randombio.com | science commentary
Wednesday, Jan 20, 2016

What went wrong in Bial's clinical trial?

L ast week a Phase 1 clinical trial in France of a new drug went horribly wrong. One subject died and several others were left in critical condition. What was this drug supposed to do, and what went wrong?

Bia 10-2474 structure
Chemical structure of Bia 10-2474

We don't know all the details yet, but here's what we know so far. The drug, code named Bia 10-2474, was an FAAH inhibitor. FAAH (fatty acid amide hydrolase) is an enzyme that degrades a class of molecules in the brain known as endocannabinoids (eCBs). Bial's patent says:

Of even greater interest is the therapeutic potential of inhibitors of anandamide (AEA) uptake as well as of inhibitors of its degradation by fatty acid amide hydrolase (FAAH) for the treatment of neuronal injury and neurodegeneration. In the latter approach, the beneficial effects of cannabinoid activation (to elevate the endocannabinoid tone) are achieved by inhibition of their degradation rather than by global cannabinoid activation, which is associated with many undesirable effects. In addition, FAAH inhibitors may provide a functionally selective way of enhancing endocannabinoid tone only in those tissues and cells with active synthesis and release of eCBs.

We will have to wait for the final report to be sure, but this suggests that they were testing it as a potential treatment for brain injury. This makes sense in terms of the importance of inflammation that occurs after traumatic brain injury.

Anandamide and cannabinoids

Cannabinoids have revolutionized our study of the brain. A general principle holds that for every drug that has a potent effect on the brain, there is a corresponding receptor. This has been demonstrated for opioids, amphetamines, benzodiazepines, the ethyl alcohol in your beer, and the rohypnol in your date's beer. All these drugs act by binding to a receptor in the brain.

THC binds to the CB1 endocannabinoid receptor. A CB1 antagonist called Rimonabant was tested as a way of blocking the acute effects of marijuana (which include depression and psychotropic effects), but was halted due to toxic effects.

Anandamide structure
Chemical structure of anandamide. The amide bond cleaved by FAAH is shown.

Marijuana has muscle-relaxing effects and anticonvulsant properties, and it can reduce the intraocular pressure of glaucoma. Oral Δ9-tetrahydrocannabinol (Dronabinol, THC) is approved for anorexia associated with weight loss in patients with HIV infection and for nausea and vomiting induced by cancer chemotherapy. There are many other isomers of THC with different potency and different DEA Schedule classification. A synthetic cannabinoid called Nabilone has similar therapeutic benefits, but fewer CNS effects.

Endocannabinoid receptors also regulate key events in neuroinflammation, so drugs binding to them are being considered for multiple sclerosis, Alzheimer's disease, and other diseases in which neuroinflammation is a feature[1].

An endocannabinoid is a lipid molecule found in the brain that normally binds to a CB receptor. The two most abundant receptors are CB1, which is mostly found in the brain and mediates the psychotropic effects of THC (Ki=10 nM), and CB2, which is less well studied. Other CB receptors bind lipids like N-palmitoylethanolamide, an anti-inflammatory lipid that is believed to act through peroxisome proliferator-activated receptor-α (PPARα).

The most well studied endocannabinoid is anandamide (N-arachidonoylethanolamide), which is a lipid that is synthesized from arachidonic acid only when needed, such as after injury. The biosynthesis uses unusual biochemical reactions. Other endocannabinoids, such as virodhamine, are also found in the brain. But it is the degradation of anandamide that is the target of Bia 10-2474.

Potential risks of elevating anandamide

Anandamide is a potent inducer of apoptosis in cells with elevated COX-2 levels [2] and thus is contemplated as an antitumor drug. The toxicity is inhibited by N-acetylcysteine, an antioxidant. Thus one could speculate that generation of reactive oxygen species might be one potential risk.

Interestingly, there is some overlap between EC receptors and vanilloid type 1 receptors, which bind to capsaicin, the ingredient in chili peppers. N-arachidonoyldopamine, for example, a molecule that structurally resembles anandamide, binds to both vanilloid and CB receptors.

