Abstract
Background/Aim: The objective of this study was to investigate the influence of digestive gastrointestinal absorption function on the pharmacokinetics of the orally-administered anticancer drug, Tegafur-gimestat-otastat potassium (TS-1), by measuring the plasma 5-fluorouracil (5-FU) concentration using stable isotope breath tests. Patients and Methods: Twenty-nine patients with progressive/recurrent digestive organ cancer were enrolled for this pharmacokinetic study, and blood samples were obtained from each patient. The area under-the-time-concentration curve between 0 and 480 min (AUC0-480 min), time-of-drug concentration peak (Tmax), maximum drug concentration (Cmax) and the half-life period (t1/2) of 5-FU were investigated. Simultaneously, a continuous 13C-acetate breath test was performed for each patient. The parameters measured with the breath test were the area under the 13CO2 excretion rate curve between 0-4 h (AUC0-4h), peak 13CO2 value and elimination rate constant (Kel) value. Results: The AUC0-8h and Cmax of 5-FU were significantly correlated with Kel (p=0.012 and p=0.024, respectively), and the 5-FU Cmax value was significantly correlated with the peak value of 13CO2 (p=0.037). Multivariate regression analysis also found the Cmax of 5-FU to be associated with Kel (p=0.0118). The Cmax and AUC0-8h of 5-FU were also significantly correlated (p<0.0001). Conclusion: The results of this study suggest that gastrointestinal absorption is closely-related to plasma 5-FU concentration after oral administration of TS-1.
5-Fluorouracil (5-FU) is widely used in the treatment of gastrointestinal carcinomas and is considered to be one of the most effective drugs against gastric cancer. However, the response rate of patients receiving 5-FU monotherapy for gastric cancer is only 10-20% (1), and the response to 5-FU-based regimens, in general, is still unsatisfactory with regard to long-term survival. Recently, TS-1, a drug comprising tegafur (1 M), gimestat (400 mM) and otastat potassium (1 M), and based on the biochemical modulation of 5-FU, was introduced as a novel oral anticancer drug. TS-1 is now considered to be a key treatment modality in the control of advanced gastric cancer (2), when given alongside an oral agent. However, the compliance of patients taking TS-1 is questionable. In an Adjuvant Chemotherapy Trial of S-1 for Gastric Cancer (ACTS-GC) trial, treatment compliance of patients who were able to tolerate TS-1 treatment for 12 months was 65.8%, with many of these (42.4%) having their dose reduced (3). The termination of administration was mainly due to a high incidence of high-grade adverse events such as hematological, mucosal, and digestive toxicity. It has been reported that there is high individual variability of 5-FU pharmacokinetics in plasma, and a close link between toxic side-effects and individual pharmacokinetic parameters has been demonstrated (4, 5). Therefore, attempts have been made to reduce the incidence of toxicity by tailoring the 5-FU dose for each individual (6, 7). Some studies attempted to detect which patients were at high risk of 5-FU-related toxicity before treatment by measuring the activity of dihydropyrimidine dehydrogenase (DPD) (8-10). However, measurement of DPD is an experimental technique only and insufficiently definitive to predict 5-FU pharmacokinetics. Recently, stable isotope breath tests have been recommended as a convenient, safe, and non-invasive method for the measurement of metabolites entering the systemic circulation from the gut. Enterically-administered 13C-acetate is particularly suited for this procedure because the appearance of 13CO2 in the breath is indicative of gastrointestinal oxidization and absorption of the substrate (11-13). In healthy individuals, 13C-acetate is rapidly absorbed by the gut and has, thus, been used for the study of gastric emptying in humans (13). Furthermore, it is principally converted into acetyl-CoA, oxidized in the Krebs cycle in many tissues and is not recycled, which constitutes an advantage over other substrates that can be labeled, such as glucose or amino acids (11). We therefore used the 13C-acetate breath test to assess gastrointestinal absorption of 5-FU. We hypothesized that high individual variability in 5-FU plasma pharmacokinetics might be related to digestive function because S-1 is absorbed from the intestinal tract. The purpose of the present study was thus to find a simple, easy, and less invasive method to anticipate 5-FU pharmacokinetics and optimize 5-FU-based treatments. For this purpose, we investigated the pharmacokinetics of orally-administered TS-1 by measuring the digestive function, using stable isotope breath tests.
