Gastric cancer is the fourth most common cancer worldwide, with 934,000 new diagnoses annually, and it is the second most common cause of cancer-related death, with 700,000 deaths annually.1 Parenteral 5-fluorouracil (5-FU) has been used since the early 1980s to treat chemonaive patients with advanced gastroesophageal cancer. In a randomized controlled trial (RCT), Cullinan et al.2 treated 151 patients with continuous intravenous infusion (CIV) 5-FU with doxorubicin and mitomycin-C (FAM), CIV 5-FU with doxorubicin (FA), or CIV 5-FU alone. Patients who received FAM had superior disease-free survival compared with patients who received 5-FU alone (P = .005), but they did not have superior overall survival (P = .17). Since that trial, several RCTs have compared parenteral 5-FU–based regimens for the treatment of gastric cancer in chemonaive patients (Table 1). In an RCT that compared 2 combinations of 5-FU/doxorubicin, with methotrexate (FAMTX) or mitomycin-C (FAM),3 patients who received FAMTX had a significantly longer median survival (P = .004). In a subsequent RCT,4 patients who received epirubicin, cisplatin (CDDP), and CIV 5-FU (ECF) had significantly superior median progression-free survival (P = .0009) and median overall survival (P = .00006) compared with patients who received FAMTX. That trial was plagued with multiple clinical trial methodology problems. Most RCTs have employed poor clinical methodology and monitoring of studies, often lacking stratifications, data monitoring committees, appropriate endpoints, inclusion of esophageal cancer, resection of localized cancer, and mixture of locally advanced versus metastatic patients. Another RCT of patients with gastroesophageal cancer showed no significant difference in response rates (RRs), progression-free survival, or overall survival among patients who received FAMTX, combination CIV 5-FU with CDDP, or combination bolus 5-FU with leucovorin (LV) and etoposide.5 Chemotherapy overall largely has been palliative, with poor response and survival rates, and without a standard regimen,6, 7 although meta-analyses have shown that chemotherapy is superior to best supportive care alone.8, 9 Recently, a combination of parenteral 5-FU, CDDP, and docetaxel was investigated for first-line treatment of metastatic gastric cancer.10, 11 Meta-analyses of adjuvant chemotherapies have yielded conflicting efficacies.12–14
Table 1. Selected Phase II or III Parenteral or Oral 5-Fluorouracil–Based Therapies for the Treatment of Advanced or Metastatic Gastric,Gastroesophageal, or Esophageal Cancer
Oral FU-based therapies are being investigated for advanced or metastatic gastric (and gastroesophageal and/or esophageal) cancer in the first-line setting (Table 1). UFT, an oral fluoropyrimidine that contains tegafur (1-[2-tetrahydrofuryl]-5-FU) and uracil, although it is not available in the United States, has been used in combination with LV,15, 16 mitomycin-C,17 and CDDP regimens combined with docetaxel18 or epirubicin.19, 20 RRs have ranged from 9% to 41%, with the median time to progression (TTP) ranging from 1.9 months to 5.6 months and the median survival ranging from 5.8 months to 15.0 months. However, a Phase III RCT showed no significant survival benefit with UFT and mitomycin-C compared with CIV 5-FU, with or without CDDP.17 Another oral 5-FU chemotherapeutic agent, S-1, which contains tegafur, 5-chloro-2,4-dihydroxypyridine, and potassium oxonate, has been used as a single agent21–23 and with CDDP.24
Capecitabine is a novel, orally administered fluoropyrimidine carbamate that is absorbed readily by the gastrointestinal tract and is metabolized by the liver, where it is converted initially to 5′-deoxy-5-fluorocytidine (5′-DFCR) and subsequently, to 5′-deoxy-5-fluorouridine (5′-DFUR).25, 26 Designed to mimic CIV 5-FU, oral capecitabine predominantly concentrates in tumor tissue (Fig. 1).26 Capecitabine currently is approved as a single agent for the adjuvant treatment of Stage III colon cancer, as first-line treatment for metastatic colorectal cancer (when fluoropyrimidine therapy alone is preferred), and for metastatic breast cancer both as a single agent (for patients who have disease that is resistant to paclitaxel and anthracyclines) and in combination with docetaxel (after failure on anthracycline-based therapy).27 This article summarizes the efficacy and safety of oral capecitabine for the treatment of gastric, gastroesophageal, and esophageal cancers.
