Abstract Purpose The aim of present study is to measure plasma clozapine (CLZ) and N-desmethyl clozapine (DMC) as biomarkers to correlate drug concentrations with the appearance of preclinical adverse hematic effects. Methods A high-performance liquid chromatographic method, using a diode-array (ultraviolet) detector, was validated to obtain reliable concentrations of CLZ and DMC, its main metabolite, in plasma of 41 schizophrenic patients taking CLZ. Blood neutrophils and leucocytes counting were concurrently assessed as a proxy to subclinical adverse reactions. Results The analytical method employed was linear, reproducible, and stable to measure concentrations of CLZ between 30 and 1000 ng/mL, while 12.5–560 ng/mL of the metabolite. The method allowed us to correlate CLZ plasma concentrations, the time taking CLZ and CLZ dose as determinants of neutrophils’ counting with a R2 = 0.447, using a multiple regression analysis model. Likewise, the correlation of leucocyte counting vs CLZ plasma levels and CLZ time, showed a R2 = 0.461. DMC correlated significantly with both neutrophils and leucocytes counting, but was excluded from the regression when CLZ concentration was included in the model. Finally, no other hematological adverse reactions were recorded. One patient presented a cardiovascular complication. The negative correlation between clozapine and neutrophil count observed in patients, suggest that CLZ itself, but not DMC, could be related to hematologic side-effects. Conclusion The findings of this study, demonstrate for the first time, that plasma levels of CLZ and time taking the drug are independent determinants of blood neutrophils and leucocytes, so the monitoring of plasma CLZ may be useful in the clinic practice to determine safe dosing of the drug.