PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of carcinLink to Publisher's site
 
Carcinogenesis. 2008 May; 29(5): 949–956.
Published online 2008 April 30. doi:  10.1093/carcin/bgn043
PMCID: PMC2902380

Dietary magnesium and DNA repair capacity as risk factors for lung cancer

Abstract

Magnesium (Mg) is required for maintenance of genomic stability; however, data on the relationship between dietary Mg intake and lung cancer are lacking. In an ongoing lung cancer case–control study, we identified 1139 cases and 1210 matched healthy controls with data on both diet and DNA repair capacity (DRC). Dietary intake was assessed using a modified Block-NCI food frequency questionnaire and DRC was measured using the host-cell reactivation assay to assess repair in lymphocyte cultures. After adjustment for potential confounding factors including DRC, the odds ratios (ORs) and 95% confidence intervals (CIs) for lung cancer with increasing quartiles of dietary Mg intake were 1.0, 0.83 (0.66–1.05), 0.64 (0.50–0.83) and 0.47 (0.36–0.61), respectively, for all subjects (P-trend < 0.0001). Similar results were observed by histology and clinical stage of lung cancer. Low dietary Mg intake was associated with poorer DRC and increased risk of lung cancer. In joint effects analyses, compared with those with high dietary Mg intake and proficient DRC, the OR (95% CI) for lung cancer in the presence of both low dietary Mg and suboptimal DRC was 2.36 (1.83–3.04). Similar results were observed for men and women. The effects were more pronounced among older subjects (>60 years), current or heavier smokers, drinkers, those with a family history of cancer in first-degree relatives, small cell lung cancer and late-stage disease. These intriguing results need to be confirmed in prospective studies.

Introduction

Lung cancer is the most common cause of cancer mortality in USA (1). Although smoking is the predominant exposure associated with this disease, dietary factors also modulate risk, but little attention has focused on the role of dietary minerals in lung cancer risk. To date, there are no published reports on dietary magnesium (Mg) intake and lung cancer risk. Two small case–control studies examined the roles of serum or tissue levels of Mg in lung cancer with mixed results (2,3).

Food sources rich in Mg are green vegetables, legumes (beans and peas), unrefined whole grains, nuts and seeds (4). Meat, fruit and dairy products have moderate Mg content (4), whereas refined foods are poor sources of Mg. Bread made from whole-grain wheat flour is a better source of Mg than bread made from refined white flour because the Mg-rich germ and bran are removed during the refining process (4).

Mg is important for maintaining the integrity of DNA. Mg cations bind to DNA and reduce the negative charge density, thereby stabilizing the structure of DNA (5). Mg is also an essential cofactor for several enzyme systems involving DNA repair such as nucleotide excision repair, base excision repair and mismatch repair (6). Several studies have reported that Mg plays a role in protection against oxidative stress (710). Further, growing evidence from studies conducted in Western societies link low intake of Mg to systemic inflammation (11,12). There is also substantive evidence that high levels of systemic inflammation (13) and proinflammatory conditions (1417) are associated with increased lung cancer risk. Therefore, we investigated the individual and joint associations between dietary Mg intake and DNA repair capacity (DRC) with risk for lung cancer in an ongoing case–control study.

Materials and methods

Study population

Our study population included 1139 newly diagnosed lung cancer patients (cases) and 1210 healthy controls with DRC and dietary data. The participants were a subset of a larger ongoing and previously described lung cancer case–control study (18), since the assay was conducted in batches on archival samples. All patients were enrolled into the study at the time they registered at the Thoracic Center at The University of Texas MD Cancer Center. The cases were all newly diagnosed patients presenting with histologically confirmed lung cancer, enrolled prior to initiation of chemotherapy or radiation therapy. There were no age, ethnic or stage restrictions. Healthy controls without a previous diagnosis of cancer were recruited from the Kelsey-Seybold clinics, the largest private multispecialty physician group of 23 clinics in the Houston area. All participants were USA residents. Controls were frequency matched to the cases by age (±5 years), sex, ethnicity and smoking status (current, former and never) (19). The overall response rate among both patients and the controls was ~75%. This research was approved by MD Anderson Cancer Center and Kelsey-Seybold Institutional Review Boards.

