Prediction of clinical drug efficacy by classification of drug-induced genomic expression profiles in vitro

Abstract

Assays of drug action typically evaluate biochemical activity. However, accurately matching therapeutic efficacy with biochemical activity is a challenge. High-content cellular assays seek to bridge this gap by capturing broad information about the cellular physiology of drug action. Here, we present a method of predicting the general therapeutic classes into which various psychoactive drugs fall, based on high-content statistical categorization of gene expression profiles induced by these drugs. When we used the classification tree and random forest supervised classification algorithms to analyze microarray data, we derived general ''efficacy profiles'' of biomarker gene expression that correlate with antidepressant, antipsychotic and opioid drug action on primary human neurons in vitro. These profiles were used as predictive models to classify naïve in vitro drug treatments with 83.3% (random forest) and 88.9% (classification tree) accuracy. Thus, the detailed information contained in genomic expression data is sufficient to match the physiological effect of a novel drug at the cellular level with its clinical relevance. This capacity to identify therapeutic efficacy on the basis of gene expression signatures in vitro has potential utility in drug discovery and drug target validation.pharmacogenomics ͉ predictive efficacy ͉ drug screening M icroarray-based gene expression patterns can be used as fingerprints of cellular physiology. The variety of cellular physiologies discernable by gene expression profile fingerprinting is expanding as an increasing range of cell types and cellular manipulations are investigated, and statistical methods of expression profile classification are refined. In yeast, distinctive profiles of genomic expression have been used to characterize cellular responses to diverse environmental transitions (1), functionally classify genetic manipulations, and discover a novel target for a drug of partially characterized function (2). In cancer studies, microarray data has been used to classify solid tumors (3), correlate tumor characteristics to clinical outcome (4), and cluster cell lines on the basis of their tissue of origin and response to drugs (5-9). In the area of toxicogenomics, large-scale gene expression analysis of toxin-treated cells and animals has yielded a highly accurate capacity to recognize the toxic potential of novel drug candidates (10-14), resulting in an increase in the efficiency of drug triage in the pharmaceutical development pipeline.Multiple statistical methods have been applied to classification and recognition of expression profiles. Supervised classification analysis methods, which can classify patterns of novel data based on prior knowledge of sample classes, include linear discriminant analysis and genetic algorithm/K-nearest neighbors (11, 15), Fisher discriminant analysis (16), support vector machines (17), neural networks (18), and tree-based analysis (19). Here, we use human primary neurons treated with multiple members of multiple classes of antidepressan...

A Data Mining Approach to In Vivo Classification of Psychopharmacological Drugs

“…This type of strategy has recently been used in a human neuronal precursor cell line as a means to discover gene-expression patterns predictive of a drug's psychoactive class. In these studies (Gunther et al, 2003(Gunther et al, , 2005 various classification algorithms were used to discover geneexpression profiles that were predictive of antidepressant, antipsychotic, and opioid drug action. Although this is a significant step forward, an in vitro system is necessarily restricted in its ability to mimic a true system pharmacology response to psychiatric drugs, and would ideally be complemented with an in vivo assay for more advanced preclinical testing.…”

“…In this instance specific mechanisms do not need to be invoked, the only criterion is that the gene(s) are highly predictive of the outcome. This type of data mining approach was recently utilized in vitro to discover geneexpression patterns that could be used for the classification of psychoactive drugs (Gunther et al, 2003(Gunther et al, , 2005. It has been proposed that the same approach would also be advantageous for in vivo analysis of behavioral data (Brunner et al, 2002;Tecott and Nestler, 2004), but no such results have been published to date.…”

Reference profiling of the genomic response induced by an antimicrotubule agent, TZT-1027 (Soblidotin), in vitro

“…20,21 Drug-to-drug comparative approaches using microarray analyses are useful for identifying drug targets; the cellular effects caused by a novel drug of incompletely characterized specificity can be matched to 'reference profiles' of the cellular effects elicited by the specific inhibition of candidate analog-sensitive drugs. 22,23 Thus, it has been proposed that phenotypic information generated by drug-induced alterations in gene expression can be matched to discrete interactions between the compound and the relevant protein targets. Using the drug-to-drug comparative approach of the microarray analysis, we obtained reference profiles of genomic expression data from cellular responses in a lung cancer cell line to antimicrotubule drugs, including five conventional agents and the mother compound D10.…”

A quantitative genomic expression analysis platform for multiplexed in vitro prediction of drug action

“…11 However, previous chemical genomic studies with mammalian cells have resolved only a single type of drug activity 9,10 or employed primary cell coculture systems apt to exhibit batch-to-batch inconsistency. 11 For the potential of this discipline to be met, it will be necessary to establish a simple cellular platform capable of evaluating multiple distinct clinical efficacies at once for each compound screened. It will also be necessary to develop a practical, quantitative means to prioritize the most promising hits for development as therapeutics.…”