VITAMINS A, E, SELENIUM AND RISK OF LUNG CANCER (BETA-CAROTENE, RETINOL, ALPHA-TOCOPHEROL, MICRONUTRIENTS

Numerous animal studies have indicated that vitamins A and E, and selenium reduce carcinogenesis. However, evidence supporting a similar protective effect in humans is controversial. To assess whether low levels of serum retinol, (beta)-carotene, vitamin E and/or selenium are associated with the subsequent development of lung cancer, a population-based, individually matched case-control study was undertaken, using serum which had been collected during a large blood collection study in Washington County, MD, in 1974. Serum levels from 99 individuals who were diagnosed with lung cancer during the subsequent nine years (1975-1983) were compared with those of 196 controls, matched on age, sex, month of blood donation, smoking history and population source. No relationship with lung cancer risk was observed for serum retinol or for serum (beta)-carotene when the latter association was examined for all cell types combined. However, a strong inverse association between low serum (beta)-carotene and squamous cell carcinoma was observed when the cell types were examined separately (R.O. = 4.30; 95% C.L. = 1.38, 13.41). Serum vitamin E also appears to be associated with a reduced risk of lung cancer. Mean vitamin E levels were lower among lung cancer cases than controls cases: 10.5 (3.2); controls: 11.9 (4.9) and a significant trend in risk by quintiles of vitamin E was observed (p = 0.04). A relative odds of 1.80 (95% C.L.: 1.13, 2.89) was detected for those whose serum vitamin E levels fell at or below the median compared to those with above median levels. In contrast to previous reports, serum selenium was found to be positively associated with lung cancer risk at higher levels (above the 67th percentile), with an elevated risk of 2.13 (95% C.L. = 1.27, 3.57). The data suggest that there may be interaction between serum selenium and vitamin E which appears to be linear.;In addition to the serum study, a food frequency dietary interview was conducted to determine how well serum vitamin A levels correlate with dietary levels. No important differences in dietary intake of vitamin A between cases and controls were observed, and there was essentially no correlation of serum retinol or (beta)-carotene levels with levels of estimated dietary intake.