| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Groupe de RMN Biomédicale, Laboratoire des IMRCP, Université Paul Sabatier, Toulouse, France
Correspondence: M. Malet-Martino, Ph.D., Groupe de RMN Biomédicale, Laboratoire des IMRCP (UMR CNRS 5623), Université Paul Sabatier, 118, route de Narbonne, 31062 Toulouse Cédex, France. Telephone: 33-561-556271; Fax: 33-561-557625; e-mail: martino{at}chimie.ups-tlse.fr
Access and take the CME test online and receive one hour of AMA PRA category 1 credit at CME.TheOncologist.com
LEARNING OBJECTIVES
Top
Learning Objectives
Abstract
5-Fluorouracil
Prodrugs of 5-FU
S-1
Conclusion
References
After completing this course, the reader will be able to:
|
ABSTRACT |
|---|
|
TopLearning ObjectivesAbstract5-FluorouracilProdrugs of 5-FUS-1ConclusionReferences |
|---|
In the first part of this review, we briefly present the metabolic pathways of 5-FU responsible for the efficacy and toxicity of this drug. This knowledge is also necessary to understand the target(s) of biomodulation.
The second part is devoted to a review of the literature on three recent prodrugs of 5-FU, i.e., capecitabine, UFT (ftorafur [FTO] plus uracil), and S-1 (FTO plus 5-chloro-2,4-dihydroxypyridine plus potassium oxonate). The pharmacological principles that have influenced the development of these new drugs and our current knowledge of the clinical pharmacology of these new agents, focusing on antitumor activity and toxicity, are presented.
The literature was analyzed until March 2002. This review is intended to be as exhaustive as possible since it was conceived as a work tool for readers wanting to go further.
Key Words. Xeloda® • OrzelTM • TS-1 • Toxicity • Phase I, II, III trials
|
5-FLUOROURACIL |
|---|
|
TopLearning ObjectivesAbstract5-FluorouracilProdrugs of 5-FUS-1ConclusionReferences |
|---|
) is an example of a rationally designed anticancer agent. The observation that rat hepatomas utilized radiolabeled uracil more avidly than nonmalignant tissues implied that the enzymatic pathways for utilization of uracil differ between malignant and normal cells [1]. Heidelberger et al. [2], thus, synthesized 5-FU in 1957 as an antimetabolite agent. In this molecule, the hydrogen atom in position 5 of uracil is replaced by the similarly sized atom of fluorine, and the molecule was designed to occupy the active sites of enzyme targets, thereby blocking metabolism in malignant cells. Although this antimetabolite is toxic, its efficacy makes it one of the most widely used agents against solid tumors. As several reviews have been devoted to 5-FU [3–13], we present briefly the main features of its metabolism, mechanism of action, and clinical uses.
|
The Anabolic Route
The anabolism of 5-FU is rather complex (Fig. 2) with various parallel reactions. Initially, 5-FU can react in the following three ways. The first, which is quantitatively the least important, leads in two stages to the formation of 5-fluoro-2'-deoxyuridine-5'-monophosphate (5-FdUMP). The other two routes both form 5-fluorouridine-5'-monophosphate (5-FUMP), which can then undergo two successive phosphorylations to give 5-fluorouridine-5'-diphosphate (5-FUDP) and then 5-fluorouridine-5'-triphosphate (5-FUTP), which can be incorporated into RNA instead of uridine-5'-triphosphate (UTP). 5-FUTP can also be conjugated to sugars giving 5-FU-nucleotide sugars (5-FUDP-sugars).
|
The Catabolic Route
After administration of 5-FU, more than 80% of the injected dose is degraded according to the scheme shown in Figure 3. The first stage of this degradation occurs very rapidly; under the action of dihydropyrimidine dehydrogenase (DPD), 5-FU is reduced to 5,6-dihydro-5-fluorouracil (5-FUH2). This first stage effectively governs the rate at which 5-FU is available for anabolism. 5-FUH2 is then cleaved to give 
-fluoro-ß-ureidopropionic acid (FUPA). A third stage leads to the formation of -fluoro-ß-alanine (FBAL), the major catabolite of 5-FU.
|
-fluoro-ß-alanine (CFBAL), three conjugates of FBAL with bile acids, two metabolites of FBAL by transamination (2-fluoro-3-hydroxypropanoic acid [FHPA] and fluoroacetate [FAC]) [14–20].
Mechanism of Action of 5-FU
As seen above, one of the routes of metabolism of 5-FU gives rise to 5-FUTP. The atom of fluorine replacing the hydrogen on position 5 of uracil is of comparable size, and during transcription, this fluoronucleotide thus mimics UTP and is recognized by RNA polymerases. This leads to the incorporation of 5-FU in all classes of RNA. It is thought that the full cytotoxicity stems from a combination of the numerous modifications of RNA due to incorporation of 5-FU rather than alteration of a single function.
The second way that 5-FU is activated, by the formation of 5-FdUMP, could be the most important. Indeed, this fluoronucleotide is an inhibitor of thymidylate synthase (TS), which is involved in the synthesis of DNA. TS is an obvious target for cytotoxic agents since thymidine is the only nucleotide precursor specific to DNA. In the presence of the cofactor 5,10-methylene tetrahydrofolate (MeTHF), serving as the methyl donor, TS and 2'-deoxyuridine-5'-monophosphate (dUMP) form a ternary complex, which enables transfer of a methyl group on carbon 5 of dUMP to form thymidine-5'-monophosphate (dTMP). Following 5-FU exposure and adequate 5-FdUMP formation, the methyl transfer does not take place because the fluorine atom in the C5 position of 5-FdUMP is much more tightly bound than hydrogen. The enzyme is then trapped in a slowly reversible ternary complex. The formation of dTMP is therefore blocked, thereby decreasing the availability of thymidine-5'-triphosphate (dTTP) for DNA replication and repair. The "genotoxic stress" resulting from TS inhibition may activate programmed cell death pathways.
Throughout the 1970s, it was believed that the anticancer effect of 5-FU was the result of the two previous mechanisms. Later, it was reported that two other mechanisms could be partly responsible for the cytotoxic effect of 5-FU. The first one is the incorporation of 5-FU into DNA. The second mechanism is the alteration of the membrane function of 5-FU-treated cells. This is related to the formation of 5-FUDP sugars and their incorporation into membranes.
The extent to which any of these pathways predominates in human tumors is unknown and is likely to vary across tumor types and with different modes and doses of drug administration. Recent studies suggest that more prolonged exposure to low doses of 5-FU leads to cell death primarily via the TS-directed mechanism, whereas bolus administration of 5-FU results primarily in an RNA-mediated process of cell death [21–23].
