Paul Knekt, Ritva Järvinen, Lyly Teppo, Arpo Aromaa, Ritva Seppänen
Affiliations of authors: P. Knekt, A. Aromaa, National Public Health Institute, Helsinki, Finland; R. Järvinen, University of Kuopio, Finland; L. Teppo, Finnish Cancer Registry, Helsinki; R. Seppänen, Social Insurance Institution, Turku, Finland.
Consistent evidence suggests that high intake of fruits and vegetables is associated with a reduced risk of lung cancer. In accordance with the hypothesis that this association is due to ß-carotene, a considerable part of the observational studies on the intake or serum concentration of ß-carotene has reported an inverse association with lung cancer risk (1). Large intervention trials testing the effect of ß-carotene supplementation that leads to high plasma ß-carotene levels have, however, failed to confirm this hypothesis (2,3). In contrast, these studies suggested an increased risk for lung cancer among the individuals receiving ß-carotene supplements. It is interesting that these intervention trials revealed an inverse association between baseline dietary intake and serum levels of ß-carotene and subsequent occurrence of lung cancer. Thus, apparently some compounds, either ß-carotene or other components highly associated with it, received at the dietary level may provide protection against lung cancer. The impact of carotenoids other than ß-carotene on lung cancer risk has seldom been studied (4,5).
The aim of this study is to reanalyze the data from the Finnish Mobile Clinic Health Examination Survey (6) to evaluate whether the inverse association observed between the intake of fruits and vegetables and lung cancer incidence could be due to the intake of some carotenoids other than ß-carotene.
During the period 1967-1972, the Finnish Mobile Clinic undertook health examinations in various parts of Finland. A total of 4545 men, 20-69 years of age and free of cancer at baseline, participated in a dietary history interview (6). The interview covered the total habitual diet of the individuals during the previous year. Inquiries on the consumption of different foods were made according to a questionnaire on more than a hundred food items and mixed food dishes commonly used at the time of the baseline study (7). Intakes of seven carotenoids from all food items were calculated with the use of a database compiled on the analyzed values of individual carotenoids on Finnish foods (8). Information on smoking habits was obtained from a questionnaire. Subjects were classified according to smoking status as never smokers, ex-smokers, smokers of cigars or pipe only, smokers of fewer than 15 cigarettes a day, and smokers of 15 or more cigarettes a day. Those who had never smoked and ex-smokers were also combined as a class of nonsmokers. During a 25-year follow-up period (1967-1991), 138 case subjects with lung cancer were diagnosed according to the nationwide Finnish Cancer Registry (9). The relative risks of lung cancer between tertiles of various dietary carotenoids were estimated by Cox's model. Adjustments were made for confounding factors, e.g., age, smoking, and other dietary factors, by including the respective factors in the model. All statistical tests were two-sided.
The intake of -carotene was inversely associated with lung cancer incidence (Table 1). The relative risk of lung cancer between the highest and lowest tertiles of -carotene intake after adjustment for age and smoking was 0.61 (95% confidence interval [CI] = 0.39-0.95). The results agree with findings from previous studies (4,5,10,11). The results are also plausible, since -carotene, because of its antioxidant or other properties (12), may play a role in cancer etiology. In contrast, dietary ß-carotene was not statistically significantly associated with lung cancer occurrence. The relative risk was 0.79 (95% CI = 0.50-1.24). The intakes of -carotene and ß-carotene were strongly associated; a total of 76% of the individuals were in the same tertile for -carotene and ß-carotene. When the two carotenoids were coadjusted, the relative risk was 0.58 (95% CI = 0.31-1.07) for -carotene and 0.97 (95% CI = 0.52-1.78) for ß-carotene, suggesting that the association with lung cancer was more closely related to the presence of -carotene but neither was statistically significant. Consideration of the interaction between the two carotenoids showed the generally reduced risk among individuals with simultaneously low (below median value of and high (above median value of 26 µg/day) -carotene intake. The relative risk for these individuals in comparison with those having a low intake of both carotenoids was 0.34 (95% CI = 0.13-0.94). The corresponding relative risk between those with simultaneously low -carotene and high ß-carotene in comparison with those having a low intake of both -carotene and ß-carotene was 0.99 (95% CI = 0.54-1.82).
The other carotenoids considered (-carotene, ß-cryptoxanthin, lycopene, or lutein and zeaxanthin) showed associations that were not statistically significant with lung cancer incidence (Table 1). Accordingly, a suggestive inverse association between the intake of -carotene and lung cancer incidence persisted after adjustment for these additional carotenoids; the relative risk was 0.58 (95% CI = 0.31-1.08). The association was likewise not due to the presence of other potentially protective nutrients, such as vitamin E, vitamin C, folate, fiber, and flavonoids, since the association also persisted after adjustment was made for them; the relative risk for -carotene was 0.60 (95% CI = 0.36-0.99).
