Quantitative structure–activity relationship (QSAR) methodology in forensic toxicology: Modeling postmortem redistribution of structurally diverse drugs using multivariate statistics

Postmortem redistribution (PMR) constitutes a multifaceted process, which renders the analytical results of drug concentrations inaccurate to be interpreted by forensic toxicologists. The aim of the present study was to evaluate whether quantitative structure–activity relationship (QSAR) methodology could serve as an effective tool to estimate the ability of drugs to redistribute across tissue barriers during postmortem period on the basis of their molecular, physicochemical and structural properties. In this aspect, multivariate data analysis (MVDA) was applied to a set of 77 structurally diverse drugs. PMR data expressed by the central:peripheral concentration ratio (C:P ratio) was taken from the literature. An adequate and robust QSAR model (R² = 0.65, Q² = 0.56, RMSEE = 0.34) was established for 59 (77%) out of 77 drugs. Although the derived QSAR model presented limited applicability, it provided an informative illustration of the contributing molecular, physicochemical and structural properties in PMR process. Drugs with strong basic properties and enhanced molecular size, flexibility, lipophilicity and number of halogens were found to be susceptible to increased PMR. Due to the high complexity of PMR process, further QSAR studies need to focus on structurally related drugs to develop more specific models, which could serve as alternative tools to evaluate PMR for different chemical classes.     

The use of a quantitative structure-activity relationship (QSAR) model to predict GABA-A receptor binding of newly emerging benzodiazepines

The illicit market for new psychoactive substances is forever expanding. Benzodiazepines and their derivatives are one of a number of groups of these substances and thus far their number has grown year upon year. For both forensic and clinical purposes it is important to be able to rapidly understand these emerging substances. However as a consequence of the illicit nature of these compounds, there is a deficiency in the pharmacological data available for these ‘new’ benzodiazepines. In order to further understand the pharmacology of ‘new’ benzodiazepines we utilised a quantitative structure-activity relationship (QSAR) approach. A set of 69 benzodiazepine-based compounds was analysed to develop a QSAR training set with respect to published binding values to GABA_A receptors. The QSAR model returned an R² value of 0.90. The most influential factors were found to be the positioning of two H-bond acceptors, two aromatic rings and a hydrophobic group. A test set of nine random compounds was then selected for internal validation to determine the predictive ability of the model and gave an R² value of 0.86 when comparing the binding values with their experimental data. The QSAR model was then used to predict the binding for 22 benzodiazepines that are classed as new psychoactive substances. This model will allow rapid prediction of the binding activity of emerging benzodiazepines in a rapid and economic way, compared with lengthy and expensive in vitro/in vivo analysis. This will enable forensic chemists and toxicologists to better understand both recently developed compounds and prediction of substances likely to emerge in the future.