Limitations of the Activity of 5-FU
The activity of 5-FU is markedly limited by its rapid degradation into 5-FUH2 via the action of the cytosolic enzyme DPD, the first enzyme in the catabolic cycle of 5-FU (Fig. 3
). It has been demonstrated that this enzyme deactivates more than 85% of the injected dose of 5-FU. Little drug is thus left for anabolism. Moreover, there is considerable variability in the activity of this enzyme between both normal and tumor tissues within one patient and among different patients [24, 25]. The bioavailability of 5-FU, especially after oral administration, is thus unpredictable. In some cases, where DPD possesses strong activity, little 5-FU is available and its efficiency is consequently low. On the other hand, if DPD possesses weak activity, levels of 5-FU are elevated, which may lead to toxicity from overdose.
There are also several other determinants of cellular sensitivity to 5-FU. Among these, the activity of enzymes involved in 5-FU anabolism, the availability of cofactors necessary for 5-FU activation (e.g., 5'-phosphoribosyl-1-pyrophosphate [PRPP]), the level of TS activity or expression, the size of intracellular reduced folate and endogenous dUMP pools, the extent of 5-FUTP incorporation into RNA and 5-FdUTP into DNA, and the intratumor activity of DPD have been identified as important in determining response to 5-FU chemotherapy in patients with solid tumors [26].
5-FU in Chemotherapy
Since its introduction more than 40 years ago, 5-FU has become a component of the standard therapy for a variety of malignancies, including gastrointestinal cancers, head and neck cancer, and breast cancer.
5-FU is a small molecule with a pKA (8.0) that should predict excellent absorption and bioavailability. However, the use of oral 5-FU was abandoned decades ago because of its irregular absorption. Plasma levels of 5-FU are quite unpredictable after oral administration with marked intra- and interindividual differences due to the variable activity of DPD, especially in the gastrointestinal mucosa [27, 28. This effectively rules out oral treatment with 5-FU.
Until recently, the standard regimen for treatment with 5-FU was 400-600 mg/m2 administered by i.v. bolus once a week for 6 weeks or for five consecutive days every 4 or 5 weeks. However, since the half-life (t
) of 5-FU is low (5 to 20 min) [29], tumor cells are only exposed to the active principle for a short time. The development of permanent venous access devices and portable infusion pumps has allowed continuous infusion of 5-FU over prolonged periods, designed to prolong exposure of cells to the drug and so confer higher activity [30, 31]. Indeed, the initial clinical results bore out this assumption and showed a twofold increase in response rate for continuous infusion compared with treatment by bolus, with a comparable survival [32]. In 1998, a meta-analysis of six randomized trials on 1,219 patients with colorectal cancers treated by either continuous i.v. infusion of 5-FU over 120 hours or i.v. bolus was published [33]. The response rate was clearly higher for the first than for the second regimen (22% versus 14%). Median survival time (MST) was comparable for the two treatments (12.1 months with bolus treatment versus 11.3 months with infusional therapy).
The toxicity profile of 5-FU is schedule dependent. For the bolus regimen, toxicities include myelosuppression, oral mucositis, and gastrointestinal disturbances (diarrhea, nausea, vomiting) due to phosphorylation of 5-FU into 5-FUMP by orotate phosphoribosyl transferase (OPRT) in the digestive tract [34]. For the continuous regimen, toxic reactions include the hand-foot syndrome (HFS, i.e., dermal pain in hands and feet), which was observed in many cases, but less hematologic and gastrointestinal toxicity. Cardiotoxicity and neurotoxicity may be observed during treatment with 5-FU (2%-5% of cases), but symptoms disappear on stopping the treatment, and resumption of treatment with a lower dose is generally well tolerated.
Treatment by continuous i.v. infusion thus presents advantages compared with treatment by bolus i.v. with respect to both response rate (RR) and toxicity, although mean survival is no better. Despite these benefits, the complex nature of the treatment (requirement for continuous infusion pumps), its high cost, and the added risks mean that it is used relatively infrequently, and i.v. bolus remains the conventional treatment.
To increase the activity of 5-FU, various researchers have proposed utilization of biochemical modulators. A biochemical modulator is a pharmacological agent designed to enhance the biological effect of the chemotherapy, either by selectively increasing the antitumor action or by selectively protecting the host. A 5-FU modulator can act at two levels: A) on the anabolic pathways, which will increase selectively the antitumor activity, or B) on catabolism to enhance bioavailability of the active principle as well as minimize toxic effects.
Several agents modulating the metabolism of 5-FU have been tested. The most efficient during treatment with 5-FU is the calcium salt of folinic acid, also called leucovorin (LV). LV is an intracellular source of reduced folates, which stabilize the ternary complex that they form with TS and 5-FdUMP, and thus increase and prolong the inhibition of TS and so enhance the efficacy of the drug. This modulator thus acts on the anabolism of 5-FU.
The modulation of 5-FU by LV has been extensively developed, and a meta-analysis of 1,381 patients in nine randomized clinical trials has clearly demonstrated its superiority over bolus 5-FU, with a response rate of 23% versus 11%. However, this improvement in RR failed to convey a significant survival advantage; MST in patients treated with 5-FU alone was 11 months compared with 11.5 months in those given 5-FU plus LV [35].
At the present time, the combination 5-FU/LV is considered the standard chemotherapy for colon cancer in both advanced disease and in the adjuvant setting. The two common regimens, which often serve as the control arms for phase III trials evaluating new agents in colorectal cancer, are A) 5-FU (370-425 mg/m2/day) + LV (low dose: 20 mg/m2/day or high dose: 200 mg/m2/day) administered by i.v. bolus during five consecutive days every 4 to 5 weeks [36], and B) 5-FU (600 mg/m2/week) + LV (500 mg/m2/week) administered by i.v. bolus during six consecutive weeks and repeated every 8 weeks [37].
Although these two regimens are associated with similar response rates and survival, the toxicities observed with each regimen differ. Dose-limiting toxicities associated with the daily administration are primarily mucositis and neutropenia (affecting 24% and 29% of patients, respectively), whereas the predominant toxicity seen in patients on the weekly regimen is diarrhea (32% of patients) [38]. Twenty percent to 30% of the patients on these 5-FU/LV regimens require hospitalization for toxicity-related events.
Conclusion
As discussed above, 5-FU is relatively efficient, but has numerous drawbacks. First of all, it is relatively toxic, causing myelosuppression and gastrointestinal disorders, due mainly to phosphorylation of 5-FU in the digestive tract. Toxicity also derives from the lack of selectivity of the drug toward tumors. Its efficiency is limited by the short t of 5-FU in plasma, by its low bioavailability due to the activity of DPD, and by the resistance to 5-FU of some tumors with strong expression of TS or low reserves of reduced folates. Its mode of administration is also a problem. Any i.v. administration, whether by bolus or continuous infusion, requires the presence of the patient in a hospital, whereas oral treatment can be taken at home with a corresponding reduction in stress. Furthermore, i.v. administration is not without risk of complications (venous thrombosis or infection around the catheter).
