Timed flat infusion of 5-fluorouracil increases the tolerability of 5-fluorouracil/docetaxel regimen in metastatic breast cancer: a dose-finding study

A dose-finding study was undertaken to determine the maximum-tolerated dose, and the recommended dose of docetaxel in combination with 12-h timed (22:00–10:00) flat infusion of 5-fluorouracil (5-FU) in metastatic breast cancer patients. This schedule seems to reduce the occurrence of stomatitis of the docetaxel and infusional 5-FU regimen.


PATIENTS AND METHODS
Patient eligibility criteria were: histologically or cytologically documented breast cancer and proven metastatic or recurrent breast cancer; measurable or evaluable metastatic disease; WBC count X4 Â 10 3 mm À3 , neutrophils X2 Â 10 3 mm À3 , platelets X100 Â 10 3 mm À3 , hemoglobin X10 g dl À1 , serum creatinine p1. 2 mg dl À1 , total bilirubin p1.5 times the upper normal limit, AST and ALT p1.5 times the upper normal limit; age between 18 and 75 years; World Health Organisation (WHO) performance status p2. The study was approved by the Local Ethics Committee. Prior chemotherapy for metastatic disease was not allowed. The exclusion criteria included: peripheral neuropathy, uncontrolled infection, diabetes and cardiac disease.
The treatment schedule consisted of a 1 h i.v. dTX (Taxotere s ) infusion on day 1 and a 12-h timed (22:00-10:00) flat i.v. 5-FU (Fluorouracil TEVA s ) infusion, over 5 days, every 21 days. Four escalation dose levels of dTX plus 5-FU were planned: 5-FU 700 mg m À2 day À1 associated to dTX 80 and 85 mg m À2 in the first two dose levels, respectively; then, 5-FU dose levels were increased to 800 and 900 mg m À2 day À1 in the other two steps. Implantation of a venous access device was required for 5-FU administration via a programmed portable pump (Cadd-Plus, Sevit) that administered 5-FU at a given constant rate for a period of 12 h. Treatment was routinely administered on an outpatient basis. The planned doseescalation strategy combined the intra-and interpatient approach (Simon et al, 1997). The MTD was defined as the dose at which at least 50% of patients developed dose-limiting toxicity (DLT). In case of DLT, as defined in the treatment plan, treatment was continued at the dose level immediately below. This dose was the RD for phase II trials. Granulocyte colony-stimulating factor (G-CSF) was administered at the first occurrence of grade (G)4 neutropenia due to its high expected rate, in order to prevent febrile neutropenia and treatment delays; in the subsequent cycles, prophylactic G-CSF treatment was administered for 5 days, after the occurrence of G4 neutropenia. Complete blood cell count was performed on days 1, 6, 8, 12, every cycle.
Dose-limiting toxicity included: febrile neutropenia requiring i.v. antibiotics or G4 neutropenia resistant to G-CSF administration (failure to recover neutrophils X1.5 Â 10 3 mm À3 on day 21 of each cycle); grade 4 thrombocytopenia; hemoglobin o6.5 g dl À1 ; grade 3 -4 nonhaematological toxicity (excluding alopecia and nausea). Toxicity was graded according to the National Cancer Institute Common Toxicity Criteria (NCI-CTC). Before entering the study, each patient underwent medical history, physical examination, complete blood cell count, serum biochemistry, computed tomography of the chest and abdomen, bone scan, an electrocardiogram and other investigations as clinically indicated. Tumour imaging was repeated every three treatment cycles.
Tumour response was assessed according to WHO response criteria. Time to disease progression and survival were assessed using the methods of Kaplan and Meier.

