![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
2 UMR U557 Inserm/U1125 Inra/EA3200 Cnam/Univ Paris 13, Bobigny, France 93017; 3 Unité de Surveillance et d'Epidémiologie Nutritionnelle; Centre de Recherche en Nutrition Humaine Ile-de-France UFR SMBH Paris 13, Bobigny, France 93017; 4 Department of Dermatology, University Hospital Erasme, Université Libre de Bruxelles, Bruxelles, Belgium 1070; 5 Biometrics and Epidemiology Unit, CE.R.I.E.S., Neuilly sur Seine, France 92521; 6 Computer Science Laboratory, Ecole Polytechnique, Université de Tours, Tours, France 37200; 7 EA 3444, Ecole de Santé Publique, Epidémiologie clinique, Faculté de Médecine, CHU Nancy, France 54035; 8 Laboratoire Lésions des Acides Nucléiques, UMR CNRS-CEA-UJF 5046, Grenoble, France 38000; and 9 EA 3677 and Centre René-Labusquière (Tropical Medicine and International Health Branch), Université Victor Segalen Bordeaux 2 and Department of Internal Medicine and Tropical Diseases, University Hospital Center, Bordeaux, France 33076
* To whom correspondence should be addressed. E-mail: hercberg{at}cnam.fr.
 |
ABSTRACT |
|---|
 TOP ABSTRACT IntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
|
Introduction |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
Numerous studies have demonstrated the role of reactive oxygen species, also called free radicals, in skin carcinogenesis and the potential protective effect of antioxidants (4). Formation of free radicals in the skin can be enhanced by UV radiation. The cutaneous system has a very efficient interlinked antioxidant defense system for counteracting UV-induced oxidative stress. However, excessive exposure to sunlight or other sources of UV light can overwhelm the skin's antioxidant capacity. A potentially interesting strategy for preventing UV exposure damage could be to boost the endogenous antioxidant system by oral intake of antioxidant vitamins and minerals. Although clinical trials have showed contradictory findings (5–7), oral antioxidant pills have been recommended for the prevention of sunburns and for their supposed photoprotective properties.
In particular, it has been suggested that nutrients such as ß-carotene, ascorbic acid, vitamin E, selenium, and zinc may prevent such harmful effects of UV exposure because of their antioxidant ability (8). Clinical trials testing the impact of supplementation with high doses of antioxidants over long periods have, however, failed to reveal beneficial effects on the incidence of SC (9,10). For example, the Nutritional Prevention of Cancer trial, a double-blind, randomized clinical trial, was designed to test whether selenium (200 µg/d) could prevent nonmelanoma SC (NMSC) in 1312 individuals with an individual history of SC recruited in the eastern United States. The initial analysis, which was conducted after 6.4 y of intervention, showed a significant inverse association between supplementation and all types of cancer incidence (10). However, a subsequent analysis found the protective effect of selenium supplementation on all types of cancer incidence to be attenuated (11). Surprisingly, the last analysis conducted after 13 y of supplementation suggested a link between selenium supplementation and an increased risk of NMSC (12).
There is some evidence that a combination of antioxidants may have a more powerful free radical scavenging effect than individual molecules, due to complementarity and synergy in their mechanism of action (13). In addition, metabolic interrelationships also exist between antioxidant nutrients with beneficial mutual protection and regeneration.
The Supplementation in Vitamins and Mineral Antioxidants (SU.VI.MAX) study was a primary-prevention trial designed to assess whether a daily supplementation with antioxidant vitamins and minerals, at nutritional doses, could reduce the incidence of the most prevalent chronic diseases in industrialized countries causing premature death, cancer, and ischemic heart disease in middle-aged men and women. A double-blind, randomized, placebo-controlled design was used to compare medium-term outcome (5–10 y) in subjects receiving antioxidant supplementation or not. The framework of the SU.VI.MAX study allowed data to be collected relating to markers of oxidative stress and thus provided an opportunity to study the relationship between antioxidant effects relating to the intervention and cancer incidence.
