Biotrial Rennes
Biotrial president Francois Peaucelle faces reporters after the death of a patient in the Rennes clinical trial.Loic Venance/Getty Images

More than four months after the death of a participant during the Rennes clinical trial, a scientific committee mandated by the French National Agency for Medicines and Health Products Safety (ANSM) has published its final conclusions.

The report (in French) described a variety of errors made by both Bial, the lab producing the drug, and Biotrial, the company which conducted the experiments. One of the most crucial issues identified is the fact that two important doses of the BIA 10‐2474 molecule tested were given in too quick a succession to the trial participants. The authors also investigated the toxicity of the molecule which played a crucial role in the fatal outcome.

Enzyme inhibitor

BIA 10‐2474 is an inhibitor of the enzyme FAAH, which plays a key role in the regulation of the endocannabinoid system. It was intended to help patients with neurodegenerative disorders and to treat neurological pains.

The report points out that the molecule's structure did not appear suspicious at first, and no indications of its toxicity could have been derived from it before the trial. "The analysis of the molecule's chemical structure does not evoke anything specific, including regarding potential toxic properties. The chemical components that constitute it are commonly found in many medicines", the experts write.

However, they explain that the molecule's action was problematic, for different reasons. They describe this action as "not very specific to the enzyme targeted", "not very progressive" - as the molecule can go from not inhibiting the enzyme to fully inhibiting it with just a small variation in the doses - and "prolonged in time" with the action of BIA 10‐2474 lasting more than 24 hours. All of these factors could have made the molecule risky to use in humans.

Human trial shortcomings

More problematic perhaps is the fact that the documents - given by Bial to gain approval for human trials - contained a number of inaccuracies and translation mistakes.

The trial itself included 128 healthy volunteers, male and female, aged between 18 and 55 years old. The report notes that these important age gaps are unusual for phase 1 clinical trials of this type. The authors were also surprised to discover that no psychological assessment and cognitive tests were planned during the trial, which is common when a experimental drug targets the central nervous system.

Beyond that, the main problem highlighted is the lack of gradual progression in the doses given. The trial was designed so that six doses (2,5 mg ; 5 mg ; 10 mg ; 20 mg ; 50 mg and 100 mg) would be tested on one group of volunteers. The logic was to multiply the quantity, by two, each time a new dose was administered. This gradual progression is, however, not applied between dosages of 20 mg and 50 mg.

The participant who died was rushed out to hospital after reaching the 50 mg dose threshold, but at the time, the scientists had not associated his symptoms to the fact he had been given the molecule. Now, the report concludes that "the symptoms presented by the volunteers can be linked to the BIA 10-2474 dose which they received daily and repeatedly".