For these reasons, one of the challenges of cancer research is the development of prodrugs of 5-FU that diminish or circumvent some of these disadvantages: reduction in toxicity by avoiding certain routes of degradation (prodrugs that are not a substrate for the enzymes of degradation) or by targeting the tumor site (prodrugs that liberate the active principle selectively in tumor cells); enhancement of activity by reducing catabolism (use of DPD inhibitors) or by increasing anabolism, and improvement in quality of life of the patient by developing oral prodrugs.
|
PRODRUGS OF 5-FU |
|---|
|
TopLearning ObjectivesAbstract5-FluorouracilProdrugs of 5-FUS-1ConclusionReferences |
|---|
The prodrugs of 5-FU are characterized by a pyrimidine ring with a fluorine atom in position 5. They differ from 5-FU in a variety of chemical alterations. Their main benefit is that of oral administration. They are designed to be well absorbed intact from the gastrointestinal tract and subsequently enzymatically converted into 5-FU in the liver or within the tumor itself, in order to expose the tumor to 5-FU for a longer time but at lower concentrations than those observed after an i.v. bolus, hence minimizing toxicity. New orally administered fluoropyrimidines thus provide protracted 5-FU delivery, which offers advantages that include schedule flexibility and reductions in professional health care resource requirements, administration costs, and toxicity-related hospitalization. These advantages may reduce the overall cost of treatment [39].
In a study of 103 patients, 89% stated a preference for oral therapy. Reasons for this choice included convenience and fewer problems than with venous access. Nevertheless, 70% of survey respondents were unwilling to accept a lower RR with oral therapy [40]. Thus, although convenience is an important potential advantage, therapeutic equivalence or superiority is required for oral agents.
Each agent has been developed according to a specification with a well-defined mechanism for liberation of the active principle. Some are designed to function alone and others require coadministration of a modulator. The aim is to mimic the pharmacokinetics of 5-FU administered by continuous i.v. infusion, not only by virtue of their chemical structure, but also by careful choice of dosage.
Prodrugs in Clinical Trials or Already Used in Clinic
Over the years, many attempts have been made to synthesize more effective fluoropyrimidine drugs (Fig. 1). 5-fluoro-2'-deoxyuridine (5-FdUrd; Floxuridine®) represents the first generation. It is much more efficiently metabolized by the liver than 5-FU, which explains why its major clinical use is hepatic arterial administration for liver metastases of colorectal cancer [41]. Even if severely compromised by the 5-FdUrd-induced hepatoxicity, it has been demonstrated in a meta-analysis of randomized trials that hepatic arterial administration of 5-FdUrd is superior, in terms of RR and survival, to i.v. treatment [42]. Recently, a high RR (74%) was observed when colorectal cancer patients with unresectable hepatic metastases were treated with concurrent systemic CPT-11 (irinotecan) and hepatic arterial infusion of 5-FdUrd and dexamethasone [43].
Ftorafur (FTO, 1-(2-tetrahydrofuryl)-5-fluorouracil, Tegafur or Futraful) and 5'-deoxy-5-fluorouridine (5'd5-FUrd, doxifluridine or Furtulon®), the second generation, were developed with the hope of permitting oral administration. The third-generation compounds include the enzymatically activated prodrug, capecitabine, and the DPD-inhibitory compounds (UFT [FTO + uracil] eniluracil, S-1 [FTO plus 5-chloro-2,4-dihydroxypyridine plus potassium oxonate], and BOF-A2). The industrial sponsor of eniluracil discontinued development of the drug in 2000, and further clinical trials of the 5-FU/eniluracil have been abandoned [44, 45]. Several reviews on the new fluoropyrimidine agents have been recently published; only some of the most recent are cited [44–59].
Capecitabine
N4-pentyloxycarbonyl-5'-deoxy-5-fluorocytidine, more commonly called capecitabine (CAP) (Fig. 1) or Xeloda®, is a cytotoxic agent that is administered orally and should be activated preferentially in tumors. This carbamate of fluoropyrimidine was synthesized in the 1990s by Japanese researchers [60] as an oral formulation designed to circumvent the unacceptable toxicity of 5'd5-FUrd. The main limitation of 5'd5-FUrd derives from its gastrointestinal toxicity, attributed to liberation of 5-FU in the small intestine under the action of thymidine phosphorylase (TP) [61], a tumor-associated angiogenesis factor [62]. CAP was thus designed as a prodrug of 5'd5-FUrd that could not be metabolized by TP in the intestine.
Indeed, after oral administration, CAP crosses the gastrointestinal barrier intact and is rapidly and almost completely absorbed [63, 64]; thus, diarrhea should not occur with its use. It is subsequently converted into 5-FU in a three-stage mechanism involving several enzymes (Fig. 4). In the first step, it is metabolized into 5'-deoxy-5-fluorocytidine (5'-dFCR) by hepatic carboxylesterase. 5'-dFCR is then deaminated into 5'd5-FUrd by cytidine deaminase, mainly localized in liver and tumor tissues. Finally, 5'd5-FUrd is transformed into 5-FU under the action of TP, an enzyme with higher activity in tumor than in normal tissues. Higher levels of 5-FU are thus produced within tumors with minimal exposure of healthy tissue to 5-FU [65]. It must be underlined that a high phosphorylase activity is good for activation of CAP to 5-FU but bad for activation of 5-FU nucleoside derivatives up to the fluoronucleotides level (Fig. 2). This reaction is in fact bidirectional, but the main direction is toward breakdown of the nucleosides to the bases [66].
|
A pharmacokinetic study was carried out to verify these expectations [67]. Nineteen patients with colorectal cancers who were undergoing elective resection of either a primary lesion or liver metastases were treated prior to surgery with 1,255 mg/m2 of CAP administered twice daily over 5 to 7 days. Analysis of samples removed on operation showed a higher concentration of 5-FU in the tumors than in adjacent healthy tissues. 5-FU concentrations in the colorectal resections were on average 3.2 times greater than in the normal bowel and 21 times greater than in plasma. In liver metastases, the ratio of 5-FU concentrations was 1.4 when metastases were compared with normal liver (not statistically significant) and 9.9 when compared with plasma. These data are explained by the fact that the activity of TP is four times higher in colorectal tumor tissue than in healthy colorectal tissue, whereas it is similar in liver metastasis and healthy liver.