Epidemiolgic and diet data

Participants were interviewed in person to obtain demographic information and smoking history. Smokers of at least 100 cigarettes in their lifetimes were classified as ever smokers; among whom, former smokers had quit smoking at least 1 year before diagnosis (cases) or before interview (controls). Race/ethnicity information was self-reported. Body mass index (BMI) was estimated from self-reported weight and height calculated as weight (kg) divided by height (m2).

Dietary data were collected from a modified version of the 135-item National Cancer Institute's Health Habits and History Questionnaire. The Health Habits and History Questionnaire includes a semiquantitative food frequency list, an open-ended food section and other dietary behavior questions. The questionnaire has been shown to be valid and reliable across various populations (20,21). Study participants were asked about their diet during the year prior to diagnosis (cases) and the year prior to study enrollment (controls). Subjects were given a one-page serving size guide, with photographs of 1/4 cup, 1/2 cup, 1 cup and 2 cups on plates and 1/2, 1 and 2 cups in bowls (Block Dietary Data Systems, Berkeley, CA). Each line item of our food frequency questionnaire (FFQ) also included four quantitative serving size choices for that line item [e.g. for corn we ask, ‘How many cups?’ and give four possible answers (1/4, 1/2, 1 and 2)]. Nutrient intake was calculated using the DIETSYS + Plus version 5.9 dietary analysis program (Block Dietary Data Systems). The DIETSYS + Plus database for the present study was expanded to include values for dietary Mg from the SR16-1, a database maintained by the US Department of Agriculture. For multiingredient food items, nutrient values were derived from recipes collected during the Continuing Survey of Food Intakes by Individuals, 1994–1996 and 1998.

Cellular DRC

The host-cell reactivation assay was used to measure nucleotide excision repair capacity in peripheral blood lymphocytes of the subjects, as described previously (22). Isolated lymphocytes were transfected with non-replicating plasmids harboring a chloramphenicol acetyltransferase gene, a bacterial drug-resistant gene. Plasmids were treated with benzo[a]pyrene diol epoxide (BPDE), an ultimate tobacco carcinogen, prior to transfection to introduce BPDE–DNA adducts as the substrate of the repair enzymes in cells. Only stimulated lymphocytes will uptake the plasmids and exhibit active nucleotide excision repair activity to repair BPDE-induced DNA adducts.

After harvesting, the cells were divided into four aliquots, each having ~1 × 106 cells, for duplicate transfections with untreated plasmids (as the baseline for comparison) and duplicate transfections with BPDE-treated plasmids. The transfections were performed with the DEAE-dextran (Pharmacia Biotech, Picataway, NJ) method (23) with ~0.25 μg of either untreated plasmids or plasmids treated with 60 μM BPDE. The chloramphenicol acetyltransferase expression was measured at the 40th h of continued culture in the same medium after transfection. The DRC value (%) is the chloramphenicol acetyltransferase expression of cells transfected with plasmids divided by cells transfected with untreated plasmids × 100%.

Statistical analysis

Pearson’s χ2 test was calculated to test differences between cases and controls by gender, ethnicity, smoking status, education, income, family history of cancer in first-degree relatives, dietary supplement use and alcohol use. Student’s t-test was calculated to test differences in mean age, years of smoking, cigarettes smoked per day, BMI [weight (kg)/height (m2)], energy-adjusted Mg intake and intake of total energy between cases and controls.

Quartiles of dietary Mg were created based on the distribution of intake in control subjects. Energy-adjusted Mg quartiles were calculated by regressing dietary Mg intake on total calories and obtaining the residuals (24). The residual value for each observation was then added to the mean dietary Mg value for our population. Multiple logistic regression analysis was performed to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for associations between dietary Mg and lung cancer, adjusting for age, gender, race, smoking status, pack-years smoked, family history of cancer, BMI, education, income, total calories and DRC. Total calories were included because food sources of Mg such as nuts are energy rich; the advantage of this model is that the full effects of calories can be observed (24). The first quartile (lowest intake) was the reference category. We tested for trends using the Wald test based on the ordinal dietary variable.

