The primary end point of the study is confirmed objective response rate (complete response [CR] and partial response [PR]). A response rate of 80 percent for cetuximab plus cisplatin and weekly 24-hour infusion of high-dose 5-fluorouracil and leucovorin (P-HDFL) chemotherapy is assumed. The Simon two-stage design will be used for P1 - P0 = 0.20. The response rates of interest are P0 = 60% and P1 = 80%. The investigators will reject cetuximab plus P-HDFL chemotherapy if the response rate is 8/13 at the first stage, and will reject the cetuximab plus P-HDFL chemotherapy if the response rate is 25/35 at the second stage. If there are more than 8 responses in 13 patients in the first stage, the study will continue to a total of 35 patients in the second stage. If there are more than 25 responses in 35 patients in the second stage, this treatment will be acceptable with a p-value of 0.05 and of 0.20. Evaluable patients for response will be those who received at least 4 doses of cetuximab (i.e. one cycle of protocol treatment). All enrolled patients will be subjected to toxicity evaluations.

The primary end point of the study is confirmed objective response rates (by RECIST, Response Evaluation Criteria in Solid Tumors). The secondary end points of the study are progression-free survival, overall survival, and treatment-related toxicities.

The analysis of response to treatment will be restricted to the eligible patients with at least one measurable lesion. The safety analysis will be restricted to the patients who received at least one cycle of the administered chemotherapy. The time-to-event end points will be estimated using the method of Kaplan and Meier and based on the intent-to-treat principle. Overall survival will be defined as the time interval between the date of study entry and the date of death. Progression-free survival will be defined as the time interval between the date of study entry and the date of disease progression or death, whichever occurred first. Duration of response will be defined as the time interval between the date of initial objective response and the date of disease progression, which is only for responders. If the event is not yet observed at the time of the last record, the patient will be censored at that time point.

Non-resectable gastric cancer is an incurable disease. Systemic chemotherapy confers prolongation of survival and improvement of quality of life. Regimens containing cisplatin and 5-fluorouracil (5-FU) are widely adopted in the world. A P-HDFL regimen, based primarily on weekly 24-hour infusion of cisplatin and high-dose 5-FU and leucovorin, is commonly used in Taiwan for patients with advanced gastric cancer; the overall response rate is around 60% (45%-76%, 95% C.I.), and the patients' compliance is excellent.

Expression of EGF and EGFR was detected in about 62.1% and 51.5% gastric cancer tissues, respectively. Tumors with simultaneous expression of EGF, TGF-alpha, EGFR and p185c-erbB-2 were associated with a high BrdU labeling index, rapid tumor growth, and an unfavorable outcome. Teramoto et al. have shown a dose-dependent growth inhibition of an anti-EGFR monoclonal antibody (MoAb 528) on EGFR-overexpressing human gastric cancer cells, both in vitro and in vivo. We have recently demonstrated that cetuximab is cytotoxic to human gastric cancer cells, and cetuximab has a chemo-sensitizing effect for cisplatin and 5-FU in human gastric cancer cells (Yeh et al, unpublished observation). We hypothesize that the combination of cetuximab with P-HDFL will further improve the efficacy of the latter on advanced gastric cancer.

This is a multi-center, prospective phase II trial. Patients with histologically proven nonresectable or recurrent/metastatic gastric adenocarcinoma, with at least one measurable lesion, no prior chemotherapy or radiotherapy, and adequate baseline organ functions will be eligible. Cetuximab (Erbitux; Merck, Germany) 400 mg/m2 will be given as an intravenous (IV) infusion, initially (i.e., day 1 of cycle 1); and then followed by weekly IV infusions of cetuximab 250 mg/m2 (i.e., days 8, 15, 22 of cycle 1, and days 1, 8, 15, 22 of cycle 2 and subsequent 28-day cycles). Cisplatin will be given as a 24-hour continuous IV infusion in a dose of 35 mg/m2/day, admixing with 5-FU 2,000 mg/m2 and leucovorin (folinic acid) 300 mg/m2, day 1 and day 8. A 24-hour continuous IV infusion of 5-FU 2,000 mg/m2 and leucovorin 300 mg/m2 will be given on day 15. The cycles will be repeated every 28 days, and the response evaluation will be performed every two cycles and at the end of protocol treatment. Confirmed objective response rate by RECIST (Response Evaluation Criteria in Solid Tumors) will be the primary end-point. Using the Simon two-stage design for phase II study (P1 - P0 = 0.20 with the response rates of interest being P0 = 60% and P1 = 80%), a total of 35 patients are planned to be enrolled in an estimated enrollment period of 12 to 18 months.

Inclusion Criteria:

1. Age 18 to 75 years
2. Histologically proven adenocarcinoma of the stomach that is nonresectable, locally advanced, or recurrent/metastatic
3. At least one measurable lesion (by RECIST, Response Evaluation Criteria in Solid Tumors), and no prior radiotherapy to the target measurable lesion
4. No prior chemotherapy for gastric cancer, but post-gastrectomy adjuvant therapy with low-dose 5-FU [e.g., 5-FU 450 mg/m2 per week] completed more than 6 months before study enrollment is acceptable
5. World Health Organization (WHO) performance status 2
6. Adequate baseline organ functions (checked within one week before entry into this study), defined as WBC count 3,000 cells/µL with neutrophils ≥ 1,500 cells/µL, platelet count 100,000 cells/µL, hemoglobin 9 g/dL, serum total bilirubin level 1.5 X UNLs (upper normal limits), serum AST and ALT 2.5 X ULNs (or serum AST and ALT 5.0 X ULNs for patients with liver metastases), serum creatinine level 1.5 X UNLs, 24-hour urine CCr 60 ml/min
7. Fasting serum triglyceride level > 70 mg/dL, which should be checked within one week before entry into this study. The lower limit for fasting serum triglyceride (70 mg/dL) is set to avoid HDFL-related hyperammonemic encephalopathy, which occurs in around 5% of Taiwanese patients.
8. Written informed consent
9. At least one month from gastrectomy, in case gastrectomy was performed; at least 2 weeks from laparotomy without resection, in case laparotomy was performed to document nonresectable status
10. Availability of tumor sample for retrospective testing of EGFR (pharmacogenomic mutation analysis and immunohistochemical staining).

Exclusion Criteria:

1. Concomitant anti-cancer biological agents, chemotherapy, or radiotherapy other than indicated in this protocol

2. CNS metastasis

3. Pregnant women, breast-feeding women, and women of child-bearing potential or fertile men without adequate contraception

4. Life expectancy less than 3 months

5. Serious concomitant illness or significant dysfunction of major organ systems which prohibit chemotherapy, such as:

   • symptomatic heart disease, including significant arrhythmias, congestive heart failure or myocardial infarction within 12 months.
   • extensive liver disease.
   • major active infection.
   • severe symptomatic pulmonary disease.

6. Concurrent or prior second malignancy (except curatively resected cervical carcinoma in situ or squamous cell carcinoma of skin).

7. Known hypersensitivity reaction to any of the components of study treatments.

8. Medical or psychological conditions that would not permit the patient to complete the study or sign informed consent

9. Significant diseases, in the investigator's opinion, which would exclude the patient from the study.

10. Legal incapacity or limited legal capacity.