The first step in degradation of endocannabinoids is hydrolysis of the C–N amide bond by fatty acid amide hydrolase (FAAH). Anandamide is rapidly degraded by FAAH when it enters the cell, producing arachidonic acid and ethanolamine. Thus FAAH inhibitors would be expected to increase the levels of anandamide and slightly reduce the amount of arachidonic acid in the brain. Most studies seem to show this effect; in mice FAAH can be genetically knocked out, which greatly increases the brain levels of anandamide, without serious adverse effects. Giving anandamide to these mice eliminated pain sensitivity and produced catalepsy, locomotor inhibition, and hypothermia that lasted 6–8 hours.[3]

Phase 1 clinical trials

The purpose of a phase 1 trial is to test a new drug on healthy volunteers to identify dose-limiting toxicity. For some diseases, such as cancer, where drugs are very often highly toxic, the patient cohort is sometimes selected from patients that are showing resistance to existing treatments. Before moving to Phase 1, a battery of tests is always done to eliminate any drugs with known toxic effects. In many cases, however, toxic effects are only detected when the drug is given to patients. For instance, in recent years it was discovered that a very small minority of patients have mutant hERG channels, which causes life-threatening arrhythmias from drugs that were previously thought safe. Thus all new drugs are tested for hERG1 channel binding and EKGs are routinely run on trial subjects to look for dangerous QT interval shortening. [4]

The most common design is called the 3 + 3 dose escalation. Three patients are started on a subtherapeutic dose that is considered safe on the basis of animal studies. If no toxicity is observed, three new patients are given a dose that is increased by 100%, then 67%, 50%, 40%, and 33%. At each stage, if one patient shows toxicity, then three more are given the same dose. Dose escalation continues until at least two of the three show dose limiting toxicities.[5]

Statistically this is one of the safest designs, but it suffers from an excessive number of escalation steps. It also assumes that the toxicity is reversible. Usually dose-limiting toxicity means the subject experiences headache, muscle pain, liver enzyme elevation, or some other adverse effect. Adverse effects (called AEs) often result from interaction with another drug the subject is taking, or they may be caused by differences in individual sensitivity.

The loss of a Phase 1 subject is a disaster, and it is unexpected from the pharmacology of these drugs. Inhibiting anandamide degradation should have been a very safe approach. We will have to wait for the autopsy results to know what happened. Clinical researchers will study the French case in intimate detail, and future clinical trials will be redesigned to accommodate the lessons learned from this tragedy.


1. Curr Drug Targets. 2016 Jan 11. The emerging role of the cannabinoid receptor family in peripheral and neuro-immune interactions. Haugh O, Penman J, Irving AJ, Campbell VA.

2. Mol Carcinog. 2012 Feb;51(2):139–149. Arachidonoyl ethanolamide (AEA)-induced apoptosis is mediated by J-series prostaglandins and is enhanced by fatty acid amide hydrolase (FAAH) blockade. Kuc C, Jenkins A, Van Dross RT.

3. Neuropsychopharmacology. 2010 Jul;35(8):1775–1787. FAAH-/- mice display differential tolerance, dependence, and cannabinoid receptor adaptation after delta 9-tetrahydrocannabinol and anandamide administration. Falenski KW, Thorpe AJ, Schlosburg JE, Cravatt BF, Abdullah RA, Smith TH, Selley DE, Lichtman AH, Sim-Selley LJ.

4. Br J Pharmacol. 2010 Jan; 159(1): 58–69. Drug-induced QT interval shortening: potential harbinger of proarrhythmia and regulatory perspectives Rashmi R Shah

5. J Natl Cancer Inst. 2009 May 20; 101(10): 708–720. Dose Escalation Methods in Phase I Cancer Clinical Trials Christophe Le Tourneau, J. Jack Lee, and Lillian L. Siu

Related articles

On the Internet, no one can tell whether you're a dolphin or a porpoise
Name and address
book reviews