This study was done to examine changes of plasma homovanillic, acid(HVA), 5-hydroxyindoleacetic acid(5-HIAA), and HVA/5-HIAA ratio during an 8-week clozapine trial and to investigate the relationship between the plasma monoamine metabolites and treatment responses. Twenty-seven chronic schizophrenic patients were treated for 8 weeks with clozapine. The psychopathology was assessed at baseline just clozapine trial and then every 2 weeks until the end of 8-week clozapine treatment using the positive and Negative Syndrome Scale(PANSS) and the Clinical Global Impression scale(CGI). The plasma HVA and 5-HIAA levels were measured also biweekly using high preformance liquid chromatography with electrochemical detection method. Plasma HVA and 5-HIAA levels were significantly decreased during a 8-week chozapine treatment, although plasma HVA/5-HIAA ratio showed no significant change. The changes of plasma HVA levels were in significant correlations with the changes of PANSS positive scores, of general psychophathology scores, and changes of total scores. The changes of plasma 5-HIAA levels were in significant correlations with the changes of PANSS negative scores. But the changes of plasma HVA/5-HIAA ratio had no significant correlation with any PANSS subscale score changes. 48% of the patients treated with clozapine was categorized as responders, who showed at least a 20% decrease in PANSS total score and a CGI severity score of mildly ill or less(< or =3) of the end pint of the study. The baseline plasma HVA levels and HVA/5-HiAA) ratio were significantly higher in responders(N=13) than in nonresponders(N=14). But no significant difference in the change of plasma HVA between responders(40.3% decrement) and nonresponders(3.1% increment). But no significant difference in the change of plasma 5-HIAA and the change of plasma HVA/5-HIAA ratio between responders and nonresponders were observed. These results suggest that the antipsychotic effect of clozapine on positive symptoms may be associated with dopaminergic blocking activity, and that on negative symptoms may be associated with serotonergic blocking activity. The baseline plasma HVA levels and the change of HVA levels from baseline may be useful predictors of treatment response with clozapine.