Systematic Approach to Organizing Structural Alerts for Reactive Metabolite Formation from Potential Drugs

Reactive metabolites are widely accepted as playing a pivotal role in causing idiosyncratic adverse drug reactions (IDR). However, much is unknown about the biological mechanisms of IDR, although the initiating event in most cases is an attachment of a reactive intermediate to macromolecules leading to immune-mediated responses. Reactive metabolites are also involved in many mutagenesis/carcinogenesis events by reacting with DNA. Drug designers thus have reasons to make large efforts to avoid making test compounds having a liability to generate reactive metabolites. In this Perspective we argue for using structural alerts (SA) as the most straightforward way to link forecasting about chemical hazards of planned test compounds to the accumulated knowledge base. Although many SAs have been widely recognized and reviewed previously, there are also a lot of observations that have no readily discernible chemical interpretation. For drug designers to benefit from all published data, the knowledge has to be organized in a way that is readily searchable starting with a query structure. We propose that an increased number of structural alerts with more details should be applied to obtain improved decision support. The association of selected SAs with reference drugs, whose proposed or hypothesized activation mechanisms build the knowledge base, should be readily available in a format comprising of small summaries with included hyperlinks for quick access to the original literature, as outlined in the TOC illustration. Since some structural alerts are present in drugs that do not cause idiosyncratic adverse reactions or drug-drug interactions, it is important to elaborate on the reasons for this discrepancy as much as possible.