ALPHA-TOCOPHEROL AND BETA-CAROTENE SUPPLEMENTS AND LUNG-CANCER INCIDENCE IN THE ALPHA-TOCOPHEROL, BETA-CAROTENE CANCER PREVENTION STUDY - EFFECTS OF BASE-LINE CHARACTERISTICS AND STUDY COMPLIANCE

Background: Experimental and epidemiologic investigations suggest that alpha-tocopherol (the most prevalent chemical form of vitamin E found in vegetable oils, seeds, grains, nuts, and other foods) and beta-car otene (a plant pigment and major precursor of vitamin A found in many yellow, orange, and dark-green, leafy vegetables and some fruit) might reduce the risk of cancer, particularly lung cancer. The initial find ings of the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study (A TBC Study) indicated, however, that lung cancer incidence was increase d among participants who received beta-carotene as a supplement. Simil ar results were recently reported by the Beta-Carotene and Retinol Eff icacy Trial (CARET), which tested a combination of beta-carotene and v itamin A. Purpose: We examined the effects of alpha-tocopherol and bet a-carotene supplementation on the incidence of lung canter across subg roups of participants in the ATBC Study defined by base-line character istics (e.g., age, number of cigarettes smoked, dietary or serum vitam in status, and alcohol consumption), by study compliance, and in relat ion to clinical factors, such as disease stage and histologic type. Ou r primary purpose was to determine whether the pattern of intervention effects across subgroups could facilitate further interpretation of t he main ATBC Study results and shed light on potential mechanisms of a ction and relevance to other populations. Methods: A total of 29 133 m en aged 50-69 years who smoked fire ol more cigarettes daily were rand omly assigned to receive alpha-tocopherol (50 mg), beta-carotene (20 m g), alpha-tocopherol and beta-carotene, or a placebo daily for 5-8 yea rs (median, 6.1 gears). Data regarding smoking and other risk factors for lung cancer and dietary factor's were obtained at study entry, alo ng with measurements of serum levels of alpha-tocopherol and beta-caro tene. Incident cases of lung cancer (n = 894) were identified through the Finnish Cancer Registry and death certificates. Each lung cancer d iagnosis was independently confirmed, and histology or cytology was av ailable for 94% of the cases. Intervention effects were evaluated by u se of survival analysis and proportional hazards models, All P values were derived from two-sided statistical tests. Results: No overall eff ect was observed for lung cancer from alpha-tocopherol supplementation (relative risk [RR] = 0.99; 95% confidence interval [CI] = 0.87-1.13; P = .86, logrank test). beta-Carotene supplementation was associated with increased lung cancer risk (RR = 1.16; 95% CI = 1.02-1.33; P = .0 2, logrank test). The beta-carotene effect appeared stronger, but not substantially different, in participants who smoked at least 20 cigare ttes daily (RR = 1.25; 95% CI = 1.07-1.46) compared with those who smo ked five to 19 cigarettes daily (RR = 0.97; 95% CI = 0.76-1.23) and in those with a higher alcohol intake (greater than or equal to 11 g of ethanol/day [just under one drink per day]; RR = 1.35; 95% CI = 1.01-1 .81) compared with those with a lower intake (RR = 1.03; 95% CI = 0.85 -1.24). Conclusions: Supplementation with alpha-tocopherol or beta-car otene does not prevent lung cancer ire older men who smoke. beta-carot ene supplementation at pharmacologic levels may modestly increase lung cancer incidence in cigarette smokers, and this effect mag be associa ted with heavier smoking and higher alcohol intake. Implications: Whil e the most direct way to reduce lung cancer risk is not to smoke tobac co, smokers should avoid high-dose beta-carotene supplementation.