In joint effects analyses for dietary Mg intake and DRC with lung cancer risk, the reference group was those who reported high dietary Mg (more than the median split for energy-adjusted Mg intake in the controls) and proficient DRC (more than the median split for DRC in the controls), because this group would be expected to be at the lowest risk. The top 10 food sources for Mg were calculated by the DIETSYS + Plus dietary analysis program. Potential interactions between dietary Mg and other risk factors for lung cancer were tested on the multiplicative scale by entering the cross product terms in the main effects multivariate models. We also conducted subgroup analyses defined by age, BMI (kg/m2), smoking status (current, former and never smokers), alcohol (non-drinkers and drinkers), years of smoking, number of cigarettes smoked per day, vitamin/mineral supplement use (yes and no), history of cancer in first-degree relatives, histology and lung cancer stage. BMI was stratified as ≤25 or >25 since a number of studies have reported that BMI >25 is associated with reduced risk of lung cancer (2527). Years of smoking were dichotomized at the median split (≤31 or >31) in the controls. Cigarettes smoked per day were dichotomized at ≤1 pack of cigarettes per day (≤20 cigarettes) or >1 pack of cigarettes per day (>20 cigarettes). Lung cancer histology was categorized as adenocarcinoma, non-small cell cancer, squamous cell carcinoma and small cell cancer. Early stage lung cancers were defined as cases with stage I and II non-small cell lung cancer and limited for small cell lung cancer. Late stage was defined as stages III and IV for non-small cell lung cancer and extensive for small cell lung cancer. Statistical analyses were performed with SAS (version 8.0; SAS Institute, Cary, NC). All statistical tests were two sided, and a P-value <0.05 was considered statistically significant.

Results

Population characteristics

Mean ages of 1139 cases and 1210 controls were 61.11 and 60.21 years, respectively (P = 0.03), well within the 5-year age matching criterion for the study (Table I). Although cases and controls were frequency matched on gender, we found a borderline significant (P = 0.06) gender difference in this subset of subjects with DRC measurement. Cases compared with controls had fewer never and former smokers but more current smokers. Compared with controls, cases reported a longer duration of smoking and a higher number of cigarettes smoked per day. Controls reported higher educational attainment and annual income than cases. Cases generally had lower BMI than controls and reported more cancers in their first-degree relatives. More controls reported alcohol drinking than cases. Total caloric intake and vitamin/mineral supplement use did not differ between the cases and controls. However, cases compared with controls had significantly lower dietary Mg intake and lower DRC (Table I).

Table I.
Demographic and other characteristics by case–control status

Main effects analysis

There was an inverse association between Mg intake and risk of lung cancer (Table II). Increased Mg intake was associated with a monotonically decreasing risk of lung cancer with 17, 36 and 53% reductions in risk by increasing quartile of intake (P-trend < 0.0001). A similar monotonic inverse association was noted in men with 6, 42 and 63% reductions in risk by increasing quartile of intake (P-trend < 0.0001), but the association in women was only borderline significant (P = 0.09). This pattern was evident across all histologic subgroups with reductions in risk ranging from 48 to 63%, with increased quartile of dietary Mg intake. Similar patterns were also noted in early (OR = 0.57; 95% CI, 0.36–0.89) and late-stage (OR = 0.42; 95% CI, 0.25–0.62) lung cancer in the highest quartile of Mg intake (Table II). However, there were many more late than early stage lung cancers.

Table II.
Adjusted ORs for dietary Mg intake and lung cancer risk

The interaction between Mg intake and DRC was significant (P < 0.001) in the multivariate model. However, when DRC was stratified into suboptimal and optimal, we found a similar magnitude of protective effects and trends with increased intake of dietary Mg (data not shown). In addition, mean DRC tended to increase as the quartile of dietary Mg consumption increased (Table III), although it tended to plateau in the third quartile in cases, and was higher in quartiles 3 and 4 compared with quartiles 1 and 2 in controls.