Our aim was to test, in the context of the SU.VI.MAX study, the impact of a combination of antioxidant nutrients on the incidence of melanoma and NMSC in a large sample of middle-aged individuals from the French general population.
|
Subjects and Methods |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
-tocopherol), 6 mg ß-carotene, 100 µg selenium (selenium-enriched yeast), and 20 mg zinc (zinc gluconate)] or a matching placebo in a single oral daily capsule. The study cohort was stratified by gender, age group, smoking habits, and residence area. Random treatment allocation was performed by block-sequence generation stratified by gender and age group. Subjects were treated throughout the follow-up period. The SU.VI.MAX study was approved by a medical ethics committee (Comité Consultatif pour la Protectiondes Personnes participant à la Recherche Biomédicale n°706) of Paris-Cochin, and the National Committee for the Protection of Privacy and Civil Liberties Comité National Informatique et Liberté (n°334641), which advocates that all medical information be confidential and anonymous. The participants signed an informed consent form allowing the investigators to perform screening tests to identify cancer or cardiovascular disease and to communicate the results to their physician.
Follow-up of the participants. The participants underwent a yearly medical visit, which consisted in alternate years either of taking a blood sample or of a clinical examination (physical examination, electrocardiogram, blood pressure measurement, visual acuity examination, anthropometrical measurements, fecal occult blood testing for subjects over 45 y, smear test for all women, and a screening mammogram for women over 50 y). Moreover, the participants were also expected to provide monthly information on treatment compliance, dietary intake, and any health event by completing computerized questionnaires. In the absence of monthly contact for >6 mo, or if the participant missed the yearly visit, an investigation was launched to determine the reasons and to resume contact. If necessary, an inquiry was conducted among neighbors and the participant's physician. Causes of death were provided by families, physicians, or hospitals. At the end of the interventional trial (September 1, 2002), vital status and possible causes of death were confirmed through the National Death Registry.
Antioxidant status.
At baseline and every 2 y, 35 mL of venous blood samples after 12-h fasting were taken from each participant. Serum antioxidant concentrations were measured on a randomized sub-sample of 1134 subjects stratified by sex, age, treatment group, and geographic location. All biochemical analyses were performed in the same reference laboratory. Vitamin C status was evaluated by serum ascorbic acid (ascorbate) determination using an automated method based on the principle of continuous flow segmented by air bubbles (16). Serum levels of retinol, ß-carotene, and -tocopherol were measured by HPLC using the Biotek-Kontron HPLC system (17). Serum levels of zinc and selenium were determined using flame atomic absorption spectrometry (Perkin Elmer 3110 for zinc and selenium and Perkin Elmer 4100 ZL for selenium) (18,19).
Endpoints of the SU.VI.MAX trial. Whatever the source of information (see paragraph above on follow-up), once an event was suspected, all relevant records, including results of diagnostic tests and procedures, were collected from physicians, hospitals, or directly from the participants. The primary outcomes of the SU.VI.MAX trial were first fatal and nonfatal major ischemic cardiovascular events and first cancer events during the follow-up. However, BCC and SCC of the skin were not considered in the cancer outcomes as defined in the study protocol. Each event was reviewed by expert committees who were unaware of the treatment assignment. In the case of cancer events, these were ascertained by pathologist reports and reviewed by a committee of oncologists [details in (14)].
Endpoints of the analysis of SC. The outcomes of the present analysis were first event of melanoma at any stage, SCC and BCC of the skin, and other types of SC (International Classification of Diseases, 10th revision, Clinical Modification, codes C43, C44, D03, D04). Moreover, all SC, which were ascertained by pathologist reports, were also reviewed by an expert committee of dermatologists, who were unaware of the treatment assignment.
Assessment of sun exposure. Two questionnaires on sun exposure were administered to participants in 1997 and 2001. The content of the 2 questionnaires was similar, with one part that investigated sun exposure and protection over the past year and another part that assessed lifetime sun exposure and protection (20). Over 64% of the questionnaires of the first survey were returned (4824 women and 3260 men). Following the next survey, we collected 1332 additional questionnaires (800 women and 532 men). The final analysis was conducted on a sub-sample of 9293 volunteers (3751 men and 5542 women) who answered the following question (global self-assessment) "How would you describe the intensity of your skin's exposure to the sun during your lifetime?" with the responses none, mild, moderate, or severe. For the purpose of this analysis, responses were dichotomized as severe vs. moderate, mild, or none. Moreover, the latitude of the place of residence was also taken into account in the analysis.
Statistical methods. All subjects who participated in the SU.VI.MAX study were evaluated in the present analysis. Statistical analyses were performed using SAS software version 8.2 (SAS Institute). Descriptive analysis was performed by gender and treatment group (UNIVARIATE and FREQ procedures). Quantitative variables were expressed as mean ± SD. Baseline characteristic means were compared using ANOVA (GLM procedure). Relationships between antioxidant levels at baseline were assessed using Pearson correlation coefficients (CORR procedure). Percentages of individuals lost to follow-up in each treatment group were compared using the Pearson test (FREQ procedure, option CHISQ) (21).