RESULTS
Fourteen patients were enrolled. A summary of baseline patient characteristics is illustrated in Table 1. A total of 88 cycles were administered, with a median of six cycles per patient (1 -12); the median number of dose-escalations per patient was 2 (1 -4).
The fourth dose level (dTX 85 mg m À2 and 5-FU 900 mg m À2 day À1 ) represented the MTD (Table 2): ten patients were treated and 27 cycles were administered. Dose-limiting toxicities were observed in five of 10 patients (50%): two patients experienced grade (G)4 diarrhoea; two patients, G3 diarrhoea; one patient, G3 mucositis (10%) and febrile G3 neutropenia. Each of the five patients received subsequent cycles at the third dose level without showing DLTs.
The third dose level (dTX 85 mg m À2 and 5-FU 800 mg m À2 day À1 ) represented the RD: 12 patients were treated and 34 cycles were administered. Dose-limiting toxicities were not observed and chemotherapy was well tolerated (Table 3): Grade 1 -2 diarrhoea in 42% of patients and 21% of cycles; G1 -2 stomatitis in 33% of patients and 53% of cycles; G 1 -2 nausea/vomiting was observed in 50% of patients and 32% of cycles. Grade 2 alopecia was present in all patients. No toxic death was observed. One case of thrombosis related to the venous access device was observed.
At the RD, G4 neutropenia was observed in 50% of patients (six of 12) and in 47% of cycles (16 of 34); G3 -4 neutropenia 79% (27 of 34). Overall, it has been observed in 36% (32 of 88) of cycles. Patients showing G4 neutropenia were all treated with G-CSF. The median time to neutrophil nadir was on day 8 and the median duration of G-CSF administration was 5 days. One patient experienced febrile G3 neutropenia at MTD.
In all, 13 patients were assessable for treatment efficacy. Two complete and six partial responses were observed giving an overall response rate of 61% (a: 0.05, CI: 728). The median time to progression was 10 months (4 -28 months) and the median overall survival was 25 months (4 to 46 þ months).

DISCUSSION
The RD for phase II studies of this dTX/5-FU schedule (dTX 85 mg m À2 , day 1; 5-FU 800 mg m À2 day À1 in 22:00 -10:00 TFI) shows a projected dose intensity equivalent to that proposed by Lortholary et al (2000) (dTX 85 mg m À2 day 1; 5-FU 750 mg m À2 day À1 5-day continuous infusion, every 3 weeks) (). These RDs were well tolerated with the use of G-CSF. No DLTs were observed in the 12 treated patients and 34 administered cycles. Nonhaematological toxicity was characterised by G1 -2 diarrhoea and stomatitis. The earlier dose-finding studies of docetaxel in combination with 5-FU in advanced breast cancer show that mucositis (stomatitis and diarrhoea) and neutropenia represented the DLTs at the MTD (Ando et al, 1998;Petit et al, 1999). In our study and in that by Lortholary et al, neutropenia was common but recovered after medical G-CSF management.
In the study by Lortholary et al, few instances of febrile neutropenia were observed and stomatitis represented the DLT at the MTD; at the RD, 40% (two of five) of patients experienced G3 dose-limiting stomatitis and G4 neutropenia was reported in four of five patients (Lortholary et al, 2000). The recently published phase II study by Lortholary confirms that: neutropenia is the most common toxic event (G3 -4, 54% of patients; febrile neutropenia, 24% of patients); stomatitis G3-4 is a frequent event (26% of patients, approximately 6% of cycles); G3 diarrhoea occurs in 7% of patients (1% cycles) (Lortholary et al, 2003). Even if in the present study the use of G-SCF did not allow to report the duration of neutropenia, no instance of febrile neutropenia was observed at the RD.
In the present study, at the RD, nonhaematological toxicity was characterised by G1 -2 diarrhoea (42% of patients and 21% of the cycles) and G1 -2 stomatitis (33% of the patients and 53% of the cycles). The present data suggest that the 22:00 -10:00 TFI of 5-FU may increase the nonhaematological tolerability of the 5-FU/dTX  Among the oral fluoropirimidine formulations, capecitabine mimicks protracted infusion of 5-FU. The present data are particularly significant to this concern in view of a potential comparison between the most tolerable and effective schedules of dTX and 5-FU or capecitabine in metastatic breast cancer.
5-Fluorouracil/docetaxel in metastatic breast cancer C Ficorella et al