Toxicogenomics – A Drug Development Perspective - Genomic Biomarkers for Pharmaceutical Development - page 134

to predict the toxicity of a candidate drug. In this manner, gene expression profiles might
be seen as unique drug 'fingerprints', the idea being that if a fingerprint closely matches
that of a similar toxicant in a database, there is a suggestion of toxicity. This approach may
provide a novel metric for the identification and ranking of the potential toxicity of candi-
date drugs, thus increasing the likelihood of success in pre-clinical and clinical development
[157] . The predictive capacity of gene expression profiling has been demonstrated in several
studies [158-160] . In fact, some pharmaceutical companies have built their own databases to
support drug development. If successful, this approach would improve the identification of
potentially toxic compounds and thus increase the likelihood of success for those candidate
drugs passing the screening. Hepato- and nephrotoxicity are two related attrition factors in
pre-clinical drug development and thus a main focus of study with toxicogenomics [29] and
[161-163] . Specific gene expression profiles identified to be associated with various hepato-
toxicants have been reported, including acetaminophen, carbon tetrachloride, and TCDD
and more [164-166] . Furthermore, these gene expression signatures may be linked to specific
biological processes, such as macrophage activators, peroxisome proliferators, and oxida-
tive stressors / reactive metabolites [167,168] . Studies have also demonstrated the capability of
toxicogenomics in the prediction of nephrotoxicity and identification of candidate toxicity-
related biomarkers in rat kidney [169,170] .
Toxicogenomics can also be used with gene expression data from a short-term in vivo study
or in vitro study to predict outcomes in long duration studies, such as chronic or carcinogenic-
ity toxicity studies [171,172] . Preliminary toxicogenomic studies suggest that gene expression
patterns are often predictive of phenotypic alterations [173] . The carcinogenic potential of
chemicals is currently evaluated with rodent life-time bioassays, which are time consuming,
and expensive with respect to cost, number of animals, and amount of compound required.
Using gene expression profiles from the livers of rats treated for up to 14 days with hepato-
carcinogens, gene expression patterns were identified and developed into classifiers to predict
a set of independent validation compounds; the prediction accuracy is reported to be up to
88% [172] . In addition, gene expression patterns were found to be more sensitive for detecting
renal toxicity than the biochemical and histopathological parameters used in conventional tox-
icology testing [18] and [174] . Predicting the long-term effects of compounds using short-term
assays is a feasible approach [140] . Therefore, it is believed that toxicogenomics could acceler-
ate the process of drug discovery and development [175] .
6.4.4 Mechanistic Toxicology with Toxicogenomics
The advent of genomic technologies has facilitated major advances in our understanding
of the molecular details of biology and holds the promise of providing new insights into elu-
cidating the molecular mechanisms of a variety of toxicities. The application of toxicogenom-
ics to mechanistic studies has played an important role when the toxicity of candidate drugs
is not associated with well-established biomarkers or significant morphological changes
[176,177] . Modern mechanistic analysis leverages the increasing sophistication of functional
genomics to interpret the biological functions of particular gene sets, understand how the
gene function contributes to a biological process, gain information on likely initiating and
controlling factors, and predict possible biological consequences of disturbed expression pat-
terns [141] . Mechanistic toxicity has application to a wide range of drug safety issues, and