Current issues of ACP Journal Club are published in Annals of Internal Medicine


β-carotene and vitamins C and E did not reduce the occurrence of colorectal adenomas

ACP J Club. 1994 Nov-Dec;121:75. doi:10.7326/ACPJC-1994-121-3-075

Related Content in the Archives
Calcium supplements reduced recurrence of colorectal adenomas

Source Citation

Greenberg ER, Baron JA, Tosteson TD, et al. for the Polyp Prevention Study Group. A clinical trial of antioxidant vitamins to prevent colorectal adenoma. N Engl J Med. 1994 Jul 21;331:141-7.



To determine the efficacy of β-carotene and vitamins C and E in preventing colorectal adenoma in patients with a previous adenoma.


2 by 2 factorial design, randomized, double-blind, controlled trial with a 4-year follow-up (Polyp Prevention Study).


6 tertiary care centers.


Eligible patients had an adenoma diagnosed within 3 months of study entry and were confirmed by colonoscopy to be free of further polyps, in good health, and < 80 years of age. Patients with familial polyposis, a history of invasive colorectal cancer, malabsorption syndromes, or conditions that might be worsened by vitamin C or E supplementation were excluded. Of 2029 eligible patients, 864 completed a 3-month placebo run-in period and were enrolled. 751 (87%) patients completed the study.


Patients were stratified by study center and randomly allocated to either placebo (n = 187); β-carotene, 25 mg/d, plus placebo (n = 184); vitamin C, 1 g/d, plus vitamin E, 400 mg/d (n = 205); or β-carotene plus vitamins C and E (n = 175). Every 6 months, patients were asked about their adherence to the treatment regimen and about any symptoms, illnesses, or hospitalizations.

Main outcome measures

Occurrence of new adenomas between follow-up colonoscopic examinations conducted at year 1 and year 4.

Main results

Between the 1st and 4th year, 279 (37%) patients developed at least 1 adenoma. The groups did not differ for number of adenomas or median size of the largest adenoma. The proportion of patients with an adenoma did not differ for the groups receiving and not receiving β-carotene (37% vs 38%, P = 0.8) or for the groups receiving and not receiving vitamins C and E (38% vs 36%, P = 0.6). The statistical models of treatment effect showed no protective effect of either β-carotene or vitamins C and E. No reduction in risk for adenoma was noted in any subgroup determined by age, sex, or number of previous adenomas.


The occurrence of new colorectal adenomas was not affected by treatment with β-carotene or vitamins C and E in patients previously diagnosed with an adenoma.

Source of funding: National Institutes of Health.

For article reprint: Dr. E.R. Greenberg, Dartmouth Medical School, Hanover, NH 03755-3861, USA. FAX 603-650-4150.


Colorectal cancer is the second leading cause of cancer mortality in the United States. Approximately 150 000 cases and 60 000 deaths occur each year. Because early lesions have an excellent cure rate, research efforts have been aimed at screening and primary prevention. Ecologic studies have shown that populations that consume large amounts of fruit and vegetables have lower rates of colorectal cancer (1). These foods are rich in antioxidant micronutrients, now touted as potential chemopreventive agents.

The strengths of the study by Greenberg and colleagues include its randomized, blinded design and complete ascertainment of end points (recurrent adenomas). Previous studies, which have reported conflicting results, are limited by small sample sizes, incomplete follow-up, and uncontrolled design. The Greenberg study is the most methodologically sound to date.

Possible reasons for the study's negative finding include a true lack of efficacy of antioxidants, an inadequate follow-up time, an insufficient duration of intervention, or an insufficient dose. Patients with cancer and those without cancer consumed similar foods—this homogeneity with regard to fat and fiber intake (known to affect cancer risk) might be a stronger factor than vitamin supplementation. Other endogenous substances in plants, such as sulforaphane and quinone reductases, may be responsible for the protective effect.

An important caveat in analyzing all polyp prevention data is the fact that adenomas are intermediate end points. They are used because these studies can be done more quickly and at less cost than those evaluating development of colorectal cancers. Despite the accepted "adenoma-carcinoma sequence" (2), few small adenomas grow into carcinoma. Antioxidants may only play a role in preventing small adenomas from growing into larger lesions with greater malignant potential (which this study could not address).

This is the second published trial to find no effect of antioxidants in preventing cancer (or a precursor) (3). At this point, patients might best be advised to adhere to a diet rich in fruit and vegetables rather than popping multivitamins while eating a double cheeseburger with fries.

Glenn Eisen, MD
University of North CarolinaChapel Hill, North Carolina, USA


1. Willett W The search for the causes of breast and colon cancer. Nature. 1989;338:389-94.

2. Hill MJ. From adenoma to carcinoma of the colorectum. Recent Results Cancer Res. 1991;122:71-84.

3. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med. 1994;330:1029-35.