RATIONALE: Drugs used in chemotherapy use different ways to stop tumor cells from dividing so they stop growing or die. Combining more than one drug may kill more tumor cells. It is not yet known which regimen of combination chemotherapy is more effective in treating patients with recurrent or metastatic colorectal cancer.

PURPOSE: Randomized phase III trial to study the effectiveness of combining fluorouracil, leucovorin, and oxaliplatin in different ways in treating patients with recurrent or metastatic colorectal cancer.

OBJECTIVES: I. Compare the effects of chronomodulated versus nonchronomodulated infusional administration of a 3 drug high-dose chemotherapy regimen on survival in patients with locoregionally recurrent or metastatic colorectal cancer. II. Assess the antitumor effect of the combination of fluorouracil, leucovorin calcium, and oxaliplatin given as first line chemotherapy in these patients. III. Assess the response rate and toxicity of this treatment in these patients. IV. Assess the quality of life of patients receiving this treatment.

OUTLINE: This is a randomized, multicenter study. Patients are stratified according to performance status (ECOG 0-1 vs 2), extent of liver involvement (none vs less than 25% vs 25% or greater), and institution. Patients are randomized into 2 arms: Arm I (chronotherapy): Patients receive a 4 day infusion of fluorouracil, leucovorin calcium, and oxaliplatin administered via pump using chronomodulated delivery rates. Arm II (fixed infusion rate): Patients receive leucovorin calcium and oxaliplatin as 2 hour concurrent infusions followed by fluorouracil as a 22 hour infusion on day 1. On day 2, patients receive a 2 hour infusion of leucovorin calcium followed by a 22 hour infusion of fluorouracil. For patients on both arms, courses repeat every 2 weeks until the occurrence of disease progression, severe toxicity, or complete remission for a minimum of 4 months. Quality of life is assessed prior to treatment and 2 weeks after courses 4 and 8. Patients are followed for survival.

PROJECTED ACCRUAL: A total of 554 patients will be accrued for this study.

DISEASE CHARACTERISTICS: Histologically proven locoregionally recurrent or metastatic adenocarcinoma of the colon or rectum Histologic or cytologic proof of adenocarcinoma consistent with colorectal cancer in a solitary lesion if plasma carcinoembryonic antigen level is no greater than normal Metastases not limited to bone, pleural effusion, or ascites Surgically resectable metastases not eligible No symptomatic brain metastasis At least one bidimensionally measurable lesion measuring at least 20 mm in one diameter outside a previously irradiated area

PATIENT CHARACTERISTICS: Age: 18 to 75 Performance status: ECOG 0-2 Life expectancy: Not specified Hematopoietic: Absolute neutrophil count at least 2,000/mm3 Platelet count at least 100,000/mm3 Hepatic: Bilirubin no greater than 3 times upper limit of normal Renal: No uncontrolled hypercalcemia Cardiovascular: No significant cardiac disease Pulmonary: No severe respiratory illness Neurologic: No peripheral sensory neuropathy Other: No uncontrolled infection or chronic disease No second malignancy except basal cell carcinoma of the skin or in situ carcinoma of the cervix

PRIOR CONCURRENT THERAPY: Biologic therapy: No prior immunotherapy for locoregionally recurrent or metastatic colorectal cancer No concurrent prophylactic growth factor Chemotherapy: No prior chemotherapy for locoregionally recurrent or metastatic colorectal cancer At least 6 months since adjuvant chemotherapy Endocrine therapy: No prior hormone therapy for locoregionally recurrent or metastatic colorectal cancer No concurrent corticosteroids Radiotherapy: See Disease Characteristics No prior radiotherapy for locoregionally recurrent or metastatic colorectal cancer Concurrent localized analgesic radiotherapy of a bone lesion permitted unless indicative of disease progression Surgery: No prior surgery for locoregionally recurrent or metastatic colorectal cancer Other: No other concurrent investigational drug