Prediction of Drug-Drug Interactions

点击翻译按钮获取中文摘要

Adverse drug reactions are the fourth leading cause of death in the US and account for 5% of hospital admissions. A large percentage of these adverse events result from drug-drug interactions and the chance of experiencing such an interaction represents one of the greatest fears of patients visiting their doctor. It is common for patients to receive 4 or more drugs simultaneously and 10 or more is common among the growing population of elderly patients in nursing homes. Despite this ubiquitous polypharmacy and social impact of drug-drug interactions there has been no structured attempt to predict and therefore manage complex multi-drug interactions. We are proposing to take the first step in remedying this shortfall by studying "ternary" drug interactions occurring within mixtures of 3 drugs. We will focus on metabolic drug interactions at the level of the CYP3A enzymes in the liver and intestinal wall because these represent the single most common cause of clinically important drug-drug interactions. We will first quantify the time course and concentration dependence of the induction of intestinal and hepatic CYP3A enzymes by the prototypical inducer, rifampin. We will use intravenous midazolam to reflect hepatic CYP3A activity and intestinal pinch biopsies to reflect intestinal CYP3A activity. This will allow us to build a predictive, physiologically based pharmacokinetic model of CYP3A induction by rifampin. In the subsequent studies we conduct ternary drug interaction studies. These studies will test the hypotheses that the effect of two CYP3A modulators given simultaneously is predictable from the individual binary interactions. We will use intravenous and oral midazolam as probes of intestinal and hepatic CYP3A activity. These ternary interactions will include combinations of inhibitors and combinations of inhibitor and the inducer, rifampin. We will develop physiologically based pharmacokinetic models of each of the drugs involved in the ternary interactions to examine the predictability of the interactions. We will also test the hypothesis that the ternary in vivo interactions can be predicted from in vitro data. The interactions between inhibitors, inducer and substrate will be quantified in subcellular fractions and cultured cells. The in vitro parameter estimates will be incorporated into our physiological pharmacokinetic models to test the predictive power and build a universal platform for complex interactions between chemicals.