INTRODUCTION AND OBJECTIVE: The frequency of agranulocytosis induced by psychoactive drugs is estimated the first year of around 0.8% under clozapine, against 0.13% under chlorpromazine (King and Wager, 1998 ). It is associated with a mortality rate of 5 to 10%, and requires heavy treatment, usually in an intensive care unit. The objective of this paper is to present a practical therapeutic answer (clozapine rechallenge with filgrastim) through a case report following a neutropenia episode preventing clozapine use. CASE AND METHODS: B.N. aged 35, native of Martinique, shows a resistant schizophrenia disorder "ultra sensitive" to clozapine. Without any treatment, after 4 years in stable clinical state under clozapine, B.N. suffered three neutropenia episodes when absorbing clozapine (2008, 2010 and 2011). First, a literature survey was conducted along with a consultation of the head of pharmacovigilance regional center and the hematology referee. Then, a 4th clozapine treatment was decided under cover of filgrastim (G-CSF), the role of which is to limit the risk of a new neutropenia. After stopping all psychoactive drugs, except morphine, the subject benefited from a first 0.3mg filgrastim injection, the day before re-introducing 25mg clozapine. Before treatment: Leucocytes=4.8 G/L while absolute neutrophils count=2.4 G/L. Filgrastim injections were carried out at a rate of two 0.3mg injections per week. Clozapine was increased to reach 25mg every 3 days and electroconvulsivotherapy continued fortnightly while supervision was double: on the first hand, daily and clinical search for an increase in body temperature and signs of treatment intolerance, and on the other hand biological surveillance with NFS three times a week besides weekly clozapinemia. The well-informed consent of the patient was obtained. RESULTS: Signs of improvement were noticed as early as the 8th day and after 8 weeks of treatment and 31 sessions of ECT, the patient was stabilized under clozapine at 300mg per day. The evolution is clearly favorable, as PANNS evolved from 158 to 90. Neutropenia episodes were not observed with a lowest measured rate of 1.9 G/L neutrophils. The filgrastim dosage was then reduced to 0.3mg per week from the 7th week onwards, along with the pursuit of a weekly NFS supervision throughout the treatment. Tolerance is satisfying, with an improvement in lipid check, glycaemia, blood pressure and QT intervals during ECG. DISCUSSION AND CONCLUSION: The B.N. case isn't an isolated one as several articles refer to filgrastim use, combined with clozapine. This confirms the role of hematopoietic cytokines (mainly G-CSF) in neutropenia episodes induced by clozapine. Filgrastim dosage appears to be an important point with regards to the risk of a new neutropenia episode. Let's mention also that it is not a harmless treatment, it could hide the occurrence of neutropenia, besides it's expensive and invasive. Clinical and biological supervision is essential as the probability of an enhanced malignant hemopathy is low but nonetheless present. We also noticed a "biased notoriety of the clozapine", with the association with other hematotoxic molecules, the existence of a circadian rhythm of neutrophils or G-CSF, along with transitional or ethnical neutropenia. These points should be discussed thoroughly before exclusively accusing clozapine; this in turn would have consequences regarding the possibility of treatment resumption. Finally, association with lithium is also an option; several cases have already been reported.