Response

Extract

We welcome the suggestions from Drs. Leo and Lieber about potential explanations for the adverse effects of beta-carotene in the ATBC Study and CARET. These studies ( 1 , 2 ) of known dosages in humans found more than one excess lung cancer per 1000 persons exposed per year; in the usual regulatory framework of the U.S. Environmental Protection Agency and the Food and Drug Administration (FDA), this incidence is the equivalent of almost 70 000 per million exposed over a presumed 70-year lifetime. Thus, the FDA, the National Toxicology Program of the National Institute of Environmental Health Sciences—National Institutes of Health, and the International Agency for Research on Cancer of the World Health Organization are all considering the evidence for carcinogenicity of this chemical agent.

The self-reported levels of alcohol intake in Finland and in the United States associated with higher β-carotene- or β-carotene-plus-retinol-related lung cancer risk in the active treatment arms of these trials ( 1 , 2 ) are much lower than the levels studied by Leo and Lieber ( 3 ) and others ( 4 ). None of our trial participants was a known alcoholic; previous liver disease was a criterion for exclusion in CARET. We have reported that there was no detectable increase in serum aspartate aminotransferase (AST) and only a modest, nonprogressive increase in alkaline phosphatase levels in the active arm. We further report now that, in the subset of CARET participants in whom AST and alkaline phosphatase levels were routinely monitored, serum concentrations of AST (but not alkaline phosphatase) in the placebo arm were statistically significantly associated with self-reported alcohol intake, whereas in the active arm there were no statistically significant associations between serum concentrations of these enzymes and alcohol intake ( Table 1 ). The concentrations of these enzymes were low relative to those reflective of liver damage; only about 1% of measured concentrations of AST and alkaline phosphatase exceeded 100 IU/L or 195 IU/L, respectively. Furthermore, while in the placebo arm there was a suggestion of increased serum β-carotene concentration with increased alcohol intake (not statistically significant), in the active arm there was a statistically significant inverse association between alcohol intake and serum β-carotene concentration. We long ago inquired of Hoffmann-La Roche (Nutley, NJ) about additives and impurities in the β-carotene beadlets and commissioned an independent toxicologic review by Dr. David Kalman of the University of Washington Department of Environmental Health. No suspect chemicals were identified.