Many phase I studies now include several subtrials that may require new risk reduction strategies.

Many phase I studies now include several subtrials that may require new risk reduction strategies, the European Medicines Agency says.

J. Ling/Wikimedia Commons

Europe overhauls rules for ‘first-in-human’ trials in wake of French disaster

The European Union is beefing up protections for volunteers in phase I clinical trials in the wake of a disastrous clinical study in Rennes, France, that resulted in the death of one volunteer and the hospitalization of five others. On 21 July, the European Medicines Agency (EMA) in London announced in a "concept paper” that it wants to improve strategies to identify and reduce risks in “first-in-human” (FIH) studies on healthy volunteers. EMA is asking for input from stakeholders.

The current guideline for FIH studies dates from 2007; it was established in the aftermath of a similar tragedy in 2006 in London, when six volunteers were hospitalized with severe adverse events after receiving a monoclonal antibody named TGN1412 for the first time.

EMA will seek in particular to reduce the risks of studies that combine a number of different subtrials. Such studies are becoming more and more common, EMA writes; for instance, the trial in Rennes included subtrials using single and multiple dose administration, as well as trials on drug interactions with food and on pharmacodynamics, the study of a drug’s biochemical and physiologic effects on the body.

This increased complexity requires a new, structured approach, EMA says, in which decisions on each new step need to be based on the data collected at the previous one. The concept paper, written by an international group of experts, says that the role of pharmacology and toxicology data in estimating the therapeutic dose, increases in dosing, and stopping criteria need to be addressed in the revised guideline. Also on the table are new instructions for decision-making processes and stopping rules, rolling review of emerging human data during the study, communication to authorities and subjects, and guidance on the type of scientific information to be included in a trial application.

EMA invites comments until 30 September, after which it will publish a draft revised guideline, probably later this year.

Many of the changes that EMA seeks are a direct response to the trial in Rennes, which international experts have criticized for its lax design. The protocol allowed administering a dose of 100 milligram over 10 days in the part of the study where the incident happened, although full inhibition of the targeted enzyme “should be achieved at doses lower than 5 mg,” according to an expert commission of the French National Agency for Medicines and Health Products Safety. The trial protocol allowed the study to move from one phase to the next without external review and to increase doses without analyzing pharmacokinetic data of the previous cohort—even if drug-related severe adverse events appeared in half of the subjects receiving the drug, named BIA 10-2474.

A report released by French health minister Marisol Touraine in May concluded that the company running the trial, Biotrial, made serious mistakes after the accident happened, but that the study design did not violate French regulations and standards. However, Touraine announced measures to improve trial safety in the future.

Clinical toxicologist Michael Eddleston from the University of Edinburgh, a co-author of a recent editorial on the Rennes trial, applauds EMA’s move to tighten the rules. However, he says the concept paper doesn’t sufficiently focus on the goal of FIH studies, which is to understand whether a drug behaves in the human body as expected based on animal studies—not to study adverse effects. “FIH studies should never be designed to test tolerance … but should be used to understand the pharmacology of the drug in humans, especially in relation to the information available from animals,” Eddleston says.  

Eddleston says the BIA 10-2474 study seemed rushed. “I can see the advantage for industry of putting [subtrials] all together into one protocol,” he wrote in an email. But to ensure safety, data must be assessed carefully as they come through, he says—not just by trial leaders but by independent reviewers.

Bial, the Portuguese company that by its own account spent millions of euros developing BIA 10-2474, has decided to abandon the drug, Portuguese newspaper Público wrote earlier this month. In June, it was revealed by the public prosecutor in Paris that the volunteer who died had an unidentified endocranial vascular pathology prior to the study. The prosecutor is still investigating whether involuntary manslaughter charges are warranted in the case.