The MEDLINE data base was searched for English-language clinical trials that were published from 1996 through October 2005 and that contained the terms “capecitabine” or “Xeloda” and “gastric,” “stomach,” “esophagus,” “esophageal,” and/or “gastro esophageal” in the title or abstract. Potentially relevant abstracts that were presented at annual meetings or gastrointestinal symposia of the American Society of Clinical Oncology and from meetings of the European Cancer Conference and the European Society of Medical Oncology dating back to the Year 2000 were examined. Only the most recent data from reported trials were included. Missing data were designated as “not available,” unless it was possible to infer or extrapolate results from other data supplied by authors.
Efficacy data extracted from trials included complete response (CR) rates, partial response (PR) rates, RRs (CRs plus PRs), stable disease rate (SD), and median survival. Safety data extracted included all reported Grade 3 and/or Grade 4 adverse events (AEs) and treatment-related mortality. If variance data were not supplied for proportions, then 95% confidence intervals (95% CIs) were calculated using the proportions method. Two-tailed Fisher exact tests with significance set at P < .05 were employed within RCTs if they were not provided by trial investigators.
The preponderance of trials of capecitabine for the treatment of metastatic gastric, esophageal, or gastroesophageal cancers have been in the first-line setting.
Several Phase II trials28–30 have evaluated the efficacy and safety of capecitabine as a single agent for the treatment of gastric cancer (Tables 2 and 3). Hong et al.28 treated patients with oral capecitabine 1250 mg/m2 twice daily for 2 weeks on/1 week off, whereas Sakamoto et al.31 treated patients with a reduced dose of capecitabine 828 mg/m2 twice daily for 3 weeks on/1 week off. CRs were infrequent. Overall activity and median survival were similar in both trials. In a trial by Koizumi et al.,30 survival was similar but low (RR, 6%; clinical benefit rate, 19%), possibly because some patients had received prior chemotherapy. Grade 3 or 4 hematologic AEs were rare. The only Grade 3 AE that was observed in >10% of patients was hand-foot syndrome (HFS), which was observed in only 1 trial.31
Table 2. Activity and Survival in Phase II and III Trials of Capecitabine-Based Therapies for Metastatic Gastric, Gastroesophageal, or Esophageal Cancers
XELOX: X (label) + oxaliplatin 85 mg/m2 D1 q3 weeks
Table 3. Summary of Grade 3 and 4 Adverse Events Occurring in ≥10% of Patients in Phase II and III Trials of Capecitabine-Based Therapies for Metastatic Gastric, Gastroesophageal, or Esophageal Cancers
To date, there have been no published trials of single-agent capecitabine therapy for the treatment of esophageal or gastroesophageal cancers.
Several Phase II trials have used the labeled dose of capecitabine (1250 mg/m2 twice daily for 2 weeks on/1 week off) with CDDP 60 mg/m2 on Day 1 in 3-week cycles for untreated patients.32–34 The combination was active, with RRs ranging from 28% to 55%. Jin35 used lower-dose capecitabine (1000 mg/m2 twice daily for 2 weeks on/1 week off) with CDDP 20 mg/m2 on Days 1 through 5 and reported a CR rate of 10%. Dose reductions were required in 35% to 63% of patients in trials that used label-dose capecitabine,32, 34 but no dose reductions were needed when lower-dose capecitabine was used.35 Similarly, Grade 3 or 4 toxicities were less frequent in trials that used lower doses of capecitabine.
Cho et al.36 used a lower dose regimen of 3 weeks of capecitabine 1000 mg/m2 twice daily 2 weeks on/1 week off and CDDP 60 mg/m2 on Day 1 in combination with epirubicin 50 mg/m2 on Day 1. In that trial, there was no increase in activity.
In a Phase II RCT, Koizumi et al.37 treated 61 patients on 4-week cycles of oral 5′-DFUR, an oral intermediate of capecitabine, at a dose of 1000 mg/m2 on Days 41 through 7, Days 111 through 14, Days 181 through 21, and Days 251 through 28 and CDDP 70 mg/m2 on Day 1 with or without mitomycin-C 7 mg/m2 on Day 2. Response rates, clinical benefit rates, and median survival were greater in patients who received mitomycin-C (25% vs. 17%, 66% vs. 41%, and 7.9 months vs. 5.9 months, respectively) but were not significant. Grade 3 or 4 toxicities were similar.
