966 Feasibility of 5-FU therapeutic monitoring

5-FV therapeutic monitoring was performed, in 26 patients with localized or disseminated epidermoïd tumour of various origin, during 64 chemotherapy cycles containing 5-FV 1000mg/m2in continuous infusion (J1–J5) and CDDP (100mg/m2J1 or 20mg/m2/j J1–J5). Blood samples were collected daily (8 a.m., 4 p.m.). 5-FV HPLC analysis used the method of Christophidis. Dose reduction of 5-FV was programmed according to the method of R. Fety using the J1–J2 and the J1–J5 5-FV area under the curve (AUC). An average of 2 cycles was administered.During the 1st cycle: J1–J2 5-FV AUC averaged 15751 μg 1−1h−1± 12309 (3902–56620) confirming the great interpatient variability. In 4 patients J1–J2 5-FV AUC 20000 μg 1−1h−1obliged to cancel chemotherapy at β. J1–J5 5-FV AUC averaged 46161 μg 1−1h−1±20020 (18380–90200). We observed a 5-FV accumulation process, characterised by an increase of daily 5-FV AUC in 18 patients. 5-FV dose reduction was scheduled in 27 cases and necessitated a further decrease during the chemotherapy cycle in 9 cases. 5-FV monitoring allowed a reduction in the toxicity which were less frequent for the cycles with J1–J2 5-FV AUC<20000μg 1−1h−1or J1–J5 5-FV AUC <30000μg 1−1h−1. Fourteen objective responses were obtained with 2 complete responses. J1–J5 5-FV AUC did not differ between responders and non responders.These time consuming techniques must find their role during more prolonged chemotherapy.