Clinical, pharmacokinetic and pharmacodynamic evaluations of metronomic UFT and cyclophosphamide plus celecoxib in patients with advanced refractory gastrointestinal cancers

Main eligibility criteria included: (1) histologically confirmed diagnosis of colorectal or other gastrointestinal adenocarcinoma with metastatic disease; (2) previous chemotherapy with fluoropyrimidines, oxaliplatin, irinotecan, when indicated; (3) measurable disease progressing during or within 3 months from the end of the treatments; (4) life expectancy greater than 3 months; (5) Eastern Cooperative Oncology Group performance status ≤ 2; (6) adequate bone marrow, renal and liver function (leukocytes ≥3,000 mm⁻³, platelet ≥100,000 mm⁻³, creatinine ≤2 mg/dL⁻¹, total bilirubin ≤ 1.5x institutional upper limit of normal, AST/ALT ≤ 5x institutional upper limit of normal). Study exclusion criteria were as follows: brain metastasis, symptomatic cardiac disease, recent myocardial infarction, active infections and inflammatory bowel disease.

Patients received on day 1 a single administration of CTX 500 mg/m² as i.v. bolus and, from day 2, 50 mg CTX p.o. once daily plus 100 mg UFT p.o. and 200 mg CXB p.o. twice a day. From day 2, the treatment was continued without interruption until either disease progression, unacceptable toxicities, deterioration of performance status, or patient refusal to continue treatment. No dose reduction for toxicities was applied. To prevent nausea and vomiting, metoclopramide 10 mg i.v plus dexamethasone 4 mg i.v. were administered before CTX i.v. chemotherapy on day 1. Loperamide 2 mg orally every 2 h and oral rehydration were prescribed in the event of delayed diarrhea. No prophylactic treatment with supportive cytokines such as G-CSF was recommended.

Pretreatment evaluation included complete history and physical examination, performance status assessment, complete blood count and differential, platelet count, complete blood profile, tumor markers, urinalysis, ECG, chest X-ray or computed tomography scan, abdominal computed tomography scan and/or sonogram, and any other appropriate diagnostic procedure to evaluate metastatic sites. During treatment, a physical examination, a complete blood cell count, blood profile, urinalysis and toxicity evaluation were performed every 3 weeks. Sites of metastatic disease were radiologically re-evaluated every 2 months, according to the RECIST criteria [21]. A chest X-ray and/or an abdominal sonogram were repeated at least every 6 months if there was no evidence of lung or abdominal disease, respectively. Toxicities were scored according to the standard NCI Common Terminology Criteria for Adverse Events (version 3.0). Duration of responses was calculated from the first day of treatment to the date of first observation of progressive disease or last examination.

The pharmacokinetic analysis of tegafur, 5-FU, 5-FUH₂, GHB and uracil were performed as previously described [22‐24] with minor modifications. Blood samples (4 ml each) for pharmacokinetic assays were taken from an indwelling i.v. cannula placed in an antecubital vein at baseline, 30 min, 1, 1.5, 2, 3 and 5 h at day 1, 28 and 56 after the beginning of UFT oral administration. Blood was centrifuged (5 min, 4,000 rpm, 4°C) to separate plasma, which was stored at −80°C and assayed within 1 week. The simultaneous assay of 5-FU and 5-FUH₂ in human plasma was performed by a validated, nonradioactive reverse-phase HPLC method with ultraviolet detection. Briefly, 0.5 ml of plasma, mixed with sodium acetate and sodium sulfate, were extracted with 7 ml of n-propyl alcohol/diethyl ether. Samples were centrifuged to separate the organic phase, which was evaporated to dryness; they were then reconstituted with 250 μl of mobile phase (50 mmol/l potassium phosphate; pH 4.0) and finally injected into the LC Module I Plus HPLC with an ultraviolet detector set at 215 nm (Waters, Milford, USA). 5-FU and 5-FUH₂ were separated on Hypersil BDS C18 stationary phase (Alltech, Deerfield, USA), eluted with 1 ml/min of mobile phase. The data analysis was performed by use of Millenium 2.1 software (Waters). Standard calibration curves were obtained by adding 5-FU and 5-FUH₂ to 0.5 ml of blank plasma obtained from healthy donors on each day of analysis, resulting in final concentrations that ranged from 0.08 to 75 μg/ml. For the analysis of tegafur (FT) and uracil (U), plasma samples (1.0 ml) were adjusted with 0.1 ml of 0.5 M NaH₂PO₄ buffer and 8 ml ethyl acetate were added. After extraction and centrifugation, the organic layer was removed and evaporated under N₂ at 50°C. The residue was dissolved in 50 μl of methanol, and 20 μl were injected into the HPLC with an ultraviolet detector set at 270 nm (Waters, Milford, USA). FT and U were separated on Hypersil BDS C18 stationary phase (Alltech, Deerfield, USA), eluted with 1 ml/min of mobile phase. The data analysis was performed by use of Millenium 2.1 software (Waters). Mobile phase was 0.01 M sodium acetate buffer (pH 4):methanol (85:15, v/v) as eluent. The retention times were 6.4, 2.7 min for FT and U, respectively. Standard calibration curves were obtained by adding FT and U to 0.5 ml of blank plasma obtained from healthy donors on each day of analysis. Sensitivity limit of quantitative analysis in plasma was 0.1 μg/ml. In order to detect GHB, 200 μl of plasma were treated with 500 μl of acetonitrile, using α -hydroxy-isovaleric acid (200 ng/ml) as internal standard. After agitation and centrifugation (9,000g for 10 min), the supernatant was collected and evaporated to dryness under nitrogen flow. The residue was derivatized by adding 50 μl N,O-bis(trimethylsilyl)trifluoroacetamide + 1% trimethylchlorosilane (BSTFA + 1% TMCS), then incubated for 30 min at 70°C. An aliquot (1 ml) of the derivatized extract was directly injected into GC/MS using a TRACE gas chromatograph equipped with a Polaris Q as mass detector and an AS2000 as autosampler (Thermo Finnigan, Rodano, Italy). The flow of carrier gas (helium, purity grade N55) through the column (Restek, Palo Alto, USA; Rtx-5MS capillary column, 30 m × 0.25 mm × 0.25 μm film thickness) was 1.0 ml/min. The injector temperature was 280°C and splitless injection was employed with a split valve off-time of 1.0 min. The column oven temperature was programmed to rise from an initial temperature of 65°C, maintained for 1 min, to 140°C at 22°C/min, then 140°C for 3 min, then to 290°C at 50°C/min and maintained at 290°C for the final 5 min. Data were recorded in full scan and ions monitored were: m/z 233, 73 and 147 and m/z 73, 145 and 219 for GHB and α-hydroxy-isovaleric acid, respectively (the underlined ions were used for quantitation).