The duration of follow-up for each participant was defined as the time from randomization until the occurrence of the first event (diagnosis of SC, death, or date of last contact). The analysis was conducted under the intention-to-treat principle.
The proportion of volunteers remaining free of event since randomization was described using Kaplan-Meier survival curves (22). Outcome analysis consisted of frequency comparisons, logrank tests (23), and Cox proportional hazard regression models (24). Cox models allowed a comparison of the effect of the different study variables, namely treatment group, age, smoking status, dwelling latitude, self-assessed lifetime sun exposure, and antioxidant status at baseline. For each outcome, we estimated hazard ratios (HR) with their 95% CI. In the next step, we tested the treatment effect controlling for variables that influenced outcome using a multivariate model to generate adjusted HR. However, as lifetime sun exposure was documented only for a sub-sample of the cohort, this item could not be included in the multivariate analysis. In addition, interaction terms between levels of each antioxidant at baseline and treatment group were also tested. Furthermore, to estimate the effect of antioxidant levels at baseline on each outcome in the antioxidant and the placebo groups, each multivariate Cox model obtained at the last step was rebuilt within each treatment group.
The number of participants was expected to allow detection of a 25% difference in incidence of the outcome (1-tailed = 5%; 1-tailed ß = 90%).
|
Results |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
|
1000 participants. In these subjects, plasma levels of the biochemical markers ß-carotene, vitamin C, and selenium significantly increased in the antioxidant group 2 and 7 y after randomization. Furthermore, the 2 treatment groups differed significantly 2 and 7 y after randomization with respect to all markers except serum zinc [details in (14,15)]. The antioxidant and placebo groups were balanced for most baseline variables, notably smoking habits, alcohol consumption, occupational status, education level, marital status, and BMI (14). The treatment groups were also balanced for self-assessed lifetime sun exposure. In women, the frequency of severe exposure was 11.2% in the placebo group and 11.1% in the antioxidant group, whereas in men, it was 11.8% in the placebo group and 12.2% in the antioxidant group.
Within each gender, the ages of the treatment groups were comparable, with women being 47.1 ± 6.6 y old and men being 51.8 ± 4.7 y old. Serum levels of antioxidants did not differ between the treatment groups at baseline, except for selenium (Table 1). However, serum levels of antioxidants at baseline differed significantly between genders. Serum ß-carotene concentrations were correlated with serum levels of vitamin C (r = 0.26; P < 0.0001) and vitamin E (r = 0.24; P < 0.0001).
|
|
The incidence of melanoma did not differ between the treatment groups in men (6 cases in the placebo group and 3 cases in the antioxidant group; P = 0.51) but was higher in the antioxidant group in women (3 cases in the placebo group and 13 cases in the antioxidant group; P = 0.01).
The incidence of NMSC did not differ between treatment groups in either men (37 cases in the placebo group and 30 cases in the antioxidant group; P = 0.39) or women (27 cases in the placebo group and 38 cases in the antioxidant group; P = 0.15).
Due to the differential effects of antioxidant supplementation according to gender, all subsequent analyses were performed by gender.
Variables associated with the incidence of all SC. According to actuarial survival analysis of the cumulative incidence of SC, the difference between the treatment groups increased regularly over time in women, with a higher incidence in the antioxidant group (logrank test, P = 0.02) (Fig. 2A). Among men, the incidence of SC did not differ between the treatment groups (logrank test, P = 0.22). This differential effect according to gender was confirmed by a Cox proportional hazard regression model, in which there was a significant interaction between gender and treatment group (P = 0.01) in addition to gender and treatment group effects (data not shown). Therefore, subsequent analyses were conducted for each gender separately. Univariate and multivariate Cox proportional hazard regression models were generated for the number of cases of all SC (Table 3). In the univariate model, antioxidant treatment group (HR = 1.70; P = 0.02), older age (HR = 1.06; P = 0.0005), and greater lifetime sun exposure (HR = 2.23; P = 0.006) were associated with a higher probability of developing SC in women. Treatment group (adjusted HR = 1.68; P = 0.02) and age (adjusted HR = 1.06; P = 0.001) were retained as independent variables associated with cancer risk in the multivariate model. In men, age was the only variable associated with increased cancer risk identified in either the univariate or the multivariate analysis (HR = 1.08; P = 0.001 in both cases).