These encouraging findings have prompted a number of phase I trials of this drug [68–72]. As LV is frequently used during treatment with 5-FU to enhance inhibition of TS, Cao et al. [73] proposed using LV to increase the efficacy of CAP. Their preclinical study confirmed this hypothesis and has led to a phase I clinical trial with LV. Initial clinical studies showed that the maximum-tolerated doses (MTDs) of CAP were schedule dependent, with higher doses possible by employing breaks of 1-2 weeks after 2 weeks of continuous therapy. The dose-limiting toxicities (DLTs) in phase I trials were similar to those observed with continuous infusion of 5-FU, i.e., mainly diarrhea and HFS, but also nausea, vomiting, mucositis, leukopenia, and abdominal pain. The occurrence of bowel toxicity was disappointing as, in theory, it should have been avoided as CAP is not metabolized in the intestinal mucosa. In a recent study on 41 patients with metastatic colorectal cancer treated with the intermittent regimen of CAP at a dose of 2,500 mg/m2/day, HFS developed in 28 patients (68%; 95% confidence interval [CI] 52%-82%), with five patients (18%) having grade 1, 20 (71%) grade 2, and three (11%) grade 3. HFS starts within the first two cycles of therapy. Dose reduction leads to complete resolution in all cases [74]. The addition of pyridoxine (
200 mg/day) for ameliorating the symptoms of CAP-induced HFS allows for the administration of higher doses of CAP [75]. "Bag balm," which is a topical petroleum-lanolin based ointment with an antiseptic ingredient, hydroxyquinoline sulfate, also improved HFS [76].
Phase II trials have evaluated both continuous and intermittent schedules of CAP, mainly in breast cancer (Table 1) [77–95]. The addition of LV resulted in a greater rate of toxicities, in particular diarrhea and HFS, without any perceptible greater antitumor efficacy. Monotherapy with CAP, given in an intermittent schedule was, thus, chosen for further development.
|
depicts the frequencies of adverse reactions of any grade that occurred in more than 15% of patients. Patients in the CAP group experienced significantly less stomatitis and alopecia of any grade compared with those in the 5-FU/LV group, whereas HFS was more frequent in patients treated with CAP. The frequency of grade 3-4 diarrhea was similar in both arms, but the 5-FU/LV regimen was associated with considerably more grade 3-4 stomatitis and neutropenia, and dose reductions were required in 35.1% of patients. HFS was rare among the 5-FU/LV patients but necessitated dose reduction in 27.3% of the CAP patients. Few patients experienced grade 4 toxicities in either group. A later onset and a lower incidence of grade 3 diarrhea, stomatitis, nausea, vomiting, alopecia, HFS, and neutropenia were observed in the CAP group. CAP patients required substantially fewer hospital visits for drug administration and spent fewer days in hospital for the management of treatment-related adverse events. In addition, CAP reduced the requirement for expensive drugs, in particular antimicrobial fluconazole and 5-HT3 antagonists to manage adverse events. Globally, the CAP treatment of colorectal cancer patients resulted in a substantial resource savings relative to the regimen of 5-FU/LV [97].
|
). Similar rates of grade 3-4 diarrhea were observed in both arms, but increased neutropenia and stomatitis were reported in the 5-FU/LV arm. However, grade 3 HFS occurred in 18% of the CAP patients versus 1% of the 5-FU/LV group. Fewer patients in the CAP group required hospitalization for adverse reactions than in the 5-FU/LV group (11.4% versus 20.4%). Dose reduction for adverse reactions was required in 40.5% of patients in the CAP group (most commonly for HFS or diarrhea) and 49.3% of patients in the 5-FU/LV group (most commonly for stomatitis or diarrhea). When data from the two clinical trials were pooled and analyzed, a statistically significant improvement in the overall RR was observed (25.7% for the CAP patients versus 16.7% for the 5-FU/LV group). This result was confirmed by an independent review committee (RR 22.4% for the CAP patients versus 13.2% for the 5-FU/LV group). In patients who had received prior adjuvant treatment, the RR with CAP was 21.1% compared with only 9.0% in patients treated with 5-FU/LV. Median time to disease progression was 4.6 months for CAP and 4.7 months for the 5-FU/LV group. The median survival was 12.9 months in the CAP group and 12.8 months in the 5-FU/LV group. Median time to treatment failure was also similar (4.2 months with CAP versus 3.6 months with 5-FU/LV). Response to treatment occurred as early in patients treated with CAP as in patients treated with 5-FU/LV (median time to response: 1.7 months versus 2.4 months, respectively). Significantly (p < 0.001) lower incidences of diarrhea (48% versus 58%), stomatitis (24% versus 62%), nausea (38% versus 47%), and alopecia (6% versus 21%) were observed with CAP compared with 5-FU/LV. The incidences of vomiting and fatigue were similar in both treatment groups. The only adverse event that occurred more frequently with CAP than with 5-FU/LV was HFS (all grades: 53% versus 6%; grade 3: 17% versus 1%). Grade 3/4 stomatitis occurred in only approximately 2% of patients receiving CAP, but was a major side effect of 5-FU/LV treatment (15%, p < 0.0001). Grade 3 or 4 neutropenia was significantly more common in the 5-FU/LV group than in the CAP group (21.1% versus 2.2%). The incidence of hyperbilirubinemia was higher in patients receiving CAP. Hospitalization for treatment-related adverse events was significantly less frequent in the CAP group than in the 5-FU/LV group (11.6% versus 18.0%, respectively; p = 0.002). Dose reductions for adverse reactions occurred less frequently with CAP than with 5-FU/LV (34% versus 42%, respectively; p < 0.004). Furthermore, dose reductions for adverse events occurred later in patients receiving CAP than in patients treated with 5-FU/LV, with a median time to dose reduction of 2.5 months versus 1.2 months [99].
The same regimen of CAP is being evaluated in a large-scale adjuvant trial, which is expected to recruit approximately 1,700 Duke’s C colon cancer patients (X-ACT study, Roche; USA). Two large randomized phase III trials have also been designed to evaluate CAP as first-line therapy in patients with previously untreated metastatic breast cancer [100].
