Patient with multiple metastases, not eligible for surgery, might not profit from intensive chemotherapy regimens. Therefore less intensive regimens focusing on survival and disease control may be a better choice for first line treatment. Therefore this study investigates the combination of capecitabine and bevacizumab versus the combination of capecitabine, bevacizumab and irinotecan. In case of progressive disease, the therapy in patients treated with capecitabine and bevacizumab is intensified by adding irinotecan. Primary endpoint is time-of-failure strategy (TFS) comparing both treatment arms.

Inclusion Criteria:

• Histologically confirmed adenocarcinoma of the colon or rectum.
• Stage IV disease.
• ECOG 0-1.
• Patients considered suitable for application of chemotherapy.
• Age 18 - 75 years.
• In- or outpatient treatment.
• Estimated life expectancy > 3 months.
• Measurable index lesion according to RECIST criteria. Evaluation of tumor manifestations ≤ 2 weeks prior to treatment start.
• Effective contraception.
• Adequate hematologic function: leukocytes >= 3000/µl, neutrophils >= 1500/µl, platelets >= 100.000/µ, and hemoglobin >= 9g/dl. Bilirubin <= 1,5x upper limit of normal (ULN). ALAT and ASAT <= 2,5x ULN, in case of liver metastases <= 5x ULN. Serum creatinine <= 1,5x ULN.
• No operations within 4 weeks prior to treatment start. No cytologic biopsies within 1 week prior to treatment start. Operation sequels need to be completely healed. Major operations must not be expected at time of study begin, except for potential secondary resection of liver metastases. In case of secondary resection of liver metastases, bevacizumab must be discontinued 6-8 weeks prior to surgery.
• No relevant toxicities due to prior medical treatment at time of study entry.

Exclusion Criteria:

• primary resectable metastases
• heart failure Grade III/IV (NYHA-classification)
• Prior treatment directed against the epidermal growth factor receptor (EGFR).
• Prior treatment with bevacizumab.
• Prior chemotherapy for colorectal cancer, except for adjuvant chemotherapy dating back > 6 months prior to study entry.
• Experimental medical treatment within 30 days prior to study entry.
• Known hypersensitivity reaction to any study medication.
• Pregnant or breast feeding women (pregnancy needs to be excluded by testing of beta-HCG).
• Known or suspected cerebral metastases.
• Clinically significant coronary heart disease, myocardial infarction within the last 12 months or high risk of uncontrolled arrhythmia.
• Acute or subacute ileus, chronic inflammatory bowel disease or chronic diarrhea.
• Abdominal or tracheo-esophageal fistulas, gastrointestinal perforation within 6 months before study entry
• Symptomatic peritoneal carcinosis.
• Severe chronic wounds, ulcera or bone fracture.
• Uncontrolled hypertension.
• Severe proteinuria (nephrotic syndrome).
• Arterial thromboembolic events or hemorrhage within 6 months prior to study entry (except tumor bleeding surgically treated by tumor resection).
• Bleeding diatheses or coagulopathy.
• Full dose anticoagulation.
• Known DPD-deficiency (special screening not required).
• Known glucuronidation-deficiency (special screening not required).
• Contraindication with irinotecan
• Medical history of other malignant disease within 5 years prior to study entry, except for basalioma, and in-situ cervical carcinoma if treated with curative intent.
• Known alcohol or drug abuse.
• Medical or psychiatric condition which contradicts participation of study.
• Limited legal capacity.