|
|
|
|
|
Discussion |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
Our results show that the effect of antioxidant supplementation on the incidence of SC varies according to gender. The incidence of all types of SC and melanomas was higher in the group of women receiving antioxidant supplementation compared with the placebo group. Such an effect was not observed in men. These results can be compared with the main results of the principal analysis of the SU.VI.MAX trial (14), which showed that antioxidant supplementation was associated with a significant reduction in the overall incidence of all-site cancers in men with no effect in women. It should, however, be noted that, in the present analysis, the incidence of SC was lower in the men receiving antioxidants, although this difference did not reach statistical significance. This may reflect underpowering of the study for the detection of differences in the incidence of relatively rare events such as SC.
Our findings may be attributed in part to gender differences in nutrient metabolism. Although the molecular basis for this metabolic difference is poorly characterized, it may result from hormone-sensitive expression of genes coding for cellular transporters of vitamins and antioxidants involved in the metabolism of carcinogens or in regulation of the cell cycle. For example, it has been hypothesized that the relative importance of low selenium status as a risk factor for cancer might differ between men and women due to sex- or gender-related factors that influence tumor biology (25).
Gender-dependent differences in the handling of antioxidants in the skin may also contribute to the effect of antioxidant supplementation on SC incidence in women. For example, women tend to have higher concentrations of antioxidants in the skin due to a higher intake of dietary vitamin C and ß-carotene and the presence of a larger reservoir of subcutaneous adipose tissue in which to store lipophilic antioxidants such as ß-carotene (26). In our study, men had significantly lower average serum levels of several antioxidants, particularly ß-carotene, than women. This may be related to the capacity of circulating ß-carotene to reflect recent or current carotenoid intake from dietary fruits and vegetables (14,27), which may be lower in men.
The discrepancy between our results and experimental data in animals that has suggested an anticarcinogenetic effect of antioxidants with respect to SC may be explained by differences in the timing of the intervention. In animal models of cutaneous carcinogenesis, the diet is enriched in antioxidants prior to irradiation with UV-A or UV-B, leading to a predominantly beneficial effect of treatment on DNA protection. In contrast, in the SU.VI.MAX trial and related primary prevention studies, antioxidants are given only after many years of exposure to sunlight or other risk factors. At this stage, it may be too late for antioxidants to prevent DNA damage, whereas increased antioxidant exposure might exert a negative influence on antitumoral immunity, angiogenesis, or apoptosis. For example, antioxidants have been reported to interfere with the ability of natural killer lymphocytes to destroy tumor cells in animal models (28). This detrimental effect late in the disease process might explain why we observed a more pronounced effect upon melanoma, which requires a longer time frame over which to develop compared with other SC types and is presumed to be mainly induced by exposure to solar radiation during infancy. For other SC types, a protective effect of antioxidants with respect to sun exposure during the time of the study could counterbalance a detrimental effect on the development of preexisting precancerous lesions induced before the trial began, which could explain the absence of effect of antioxidant supplementation observed in our study.
Previous observational surveys, which have evaluated the relationship between antioxidant status, estimated by serum biomarkers, and the risk of SC have yielded contradictory results. Some studies reported a negative association between serum levels of -tocopherol, carotenoids, or selenium and the risk of melanoma or NMSC (29–31), whereas others did not find any such protective effect (32–35). In contrast, a recent Italian study of a community-based cohort reported that the incidence rate of melanoma was nearly 4 times higher in individuals exposed to a high selenium intake (provided by tap water) than in unexposed individuals (36). Several large randomized trials evaluating the impact of ß-carotene supplementation did not reveal any beneficial effect on the development of SC in community-based populations (9,32–34,37,38) or in patients with antecedents of NMSC (39).
In addition, some trials have evaluated gender effects with respect to the potential benefits of dietary supplementation with individual antioxidants. For example, in the Nutritional Prevention of Cancer study, the initial data analysis showed that the beneficial effect of supplementation with selenium on the incidence of all types of cancer was restricted to men, specifically to those with the lowest baseline serum selenium concentrations (12). This differential effect of selenium supplementation according to gender is consistent with the results of previous case-control studies (40) and prospective studies conducted on all types of cancers (41).