Since the conversion of CAP to 5-FU is mediated by TP, the combination of CAP with agents that upregulate TP concentrations in human cancer xenograft models, such as paclitaxel and docetaxel [101], cyclophosphamide (CP) [102], and x-ray irradiation [103], offers the potential to improve efficacy further. Moreover, a study on breast cancer models reported that the most potent and synergistic activity was observed when docetaxel was given on day 8 [104]. Numerous phase I trials with combinations containing CAP and other anticancer agents (taxanes, oxaliplatin, CPT-11, vinorelbine, gemcitabine, R115777 [a farnesyl transferase inhibitor], CI-994 [a histone deacetylase inhibitor], or epirubicin [combined with docetaxel, cisplatin, or CP]) or CAP combined with radiotherapy (RT) have been investigated recently [105–129]. DLTs were mainly HFS and neutropenia with taxanes (paclitaxel at 175 mg/m2 or docetaxel at 75 or 100 mg/m2 administered i.v. every 21 days) [105–109], diarrhea with oxaliplatin (120 or 130 mg/m2 administered i.v. every 21 days) [110–113], diarrhea and neutropenia with CPT-11 (70-300 mg/m2 with various schedules of administration) [114–116], myelotoxicity and stomatitis with gemcitabine (1 g/m2) [119–120], neutropenia, diarrhea, and HFS with RT (50.4 Gy) [123–126], neutropenia with docetaxel and epirubicin (both injected at 75 mg/m2 on day 1 every 21 days) [127], and neutropenia and stomatitis with epirubicin and cisplatin or CP [128–129]. The phase II trials are reported in Table 2 [130–140].
|
The approved dose of CAP (2,500 mg/m2/day as intermittent regimen) leads to unacceptable toxicities in many patients with metastatic breast cancer. A recent retrospective analysis suggests a standard starting dose of 2,000 mg/m2/day. If patients still have toxicities, individualization of dosing is necessary [146]. Moreover, the dose of CAP can be reduced to minimize toxicity without compromising efficacy. The impact of dose reduction was evaluated retrospectively from four phase II trials (2,510 mg/m2/day, intermittent regimen) including 321 patients with advanced or metastatic breast cancer. One hundred thirty-one patients with dose reductions were compared with 190 patients with no dose reductions. There were little differences in duration of response (220 days versus 211 days), time to treatment failure (234 days versus 218 days), and survival time (350 days versus 243 days) [147].
In breast cancer patients with brain metastases, a drug interaction between CAP and the anticonvulsant phenytoin led to cerebellar symptoms indicative of phenytoin intoxication. CAP downregulates the isoenzyme P4502C9, which plays a major role in the hepatic degradation of phenytoin. A frequent monitoring of the phenytoin level is thus imperative [148].
The polymorphism of the human TS gene determines response to CAP in advanced colorectal cancer [149]. The human TS gene promoter is polymorphic, having either double or triple repeats of a 28-base-pair sequence. Colon cancer patients homozygous for the triple tandem repeats (L/L) have significantly higher TS mRNA, which is associated with lower RR to 5-FU and poorer clinical outcome compared with those homozygous for the double repeat variant (S/S). In a study on 22 patients, 80% of individuals with the S/S variant responded to CAP, compared with 10% and 14% of those with the S/L and L/L variants, respectively. Genotyping patients for the TS polymorphism may thus be useful in identifying colon cancer patients who are more likely to respond to CAP treatment.
UFTTM and OrzelTM
In an effort to improve the therapeutic index of FTO, Japanese researchers [150] prepared a special formulation of FTO, called UFTTM (Fig. 1), a mixtue of FTO and uracil (U) in molar proportions of 1:4. This ratio was chosen based on preclinical models that suggested maximal tumor selectivity with these relative concentrations. U acts as a modulator on the catabolism of 5-FU. Being a natural substrate for DPD, it competes with 5-FU. Since its molar concentration is much higher than that of 5-FU (in the form of FTO), little 5-FU is degraded by DPD, and thus, is more available to the anabolic route of activation. The coadministration of FTO (slow liberation of 5-FU in the organism) with U (reduced degradation of 5-FU) was designed to produce a constant reserve of 5-FU and its active metabolites and to minimize production of inactive and potentially toxic metabolites [151]. In 10 patients receiving equimolar dosages of 5-FU by continuous infusion (250 mg/m2/day over 5 days) or oral UFT (370 mg/m2/day over 28 days), similar systemic 5-FU exposure (measured by the area under the concentration curve [AUC]) was obtained [152]. This study also demonstrated a prolonged elimination t of 5-FU after oral UFT administration.
It has been reported recently that UFT, partly by maintaining a high and long-lasting blood level of 5-FU and partly through its metabolites (
-hydroxybutyric acid and -butyrolactone), had a stronger angiogenesis inhibitory effect in a murine renal carcinoma cell line than 5-FU or 5'd5-FUrd [153]. While the clinical relevance of this observation remains to be determined, it might be another mechanism of action of this compound.
Of the newer oral fluoropyrimidines, UFT has been studied the most extensively. Japanese investigators studied UFT about 20 years ago, but interpretation of the data was difficult because dissimilar methodologies, criteria, and evaluation standards were used [56]. Results of these studies indicated that UFT was well tolerated and had significant activity in patients with colon, stomach, and breast cancers [154]. Since the initial studies using UFT alone have produced encouraging results, researchers have proposed using an additional modulator, such as LV, to act on the anabolism of 5-FU by increasing available reduced folates and thereby stabilizing the binding of 5-FdUMP to TS. OrzelTM is the trade name for the combination UFT/LV. It is composed of UFT capsules copackaged with LV tablets.
Phase I trials [155–161] have established that the maximal dosage of UFT administered concomitantly with LV over 14 to 28 days in divided doses ranges from 350 to 400 mg/m2/day, which will thus be employed in the phase II trials. The main toxic reactions at the MTD were gastrointestinal (diarrhea, nausea, vomiting). The occurrence of these toxic effects correlated significantly with the maximum plasma concentration and AUC0-6h of 5-FU [162]. The HFS, characteristic to continuous i.v. infusion of 5-FU, was not observed.
UFT/LV in Colorectal and Rectal Cancers
Phase II Trials in Colorectal Cancer
Phase II trials in advanced colorectal cancer have shown that an oral regimen of UFT/LV is well tolerated and is at least as efficient as treatment with 5-FU administered by continuous i.v. infusion with or without LV [162–174]. The objective RR in these studies ranged from 20%-40%, and median overall survival was around 13 months. The most frequent severe toxicities were diarrhea and nausea/vomiting. The absence of objective responses in patients refractory to previous i.v. bolus 5-FU suggests that UFT is cross-resistant with bolus 5-FU [160, 169]. For elderly patients, the RRs were lower than in the general population [167, 168].