Table III.
Energy-adjusted quartiles of dietary Mg intake and mean DRC (±SD) in cases and controls

Joint effects analysis

Compared with subjects with high dietary Mg + proficient DRC, there were monotonic 37, 54 and 136% increased risks of lung cancer, for subjects who had high dietary Mg + suboptimal DRC, low dietary Mg + proficient DRC and low dietary Mg + suboptimal DRC, respectively (P-trend < 0.0001) (Table IV). In the low dietary Mg + suboptimal DRC group, the effects were more pronounced in men (~200% increase risk) than women (~80% increase risk).

Table IV.
OR (95% CI) of lung cancer by joint effects of Mg intake and DRC

Subgroup analysis

Similar patterns were evident across all the subgroups analyzed (Table V) with the highest risk observed in the low dietary Mg + suboptimal DRC. The risks were more pronounced in older subjects (OR = 3.0; 95% CI, 2.13–4.23), those with lower BMI (OR = 2.77; 95% CI, 1.82–4.23), current smokers (OR = 3.88; 95% CI, 2.46–6.14), those with longer duration of smoking (OR = 2.90; 95% CI, 2.00–4.20) and heavy smokers (OR = 2.73; 95% CI, 1.79–4.15).

Table V.
OR (95% CI) of lung cancer by joint effects of Mg intake and DRC defined by selected variables

Risk estimates were generally similar among users as well as non-users of vitamin/mineral supplements within groups of Mg–DRC. For alcohol use, the risk estimates were higher in drinkers versus non-drinkers. Higher risks were also observed for those with a family history of cancer in first-degree relatives (OR = 2.78; 95% CI, 2.01–3.85). For histologic types of lung cancer, in the low Mg + suboptimal DRC group, the risk was more pronounced in small cell lung cancer (OR = 3.30; 95% CI, 1.69–6.46) even though the number of cases with small cell lung cancer were much smaller than the other histologic types (Table V). Similarly, the risk was more pronounced in late-stage lung cancer (OR = 2.57; 95% CI, 1.91–3.44).

Food contributors to Mg intake

We also evaluated the top food contributors to Mg intake in our population (data not shown). Dark bread, coffee, bananas, milk, nuts and cereals were the major sources of reported Mg intake in patients and controls. When we analyzed these items as risk factors, as expected, we found a significant inverse trend with increased intake of dark bread, banana and nuts with lung cancer risk (data not shown).

Discussion

In this large case–control analysis of DRC using the host-cell reactivation assay, increasing dietary intake of Mg was associated with reduced lung cancer risk ranging from 17 to 53%. In joint effects analysis, subjects with low Mg + suboptimal DRC were at the highest risk for lung cancer. To the best of our knowledge, this is the first study to report that dietary Mg intake is inversely associated with lung cancer risk, that DRC is lower in strata of low Mg intake and that low dietary Mg + suboptimal DRC together are associated with substantial increased risk.

The current research on Mg and lung cancer, including the statistical modeling, is based on biological plausibility. The potential mechanisms by which Mg may protect against lung cancer includes its role in maintaining genetic stability (6), regulation of cell proliferation (28), protection against inflammation (11,12), maintenance of lung function (29) and protection against oxidative stress (8,9). All these functions of Mg are important in maintaining the integrity of the cell and in the prevention of lung and other cancers. Likewise, we have shown previously that increasing levels of DRC were associated with reduced risk for lung cancer (P-trend < 0.0001) among men and women (data not shown) (22).

In our study, there were substantial increases in lung cancer risk among subjects with low Mg intake + suboptimal DRC compared with high Mg intake + proficient DRC. In particular, these joint effects were more pronounced in subjects who were older, current smokers, had longer duration of smoking, smoked more cigarettes per day, had a family history of cancer in first-degree relatives or were alcohol drinkers, but no difference was evident by vitamin/mineral supplement use. Unfortunately, dosage and frequency of supplementation were unavailable for analysis. The higher risk observed for thinner (BMI ≤ 25) compared with heavier subjects (BMI > 25) (Table V) in the low Mg-suboptimal DRC group was similar to findings from earlier research (26,27) and may reflect severity of disease.

