Aims: Liver cytochromes (CYPs) play an important role in drug metabolism but display a large interindividual variability resulting both from genetic and environmental factors. Most drug dose adjustment guidelines are based on genetics performed in healthy volunteers. However, hospitalized patients are not only more likely to be the target of new prescriptions and drug treatment modifications than healthy volunteers, but will also be more subject to polypharmacy, drug-drug interactions, or to suffer from disease or inflammation affecting CYP activities.

Methods: We compared predicted phenotype based on genetic data and measured phenotype using the Geneva cocktail to determine the extent of drug metabolizing enzyme variability in a large population of hospitalized patients (>500) and healthy young volunteers (>300). We aimed to assess the correlation between predicted and measured phenotype in the two populations.

Results: We found that, even in cases where the genetically predicted metabolizer group correlates well with measured CYP activity at group level, this prediction lacks accuracy for the determination of individual metabolizer capacities. Drugs can have a profound impact on CYP activity, but even after combining genetic and drug treatment information, the activity of a significant proportion of extreme metabolizers could not be explained.

Conclusions: Our results support the use of measured metabolic ratios in addition to genotyping for accurate determination of individual metabolic capacities to guide personalized drug prescription.