Read our COVID-19 research and news.

A botched clinical trial in France led last year to the hospitalization of six men including Stephane Schubhan, who suffered from neurological problems.

Jean-Francois Monier/Stringer/AFP/Getty Images

New clues to why a French drug trial went horribly wrong

Scientists are one step closer to understanding how a clinical trial in France killed one volunteer and led to the hospitalization of five others in January 2016. A new study shows that the compound tested in the study, BIA 10-2474, has effects on many other enzymes in addition to the one it was supposed to inhibit. These “off-target” effects might explain why the drug caused side effects ranging from headaches to irreversible brain damage.

“We suspected that BIA 10-2474 was a bad compound—now we know for sure,” says neuropharmacologist Daniele Piomelli from the University of California, Irvine, who was not involved in the new study.

The molecule was produced by Portuguese drugmaker Bial and tested in healthy human volunteers in a phase I study by the French contract research company Biotrial in Rennes. BIA 10-2474 inhibits an enzyme called fatty acid amide hydrolase (FAAH), which breaks down endocannabinoids in the brain. Previous research had suggested that FAAH inhibitors might help treat anxiety, chronic pain, or neurodegenerative disorders such as Parkinson’s disease. Although other drug developers, including Pfizer, had already dropped FAAH inhibitors because of disappointing efficacy studies, most of the molecules were shown to be safe, as the U.S. Food and Drug Administration confirmed in August 2016

But that was not true for BIA 10-2474. Neuroscientist Steven Kushner of Erasmus University Medical Center in Rotterdam, the Netherlands, together with chemical biologist Mario van der Stelt from nearby Leiden University and colleagues at several other institutes, set out to find out why.

The researchers used a technique called activity-based protein profiling, which allowed them to screen the molecule’s activity against a very large group of enzymes in living human cells. They found that at higher concentrations, the Bial drug disrupted the activity of several lipases, enzymes that break down fatty acids; the Pfizer drug, by contrast, did not. One of the off-target enzymes is called PNPLA6; previous studies have linked defects in the gene encoding for PNPLA6 to rare neurological disorders. The results suggest that BIA 10-2474 may disrupt how neurons in the cerebral cortex metabolize lipids, the team reports today in Science.

The researchers can’t be sure that the off-target effects actually caused the brain damage seen in the volunteers, however. “We do not have evidence for a causal relationship yet,” Van der Stelt says. One way to find out might be to analyze samples of the deceased volunteer’s brain.

The off-target effects can be species-dependent, says Van der Stelt, which could explain why studies in rats and mice did not identify the drug’s danger. But Bial could have known about the risk if it had thoroughly screened for off-target effects in human cells, like the new study does, Van der Stelt says; that’s what Pfizer did with its FAAH inhibitor. “If Bial had done the same, they might have reached a different conclusion,” Van der Stelt says.  

Jürg Gertsch, a biochemist at the University of Bern, agrees; he says it’s “unbelievable” that Bial did not do a more extensive study of the drug’s effects before giving it to humans. Gertsch has himself investigated the effects of BIA 10-2474 in human blood cells; that study has yet to be published.

“The new study does not provide an explanation for the toxicity observed with the compound BIA 10-2474, but does highlight its promiscuous nature,” says Pomelli, who calls the decision to move forward with the human trial “deeply misguided.” He says the doses given to the volunteers in the group where the accident occurred were also unnecessarily high, as full inhibition of FAAH occurred at a much lower level.

Bial welcomes “any study” that could help shed light on the incident, a spokesperson says. They add that results from the company’s own investigation largely align with the new paper, but that in Bial’s opinion, they are unlikely to explain the neurological effects seen in the study.

Although Bial and Biotrial have been heavily criticized for the study, French authorities have concluded that the companies did not violate clinical trial regulations. In the wake of the case, the European Medicines Agency is developing stricter rules for “first-in-human” studies.