There are several methodological limitations to the present study. For example, BCC occurrence was not considered as a primary endpoint within the framework of the SU.VI.MAX core trial and our analysis thus corresponds to a post hoc analysis. In addition, the pertinence of the analysis of the incidence of melanoma is limited by the small number of cases.
In conclusion, our findings suggest that dietary supplementation with vitamins and trace element antioxidants may not always provide beneficial effects. This issue is important, given the vast quantities of antioxidant pills that are sold in certain countries. This is particularly true among sunseekers and women in northern countries, where the use of such pills is reputed to prevent solar damage to the skin. In this respect, our study indicates that regular intake of such nutrients, especially at doses taken by consumers of supplements in northern countries, may be associated with harmful effects. Clearly, high doses of specific nutrients given to middle-aged adults over a period of several years do not have an equivalent biological impact to the lifelong intake of a broad palette of nutrients from a well-balanced diet. It therefore appears crucial to better define the safety profile of such nutrients to be able to assess accurately the risk-benefit relationship associated with their use as dietary supplements.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
10 Abbreviations used: BCC, basal cell carcinoma; HR, hazard ratio; MSC, melanoma skin cancer; NMSC, nonmelanoma skin cancer; SC, skin cancer; SCC, squamous cell carcinoma; SU.VI.MAX, Supplémentation en Vitamines et Minéraux Antioxydants study.
Manuscript received 28 February 2007. Initial review completed 21 March 2007. Revision accepted 22 June 2007.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
1. Urbach F. Incidence of nonmelanoma skin cancer. Dermatol Clin. 1991;9:751–5.[Medline]
2. English DR, Armstrong BK, Kricker A, Fleming C. Sunlight and cancer. Cancer Causes Control. 1997;8:271–83.[Medline]
3. Preston DS, Stern RS. Non melanoma cancers of the skin. N Engl J Med. 1992;327:1649–62.[Medline]
4. Sander CS, Chang H, Hamm F, Elsner P, Thiele JJ. Role of oxidative stress and the antioxidant network in cutaneous carcinogenesis. Int J Dermatol. 2004;43:326–35.[Medline]
5. Steenvoorden DP, van Henegouwen GM. The use of endogenous antioxidants to improve photoprotection. J Photochem Photobiol B. 1997;41:1–10.[Medline]
6. Fuchs J, Kern H. Modulation of UV-light-induced skin inflammation by D-alpha-tocopherol and L-ascorbic acid: a clinical study using solar simulated radiation. Free Radic Biol Med. 1998;25:1006–12.[Medline]
7. Werninghaus K, Meydanu M, Bhawan J, Margolis R, Blumberg JB, Gilchrest BA. Evaluation of the photoprotective effect of oral vitamin E supplementation. Arch Dermatol. 1994;130:1257–61.
8. Byers T, Perry G. Dietary carotenes, vitamin C, and vitamin E as protective antioxidants in human cancers. Annu Rev Nutr. 1992;12:139–59.[Medline]
9. Schaumberg DA, Freiling UM, Rifai N, Cook N. No effect of beta-carotene supplementation on risk on nonmelanoma skin cancer among men with low baseline plasma beta-carotene. Cancer Epidemiol Biomarkers Prev. 2004;13:1079–80.
10. Clark LC, Combs GF Jr, Turnbull BW, Slate EH, Chalker DK, Chow J, Davis LS, Glover RA, Graham GF, et al. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. A randomized controlled trial. Nutritional Prevention of Cancer Study Group. JAMA. 1996;276:1957–63.
11. Duffield-Lillico AJ, Reid ME, Turnbull BW, Combs GF Jr, Slate EH, Fischbach LA, Marshall JR, Clark LC. Baseline characteristics and the effect of selenium supplementation on cancer incidence in a randomized clinical trial: a summary report of the Nutritional Prevention of Cancer Trial. Cancer Epidemiol Biomarkers Prev. 2002;11:630–9.
12. Duffield-Lillico AJ, Slate EH, Reid ME, Wilkins PA, Combs GF Jr, Park HK, Gross EG, Graham GF, Stratton MS, et al. Nutritional Prevention of Cancer Study Group. Selenium supplementation and secondary prevention of nonmelanoma skin cancer in a randomized trial. J Natl Cancer Inst. 2003;95:1477–81.
