There is no clear standard of care for metastatic stomach or esophageal cancer in the United States. The purpose of this research study is to determine the differences between two regimens of chemotherapy; Arm A: PCA (Cisplatin, Irinotecan and Bevacizumab) and Arm B: TPCA (Docetaxel, Cisplatin, Irinotecan and Bevacizumab). Docetaxel, Cisplatin, and Irinotecan are traditional chemotherapy drugs. Bevacizumab is an antibody (a protein that attacks a foreign substance in the body). Bevacizumab is believed to stop the formation of new blood vessels that carry nutrients to tumors. Both of the chemotherapy regimens (PCA and TPCA) have been studied in patients with esophageal and gastric cancer, and we are trying to determine if one regimen will keep your cancer from growing and improve how long you can live.

OBJECTIVES:

Primary

* To evaluate progression-free survival at 7 months in metastatic esophageal and gastric patients treated with either PCA or TPCA

Secondary

• To determine overall survival
• To determine the response rate (RECIST) in measurable disease patients
• To evaluate type and severity of toxicities associated with each regimen

Exploratory:

• To correlate expression of tumoral and serum VEGF with response and survival
• To correlate TGF alpha levels and tumor microvessel density with clinical activity of the combination of PCA or TPCA
• To examine circulating endothelial cells (CECs) as surrogate markers of antitumor activity of bevacizumab
• To explore if 7/7 and 7/6 UGT1A1 polymorphisms correlate with grade III/IV irinotecan-related diarrhea and neutropenia when irinotecan is given at relatively low dose to patients with esophageal and gastric cancer

DESIGN:

This trial was designed to compare 7-month progression-free survival between arms. The hypothesis was that TPCA would have superior outcome over PCA (70% vs 50%). With 40 eligible patients per arm followed for 1 year there was 80% power to detect a hazard ratio of 0.48 using the log-rank test at a one-sided type I error rate of 5%. Stratification factors were ECOG performance status 0/1 vs 2 and site of primary tumor (gastric vs GE junction/esophageal).

Inclusion Criteria:

• Histologically confirmed, unresectable esophageal, GE junction or gastric adenocarcinoma (including adenosquamous, or undifferentiated carcinoma). Measurable disease is not required.
• 18 years of age or older
• ECOG Performance Status=2
• Life expectancy of 12 weeks or greater
• Adequate bone marrow, renal and liver function as outlined in the protocol.
• Men and women of childbearing potential must use adequate contraception

Exclusion Criteria:

• Prior chemotherapy (except as part of pre- or post-operative therapy, completed at least 1 prior to start of this protocol).
• Squamous cell carcinoma histology of esophageal, GE junction or gastric tumor
• Known history of allergy or hypersensitivity to Chinese hamster ovary products, polysorbate 80, or any of the study drugs
• Treatment or planned participation in an experimental drug study within 4 weeks of C1 D1. Concurrent use of herbal medications or other alternative therapies
• Major surgical procedures, such as fine needle aspirations, port-a-cath placement, or core biopsies, within 7 days of cycle 1 day 1
• Palliative radiation to 25% or less of bone marrow, must be completed > 2 weeks prior to day 1, palliative radiation to > 25% of bone marrow, must be completed > 4 weeks prior to day 1
• Myocardial infarction, unstable angina, CVA or TIA or other thrombotic event in the past six months
• Inadequately controlled hypertension (defined as systolic blood pressure of >150mmHg and/or diastolic blood pressure of > 100mmHg). Initiation of antihypertensive medication is recommended, however adequate control of blood pressure must be documented prior to C1 D1
• No history of prior hypertensive crisis or hypertensive encephalopathy
• NYHA Grade II or greater congestive heart failure
• Clinically significant peripheral vascular disease
• Active bleeding from primary tumor
• Evidence of bleeding diatheses or coagulopathy (other than deep venous thrombosis, portal vein thrombosis, pulmonary embolism, or atrial fibrillation). Patients on therapeutic anticoagulation may be enrolled provided they have been clinically stable on anticoagulation for a least 2 weeks prior to C1 D1.
• Uncontrolled serious medical or psychiatric illness
• Uncontrolled diarrhea
• Peripheral neuropathy
• No known brain or other CNS metastasis by history or clinical examination
• Other active malignancy other than non-melanoma skin cancer or in-situ cervical carcinoma. A resected or previously treated cancer (other than in-situ carcinoma) must have demonstrated no evidence of recurrence for at least 3 years
• Urine protein:creatinine ratio 1.0 or greater at screening
• History of abdominal fistula, gastrointestinal perforation, or intra-abdominal abscess with 6 months of C1 D1
• Serious, non-healing wound, ulcer or bone fracture
• Pregnant or breast feeding
• Inability to comply with study and/or follow-up procedures
• History of HIV seropositivity, hepatitis C virus, acute or chronic hepatitis B, or other serious chronic infection