Individual plasma concentration profiles of tegafur and its catabolites were fitted according to a two-compartment model by use of nonlinear least squares regression analysis (MwPharm software, version 3.60; MediWare, Groningen, The Netherlands). The area under the curve (AUC) of tegafur, 5-FU, 5-FUH₂, GHB and uracil was calculated by the trapezoidal method for the area from time 0 to the time of the last measurable concentration. The maximum plasma concentration (C_max) and time to reach C_max (T_max) were identified from the inspection of tegafur and its catabolite concentration–time plots.

Before drug administration and at day 28, 56, 84 and 112, 10 ml of blood were drawn from the antecubital vein of patients. PBMCs were collected as previously published [7]; the resulting pellet was immediately frozen in liquid nitrogen and stored at −80°C. As previously described [25], RNA was reverse transcribed and the resulting cDNA was diluted and then amplified by QRT-PCR with the Applied Biosystems 7900HT sequence detection system. CD133 validated primer were purchased from Applied Biosystems (Assay ID Hs00195682_m1). The PCR thermal cycling conditions and optimisation of primer concentrations were followed as per manufacturer’s instructions. Amplifications were normalized to GAPDH and the quantitation of gene expression was performed using the ΔΔC _t calculation; the amount of CD133, normalized to the endogenous control and relative to the calibrator (PBMC sample at day 0), is given as \( 2^{{ - \Updelta \Updelta C_{t} }} \). The data are presented as the percentage of \( 2^{{ - \Updelta \Updelta C_{t} }} \) at day 0 (before the beginning of metronomic schedule).

Plasma samples obtained at the same days of PBMC collection were assessed for VEGF, TSP-1 and sVE-C levels using commercially available ELISA kits. Each sample was assayed for human VEGF and TSP-1 concentrations by the ELISA Kit Quantikine^® (DVE00 and DTSP10, R&D Systems, Minneapolis, MN, USA) and for soluble VE-cadherin by Instant ELISA Kit (Bender Medsystems, Wien, Austria). Measurements were performed by the microplate reader Multiskan Spectrum (Thermo Labsystems, Milan, Italy) set to 450 nm (with a wavelength correction set to 540 nm).

The primary objective of the study was to evaluate the percentage of patients not progressed within 2 months from the beginning of metronomic CTX plus UFT and celecoxib regimen. In phase II studies of chemotherapy administered for palliation in patients with mCRC or with gastrointestinal tumors, already treated with standard chemotherapy treatments, a rate of approximately 20% of patients free from progression within 2 months of treatment was generally observed. Our study of metronomic UFT plus CTX and CXB aimed to achieve an increase from 20 to 40% in the proportion of patients not progressed at 2 months from starting treatment.

According with the single-stage design described by Fleming and A’Hern, choosing a parameter P0 (percentage of patients free from progression at 2 months: null hypothesis) = 0.20, and P1 (proportion of patients free from progression at 2 months: alternative hypothesis) = 0.40, and considering the errors α and β of 0.10 and 0.10, the study required the enrollment of at least 36 evaluable patients. Study treatment was considered promising when at least 11 patients were progression free at 2 months. Progression free survival (PFS) and overall survival (OS) were calculated from the date of progression or death/loss to follow-up, respectively, using the Kaplan–Meier method. The log-rank test was used to compare survival between patients having stable disease (SD) and progressive disease (PD). Statistical analysis by ANOVA, followed by the Student–Newman–Keuls test, was used to assess the statistical differences of pharmacokinetic and pharmacodynamic data. Correlations between pharmacokinetic parameters were investigated by linear regression analysis. Cut off values for 5-FU C_max and AUC parameters were found with nonparametric receiver operating characteristic (ROC) analysis, performed to assess the accuracy of pharmacokinetic parameters to discriminate between stable and progressive disease groups of patient. Statistical analyses were performed using the GraphPad Prism software version 5.0 (GraphPad Software Inc., La Jolla, CA, USA).