13. Hercberg S, Galan P, Preziosi P, Alfarez MJ, Vasquez C. The potential role of antioxidant vitamins in preventing cardiovascular diseases and cancers. Nutrition. 1998;14:513–20.[Medline]
14. Hercberg S, Galan P, Preziosi P, Bertrais S, Mennen L, Malvy D, Roussel AM, Favier A, Briancon S. The SU.VI.MAX study: a randomised, placebo-controlled trial of the health effects of antioxidant vitamins and minerals. Arch Intern Med. 2004;164:2335–42.
15. Hercberg S, Preziosi P, Briancon S, Galan P, Triol I, Malvy D, Roussel AM, Favier A. A primary prevention trial using nutritional doses of antioxidant vitamins and minerals in cardiovascular diseases and cancer in a general population: the SU.VI.MAX study: design, methods and participant characteristics. Control Clin Trials. 1998;19:336–51.[Medline]
16. Bourgeois CF, Chartois RR, Coustans MF, George PR. Automated determination of vitamin C in foodstuffs and biological tissues. Analusis. 1989;9:519–25.
17. Vuilleumier JP, Keller HE, Gysel D, Hunziker F. Clinical chemical methods for the routine assessment of the vitamin status in human populations. Part I: the fat-soluble vitamins A and E, and beta-carotene. Int J Vitam Nutr Res. 1983;53:265–72.[Medline]
18. Arnaud J, Bellanger J, Bienvenu F, Chappuis P, Favier A. Recommended method for assaying serum zinc with flamme atomic absorption. Ann Biol Clin (Paris). 1986;44:77–87.[Medline]
19. Nève J, Chamart S, Molle L. Optimisation of a direct procedure of selenium in plasma and erythrocytes using Zeeman effect atomic absorption spectrometry. In: Bratter P, Schramel P, editors. Trace element analytical chemistry in medicine and biology. 4th vol. Berlin: Walter de Gruyter; 1987. p. 349–58.
20. Guinot C, Malvy D, Latreille J, Preziosi P, Galan P, Vaillant L, Tenenhaus M, Hercberg S, Tschachler E. Sun exposure behaviour of a general adult population in France. In: Monduzzi, editor. Skin and environment: perception and protection. 10th EADV Congress. Munich 2001, Bologna: p. 1099–106.
21. Armitage P. Tests for linear trends in proportion and frequencies. Biometrics. 1955;11:375–86.[Medline]
22. Borgan O. Kaplan-Meier estimator. In: P. Armitage and T. Colton, editors. Encyclopedia of biostatistics. 3rd vol. Chichester (UK): John Wiley & Sons; 1998. p. 2154–60.
23. Keiding N. Logrank test. In: P. Armitage and T. Colton, editors. Encyclopedia of biostatistics. 3rd vol. Chichester (UK): John Wiley & Sons; 1998. p. 2350–1.
24. Sasieni P. Cox regression model. In: P. Armitage and T. Colton, editors. Encyclopedia of biostatistics. 1st vol. Chichester (UK): John Wiley & Sons; 1998. p. 1006–20.
25. Waters DJ, Chiang EC, Cooley DM, Morris JS. Making sense of sex and supplements: differences in the anticarcinogenic effects of selenium in men and women. Mutat Res. 2004;551:91–107.[Medline]
26. Kaplan LA, Lau JM, Stein EA. Carotenoid composition, concentrations, and relationships in various human organs. Clin Physiol Biochem. 1990;8:1–10.[Medline]
27. McShane LM, Clark LC, Combs GF Jr, Turnbull BW. Reporting the accuracy of biochemical measurements for epidemiologic and nutrition studies. Am J Clin Nutr. 1991;53:1354–60.
28. Riondel J, Glise D, Fernandez-Carlos T, Favier A. In vitro comparative study of cytolysis mediated by natural killer cells towards malignant cells preincubated with antioxidants. Anticancer Res. 1998;18:1757–63.[Medline]
29. Clark LC, Graham GF, Crounse RG, Grimson R, Hulka B, Shy CM. Plasma selenium and skin neoplasms: a case-control study. Nutr Cancer. 1984;6:13–21.[Medline]
30. Karagas MR, Greenberg ER, Nierenberg D, Stukel TA, Morris JS, Stevens MM, Baron JA. Risk of squamous cell carcinoma of the skin in relation to plasma selenium, alpha-tocopherol, beta-carotene, and retinol: a nested case-control study. Cancer Epidemiol Biomarkers Prev. 1997;6:25–9.
