Prediction of toxicity

There do not appear to be any published studies to date of ANNs being used for the prediction of drug toxicity, although they have been used for the prediction of toxicity of chemicals such as pesticides [68, 69]. 

Greene N. Computer systems for the prediction of toxicity an update. Adv Drug Deliv Rev 2002 54 417-31. 

Barratt MD, Rodford RA. The computational prediction of toxicity. Curr Opin Chem Biol 2001,5 383-8. 

Barratt, M.D. Rodford, R.A. (2001) The Computational Prediction of Toxicity. Current Opinion in Chemistry and Biology, 5, 383-388. 

There are software that use more approaches for the prediction of toxicity expert systems, QSAR, and read-across (http //www.insilico.eu/use-qsar.html). 

DEREK Expert system for the prediction of toxicity (genotoxicity, carcinogenicity, skin sensitization, etc.)... 

Dobler, M., Till, M.A., and Vedani, A. From crystal stmctures and their analysis to the in silico prediction of toxic phenomena. Helv. Chim. Acta 2003, 86, 1554-1568. 

QSAR are useful In the design of pesticides and medicinal drugs, and In environmental problems such as the prediction of toxicity and blodegradablllty. An empirical relationship can be properly used only for Interpolation whereas one based solidly on well-established theory can be used at least to some extent for extrapolation as well. It seems of real Importance, then, to determine the nature and slgmiflcance of steric and bulk parameters In QSAR. 

There are a number of key features essential for a cell-based model to be effectively predictive of toxicity (Table 14.3). First, as indicated above, it must be sensitive enough to reflect early, sublethal injury and not merely cell death. 

A variety of different approaches to the prediction of toxicity have been developed under the sponsorship of the Predictive Toxicology Evalnation project of the National Institnte of Environmental Health Sciences. The widespread application of compnta-tional techniqnes to stndies in biology, chemistry, and environmental sciences has led to a qnest for important, characteristic molecnlar parameters that may be directly derived from these compntational methods. Theoretical linear solvation energy relationships combine compntational molecular orbital parameters with the linear solvation energy relationship of Kamlet and Taft to characterize, nnderstand, and predict biological, chemical, and physical properties of chemical componnds (Eamini and Wilson, 1997). 

Troester MA, Hoadley KA, Parker JS et al. Prediction of toxicant-specific gene expression signatures after chemotherapeutic treatment of breast eell lines. Environ Health Perspect 2004 112 1607-1613. Yang Y, Blomme EA, Waring IF. Toxicogenomics in drug discovery from preclinical studies to elin-ical trials. Chem Biol Interact 2004 150 71-85. 

Adverse effects from aminoglycoside are both time- and concentration-dependent. Toxicity is unlikely to occur until a certain threshold concentration is reached, but once that concentration is achieved the time beyond this threshold becomes critical. This threshold is not precisely defined, but a trough concentration above 2 mcg/mL is predictive of toxicity. At clinically relevant doses, the total time above this threshold is greater with multiple smaller doses of drug than with a single large dose. 

In some cases, physical and chemical properties are highly predictive of toxicity. A good example is acute aquatic toxicity, which is discussed in detail later in this chapter. However, some biological interactions that lead to toxicity, such as receptor binding, are highly specific and in those cases it is less likely that straightforward relationships between properties and toxicity can be found. A common example is the difference in toxicity of two enantiomers, such as the R- and S-enantiomers of thalidomide. 

Assay response could lead to prediction of toxicity in vivo which metabolic stability considerations could modify... 

Prediction of toxicity from knowledge of the properties of substances such as the octanol-water partition coefficient (Ki m) that determine toxicity through basic mechanisms such as... 

The prediction of toxicity through the use of quantitative structure-activity relationships ([Q]SAR) is the most basic of effect extrapolations and is applied when no toxicity or response data have been measured for the substances in question. (Q)SAR... 

Computer Models Provide a Prediction of Toxicity and Fate... 

Approximately 100,000 separate chemicals may be released into the environment annually it is frightening to consider that reliable toxicity data exist for only a tiny proportion of these chemicals, probably less than 5%. The percentage of chemicals with a complete set of reliable toxicity data (i.e., across a broad spectrum of environmental and human health effects) is considerably less than 5%. Computer-aided prediction of toxicity has the capability to assist in the prioritisation of chemicals for testing, and for predicting specific toxicities to allow for labeling. Chapter 19 describes these activities in more detail. As the reliability of models for toxicity prediction increases, there will undoubtedly be increased use for the filling of data gaps. 

This book describes numerous methods for the prediction of toxicity (Chapters 8, 9,12, and 13), environmental fate (Chapters 14 to 16), and the effects of chemicals in humans (Chapters 8, 10, 11, 13, and 17). In addition to those models reviewed in these chapters there are many more available in the open literature (see the next section). Despite these predictive models, animal tests are still being performed to assess toxicity and fate. The question then becomes, How are we to use predictions The simple answer to this question is cautiously . 

As described in Chapter 9 there are an increasing number of commercial toxicological prediction systems available. Naturally these have been designed to be user friendly most run under Microsoft Windows and use the Simplified Molecular Input Line Entry System (SMILES) as the molecular input. It is therefore possible to obtain a prediction of toxicity instantaneously, and often this may be performed for large numbers of compounds. There is a great temptation to use predicted toxicities at face value (i.e., if a compound is predicted to be non-toxic then it must be non-toxic). This simplistic use of predicted values should be avoided at all costs. Ideally, there are a number of criteria that should be applied when predicting toxicity. It is essential that a trained expert uses the predictive system. The user should be an expert both in the endpoint being predicted and the use of the predictive system. 

---