Clozapine and therapeutic drug monitoring: is there sufficient evidence for an upper threshold?

Clozapine levels are advocated in the monitoring of patients on this drug and have now been used for a number of years. A safety-related threshold has also been proposed, as well as therapeutic lower and upper thresholds. While there has been reasonable consensus regarding a lower therapeutic threshold, this is not the case for the upper thresholds. Our aim was to review available evidence related to upper thresholds. We carried out an electronic search of different databases and a manual search of articles between 1960 and 2011, cross-referencing the following terms with clozapine-interactions, monitoring, pharmacokinetics, plasma levels, serum levels, and toxicity. Sixty-nine articles met our search criteria and these could be divided into reviews (11), studies (24), and case reports (35). Study quality was evaluated, and none met criteria for a prospective, randomized controlled trial specifically addressing higher plasma levels, e.g., >500 ng/ml. Case reports emphasize in particular the impact of interactions, e.g., antidepressants and smoking. There is clear evidence indicating a dose-related increased risk of seizures, at least to 500-600 mg/day, but a lack of data to suggest such a relationship between plasma levels, dose, and side effects linked to safety, e.g., seizures, myocarditis, and agranulocytosis. The very limited evidence addressing an upper threshold related to clinical response suggests a "ceiling effect" in the range of 600-838 ng/ml. It appears that the current safety-related threshold is not supported by evidence. There may be an upper threshold for clinical response, beyond which chance of response falls off, although further studies are warranted.