Evans et al.38 reported a Phase I/pharmacokinetic trial of escalating doses of oral capecitabine at doses of 500 mg/m2, 825 mg/m2, 1000 mg/m2, and 1250 mg/m2 twice daily for 14 days with fixed-dose epirubicin (50 mg/m2) and CDDP (60 mg/m2) in 3-week cycles in untreated patients. The recommended dose of capecitabine was 1000 mg/m2 twice daily.
In a Phase II trial, Lee et al.39 reported on 20 patients who received label-dose capecitabine and CDDP. Activity and toxicity were similar to those seen in patients with gastric cancer.32–34
Capecitabine-Taxane Combination Therapies
Gastric and gastroesophageal cancer
Tables 2 and 3 summarize the trials of combination of capecitabine and a taxane.40–46 The labeled dose of capecitabine and docetaxel 75 mg on Day 1 in 3-week cycles44, 45 and a variety of lower doses of capecitabine with various doses and schedules of docetaxel or paclitaxel40–43, 46 have been reported, with RRs rates ranging from 37% to 67%. In those trials, the median TTP and overall survival ranged from 4.5 months to 5.3 months and from 10.5 months to 17.2 months, respectively. Generally, capecitabine-taxane regimens were well tolerated. High incidence rates of Grade 3 HFS (23-52%) were reported in trials that used a capecitabine dose ≥1000 mg/m2 twice daily with docetaxel.41, 44, 45 A high incidence of Grade 3 or 4 neutropenia (41% and 15%, respectively) was observed in 2 trials44, 46 but not in others.
Other agents also have been added to combinations of capecitabine and a taxane. In a Phase II trial, Damstrup et al.47 reported on 36 patients with advanced gastric or gastroesophageal cancer who received capecitabine 1000 mg/m2 twice daily, docetaxel 60 mg/m2, and carboplatin (area under the curve = 5) in 4-week cycles. In that trial, the RR was 62%, and the median survival was 11.3 months. The regimen was well tolerated, and ≤6% of patients had Grade 3 or 4 toxicity.
The intermediate catabolite of capecitabine, 5′-DFUR, also has been used in combination with paclitaxel. In a Phase II trial,48 72 patients received 5′-DFUR at a dose of 533 mg/m2 on Days 1 through 5 with paclitaxel 80 mg/m2 on Days 1, 8, and 15 in 4-week cycles. The regimen was well tolerated and had the expected activity in this setting.
Lorenzen et al.49 treated 24 patients (16 chemonaive patients and 8 patients who received prior chemotherapy for metastatic disease) with capecitabine at a dose of 1000 mg/m2 twice daily on Days 1 through 14 and docetaxel 75 mg/m2 on Day 1 of 3-week cycles. The RR was 46%. However, 41% of patients in that study required a dose reduction, and there were 2 treatment-related deaths, a high incidence of Grade 3 or 4 neutropenia (42%), and a high incidence of Grade 3 HFS (29%).
Capecitabine-Irinotecan Combination Therapies
Gastric and gastroesophageal cancer
In an ongoing Phase II trial, Kattan et al.50 treated 21 previously untreated patients with capecitabine 625 mg/m2 twice daily on Days 1 through 14 and irinotecan 80 mg/m2 on Days 1, 8, and 15 in 4-week cycles (XELIRI) (Tables 2 and 3). The RR was 43%, and the regimen was well tolerated. Jackson et al.51 treated 32 chemonaive patients with capecitabine 1000 mg/m2 twice daily on Days 1 through 9 and irinotecan 150 mg/m2 on Day 1 of 2-week cycles. That regimen resulted in frequent dose reductions and Grade 3 toxicities.
Assersohn et al.52 treated 25 patients with capecitabine 1000 mg/m2 twice daily on Days 11 through 14 and irinotecan 250 mg/m2 on Day 1 of 3-week cycles. The RR was 14%, and there was a 32% rate of Grade 3 or 4 neutropenia.
To date, there have been no reported clinical trials of XELIRI combination therapies for the treatment of metastatic esophageal cancer.
