Translation of off-target effects: prediction of ADRs by integrated experimental and computational approach

Abstract

Adverse drug reactions (ADRs) are associated with most drugs, often discovered late in drug development and sometimes only during extended course of clinical use. They are linked either to the therapeutic target or pathway, or could emerge as the consequence of known or unknown off-target effect(s) of a drug or drug combinations. ADRs are a major burden on patients, medical professionals and the society in general. Discovery of intolerable ADRs during clinical trials significantly contributes to high attrition rates with associated rising costs. Thus, prediction of ADRs at the early stage of drug discovery is an emerging approach; however, it remains a challenging task to identify the mode of action of drug candidates which might lead to ADRs. We review here the implementation of in vitro and in silico tools streamlined for the prediction of ADRs as early as the target/lead identification and lead optimization phases of the drug discovery process. This integrated approach has been developed during the past decade by both academic institutions and the pharmaceutical industry with the aim to provide toxicological analysis, assessment and ranking of drug candidates on a broad scale. The major aim is to be able to mitigate targets associated with ADRs earlier and guide chemistry to address the therapeutic and side effects in parallel. The major components of this effort are (1) experimental approach: early in vitro safety profiling linked to (2) computational toxicology algorithms and models utilizing statistics, data mining, cheminformatics and system biology. The third component embraces the translational aspect for clinical ADRs, which includes in vivo exposure. In this review we focus on the prediction of the integrated molecular network approach.