Promises and Pitfalls of Quantitative Structure−Activity Relationship Approaches for Predicting Metabolism and Toxicity

Abstract

The description of quantitative structure-activity relationship (QSAR) models has been a topic for scientific research for more than 40 years and a topic within the regulatory framework for more than 20 years. At present, efforts on QSAR development are increasing because of their promise for supporting reduction, refinement, and/or replacement of animal toxicity experiments. However, their acceptance in risk assessment seems to require a more standardized and scientific underpinning of QSAR technology to avoid possible pitfalls. For this reason, guidelines for QSAR model development recently proposed by the Organization for Economic Cooperation and Development (OECD) [Organization for Economic Cooperation and Development (OECD) (2007) Guidance document on the validation of (quantitative) structure-activity relationships [(Q)SAR] models. OECD Environment Health and Safety Publications: Series on Testing and Assessment No. 69, Paris] are expected to help increase the acceptability of QSAR models for regulatory purposes. The guidelines recommend that QSAR models should be associated with (i) a defined end point, (ii) an unambiguous algorithm, (iii) a defined domain of applicability, (iv) appropriate measures of goodness-of-fit, robustness, and predictivity, and (v) a mechanistic interpretation, if possible [Organization for Economic Cooperation and Development (OECD) (2007) Guidance document on the validation of (quantitative) structure-activity relationships [(Q)SAR] models. The present perspective provides an overview of these guidelines for QSAR model development and their rationale, as well as the promises and pitfalls of using QSAR approaches and these guidelines for predicting metabolism and toxicity of new and existing chemicals.

Application of quantitative structure–activity relationships for modeling drug and chemical transport across the human placenta barrier: a multivariate data analysis approach

“…This methodology has been widely accepted to evaluate and predict the activity of drug molecules against a therapeutic target, as well as the toxicity-risk assessment of drugs and chemicals, supporting the reduction, refi nement and/or replacement of experimental studies (Vedani et al, 2005(Vedani et al, , 2006Zvinavashe et al, 2008). Relevant software packages are currently available which permit the calculation of a large number of descriptors to be used in QSAR studies as an alternative strategy for the establishment of predictive models.…”

Predicting Drug Safety: Next Generation Solutions

“…In some cases the chemistry involved in building the models remains undisclosed and therefore the predictive algorithms exist as "black box" tools. QSAR modeling is frequently used in drug toxicity prediction, with acknowledgement that this type of prediction does have both promises and pitfalls [4]. It has become increasingly recognized that QSAR models should be structured to provide: a defined endpoint; an unambiguous algorithm; a defined domain of applicability; appropriate measures of goodness-of-fit, robustness, and measures of predictability; and a mechanistic correlation.…”

“…It has become increasingly recognized that QSAR models should be structured to provide: a defined endpoint; an unambiguous algorithm; a defined domain of applicability; appropriate measures of goodness-of-fit, robustness, and measures of predictability; and a mechanistic correlation. Ideally, the choice of chemical descriptors, would take into account the mechanism of action and most desirably the rate limiting step in the endpoint and/ or the biological process being modeled [4].…”

In SilicoADME/Tox Predictions

“…The latter usually uses machinelearning approaches (e.g., neural networks, support vector machines) that can yield models with higher predictive accuracy as compared to rule-based strategies. This improvement is often costly since the chemical compound is retained (or rejected) in (from) the collection with little or no possibility of understanding why [25,208]. This lack of interpretability hinders chemists' efforts to identify the causes and solutions of the problem.…”