Capecitabine-Oxaliplatin Combination Therapies
Gastric, gastroesophageal, and esophageal cancer
Capecitabine 1000 mg/m2 twice daily on Days 1 through 14 (a dose that subsequently was reduced to 850 mg/m2 twice daily after 4 treatment-related deaths) with oxaliplatin 130 mg/m2 on Day 1 in 3-week cycles (XELOX) has been used to treat patients with metastatic gastric, gastroesophageal, or esophageal cancer (Jatoi et al.53) or exclusively in patients with metastatic gastric cancer (Park et al.54) in the first-line setting. The RR was 65% in the trial by Park et al.54 versus 35% in the trial by Jatoi et al.53 Patients with gastric cancer have a lower tolerance to chemotherapy, which is predictable based on experience with other chemotherapy regimens. In the trial by Park et al.,54 there were 3 treatment-related deaths and Grade 3 or 4 diarrhea (32%), nausea (27%), asthenia (23%), and emesis (21%).
XELOX also has been combined with other chemotherapeutic agents. In an ongoing Phase III RCT, Sumpter et al.55 treated 204 untreated patients with 3-week cycles of epirubicin 50 mg/m2 as an intravenous bolus with 1) CDDP 60 mg/m2 every 3 weeks and CIV 5-FU 200 mg/m2 per day (ECF); 2) oxaliplatin 130 mg/m2 every 3 weeks and 5-FU 200 mg/m2 per day; 3) CDDP 60 mg/m2 every 3 weeks and oral capecitabine (500 mg/m2 escalated to) 625 mg/m2 twice daily continuously; or 4) oxaliplatin 130 mg/m2 every 3 weeks and oral capecitabine (500 mg/m2 escalated to) 625 mg/m2 twice daily continuously. All regimens were active (RR range, 31-48%). The RR was greater in capecitabine-treatment arms than in 5-FU-treatment arms (41% vs. 36%, respectively). These are preliminary results, and no survival data are available to date. All 4 regimens were well tolerated.
Petrovic et al.56 treated 17 patients with label-dose capecitabine and oxaliplatin 85 mg/m2 on Day 1 of 3-week cycles. The RR was 25%, and tolerance to the regimen was fair.
Identifying the patient populations that are most likely to respond to treatment or develop significant toxic effects is essential. Thymidine phosphorylase (TP) and/or dihydropyrimidine (DPD) have been associated with tumor response to capecitabine therapy in other cancers.57–66 The data in gastric cancer patients are limited. Koizumi et al.67 stained primary gastric cancers for anti-TP and anti-DPD antibodies in 24 patients who were participants in a Phase II trial of 4-week cycles of capecitabine 828 mg/m2 twice daily on Days 1 through 21. The RR in that trial was 44% among 18 patients with TP-positive primary tumors, 0% among 6 patients with TP-negative tumors, and 67% (6 of 9 patients) in patients with TP-positive/DPD-negative primary tumors (P = .021), indicating that the TP:DPD ratio may provide useful prognostic information. Nishina et al.68 investigated the use of the TP:DPD ratio for predicting clinical responses in 22 patients who received oral 5′-DFUR. By using the median TP:DPD ratio (1.9) as a cut-off level, patients who had a high TP:DPD ratio had a significantly greater RR than patients who had a low TP:DPD ratio (64% vs. 9.1%, respectively; P = .024) and a greater median survival (9.9 months vs. 6.0 months, respectively; P = .047). However, severe AEs did not correlate with the TP:DPD ratio.
Because it mimics CIV 5-FU and has radiosensitizing properties, capecitabine has been used in conjunction with radiotherapy in the adjuvant setting. In an ongoing Phase I/II trial, Boot et al.69 treated 26 patients with T2-T4,N0-N3,M0 gastric cancer with 45 grays (Gy) (in 25 fractions of 1.8 Gy) in conjunction with oral capecitabine escalated from 600 mg/m2 to 800 mg/m2 twice daily and observed no Grade 3 or 4 AEs. In another ongoing Phase I/II trial, Jansen et al.70 treated patients with capecitabine 1000 mg/m2 twice daily for 2 weeks, followed by 45 Gy (in 25 fractions) with CDDP escalated from 3 mg/m2 to 6 mg/m2 at 1 hour preradiotherapy and capecitabine escalated from 250 mg/m2 to 800 mg/m2 twice daily on radiotherapy days. Sixteen patients on that trial had completed therapy without treatment-related mortality. In a small Phase II trial, Kwon et al.71 treated 21 patients with Stage IIIA to IV, M0 gastric cancer after curative resection with 4 cycles of CIV 5-FU 1000 mg/m2 on Days 1 through 5, CDDP 60 mg/m2 on Day 1, followed by 45 Gy (1.8 Gy per day) with capecitabine 825 mg/m2 twice daily throughout radiotherapy. The recurrence rate was only 10% after a 14.2-month median follow-up. There were no treatment-related deaths, and the most common Grade 3 or 4 AEs were neutropenia (19%), thrombocytopenia (5%), fatigue (5%), and nausea/emesis (5%). These preliminary results indicate that radiotherapy regimens appear to be well tolerated.
