Can extremely low or high morphine formation from codeine be predicted prior to therapy initiation?

Activation of codeine by O-demethylation into morphine is a prerequisite for its analgesic effects and severe toxicity. Identifying patients in whom morphine is formed either at extremely low or at extremely high amounts may improve efficacy and safety of codeine therapy. To assess how well this identification is possible, we compared the performance of current CYP2D6 phenotype association systems (traditional genotype-based classification, a recently proposed CYP2D6 activity score, and the plasma dextromethorphan metabolic ratio) in 57 healthy Caucasians after oral administration of 30 mg dextromethorphan hydrobromide or 50 mg codeine. Most subjects (87.5%) at the lower 15% of morphine formation from codeine and thus likely to not to respond to codeine therapy were correctly identified by CYP2D6 genotype- or phenotype-based systems. In contrast, in subjects at the upper 15% of morphine formation being at risk for opioid toxicity, CYP2D6 genotyping predicted only the 50% who carried gene duplication, whereas dextromethorphan-based phenotyping identified 67.5% of the subjects with high morphine formation. However, satisfactory prediction (87.5%) of high morphine formation was only achieved when combining genotyping with phenotyping. In conclusion, insufficient morphine formation from codeine and thus likely failure of analgesia can currently be well predicted. However, to make codeine therapy safe, extremely high morphine formation has to be predicted as well, which has to be obtained at the effort of combining genotyping with phenotyping.