31. Reinhold U, Biltz H, Bayer W, Schmidt KH. Serum selenium levels in patients with malignant melanoma. Acta Derm Venereol. 1989;69:132–6.[Medline]
32. Coates RJ, Weiss NS, Daling JR, Morris JS, Labbe RF. Serum levels of selenium and retinol and the subsequent risk of cancer. Am J Epidemiol. 1988;128:515–23.
33. Breslow RA, Alberg AJ, Helzlsouer KJ, Bush TL, Norkus EP, Morris JS, Spate VE, Comstock GW. Serological precursors of cancer: malignant melanoma, basal and squamous cell skin cancer, and prediagnostic levels of retinol, beta-carotene, lycopene, alpha-tocopherol, and selenium. Cancer Epidemiol Biomarkers Prev. 1995;4:837–42.[Abstract]
34. Fung TT, Hunter DJ, Spiegelman D, Colditz GA, Speizer FE, Willett WC. Vitamins and carotenoids intake and the risk of basal cell carcinoma of the skin in women (United States). Cancer Causes Control. 2002;13:221–30.[Medline]
35. Dorgan JF, Boakye NA, Fears TR, Schleicher RL, Helsel W, Anderson C, Robinson J, Guin JD, Lessin S, et al. Serum carotenoids and alpha-tocopherol and risk of nonmelanoma skin cancer. Cancer Epidemiol Biomarkers Prev. 2004;13:1276–82.
36. Vinceti M, Rothman KJ, Bergomi M, Borciani N, Serra L, Vivoli G. Excess melanoma incidence in a cohort exposed to high levels of environmental selenium. Cancer Epidemiol Biomarkers Prev. 1998;7:853–6.[Abstract]
37. Green A, Williams G, Neale R, Hart V, Leslie D, Parsons P, Marks GC, Gaffney P, Battistutta D, et al. Daily sunscreen application and beta carotene supplementation in prevention of basal-cell and squamous-cell carcinomas of the skin: a randomized controlled trial. Lancet. 1999;354:723–9.[Medline]
38. Frieling UM, Schaumberg DA, Kupper TS, Muntwyler J, Hennekens CH. A randomized, 12-year primary-prevention trial of beta carotene supplementation for nonmelanoma skin cancer in the Physician's Health Study. Arch Dermatol. 2000;136:179–84.
39. Greenberg ER, Baron JA, Stukel TA, Stevens MM, Mandel JS, Spencer SK, Elias PM, Lowe N, Nierenberg DW, et al. A clinical trial of beta carotene to prevent basal-cell and squamous-cell cancers of the skin. The Skin Cancer Prevention Study Group. N Engl J Med. 1990;323:789–95.[Abstract]
40. Kok FJ, de Bruijn AM, Hofman A, Vermeeren R, Valkenburg HA. Is serum selenium a risk factor for cancer in men only? Am J Epidemiol. 1987;125:12–6.
41. Knekt P, Aromaa A, Maatela J, Alfthan G, Aaran RK, Hakama M, Hakulinen T, Peto R, Teppo L. Serum selenium and subsequent risk of cancer among Finnish men and women. J Natl Cancer Inst. 1990;82:864–8.
This article has been cited by other articles:
|
|
 |
|
  K. Park, L. Harnack, and D. R. Jacobs Jr. Trends in Dietary Supplement Use in a Cohort of Postmenopausal Women From Iowa Am. J. Epidemiol., April 1, 2009; 169(7): 887 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Green, M. C. Hughes, T. I. Ibiebele, G. M. Williams, J. C. van der Pols, and J.-P. Ortonne Antioxidant Supplementation and Risk of Skin Cancers J. Nutr., May 1, 2008; 138(5): 978 - 978. [Full Text] [PDF]
|
|
|
|
|
|
K. Ezzedine, D. Malvy, S. Briancon, and S. Hercberg Reply to Dr. Green et al. J. Nutr., May 1, 2008; 138(5): 979 - 979. [Full Text] [PDF]
|
|
|
|
|
|
G. N. Schrauzer Antioxidant Supplementation Increases Skin Cancer Risk, or, Why Zinc Should Not Be Considered an Antioxidant J. Nutr., April 1, 2008; 138(4): 820 - 820. [Full Text] [PDF]
|
|
|
|
|
|
K. Ezzedine, S. Hercberg, A. Favier, and D. Malvy Reply to Dr. Schrauzer J. Nutr., April 1, 2008; 138(4): 821 - 822. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
