Dogs died in studies of drug that killed man in French trial

BMJ 2016; 352 doi: (Published 29 February 2016) Cite this as: BMJ 2016;352:i1228
  1. Nigel Hawkes
  1. London

Several dogs died in preclinical studies of the experimental drug that killed a human volunteer in a French trial, the French medical safety agency ANSM has confirmed.

But the expert committee the agency established to review the trial’s conduct concluded that these deaths were to be expected in animal testing designed to establish a safe dose for human administration. The protocol for the first in human trial recorded that this “no observed adverse effect level” was determined for mice, rats, dogs, and monkeys.

As part of these studies, ANSM said, “very high doses are administered to determine the maximum tolerated doses, which can be lethal in animals.” The 12 strong expert committee reviewed all the animal data, including those on the deaths of the dogs, and concluded that they met the required standards to allow the experimental drug to be tested in human volunteers.

Neither the protocol nor the account so far given by ANSM has revealed what postmortem examinations were made of the experimental animals, including brain histology, but more details have been sought from the Portuguese developers of the drug, Bial, to “deepen and consolidate” the expert committee’s analysis. Full findings are promised at the end of this month.

Two hypotheses are considered by the committee to be the most likely explanation for the brain damage that killed one of six volunteers in the trial.1 One is an “off-target” effect in which the experimental molecule, designed to act on the brain’s cannabinoid receptors, has additional and unexpected effects on another part of the brain, and the other is the possibility that metabolites from the tested products were responsible for the effect.

The brief account of the preclinical studies included in the protocol said that in all the animal species the test substance, BIA 10-2474, undergoes extensive metabolism.

The committee has found a consistent pattern in the five volunteers who experienced ill effects, which it describes as “a purely cerebral neurological symptomology with very fast appearance, relatively consistent among volunteers but with great variability in severity and progression.” Magnetic resonance imaging has shown tissue damage of varying severity but with very unusual topography that, like the symptoms, was consistent among the volunteers.

It is still not clear how severe the damage has been in the surviving volunteers. The magazine Le Figaro quoted Giles Edan, the neurologist who is treating them at Rennes Hospital, as saying, “Four patients of the surviving five have neurological problems. Three have symptoms so severe that it leads us to worry that they will be disabled for the rest of their life. However, this prognosis is still not definitive.”2

Le Figaro has also reported that the unusual location of the brain damage is at the base of the skull, but its efforts to pin it down more precisely were declined by ANSM on the grounds of patient confidentiality. Dominique Martin, director of the agency, has also rejected calls for more details to be given of the background to the trial. “We have given all the information we could. It is an issue of industrial property,” she said.


View Abstract

Log in

Log in through your institution


* For online subscription