Kwon et al.72 also conducted an RCT comparing chemotherapy with radiotherapy (45 Gy) and CIV 5-FU 1000 mg/m2 on Days 1 through 5 or with oral capecitabine 825 mg/m2 twice daily throughout radiotherapy or chemotherapy alone in 81 patients with postoperative, Stage IIIA or IV-M0 disease. No firm conclusions could be drawn because of the small number of patients in that trial; however, there were no significant differences among treatment groups with regard to 2-year disease-free survival (70% vs. 60%, respectively) or overall survival (85% vs. 79%, respectively). However, significantly more Grade 3 and 4 neutropenia was observed with intravenous chemotherapy (49% vs. 23%, respectively; P = .03).
Doxifluridine also has been evaluated as a single-agent in the adjuvant setting. In a Phase III RCT, Takiguchi et al.73 compared oral 5′-DFUR 460 mg/m2 per day with oral 5-FU 115 mg/m2 per day in 482 patients with postoperative Stage II or III gastric cancer. Although the overall 5-year survival was not significantly greater with 5′-DFUR than with oral 5-FU, patients with Stage IIIB disease who received 5′-DFUR had superior 5-year disease-free survival (56.4% vs. 31.4%; P = .023). Multivariate analysis showed that patients with serosal invasion fared better with 5′-DFUR than with oral 5-FU (risk ratio, 1.65; 95% CI, 1.11-2.44).
Ongoing trials are primarily Phase II investigations of capecitabine combination therapies for treatment in the metastatic setting, largely using capecitabine and taxane74–76 or XELOX77, 78 and also using combination therapies of XELOX with irinotecan,79 capecitabine with CDDP and epirubicin,80 and capecitabine with matuzumab,81, 82 which is a monoclonal antibody specific for the epidermal growth factor receptor (EGFR) (Table 4). All but 1 trial is intended for patients who have not received prior chemotherapy for metastatic disease. Agents that target vascular endothelial growth factor (bevacizumab) or EGFR (e.g., erlotinib), which are in early stages of development for the treatment of gastric cancer,83, 84 may yield dynamic, effective therapies when used in combination with capecitabine.
Table 4. Ongoing Trials of Capecitabine for the Treatment of Gastric, Gastroesophageal, or Esophageal Cancers
NCCTG indicates North Central Cancer Treatment Group; NCI, National Cancer Institute; XT, capecitabine and taxane; BID, twice daily; D, day(s); q, every; CA, cancer; GE, gastroesophageal; QOL, quality of life; XELOX, capecitabine and oxaliplatin; CEX, carboplatin, epirubicin, and capecitabine; MTD, maximum tolerated dose; AUC, area under the curve; EMD, EMD Pharmaceuticals; CDDP: cisplatin.
Superscript numbers indicate the list of references.
Gastric, GE, or esophageal CA; metastatic or recurrent; 1st line
Safety and tolerability
Oral fluoropyrimidines such as S-1 and capecitabine are being investigated actively for the treatment of gastroesophageal cancer in the first-line metastatic and adjuvant settings. Combination therapies include capecitabine with CDDP, taxanes, irinotecan, and oxaliplatin. Phase II trials of capecitabine in combination with CDDP and with docetaxel have shown favorable efficacy data compared with Phase III trials of parenteral 5-FU combinations in the first-line, metastatic setting. Capecitabine-based regimens mostly are well tolerated. Additional Phase III trials are warranted to demonstrate the comparative efficacy and safety of capecitabine-based combination therapies in the metastatic and adjuvant settings. Oral fluoropyrimidines are emerging chemotherapy agents for patients with gastric, gastroesophageal, and esophageal cancers.
The author acknowledges Nelson Erlick, DPM, MS, for his editorial assistance in the preparation of this article.