Phase III Trials in Colorectal Cancer
The above results led to two randomized phase III trials comparing oral treatment of UFT/LV (300 mg/m2/day in three daily doses for 28 days every 5 weeks in combination with 75 or 90 mg/day of oral LV, the doses found to lead to maximum LV bioavailability) with i.v. administration of 5-FU/LV (425 mg/m2/day of 5-FU + 20 mg/m2/day of LV administered by i.v. bolus during 5 days every 4 or 5 weeks) in patients with metastatic colorectal cancer [175, 176]. The overall RRs were not significantly different in both studies. In the study of Pazdur et al. [175], which enrolled 816 patients, RR was 12% in the UFT/LV arm and 15% in the 5-FU/LV arm. For the 380 patients evaluated in the study of Carmichael et al. [176], RR was 11% in the UFT/LV arm and 9% in the 5-FU/LV arm. In the latter study, the median time to progression was 3.4 months and 3.3 months in the UFT/LV arm and the 5-FU/LV arm, respectively. The MSTs were 12.2 months (UFT/LV arm) versus 11.9 months (5-FU/LV arm) in the Carmichael et al. study, and 12.4 months (UFT/LV arm) versus 13.2 months (5-FU/LV arm) in the Pazdur et al. study. However, UFT/LV was associated with a significantly better safety profile compared with 5-FU/LV. UFT/LV treatment resulted in fewer episodes of febrile neutropenia and documented infection. Significant HFS was not observed with UFT/LV, whereas bilirubin was increased more often in the UFT/LV-treated patients. Both studies concluded that the oral treatment was accompanied by lower toxicity and was thus an acceptable alternative to treatment with i.v. 5-FU/LV.
A recent study examined patient preference for oral UFT/LV or i.v. 5-FU/LV chemotherapy in metastatic colorectal cancer. Thirty-seven previously untreated patients were randomized to start treatment with either oral UFT/LV (300 mg/m2/day of UFT plus 90 mg/day of LV, both in three daily doses for 28 days every 5 weeks) or bolus i.v. 5-FU/LV (425 mg/m2/day 5-FU plus LV 20 mg/m2/day for 5 days every 4 weeks). For the second treatment cycle, patients were crossed over to the alternative treatment regimen. Prior to the first and after the second therapy cycle, patients were required to complete a therapy preference questionnaire. Eighty-four percent of patients (n = 31) preferred oral UFT to i.v. 5-FU because they experienced less stomatitis and diarrhea, could take their medication at home, and preferred pill to injection. The authors also showed that oral UFT/LV led to prolonged 5-FU exposure that was comparable with continuous i.v. treatment [177].
Adjuvant Therapy in Colorectal Cancer
Oral UFT or UFT/LV were also explored in the adjuvant setting of colon cancer. Four hundred seventy-six patients were given either i.v. mitomycin C (MMC; 6 g/m2 on the day of and the day after surgery) or the same dose of MMC plus oral UFT (400 mg/day for 1 year). The disease-free survival rate of the UFT group was significantly higher than that of the control, though no significant difference in the overall survival rate was observed [178].
A clinical trial of oral UFT (600 mg/day for 5 days followed by 2 drug-free days, for a year) as adjuvant chemotherapy for resected colorectal cancer was conducted on 87 patients. This schedule was well tolerated and was favorable in terms of compliance [179]. A recent study on the pharmacokinetics of 5-FU during the 2 days off UFT showed that the 5-FU concentrations in the tumor were maintained at relatively high levels until 48 hours after the final dose of UFT, while serum 5-FU concentrations decreased to very low levels within 24 hours. These data suggest that the two off-drug days reduced the adverse reactions and improved drug tolerance without compromising antitumor efficacy [180].
Espinosa et al. [181] evaluated 269 patients with operated Dukes’ stage B2-C colon cancer. The UFT/LV regimen was oral UFT 390 mg/m2/day for 14 days every 4 weeks and i.v. LV (500 mg/m2) on day 1 followed by oral LV (24 mg/day) on days 2-14. Diarrhea was the primary side effect. After a median follow-up of 36 months, the disease-free survival was 83% for stage B2 and 62% for stage C. The overall survival rates were 94% and 87%, respectively.
The National Surgical Adjuvant Breast and Bowel Project Protocol C-06 is a randomized comparison of the relative efficacies of i.v. 5-FU/LV versus oral UFT/LV. The 5-FU/LV regimen was given for three cycles and consisted of a 2-hour infusion of LV (500 mg/m2) and, 1 hour after beginning LV, an i.v. bolus of 5-FU (500 mg/m2), both given weekly for 6 weeks. The oral UFT/LV regimen consisted of five cycles of UFT (300 mg/m2/day) in three daily doses and LV (90 mg/day) for 28 days every 5 weeks. An interim analysis on 473 patients indicated that both regimens were well tolerated and had similar toxicity profiles [182].
A randomized prospective trial was performed to determine the efficacy of preoperative and postoperative adjuvant oral UFT, administered with MMC after resection for advanced colorectal cancer. The patients received oral UFT (400 mg/day) for 7 days prior to surgery and, after surgery, were randomly assigned to group A, receiving MMC postoperatively, or group B, receiving MMC and oral UFT (400 mg/day) for 1 year. The survival results revealed no significant difference between groups A and B. However, in patients with nuclear DNA aneuploid tumors, the hematogenous recurrence rate after curative surgery was lower in group B than in group A [183].
A recent study on 55 patients compared the effects of sequential methotrexate (MTX; 100 mg/m2) and 5-FU (600 mg/m2) followed by LV rescue (treatment MFL), as an adjuvant chemotherapy, with a combination of MMC (12 mg/m2 intraoperatively, then 6 mg/m2 every other week after surgery for 2 months) and oral UFT (375 mg/m2/day) for at least 3 years. The overall survival and disease-free survival rates were significantly higher and the recurrence rate significantly lower in the MFL group, demonstrating the superiority of MFL therapy for improving postsurgical survival [184].
It recently has been demonstrated that postoperative adjuvant immunotherapy using protein-bound polysaccharide K (PSK) and UFT was highly effective in preventing recurrence of colorectal cancer. Two hundred seven patients were assigned to two groups. Group A received oral PSK (3 g/day) and UFT (300 mg/day) while group B received only UFT (300 mg/day), for 2 years after surgery. Three years after surgery, disease-free survivals were significantly different (80.5% and 68.7% for groups A and B, respectively, p = 0.021), whereas overall survival times were not (87.2% and 80.6% for groups A and B, respectively, p = 0.247) [185].
The Future in Colorectal Cancer
Phase I-II trials are being conducted in which UFT alone or UFT/LV is combined with other agents to find more effective regimens for advanced metastatic colorectal cancer that incorporate both i.v. and oral drugs (Table 3).
|
PMC was also used as postoperative adjuvant treatment for hepatic colorectal metastases. Sixty-six patients were divided into two groups after hepatectomy. Group A (n = 38) received a hepatic arterial infusion of 5-FU (600 mg/m2/day) for two consecutive days per week and an oral administration of UFT (400 mg/day) for 5-7 days per week, repeated 10 times. Group B (n = 28) received oral UFT (400 mg/day) for 6 months. Cumulative 5-year survival rates after hepatectomy were 62% in group A and 27% in group B (p = 0.00001). Median hepatic recurrence-free times were 31.6 months in group A and 18.4 months in group B (p = 0.0004). Main cause of death was lung recurrence in 21% of group A and hepatic recurrence in 57% of group B [200].