Patients and Methods
Patients with progressive/recurrent upper digestive cancer were enrolled for this pharmacokinetic study from April 2007 to March 2008. Eligibility was as follows: 20 years or older with histologically-confirmed metastatic or recurrent solid tumor and previously receiving S-1; World Health Organization (WHO) performance status of 0-2; no history of chemotherapy within 4 weeks; adequate bone marrow function (neutrophil count of 1.5×109 /l or more; platelet count of 100×109 /l or more); adequate liver function (plasma bilirubin level <3.0 mg/dl; transaminase level <2 times the normal upper limit); and adequate renal function (plasma creatinine level of <2.0 mg/dl). All patients provided written informed consent to participate in this study. The study protocol was approved by the Institutional Review Board of Kawasaki Medical School. For sixteen patients, TS-1 was given per os twice daily for 28 consecutive days, followed by two weeks of rest. The dose of TS-1 was based on the body surface area (BSA) of the individual. The dose was 80 mg/day for patients with a BSA of less than 1.25 m2, 100 mg/day for those with a BSA of 1.25 to 1.5 m2, and 120 mg/day for those with a BSA of more than 1.5 m2. For thirteen patients, TS-1 was given per os twice daily for 14 consecutive days, followed by one week of rest. The dose of TS-1 was based on the BSA of the individual. The dose was 60 mg/day for patients with a BSA of less than 1.25 m2, 80 mg/day for those with a BSA of 1.25 to 1.5 m2, and 100 mg/day for those with a BSA of more than 1.5 m2.
Pharmacokinetic analysis of 5-FU. Blood samples for pharmacokinetic analysis were obtained on the seventh day of TS-1 treatment at 2, 4, 6 and 8 h after administration of TS-1. The samples were centrifuged immediately and plasma was stored at under 80°C until analysis. The plasma levels of 5-FU were assessed by gas chromatography-mass spectrometry (GC-MS) (14). GC-MS was carried out using the Trace GC and Trace MS with an Xcalibur (Ver. 1.2) control system (Thermo Electron K.K., Yokohama, Japan). The methodology of GC-MS has been previously described in detail (14).
The area under the time concentration curve between 0 and 480 min (AUC0-8h), time of drug concentration peak (Tmax), maximum drug concentration (Cmax) and the half-life period (t1/2) of 5-FU were calculated with the linear trapezoidal rule (until the peak plasma concentration) and the linear-log trapezoidal rule (until the last measurable concentration) using the Microsoft Excel-based MOMENT program (15) for moment analysis.
13C–Acetate breath test. All examinations were performed after an overnight fast (although participants were permitted to drink water freely until 30 min before examination). All participants ingested a 13C-labeled meal during the first 5 min of the examination and then rested in a reclined position. The meal was 200 kcal/200 ml liquid meal (Racol, Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) labeled with 100 mg 13C-sodium acetate (Cambridge Isotope Laboratories Inc., Cambridge, MA, USA). The breath test was performed according to the Japan Society of Smooth Muscle Research (JSSMR) protocol (16). A continuous 13C breath test (Breath ID System; Exalenz Bioscience Ltd., Modiin, Israel) was performed. Breath samples were obtained automatically and continuously through an exclusive nasal cannula for 4 h after ingestion of the labeled meal. For patients with gastrectomy or esophagectomy, these studies were carried out two weeks after surgery. Data were analyzed using Oridion Research Software β version (Oridion Medical Ltd, Jerusalem, Israel). The breath data were analyzed according to the conventional method of Ghoos et al. (17). After mathematical analysis, the area under the 13CO2 excretion rate curve (% dose/h) (AUC0-4h), peak value of 13CO2 (fixed by BSA), peak value of 13CO2, time of peak 13CO2 and t1/2 were calculated. The elimination rate constant (Kel) was estimated from the slope of the stable tail of a semi-logarithmic plot of 13CO2 excretion rate between 2 and 4 h using the least squares algorithm (17-21).
Clinical characteristics of 29 patients, who had evaluation of 5-FU plasma concentration.
Statistical analysis. Data are expressed as the mean±SD. Correlation of the 5-FU pharmacokinetic parameters with possible related factors was examined by ANOVA. Correlations were considered statistically significant when the two-tailed p-value was less than 0.05. The correlations were also analyzed by multivariate linear least-squares regression analysis. The estimated model was considered significant when the two-tailed p-value obtained by ANOVA was less than 0.05. Factors in the model were significantly associated with a variable when the two-tailed p-value was less than 0.05. All analyses were carried out with JMP version 9 software (SAS Institute, Cary, NC, USA).
Concentration-time profile of plasma 5-FU concentration after oral administration of TS-1. Results are the mean±SD (n=29).
Pharmacokinetic parameters of oral administration of TS-1.
