Assessment of a Candidate Marker Constituent Predictive of a Dietary Substance-Drug Interaction: Case Study with Grapefruit Juice and CYP3A4 Drug Substrates

Dietary substances, including herbal products and citrus juices, can perpetrate interactions with conventional medications. Regulatory guidances for dietary substance-drug interaction assessment are lacking. This deficiency is due in part to challenges unique to dietary substances, a lack of requisite human-derived data, and limited jurisdiction. An in vitro-in vivo extrapolation (IVIVE) approach to help address some of these hurdles was evaluated using the exemplar dietary substance grapefruit juice (GFJ), the candidate marker constituent 6',7'-dihydroxybergamottin (DHB), and the purported victim drug loperamide. First, the GFJ-loperamide interaction was assessed in 16 healthy volunteers. Loperamide (16 mg) was administered with 240 ml of water or GFJ; plasma was collected from 0 to 72 hours. Relative to water, GFJ increased the geometric mean loperamide area under the plasma concentration-time curve (AUC) significantly (1.7-fold). Second, the mechanism-based inhibition kinetics for DHB were recovered using human intestinal microsomes and the index CYP3A4 reaction, loperamide N-desmethylation (K-I [concentration needed to achieve one-half k(inact)], 5.0 +/- 0.9 mu M; k(inact) [maximum inactivation rate constant], 0.38 +/- 0.02 minute(-1)). These parameters were incorporated into a mechanistic static model, which predicted a 1.6-fold increase in loperamide AUC. Third, the successful IVIVE prompted further application to 15 previously reported GFJ-drug interaction studies selected according to predefined criteria. Twelve of the interactions were predicted to within the 25% predefined criterion. Results suggest that DHB could be used to predict the CYP3A4-mediated effect of GFJ. This time- and cost-effective IVIVE approach could be applied to other dietary substance-drug interactions to help prioritize new and existing drugs for more advanced (dynamic) modeling and simulation and clinical assessment.