’re willing to volunteer for and what not, according to safety concerns," Turner says. "It’s a better guard against higher risk level studies to have professional subjects review the material."

The other reason it might not have a major impact on the industry is because this is a very rare problem with phase I studies, Turner adds.

"But if they cannot define a reason or problem that caused this outcome, then it might result in some second thoughts, second guessing," Turner says.