Results
Twenty-nine cases with progressive or recurrent digestive cancer were enrolled in the present study between April 2007 and March 2008. The primary lesion was stomach cancer in 22 patients and esophageal cancer in the remaining seven. The median age was 67±8.4 years (range=54-87 years). Patients' characteristics are summarized in Table I. Most patients had an (ECOG) performance status 0 or 1. The median creatinine clearance (Ccr) calculated with the Cockcroft-Gault equation was 95.4 ml/min (range=41.1-130.5 ml/min).
Pharmacokinetic analysis of 5-FU in TS-1. The pharmacokinetic parameter profiles of plasma 5-FU concentration are shown in Figure 1 and Table II. The mean AUC0-8h of 5-FU was 636.69±333.15 ng/ml h (range=130.8-1175.1 ng/ml h). The mean Cmax of 5-FU was 128.47±63.95 ng/ml (range=32.0-246.7).
Pharmacokinetic parameters of 13C-acetate on the basis of time-related 13CO2 concentration in breath, following oral administration of 13C-acetate.
13C–Acetate breath test. 13CO2 excretion rate (% dose/h) profiles of 13C-acetate breath test are shown in Figure 2, with each parameter summarized in Table III. The mean AUC0-4h for the 13CO2 excretion rate was 55.81±17.24% dose/h (range=32.93-89.35%). The mean peak 13CO2 was 71.55±17.20% dose/h (range=42.11-112.1% dose/h). The mean Kel value was 0.343±0.13 (range=0.097-0.516).
Profile of respiratory concentrations of 13CO2 (Δ13C) with time after oral administration of 13C-Acetate. Results are the mean±SD (n=29).
Correlation between 5-FU pharmacokinetics and breath test parameters. The relationship between 5-FU pharmacokinetics and breath test parameters are shown in Table IV. The Cmax of 5-FU was found to be significantly associated with Kel (ANOVA, R2=0.1749, p=0.024) (Figure 3a). The AUC0-8h of 5-FU was also found to be significantly associated with Kel (ANOVA, R2=0.2118, p=0.012) (Figure 3b), and the 5-FU Cmax was associated with the peak value of 13CO (ANOVA, R2 2 =0.1515, p=0.0369) (Figure 3c). Furthermore, correlation between the AUC0-8h and Cmax of 5-FU was found to be significant (co-efficient of correlation, R2=0.7426, p<0.0001) (Figure 4).
Multivariate regression analysis indicated that the Kel value could be a significant predictor of 5-FU Cmax (ANOVA, R2=0.370, p=0.0083) (Table V) and AUC0-8h (ANOVA, R2=0.270, p=0.0456) (Table VI).
Discussion
TS-1 is a key drug in gastric cancer therapy, but compliance is poor because of the high incidence of severe adverse events associated with this drug. There is a high individual variability of 5-FU pharmacokinetics in plasma, and a close link between toxic effects and individual pharmacokinetic parameters has been demonstrated (4, 5). Therefore, we investigated the use of a breath test to monitor the pharmacokinetics of 5-FU after TS-1 administration, without having to resort to the invasive use of frequent blood collection. Some authors have reported attempts to reduce the incidence of toxicity by adjusting 5-FU dosing for individual patients (6, 7). Some studies have identified patients at high risk of 5-FU-related toxicity prior to treatment by measuring the activity of DPD. This enzyme is subject to a genetic polymorphism and exhibits a wide range of individual variation (9, 10). The clinical importance of DPD was shown by the finding that patients with deficient levels of DPD activity in their peripheral blood mononuclear cells, were more likely to suffer severe or even lethal toxicity following 5-FU therapy (22). Detecting patients with severe DPD deficiency has thus become a means of avoiding serious adverse effects of 5-FU. However, even in patients where DPD activity is not deficient, there is still wide inter-individual variability in absolute DPD activity (9) and individuals with normal enzyme activity can still be diagnosed with high plasma 5-FU levels, resulting in increased toxicity. Indeed, we have investigated DPD activity in peripheral blood mononuclear cells that exhibited a wide individual variability (data not shown) and found no relationship between plasma 5-FU pharmacokinetics or toxicity and DPD activity. It can, thus, be concluded that factors other than the DPD status contribute to 5-FU metabolism and toxicity. The most accurate means of providing individualized 5-FU therapy is to measure the plasma 5-FU concentration in every patient using GS-MS, a known reliable technique (23). However, frequent blood collection is a burden on patients and is labor-intensive and expensive at present. Moreover, hemodynamic tests' results are usually obtained only after an adverse event has occurred. We hypothesized that because TS-1 is an orally-administered drug, the plasma concentration of 5-FU after TS-1 administration might be related to the absorptive activity of the intestine. To evaluate the gastrointestinal absorptive function, we used the 13C-acetate BreathID system (Exalenz BioScience). This test is non-invasive and permits real-time analysis, while also reducing examination time and alleviating patient discomfort. Acetate is a short-chain fatty acid that is readily absorbed by the intestinal mucosa, is metabolized by nearly all body tissues, and has very limited non-oxidative metabolism in humans, in contrast to other substrates such as glucose or amino acids, which can be stored either directly or after transformation into other compounds. 