RATIONALE: Screening tests may help doctors detect cancer cells early and plan more effective treatment for colorectal cancer.

PURPOSE: Randomized screening trial to compare the effectiveness of fecal occult blood testing with that of DNA-based testing of stool and blood in identifying colorectal cancer.

Primary objectives:

1. To compare the performance characteristics (sensitivity, specificity, predictive values) of the fecal MTAP and most widely-used fecal occult blood test (Hemoccult) for identification of screen-relevant colorectal neoplasia (curablestage cancer and advanced adenomas with high-grade dysplasia/carcinoma in situ or size ≥ 1.0 cm.)
2. To evaluate the necessity of formal pretest preparation for MTAP by comparing the specificity of both the MTAP and Hemoccult tests in subject groups randomized to pre-test preparation versus no pre-test preparation.

Secondary objectives:

1. To compare detection rates of colorectal neoplasia by the fecal MTAP alone with those by flexible sigmoidoscopy (distal 60 cm of colonoscopic examination to serve as surrogate) and by the combination of sigmoidoscopy plus Hemoccult.
2. To characterize and compare the pathological and molecular features of screen-relevant colorectal neoplasms detected and not detected by the fecal MTAP.
3. To explore the sensitivity and specificity of the MTAP applied to plasma for the detection of screen-relevant colorectal neoplasia.
4. To maintain a specimen bank comprising stools and blood (plasma) from all subjects and tissue from screen-relevant neoplasms.

OUTLINE: This is a randomized, multicenter study. Participants are stratified according to age (50-64 [closed to accrual as of 6/5/03] vs 65-80), gender (male vs female), and participating center. Participants are randomized to one of two screening arms.

• Arm I: Participants eat no red meat and take no nonsteroidal anti-inflammatory drugs (NSAIDs) and no vitamin C or multivitamins for 3 days prior to and during stool sample collection. Participants collect stool samples 3 different times and perform fecal occult blood (FOB) test smears from each stool. After each collection, participants ship the whole stool and FOB test smear to their participating center for blinded multitarget DNA-based assay panel (MTAP) testing.
• Arm II: Participants take no vitamin C or multivitamins for 3 days before and during stool sample collection. Participants collect stool samples and FOB test smears and samples are tested as in arm I.

Within 2 months after stool sample collection, participants have their blood drawn for additional MTAP testing and undergo colonoscopy.

PROJECTED ACCRUAL: A total of 4,000 participants (2,000 per arm) will be accrued for this study.

Required Characteristics:

1. ≥ 65 and ≤ 80 years of age.
2. Females must be postmenopausal (absence of menstrual periods for at least one year; patients on regular hormone replacement therapy; surgical intervention).

Contraindications:

1. FOBT screening ≤1 year prior to randomization.

2. Structural colorectal evaluation (i.e. colonoscopy, colon x-ray, or sigmoidoscopy) ≤10 years prior to randomization.

3. Overt rectal bleeding (hematochezia or melena) ≤1 month prior to randomization.

4. Prior colorectal resection for any reason.

5. Inability to stop therapeutic doses of NSAIDs (prophylactic doses of aspirin (≤325 mg) allowed [121] and Cox2 inhibitors (i.e. Celebrex, Vioxx) allowed.)

6. Coagulopathy or inablitity to discontinue anticoagulants (discontinuation must be superviesed by a physician).

7. Aerodigestive cancer ≤5 years prior to randomization.

8. Contraindications to colonoscopy (e.g., serious cardiopulmonary disease).

9. High-risk conditions for colorectal cancer (familial adenomatous polyposis, hereditary nonpolyposis colorectal cancer syndrome, other hereditary cancer syndromes, prior colorectal cancer or adenoma, inflammatory bowel disease, and

   • 2 first-degree relatives with colorectal cancer).

10. Chemotherapy ≤ 3 months prior to registration.

Patients aged 65 -80 years