Rectal Cancer
Preoperative protracted i.v. infusions of 5-FU with concurrent pelvic RT are commonly used to treat rectal carcinoma [201]. Preliminary results suggest that UFT/LV combined with RT has similar efficacy, improved tolerability, and enhanced quality of life in comparison with continuous 5-FU [202–207].
UFT in Gastric Tumors
Several phase II trials were conducted to evaluate UFT alone or in combination with other agents in gastric cancer (Table 4) [163, 208–219].
|
In another small randomized trial, UFT/MMC was compared with 5-FU/MMC. RRs were 23% for the first regimen and 7% for the second [221].
UFT has also been used in adjuvant therapy after resection for gastric cancer. A slight difference in 5-year survival was reported when patients were treated for 2 years with oral FTO or UFT at a dose of 600 mg/day (52.1% versus 68.7%, p = 0.04) [222]. One thousand four hundred ten patients from 180 Japanese institutions were allocated into either a low-dose group, receiving MMC (8 mg/m2 on the day of surgery) and three capsules of UFT (300 mg in FTO) daily for 6 months, or a high-dose group, receiving MMC (8 mg/m2 on the day of surgery and in weeks 4, 10, 16, and 22 after surgery) and six capsules of UFT (600 mg in FTO) daily for 6 months. Although the number of adverse events was higher, a better prognosis is expected with the high-dose regimen [223].
UFT in Non-Small Cell Lung Cancer
UFT has been used in non-small cell lung cancer in patients with advanced disease and in the adjuvant setting. Several phase II studies (Table 5) [224–236] showed that the combination UFT/cisplatin was comparable with other currently used regimens for this disease.
|
To examine whether the efficacy of postoperative oral administration of UFT may be influenced by incidence of apoptosis or apoptosis-related gene status (p53 and bcl-2), 44 patients were treated with UFT (300-400 mg/day) 1 month after surgery, whereas 118 were not. For patients with a higher apoptotic index (
10.9), the 5-year survival rate of the UFT group was significantly higher than that of the control group (83% and 68%, respectively), whereas there was no difference for patients with lower apoptotic index (<10.9). Similarly, UFT was effective for patients without p53 aberrant expression (5-year survival rate of the UFT group was 95% versus 74% for the control group), but not for patients with p53 aberrant expression. Bcl-2 status did not influence the efficacy of UFT [243].
UFT in Breast Cancer
Several phase I and I/II trials were conducted with UFT in combination with other anticancer drugs in breast cancer therapy [244–252]. The DLTs were mainly neutropenia and diarrhea when UFT was combined with LV (90 mg/day) and paclitaxel [244–246, 251], oral CP [252], or vinorelbine [249, 250], with LV, paclitaxel, and doxorubicin [247], or with LV, epirubicin, and CP [248]. The phase II studies on patients treated with UFT alone or in combination with other treatments are presented in Table 6 [253–259].
|
More recently, a randomized trial compared the efficacy of oral UFT (350 mg/m2/day for 14 days; 31 patients) with i.v. 5-FU (500 mg/m2/day on days 1 and 8; 31 patients), both treatments being combined with doxorubicin (50 mg/m2 i.v. on day 1) and CP (500 mg/m2 i.v. on day 1). No statistical difference in overall RR was seen (48% in the UFT arm and 35% in the 5-FU arm). The MDR was 16 weeks for both arms. The median overall survival was 12 months for the UFT arm and 11 months for the 5-FU arm. Toxicity profiles (alopecia, anemia, leukopenia, thrombocytopenia, and diarrhea) were similar in both groups. Anemia and stomatitis were significantly more common in the 5-FU arm [261].
In the adjuvant setting, Spanish investigators have conducted two trials [262]. In the first one, 187 premenopausal women were randomly assigned to either oral UFT (400 mg/day for 6 months) and oral prednimustine (60 mg/m2 for 7 consecutive days, every 28 days in six cycles) or six cycles of the standard therapy CMF (CP 600 mg/m2, MTX 40 mg/m2, 5-FU 600 mg/m2, every 4 weeks). Disease-free survival and overall survival were similar in both arms. Toxicities were mild (nausea/vomiting, alopecia) and slightly worse in the CMF arm. In the second trial, 222 postmenopausal patients received 20 mg/day of tamoxifen for 1 year or the same dose of tamoxifen plus UFT at 400 mg/day for 6 months. The disease-free survival and overall survival rates were equal in both arms, but were longer in patients with five or more axillary-involved nodes treated with tamoxifen plus UFT. Toxicities were negligible in both arms.
A large Japanese trial [263] evaluated the combination of UFT, MMC, and tamoxifen as postoperative adjuvant therapy in the treatment of patients with stage II, estrogen receptor (ER)-positive primary breast cancer. All patients received 13 mg/m2 of i.v. MMC on the day of surgery. Patients were then randomly allocated to either oral tamoxifen (20 mg/day, 14 days after surgery for 2 years; 213 evaluable patients; group A) or oral UFT plus tamoxifen (400 mg/day and 20 mg/day, respectively; 223 evaluable patients; group B). There was no difference in the 5-year survival rate (93% for group A and 91% for group B). However, there was a significant advantage for the UFT plus tamoxifen group in the 5-year relapse-free survival rate (83% for group A and 91% for group B). In another large Japanese trial, a benefit was also observed in premonopausal ER-positive patients receiving a similar regimen (30 mg/day of tamoxifen, 300 mg/day of UFT) [264].
UFT in Head and Neck Cancer
The results of the phase II studies of UFT in head and neck cancer are shown in Table 7 [265–271]. In one study, 398 patients had a standard surgical therapy for stage III-IV head and neck cancer with or without adjuvant oral UFT (300 mg/day for 1 year). There was no significant difference in overall survival or relapse-free survival between arms. However, the addition of UFT improved distant metastasis of disease [272].
|
One study compared the combination UFT/RT with RT alone and showed significantly longer disease-free survival and overall survival for the group of patients treated concomitantly with UFT and RT [274]. More recently, patients with unresectable stage III-IV head and neck carcinoma received two cycles of neoadjuvant chemotherapy (cisplatin [25 mg/m2 for 5 days], LV [500 mg/m2 for 5 days], and 5-FU [800 mg/m2] for 4 days every 3 weeks) followed by radiochemotherapy (carboplatin [300 mg/m2 every 3 weeks], UFT [600 mg/day], and RT [65-72 Gy]). Higher complete and partial RRs were obtained after both steps of treatment (complete RR 69%, partial RR 10%) than after neoadjuvant chemotherapy alone (29%, 56%) [275].