13C-Acetate was selected for this study because it is rapidly extracted from the systemic circulation, converted into acetyl-CoA and finally oxidized to CO2 in most cells. Oxidation is the essential metabolic fate of absorbed acetate because there is no significant pathway for its storage in the human body (12). Several reports have described the use of a 13C-5-FU breath test to assess DPD activity and to predict susceptibility to 5-FU toxicity (24,25). We have previously investigated the relationship between 5-FU concentration and renal function, hepatic function and DPD status but found no significant associations (data not shown). Most importantly, we believe that the obtained data should be related to plasma 5-FU concentration after TS-1 administration. In our study, the ‘time-of-peak 13CO2’ parameter is equivalent to the Tmax parameter in the previous Japan Society of Smooth Muscle Research (JSSMR) protocol (16). Indeed, in the 13C-Acetate breath test, the time-of-peak 13CO2 value of 32.24±26.78 min was within the Tmax reference range (23.3-64.5 min) described in the JSSMR protocol (16). This suggests that the breath test performed in patients two weeks after surgery is appropriate for examining gastrointestinal absorption. We included mainly postoperative patients in the 5-FU pharmacokinetic study. In these patients, the storage function of the stomach is lost or reduced. The tegafur component of TS-1, which is the precursor of 5-FU, flows rapidly into the small intestine after ingestion and is absorbed from there. Some previous studies have reported that AUC and Cmax values of tegafur are higher in patients with a history of total gastrectomy (26, 27), although we have found no correlation between AUCs for 5-FU and tegafur (28), which supports the conclusion that gastrectomy does not affect 5-FU pharmacokinetics (28). We report here the relationship between plasma 5-FU pharmacokinetics and gastrointestinal absorptive function using the stable isotope breath test (BreathID system) for the first time. We showed that the Kel value is significantly related to 5-FU Cmax and AUC, suggesting that the 13C-Acetate breath test might give an accurate reflection of 5-FU pharmacokinetics. However, order to confirm in greater detail that the 13C-Acetate breath test can be employed to detect 5-FU catabolic deficiencies and optimize 5-FU-based treatments, further examination in greater numbers of patients is necessary. We believe that achieving an adequate plasma concentration of 5-FU after TS-1 administration will reduce the incidence rate of adverse events and consequently enhance the anticancer effect because of improved compliance with the treatment. The data shown here for plasma 5-FU concentration differ by approximately 2- to 4-fold from our previous study (28). We believe that this difference is due to basing the dose of TS-1 on the patient's body surface area. An alternative method of predicting an adequate plasma concentration of 5-FU after TS-1 administration is to periodically measure the 5-FU plasma concentration in every patient using GS-MS.
Correlation between the pharmacokinetics of 5-fluorouracil (5-FU) and possible prognostic factors.
a: Correlation between (Cmax) of (5-FU) and (Kel): correlation co-efficient, R2=0.175, p=0.024. b: Correlation between (AUC0-8h) of 5-FU and Kel: correlation co-efficient, R2=0.212, p=0.012. c: Correlation between Cmax of 5-FU and peak 13CO2 value: correlation co-efficient, R2=0.152, p=0.037.
In conclusion, gastrointestinal absorptive functionality, especially that defined by the Kel value, is closely related to the plasma 5-FU concentration after oral administration of TS-1. However, this method for predicting plasma 5-FU concentration requires refinement and further study before being widely employed in the clinic. We intend to continue developing this predictive method to facilitate the achievement of an adequate plasma 5-FU concentration after TS-1 administration.
Correlation between (AUC0-8h) and (Cmax) of 5-FU: correlation co-efficient, R2=0.7426, p<0.0001.
Multivariate regression analysis of the relationship between the pharmacokinetics of 5-fluorouracil (5-FU) (Cmax) and possible prognostic factors.
Multivariate regression analysis of the relationship between the pharmacokinetics of 5-fluorouracil (5-FU) (AUC) and possible prognostic factors.
Footnotes
-
Conflicts of Interest
The Authors declare no conflicts of interest.
- Received August 6, 2012.
- Revision received October 9, 2012.
- Accepted October 12, 2012.
- Copyright© 2012 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.