Can the fatal toxicity of antidepressant drugs be predicted with pharmacological and toxicological data?

Antidepressant drugs are among the most common drugs involved in fatal poisoning and large variations between antidepressant drugs have been noted. Despite the fact that a large number of studies have calculated a fatal toxicity index (FTI) for antidepressants, no serious attempts have been made to compare the differences in fatal toxicity against known pharmacological and toxicological differences in receptor affinity. It is potentially from such data that screening of drugs during their pre-clinical development can be facilitated. We examined correlations between the FTI and noradrenaline (norepinephrine)/serotonin (5-hydroxytryptamine; 5-HT) reuptake inhibition selectivity, the dose that is lethal to 50% of animals (LD50), lipid solubility, and antagonist activity at cholinergic, histaminergic, alpha-adrenergic and gamma-aminobutyric acid (GABA)A receptors or sodium and potassium channel blocking effects. We obtained data on the number of fatal poisonings between 1983 and 1992 in England and Wales caused by a single antidepressant drug from the Department of Health in the UK. This number was divided by the number of prescriptions in England for these drugs over this time to derive a FTI of deaths per million prescriptions. The highest FTIs were for amoxapine, viloxazine, desipramine and dothiepin. Lofepramine, paroxetine and fluoxetine had very low FTIs. Using Poisson regression, there was a significant positive relationship between the FTI of antidepressant drugs and their lethal toxicity in animals, and measures of their cardiac effects. The relative noradrenaline/serotonin reuptake inhibition, lipid solubility and their potency at histamine H1, muscarinic and alpha 1-adrenergic receptors had no substantial association with the FTI. Limited data suggest that some cardiac effects and potency as a GABAA antagonist may be important predictors of significant toxicity. Further data using standardised bio-assays are needed to compare the direct cardiac effects of antidepressants. Thus, the best current pre-clinical indicator of fatal toxicity in humans is the LD50 in animal studies. Clearly, there are humane and practical reasons for developing a better pre-clinical indicator of toxicity in overdose for this rapidly expanding group of drugs.