A prospective randomized study was conducted to evaluate the benefit of adjuvant levamisole/UFT chemotherapy in head and neck squamous cell carcinoma. A trend of better distant control was observed in the treated group (n = 29) with respect to the control group (n = 34) ( p = 0.06). However, there was no statistically significant improvement in overall long-term survival rates (57% and 39% for patients with and without adjuvant chemotherapy, respectively) (p = 0.207) [276].
The correlation of the responsiveness of patients with oral squamous cell carcinoma to UFT with the intratumoral levels of DPD and TS mRNA was examined. Biopsy specimens were obtained before treatment, which consisted of oral administration of UFT, cobalt-60 irradiation, and injection of an immunopotentiator (OK-432). Intratumoral levels of DPD mRNA in patients who showed complete or partial responses were significantly lower than those in patients who showed no change. However, they did not correlate with local recurrence of the tumor. On the other hand, intratumoral TS mRNA levels did not correlate with any clinicopathological parameters [277].
UFT in Other Tumors
Pancreatic Cancer
A retrospective analysis of postoperative chemotherapy with UFT in 78 patients showed a significant difference in the MST between the UFT group (204 days) and non-UFT group (123 days) [278]. Recently, a phase I trial examined the use of UFT/LV with conventional RT (45 Gy) in 12 patients. MST was 8 months showing that this treatment is a viable option and compares favorably with continuous-infusion regimens [279]. The combination gemcitabine/UFT/LV (Table 7) is an adequate palliative therapy for this disease as it is moderately active and its toxicity is acceptable [268, 269].
Prostate Cancer
A phase II trial of UFT/LV in 28 patients with hormone-refractory prostate cancer demonstrated a level of activity similar to that obtained with i.v. 5-FU [280]. The combinations of UFT, CP, and estramustine or dexamethasone were reported as active and well-tolerated regimens (Table 7).
Bladder Cancer
A randomized, prospective trial of prophylactic oral UFT (300-400 mg/day for 2 years) for patients with superficial bladder cancer indicated that the recurrence rate was significantly lower in the UFT group (26% versus 43% for control). In particular, UFT was effective in preventing recurrence in patients with single tumors or small lesions. Gastrointestinal toxicities occurred in less than 10% of patients [281].
In conclusion, UFT combined with LV or other cytotoxic agents has shown equivalent efficacy for several advanced carcinomas in comparison with standard therapies. However, its advantages may be in ease of administration and absence of infusion-related side effects, including infection and bleeding.
Of note, UFT should not be given with food, which decreases the systemic exposure to 5-FU [282] and has a lethal interaction with sorivudine, a new antiviral drug for herpes zoster. A major metabolite of sorivudine, (E)-5-(2-bromovinyl)-uracil, formed by gut flora, is absorbed through the intestinal membrane and reduced in the liver by DPD, giving a reactive intermediate that is a suicide inhibitor of the enzyme. The DPD inactivation leads to extremely high concentrations of 5-FU, resulting in severe gastrointestinal and myelotoxicities then death [283].
UFT/LV has been submitted to the FDA for approval as a first-line treatment of advanced colorectal cancer. As of this writing, approval by the FDA was still pending because the overall role of UFT/LV alone remains to be established, since the first-line therapy of advanced colorectal cancer is changing to the combination of 5-FU/LV/CPT-11 [284]. In Europe, UFT/LV has been approved as a first-line treatment of metastatic colorectal cancer.
|
S-1 |
|---|
|
TopLearning ObjectivesAbstract5-FluorouracilProdrugs of 5-FUS-1ConclusionReferences |
|---|
). S-1 is a combination of a prodrug of 5-FU, FTO, and two compounds, 5-chloro-2,4-dihydroxypyridine (CDHP; gimestat) and potassium oxonate (OXO; otastat). It was developed in 1996 by Japanese workers in an attempt to circumvent the adverse reactions of FTO, but without compromising efficacy [285]. In this combination, FTO provides stable and prolonged liberation of 5-FU, equivalent to a continuous i.v. infusion of this drug. Neither CDHP nor OXO has any antitumor activity itself, and they act solely as modulators of 5-FU, designed to act at different steps in the metabolism of 5-FU. CDHP is a potent and reversible inhibitor of DPD, thereby prolonging high 5-FU concentration in the circulation. In vitro, CDHP is 180 times more potent than U, another reversible inhibitor of DPD [286]. Animal studies have demonstrated an elevation and a stability over 12 hours in levels of 5-FU in plasma and tumors after coadministration of FTO with CDHP, or S-1 [285, 287, 288]. OXO is employed to limit the gastrointestinal toxicity of FTO. It has already been pointed out that this toxicity stems from the phosphorylation of 5-FU within the digestive tract by OPRT. OXO accumulates in gastrointestinal tissues more than in other tissues or in the blood of normal or tumor-bearing rats [289, 290]. OXO competitively inhibits this enzyme, which converts 5-FU to 5-FUMP. The levels of 5-FUMP are decreased by 69% in the small intestine, 2% in Yoshida sarcoma, and 0% in bone marrow. The levels of 5-FU incorporated into RNA are thus decreased by 71% in the small intestine, 22% in Yoshida sarcoma, and 0% in bone marrow [289]. OXO also protects the activity of TS by decreasing 5-FdUMP via 5-FUMP in gastrointestinal tissue, which leads to a reduction in gastrointestinal toxicity [291].
The optimal molar ratio of the three constituents (FTO/ CDHP/OXO) of this combination is 1:0.4:1. With this formulation, an optimal inhibition of the transformation of 5-FU in the intestine is obtained without affecting its activation in the tumor [285]. Researchers in Japan conducted preclinical evaluation of S-1 and demonstrated its antitumor activity in experimental models of rodent tumors and human xenografts [287, 288, 292–296]. S-1 significantly inhibited tumor growth in rats with subcutaneous Yoshida sarcoma, and in nude rats and mice implanted with human colon, stomach, head and neck, and breast cancer cell lines. S-1 achieved induction of high and sustained apoptosis. S-1 also showed a significant antimetastatic effect on liver metastasis. The animal studies also confirmed that the gastrointestinal toxicity of S-1 was low because of the protection afforded by OXO [293]. Recently, a potentiated antitumor effect in Yoshida sarcoma-bearing rats was demonstrated when S-1 was combined with cisplatin [297].
Phase I studies of S-1 have been conducted in Japan, Europe, and the United States [298–301], and Phase II trials have been conducted in Japan (Table 8) [302–318]. S-1 achieved similar responses to those of conventional treatments, with a lower incidence of grade 3 or 4 toxicities. The DLT of S-1 was myelosuppression in Japan and diarrhea in Western countries. This difference remains unexplained, although it seems that the conversion of FTO to 5-FU occurs more slowly in Asians than in other ethnic groups [299,
