Polyp Prevention Trial

OBJECTIVES:

The primary objective of the Polyp Prevention Trial (PPT) is to determine whether a low fat, high fiber, high vegetable and fruit eating plan will decrease the recurrence of adenomatous polyps of the large bowel. Secondary objectives of the PPT include 1) evaluating the effectiveness of the intervention program with respect to participant achievement of dietary goals; 2) examining the relation of dietary change and biochemical markers in blood; and 3) assessing the impact of the intervention on quality of life indicators.

BACKGROUND:

Nearly 60,000 men and women will die from large bowel cancer in the United States this year, making it the second leading cause of death from malignant disease in this country. The evidence that diet plays a key role in large bowel carcinogenesis is strong and growing. The PPT was undertaken because it is unlikely, with respect to diet and large bowel cancer, that any combination of further animal research, clinical investigations employing non-neoplastic endpoints, or observational epidemiologic studies would be sufficiently persuasive to influence public health policy.

A large body of ecologic, analytic epidemiologic, human metabolic, and animal experimental data suggests that three dietary factors increase the risk of large bowel cancer: high dietary fat, low dietary fiber, and low vegetable and fruit intake. Previous dietary intervention studies have generally focused on a single nutrient. The PPT is unique in that three dietary factors are being investigated simultaneously through intervention with a realistic, comprehensive dietary pattern change. NCI investigators adopted this multiple-goal dietary intervention strategy in the PPT for several reasons: people eat foods, not isolated nutrients; there may be unknown protective food constituents not encompassed in a single-nutrient intervention; and there may well be important synergistic effects of various food components best captured in a comprehensive, multifactor dietary pattern. Moreover, by embracing three promising dietary hypotheses simultaneously, the PPT intervention is intended to maximize the possibility of reducing polyp recurrence. The direct demonstration of any effect of diet on the large bowel neoplastic process would be a major advance.

Large bowel adenomas (polyps) present a unique opportunity to conduct an intervention trial because of the high prevalence of these lesions in the general population (over 30% in adults over 50 years of age), the high polyp recurrence rate (at least 10% annually) in those who have undergone adenoma removal, and the strong link between adenomas and cancer (the so-called adenoma-carcinoma sequence). It is generally accepted that large bowel adenomas are an obligate precursor lesion for most large bowel cancers. An intervention reducing the recurrence of large bowel polyps would thus have a strong likelihood of reducing the incidence of large bowel cancer.

DESIGN:

The PPT is a randomized, controlled trial being carried out at eight Clinical Centers in the United States. The Data and Nutrition Coordinating Center (DNCC) is Westat, Inc., Rockville, Maryland. The target sample size of 2000 permits the detection with 90% power of a reduction of 24% in the polyp recurrence rate.

The dietary goals for the intervention arm are 20% of calories from fat, 18g fiber/1000 kcal, and 5-8 servings of fruits and vegetables (the exact number based on caloric intake). The usual (control) diet, based on data from national surveys, is expected to comprise approximately 35% of calories from fat, 10-15 g per day of dietary fiber, and 3.5 servings of vegetables and fruits daily.

The overall strategy for the intervention program is to reduce calories from fat and to replace them with calories from fruits vegetables and grains thus increasing dietary fiber intake. The intervention program, patterned after the approach used in the feasibility phase of the Women's Health Trial, integrates nutrition education and behavior modification techniques. Although initial counseling sessions are carried out one-on-one by the PPT nutritionists, most intervention group members participate in group counseling after the first year.

The control group is not offered a nutrition intervention program since the general strategy adopted for this group will be minimum interference with customary diets while collecting nutritional data considered necessary for appropriate comparison with the nutrition intervention group. Subjects in the control group are expected to maintain their usual diet.

Three different dietary assessment instruments are used in the PPT: a modified Block/NCI Food Frequency Questionnaire (FFQ), Four Day Food Records (4DFR), and 24 Hour Recalls.

Participants undergo colonoscopy again at one (T1) and four (T4) years into the study. Over 280 endoscopists participating in the PPT carry out the T1 and T4 colonoscopies. Although the recurrence of one or more adenomas is the primary endpoint of the study, it will also be possible to relate the dietary intervention to number, size, and histotype of polyps. Because some polyps (10-15%) tend to be missed at baseline, the primary analytic interval (on which sample size calculations were based) is T1 through T4.

The presence of one or more adenomas at the qualifying colonoscopy--a key eligibility factor for PPT--was determined by Clinical Center trial pathologists who had been previously oriented by the central pathologists to PPT pathologic criteria. The number, size, and location of adenomas at baseline was determined by the participating endoscopists. To standardize the histologic diagnoses used in endpoint ascertainment, the PPT requires that histologic type and extent of atypia be determined by the central pathologists for all lesions removed at baseline, T1, and T4.

Blood specimens are collected on all participants annually; these are analyzed on a 20% participant sample (the same individuals for whom the diet records are analyzed) for lipids, carotenoids, and vitamins A and E. Blood specimens are also available for other analyses including hormone and molecular genetic analysis.