Title

 

Authors

T. Di Desidero¹, G. Allegrini², M. T. Barletta³, A. Fioravanti¹, P. Orlandi¹, B. Canu¹, S. Chericoni⁴, A. Di Paolo¹, F. Loupakis³, G. Masi³, M. Giusiani⁴, M. Del Tacca¹, R. S. Kerbel⁵, A. Falcone³, R. Danesi¹, G. Bocci¹.
¹Div. of Pharmacology, Department of Internal Medicine, University of Pisa, Italy, ²Div. of Medical Oncology, Azienda USL 5, Pisa, Pontedera, Italy, ³Dept. of Oncology, Transplants and New Technologies in Medicine, University of Pisa, Italy, ⁴Div. of Legal Medicine, Dept. of Neuroscience,University of Pisa, Italy, ⁵Sunnybrook Health Sciences Centre, Molecular and Cellular Biology Research, Toronto, Canada.

 

Abstract

Recent preclinical and phase II clinical trial results indicate promising activity of the concurrent daily low-dose metronomic chemotherapy using 5-FU prodrugs (e.g. UFT or capecitabine) with cyclophosphamide, alone or in combination with an agent such as bevacizumab for the treatment of metastatic breast cancer. We decided to test clinically this type of metronomic chemotherapy combination with celecoxib for the treatment of heavily pre-treated advanced GI malignancies. Metronomic UFT, CTX and CXB (CTX 500mg/mq² i.v bolus at day 1; by day 2, CTX p.o. 50 mg daily, UFT p.o. 100 mg/twice a day and CXB p.o. 200 mg/twice a day)along with investigating the pharmacokinetic (PK) parameters of UFT and their metabolites, and modulation of CD133 gene expression, as well as VEGF, TSP-1 and VE-cadherin plasma levels as possible pharmacodynamic markers of the therapy. Thirty-eightpatients (30 colorectal, 2 pancreatic, 3 biliary tracts, 2 gastric, 2 liver) entered the study. Patients characteristics were: M/F = 25/13, median age = 71 years,   PS 0/1/2 = 13/24/1; sites of disease (single/multiple) = 7/31; median of previous lines of chemotherapy: 3 (range 1-5). Drug concentrations were investigated by HPLC, whereas plasma levels of VEGF, TSP-1 and VE-cadherin by ELISA. Thirty-eight patients were assessable for response: 17 patients (45%) obtained stable disease with a median duration of  5.8 ms (range, 4.2–7.4). After a median follow-up of 18.3 ms (95% confidence interval (CI), 10-26.6 ms), median progression free survival (PFS) and median overall survival (OS) were 2.7 ms (95% CI, 1.6-3.9 ms) and 7.1 ms (95% CI, 4.3-9.9 ms), respectively. No toxicities of grade >1 (NCI scale) were observed. Both VEGF and VE-cadherin plasma levels were significantly greater in the progressive disease (PD) group when compared to the stable disease (SD) group. CD133 gene expression (assessed by real time PCR) increased during the treatment in the PD group of patients. The PK analysis of 27 pts (13/14, SD/PD) after the first intake of UFT revealed a significant difference between the PD and SD groups at day 1 in 5-FU (the main UFT active metabolite) AUC (0.997±1.271 vs. 2.765±1.709h·μg/ml, respectively, p<0.05) and Cmax (0.453±0.573 vs.1.134±0.749μg/ml respectively, p<0.05). Nonparametric receiver operating characteristic (ROC) analysis was used to find cut off values for each PK parameter.In particular, the 5-FU AUC and Cmax greater than 1.313 h·μg/ml and 0.501 μg/ml, predicts stabilization of disease and a prolonged PFS and OS.These results indicate that metronomic UFT, CTX with CXB in heavily pre-treated GI patients was feasible and clinically active. Significant differences in PK and PD parameters were found between responders and non-responders; the calculated cut-off  PK values could be particularly important for future clinical studies for this regimen.