Review
Informatics
Computational toxicology in drug development

https://doi.org/10.1016/j.drudis.2007.12.007Get rights and content

Computational tools for predicting toxicity have been envisaged for their potential to considerably impact the attrition rate of compounds in drug discovery and development. In silico techniques like knowledge-based expert systems (quantitative) structure activity relationship tools and modeling approaches may therefore help to significantly reduce drug development costs by succeeding in predicting adverse drug reactions in preclinical studies. It has been shown that commercial as well as proprietary systems can be successfully applied in the pharmaceutical industry. As the prediction has been exhaustively optimized for early safety-relevant endpoints like genotoxicity, future activities will now be directed to prevent the occurrence of undesired toxicity in patients by making these tools more relevant to human disease.

Section snippets

Computational methods applicable for toxicity prediction

In silico techniques for the prediction of toxicological endpoints are extremely appealing because of their expeditious return of results and their inexpensiveness. Moreover, these techniques can be used in a very early phase of drug discovery even before the molecule is synthesized. Numerous commercially available and free web-based programs for toxicity prediction are available; some of them are listed and briefly described in Table 1.

In silico prediction methods that are widely used in the

Application of SAR models in cardiovascular safety pharmacology

Some recent withdrawals of otherwise successful drugs from the market received particular attention owing to their rare induction of potentially life-threatening ventricular tachyarrhythmias of the Torsades de Pointes (TdP) type [37]. In fact most, if not all, of the non-cardiovascular agents associated with a torsadogenic liability prolong the QT-interval by blocking the rapidly activating component of the delayed rectifier potassium current, termed IKr. The ion channel protein is encoded by

Drug bioactivation and hepatotoxicity

Major reasons for drug failure are adverse events in man with some toxicities appearing only during the post-approval period of a drug. Serious adverse drug reactions are believed to be one of the leading causes of death in the United States and are estimated to have occurred in over two million patients in 1994 with more than 100 000 fatalities. Hepatotoxicity has been identified as the major safety concern for discontinuation of clinical trials and either post-approval withdrawal [45] or

In silico screening for drug-induced phospholipidosis

Phospholipidosis describes the intracellular accumulation of various phospholipids reflecting a disorder in phospholipid storage in the lysosomes. Drug-induced phospholipidosis was first reported in 1966 when Greselin [51] observed an increased number of foam cells in the rat lung after the application of a cholesterol metabolism inhibitor. Since then, a number of drug-induced phospholipid disorders have been described in animals and humans for a wide variety of pharmacological compounds like

Predicting non-DNA reactive genotoxic activity of kinase inhibitors in early drug development

Identification of genotoxic liabilities is one of the key functions in preclinical safety assessment of drug discovery. The main reason is to judge any relevant mechanisms leading to mutations as part of the initiating process for carcinogenesis. It is inherent in this area of safety assessment that human data are normally lacking.

Direct-acting genotoxins can be predicted with a very high concordance, as high-quality and comprehensive databases are available for nearly all kinds of genotoxicity

Conclusions and future trends

In general, drug-induced ADRs can be classified in (a) direct-acting mechanisms, which are often triggered by bioactivation of the parent drug to toxic reactive, typically electrophilic metabolites capable of covalent binding to cellular macromolecules and (b) pharmacology-related undesired effects (primary target or cross-reactivities). It has been shown in the present review that computational methods can be successfully applied in early drug development and SARs can be constructed quite

Acknowledgements

The authors thank Dr Manfred Kansy for helpful scientific discussions and Dr Jacqueline Gillis for carefully reviewing the manuscript.

References (73)

  • P.B. Bennett et al.

    Trends in ion channel drug discovery: advances in screening technologies

    Trends Biotechnol.

    (2003)
  • F. Ballet

    Hepatotoxicity in drug development: detection, significance and solutions

    J. Hepatol.

    (1997)
  • L.P. Grazette

    Inhibition of ErbB2 causes mitochondrial dysfunction in cardiomyocytes: implications for herceptin-induced cardiomyopathy

    J. Am. Coll. Cardiol.

    (2004)
  • P. Lenart

    The small-molecule inhibitor bi 2536 reveals novel insights into mitotic roles of polo-like Kinase 1

    Curr. Biol.

    (2007)
  • S.Y. Wu

    Discovery of a novel family of CDK inhibitors with the program LIDAEUS: structural basis for ligand-induced disordering of the activation loop

    Structure (Camb.)

    (2003)
  • N.M. Heron

    SAR and inhibitor complex structure determination of a novel class of potent and specific Aurora kinase inhibitors

    Bioorg. Med. Chem. Lett.

    (2006)
  • White paper produced by the FDA (2004) Innovation/stagnation: challenge and opportunity on the critical path to new...
  • P.N. Judson

    Expert systems

  • V.K. Gombar

    Computational approaches for assessment of toxicity: a historical perspective and current status

  • R.D. Benz

    Toxicological and clinical computational analysis and the US FDA/CDER

    Exp. Opin. Drug Metabol. Toxicol.

    (2007)
  • E.J. Matthews

    Use of toxicological information in drug design

    J. Mol. Graph. Model.

    (2000)
  • P.N. Judson

    Using computer reasoning about qualitative and quantitative information to predict metabolism and toxicity

  • G. Klopman

    MULTICASE. 1. A hierarchical computer automated structure evaluation program

    Quant. Struct.—Activity Relation.

    (1992)
  • K. Enslein

    Computation of developmental toxicity potential by QSTR models in the TOPKAT program

    Adv. Anim. Altern. Safety Efficacy Test.

    (1998)
  • Y-T. Woo et al.

    OncoLogic: a mechanism-based expert system for predicting the carcinogenic potential of chemicals

    Predict. Toxicol.

    (2005)
  • C. Helma

    Lazy structure–activity relationships (lazar) for the prediction of rodent carcinogenicity and Salmonella mutagenicity

    Mol. Divers.

    (2006)
  • G. Roberts

    LeadScope: software for exploring large sets of screening data

    J. Chem. Inf. Comput. Sci.

    (2000)
  • D.E. Johnson

    Future of computational toxicology: broad application into human disease and therapeutics

    Computat. Toxicol.

    (2007)
  • M.P. Smithing et al.

    HazardExpert. An expert system for predicting chemical toxicity

    ACS Symposium Series 484

    Food Saf. Assess.

    (1992)
  • D.F.V. Lewis

    COMPACT: a structural approach to the modelling of cytochromes P450 and their interactions with xenobiotics

    J. Chem. Technol. Biotechnol.

    (2001)
  • V. Poroikov et al.

    PASS: prediction of biological activity spectra for substances

    Predict. Toxicol.

    (2005)
  • A. Cheng et al.

    In silico models for the prediction of dose-dependent human hepatotoxicity

    J. Comput. Aid. Mol. Des.

    (2003)
  • P. Japertas

    In silico technology for identification of potentially toxic compounds in drug discovery

  • S. Ekins

    Algorithms for network analysis in systems—ADME/Tox using the MetaCore and MetaDrug platforms

    Xenobiotica

    (2006)
  • T. Okada

    Characteristic substructures and properties in chemical carcinogens studied by the cascade model

    Bioinformatics

    (2003)
  • A. Vedani

    The challenge of predicting drug toxicity in silico

    Basic Clin. Pharmacol. Toxicol.

    (2006)
  • Cited by (0)

    View full text