Cigarette smoke, an established risk factor for lung cancer, is an important source of reactive oxygen species in the lungs; thus individuals with low Mg intake and suboptimal DRC may be unable to combat carcinogenic effects from cigarette smoke. Alcohol can also act as a pro-oxidant in lung tissue (30). Thus, Mg may function by counteracting oxidative stress induced by alcohol consumption. In animal studies, dietary Mg restriction has significantly increased lipid peroxidation (710), reduced the activity of superoxide dismutase and catalase (8) and upregulated genes associated with oxidative stress (31).

Two prospective studies reported that dietary Mg intake was inversely associated with risk for colorectal cancer in women (32,33). Furthermore, high compared with low serum levels of Mg was associated with significantly lower all-cause cancer mortality in a cohort study (34).

We have reported previously that the dietary trace metals, zinc, copper and selenium were associated with reduced risk of lung cancer at the highest quartile of intake (18). The Mg effect is much stronger than the associations for zinc and selenium, but less than the associations for dietary copper and lung cancer. When individual addition of dietary zinc, copper, carotenoids (α-carotene, β-carotene, β-cryptoxanthin, lutein–zeaxanthin and lycopene) and folate were made to model 2 (Table II), the magnitude and direction of the association with dietary Mg intake and lung cancer risk remained similar.

When we assessed top food contributors of Mg in our population as risk factors, as expected, we found a significant inverse trend with increased intake of dark bread, banana and nuts with lung cancer risk (data not shown). While these results from food contributors of Mg validate our findings regarding Mg and lung cancer risk, it is also possible that it may not be dietary Mg that is etiologically important, but something else in high Mg foods.

Emphysema, which is strongly influenced by smoking (35), is a chronic inflammatory condition (36) and a risk factor for lung cancer (37). There are experimental (38) and epidemiological (11,12) evidence that Mg has anti-inflammatory properties. For example, in the Women’s Health Study (11) and National Health and Nutrition Examination Survey (1999–2000) (12), Mg intake was inversely associated with C-reactive protein levels. In stratified analysis, we observed a significant protective trend for lung cancer with increasing intake of dietary Mg only in subjects who did not report a diagnosis of emphysema (data not shown). Chronic inflammation in emphysema possibly results in a cycle of lung injury and repair that may overwhelm the anti-inflammatory effects of dietary Mg. Nonetheless, the numbers of participants with reported emphysema were too few for meaningful interpretation of the risk estimates from joint effects analysis by the Mg–DRC groups.

We recognize that our study would be strengthened by more objective assessment of Mg status such as serum or intracellular measurements. Unfortunately, biological samples were unavailable for measurement of Mg in our study. However, since our cases were diagnosed at MD Anderson Cancer Center, we were able to identify serum Mg values for most of the cases from medical records. In the parent study, serum Mg values were available for 1103 lung cancer patients with information on clinical stage of the disease, and we found that serum Mg values did not differ between subjects with early stage (n = 289; mean Mg = 1.90 mg/dl) versus late-stage (n = 814; mean Mg = 1.92 mg/dl) lung cancer. In our subset of cases with DRC measurements (n = 1139), serum Mg values were available for 627 patients, and mean serum Mg were as follows: late-stage (n = 528; mean Mg = 2.04 mg/dl) versus early stage lung cancer (n = 199; mean Mg = 1.95 mg/dl).

Like all case–control studies, our study raises concern about recall bias. In an attempt to reduce systematic errors in reporting of dietary intake, the cases were asked to recall their diet the year prior to diagnosis and controls were asked about their diet during the year prior to enrollment into the study. In the current analysis, total calories, a marker of total intake, did not differ between cases and controls. Further, in our study, the control population consumed comparable daily mean dietary Mg intakes to values reported by National Health and Nutrition Examination Survey, 1999–2000, a national sample of USA population (39). Nevertheless, underreporting or overreporting problems associated with certain types of food cannot be totally ruled out.