Intake of fruits and root vegetables was inversely associated with lung cancer incidence (Table 1). The relative risks were 0.58 (95% CI = 0.37-0.93) and 0.56 (95% CI = 0.36-0.88), respectively, after adjustment was made for age and smoking status. Adjustment for intake of -carotene, ß-carotene, the other carotenoids, vitamin C, vitamin E, flavonoids, fiber, and folate did not notably influence this association with root vegetables: The relative risks for intake of root vegetables and fruits were 0.53 (95% CI = 0.30-0.94) and 0.66 (95% CI = 0.36-1.23), respectively. About 90% of the dietary -carotene and 50% of the dietary ß-carotene were derived from carrots, whereas the other root vegetables (beetroot, rutabaga, turnip, celeriac, and radish) were poor sources of carotenoids. The relative risk of lung cancer after adjustment for age and smoking status for intake of carrots was 0.60 (95% CI = 0.39-0.94); after further adjustment for -carotene and ß-carotene, it was 0.61 (95% CI = 0.33-1.12). For the other root vegetables, the corresponding values were 0.73 (95% CI = 0.49-1.10) and 0.88 (95% CI = 0.54-1.45), respectively. Although adjustment for vitamin C, vitamin E, and other potentially protective substances in fruits and vegetables did not notably alter the association between -carotene and lung cancer, the fact that the inverse association between carrot intake and lung cancer incidence persisted after adjustment for intake of -carotene and ß-carotene supports the hypothesis that the association is probably due to some other unrecognized substance. The conclusion for intake of fruits and vegetables is similar; the relative risk of lung cancer between the highest and lowest tertiles of intake of these foods after adjustment for age and smoking status was 0.60 (95% CI = 0.38-0.96) and, after further adjustment for all single carotenoids considered, it was 0.59 (95% CI = 0.31-1.10).
The intake of various carotenoids was more strongly associated with lung cancer incidence in nonsmokers than in smokers (Table 2). The relative risks varied between 0.33 and 0.46 among nonsmokers and, with the exception of -carotene, between 0.82 and 1.29 among smokers. It was also apparent that nonsmokers consumed more carotenoids than smokers. Nonsmokers had statistically significantly higher intakes of -carotene, ß-carotene, ß-cryptoxanthin, and lutein and zeaxanthin. The possibility thus cannot be ruled out that the association observed between -carotene and lung cancer incidence is due to residual confounding by smoking status. Since the reliability of smoking status in the present population was 0.70 (13), the possibility exists that the association observed is due to lower -carotene intake among unreported smokers in the nonsmoking group.
In summary, this study suggests that -carotene rather than ß-carotene may be a substance associated with a reduced risk of lung cancer. The possibility, however, remains that this association reflects the effect of some other unidentified substance in plant foods or that it is due to incomplete control for smoking habits or to some other factors associated with intake of fruits and vegetables. Further observational studies analyzing the effects of -carotene and other carotenoids on lung cancer incidence are therefore warranted.
Supported by a grant from the Finnish Cancer Foundation.
1 World Health Organization. Carotenoids. IARC handbooks on cancer prevention. Vol 2. Lyon (France): International Agency for Research on Cancer; 1998.
2 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.
3 Omenn GS, Goodman GE, Thornquist MD, Balmes J, Cullen MR, Glass A, et al. Risk factors for lung cancer and for intervention effects in CARET, the Beta-Carotene and Retinol Efficacy Trial. J Natl Cancer Inst 1996;88:1550-9.
4 Ziegler RG, Colavito EA, Hartge P, McAdams MJ, Schoenberg JB, Manson TJ, et al. Importance of alpha-carotene, beta-carotene, and other phytochemicals in the etiology of lung cancer. J Natl Cancer Inst 1996;88:612-5.
5 Comstock GW, Alberg AJ, Huang HY, Wu K, Burke AE, Hoffman SC, et al. The risk of developing lung cancer associated with antioxidants in the blood: ascorbic acid, carotenoids, alpha-tocopherol, selenium, and total peroxyl radical absorbing capacity. Cancer Epidemiol Biomarkers Prev 1997;6:907-16.
6 Knekt P, Jarvinen R, Seppanen R, Rissanen A, Aromaa A, Heinonen OP, et al. Dietary antioxidants and the risk of lung cancer. Am J Epidemiol 1991;134:471-9.
7 Jarvinen R, Seppanen R, Knekt P. Short-term and long-term reproducibility of dietary history interview data. Int J Epidemiol 1993;22:520-7.
8 Heinonen M. Carotenoids and retinoids in Finnish foods and the average diet [Ph.D. dissertation]. EKT series 811. Helsinki (Finland): Univ. of Helsinki, Department of Food Chemistry and Technology; 1990.
9 Teppo L, Pukkala E, Hakama M, Hakulinen T, Herva A, Saxen E. Way of life and cancer incidence in Finland. A municipality-based ecological analysis.Scand J Soc Med Suppl 1980;19:1-84.
10 Candelora EC, Stockwell HG, Armstrong AW, Pinkham PA. Dietary intake and risk of lung cancer in women who never smoked. Nutr Cancer 1992;17:263-70.
11 Le Marchand L, Hankin JH, Kolonel LN, Beecher GR, Wilkens LR, Zhao LP. Intake of specific carotenoids and lung cancer risk [published erratum appears in Cancer Epidemiol Biomarkers Prev 1994;3:523]. Cancer Epidemiol Biomarkers Prev 1993;2:183-7.
12 Krinsky NI. Actions of carotenoids in biological systems. Annu Rev Nutr 1993;13:561-87.
13 Heliovaara M, Aho K, Aromaa A, Knekt P, Reunanen A. Smoking and risk of rheumatoid arthritis. J Rheumatol 1993;20:1830-5.
Source: Journal of the National Cancer Institute, Vol. 91, No. 2, 182-184, January 20, 1999