Clozapine:norclozapine ratio as a marker of metabolic risk and therapeutic response and effect of valproate on clozapine levels.
Rask Susanna Maria SM, Lamminmäki Miina M, Kampman Olli O, Hämäläinen Mari M et al.
Clozapine is the most effective antipsychotic in treatment-resistant schizophrenia. Clozapine causes weight gain and increased diabetes risk. The role of its metabolite norclozapine and clozapine:norclozapine ratio (CLZ/NCLZ) in metabolic effects has been debated. Also, the role of norclozapine in efficacy of clozapine treatment has been an area of interest. Valproate raises CLZ/NCLZ, but its impact on clozapine levels has been unclear. The aims of this study were to (1) examine associations between CLZ/NCLZ and metabolic markers, (2) assess whether concomitant valproate is associated with clozapine and norclozapine serum concentrations and CLZ/NCLZ and (3) evaluate whether CLZ/NCLZ is associated with treatment response, using antipsychotic or mood stabilizer augmentation as a proxy for treatment resistance. We studied 237 psychosis patients on clozapine, with weight gain data for 228 patients and serum levels and metabolic markers for 190 patients. The studied outcomes were weight gain, hyperglycaemia or diabetes and dyslipidaemia. Effects of clozapine concentration/dose (C/D), norclozapine C/D and CLZ/NCLZ on outcomes were analysed, adjusting for age, sex and smoking. CLZ/NCLZ was compared between augmented and clozapine-only patients to assess treatment response. Lower CLZ/NCLZ ratios did not predict weight gain, hyperglycaemia or dyslipidaemia. CLZ/NCLZ was not associated with polypharmacy. Valproate showed a non-significant increase of clozapine C/D by 21.8% (95% confidence interval [CI] -0.3% to 49.0%, p = .054) and a significant increase in CLZ/NCLZ by 76.1% (95% CI +58.8% to +95.2%, p < .001). CLZ/NCLZ is not linked to metabolic outcomes or treatment response in our study. Valproate elevated CLZ/NCLZ significantly. We observed higher clozapine serum levels among patients receiving valproate, but the association was not statistically significant.