Cancer chemoprevention by supplemental carotenoids in animals and humans.

Experiments were carried out in mice demonstrating that dietary carotenoids (beta-carotene or canthaxanthin), starting before cancer initiation and continuing throughout the experiment, have a protective effect against indirect skin carcinogenesis induced by benzo[a]pyrene +/- UVA and breast cancer induced by 8-methoxypsoralen + UVA. Experiments in rats demonstrated that carotenoids also prevent the direct gastric carcinogenesis induced by N-methyl-N'-nitro-nitroso-guanidine. Recently, prevention by beta-carotene against colon cancer induced in mice by dimethylhydrazine, another indirect carcinogen, was confirmed by others. The prospects for carotenoid intervention with humans were based on their antitumorigenic effect, which is quite independent of pro-vitamin A activity, their lack of toxicity even after prolonged administration, and their immunostimulating activity. These facts helped to build up a rationale predicting that any epithelial cancer, after radical surgery, can be chemoprevented with supplemental carotenoids. Thus, it is expected that the remaining initiated epithelial tissue will be protected by quenching oxygen radical formation, against the onset of a second primary malignancy. This type of prevention can be envisaged in organs like the lung, urinary bladder, breast, stomach, and colon-rectum. At present, human intervention protocols with a randomized drug/placebo method are underway under the supervision of the Centro Tumori of Pavia to chemoprevent with beta-carotene second primary lung or bladder cancer after radical surgery. Preliminary observations regarding findings in humans without randomization (1980-1988) in Pavia are also reported here. This consisted of chemoprevention with beta-carotene plus canthaxanthin against recurrence of different epithelial malignancies after radical treatment (surgery +/- chemoradiotherapy). None of the 11 cases recruited, on the basis of radical nature of treatment and patient adherence, have shown any recurrence beyond their expected disease-free intervals.