The FFQ is practical for large epidemiology studies such as ours, but its use may introduce measurement errors (40,41), leading to biased estimates of the association between a given dietary factor and cancer. It has been argued that because of misclassification errors, the FFQ is not always able to detect weak associations (42,43), thereby attenuating the true association. In an effort to improve the accuracy, our interviewers were trained in FFQ administration, whereas FFQ responses were reviewed by staff nutritionists; and portion sizes were assessed with visual aids. Despite the above-mentioned limitations, the FFQ has been shown to reliably classify individuals by quartile of intake (44). Among the top 10 food contributors to Mg (data not shown), no single food was a major contributor of Mg content of the diet, but several foods contributed small amounts of intake. Our participants resided in the Houston area; and although the DIETSYS + Plus database constitutes a wide cross section of foods, we did not have data to compare Mg composition of foods in the Houston area.

Another important limitation of this study is the lack of information on the frequency and duration of use of dietary supplements containing Mg. It is well known that several multivitamin/mineral supplements with different brand names contain varying amounts of Mg. For these reasons, the current analysis is focused only on dietary Mg intake.

Regarding the DRC assay, the issues of intra- and interassay variation and reproducibility have been addressed in earlier publications of the assay development (4547). The DRC assay uses primary, viable lymphocytes that are obtained from a frozen sample of lymphocytes that were isolated from ~20 cc fresh blood sample, enabling us to perform the assay only in duplicates per sample. Therefore, it is not possible at this time to generate the coefficient of variation for each sample with triplicates. Also, the blood was drawn one time after cancer diagnosis but before any treatment and any blood drawn after the treatment may not be comparable with the one drawn before the treatment.

Dietary minerals in lung cancer risk remain an understudied area of research. Our findings suggest that Mg from food sources in the typical USA diet offers protection against lung cancer and that Mg intake and DRC together modulate risk. These intriguing results need to be confirmed in prospective studies. Assessment of the role of Mg intake and tissue levels and lung cancer mortality would complete the evidence-based research picture for this essential mineral.

Funding

Flight Attendant Medical Research Institute and Public Health Service (CA 55769 and CA 86390 to M.R.S.) from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services; Lung Specialized Program of Research Excellence (CA70909 to M.R.S.).

Acknowledgments

All study funding was provided through public grants for scientific research. No funding organization had any role in the design and conduct of the study; the collection, analysis and interpretation of the data; or the preparation, review or approval of the manuscript.

Conflict of Interest Statement: None declared.

Glossary

Abbreviations

BMI
body mass index
BPDE
benzo[a]pyrene diol epoxide
CI
confidence interval
DRC
DNA repair capacity
FFQ
food frequency questionnaire
Mg
magnesium
OR
odds ratio

References

1. Edwards B, et al. Annual report to the nation on the status of cancer, 1975–2002, featuring population-based trends in cancer treatment. J. Natl Cancer Inst. 2005;97:1407–1427. [PubMed]
2. Diez M, et al. Serum and tissue trace metal levels in lung cancer. Oncology. 1989;46:230–234. [PubMed]
3. Sartori S, et al. Serum and erythrocyte magnesium concentrations in solid tumors: relationship with stage of malignancy. Magnes. Res. 1992;5:189–192. [PubMed]
4. Fleet J, et al. Magnesium. In: Bowan BA, Russell RM, editors. Present Knowledge in Nutrition. 8th edn. Washington, DC: ILSI Press; 2001. pp. 292–301.
5. Anastassopoulou J, et al. Magnesium-DNA interaction and the possible relation of magnesium to carcinogenesis. Irradiation and free radicals. Crit. Rev. Oncol. Hematol. 2002;42:79–91. [PubMed]
6. Hartwig A. Role of magnesium in genomic stability. Mutat. Res. 2001;475:113–121. [PubMed]
7. Kumar B, et al. Magnesium deficiency-related changes in lipid peroxidation and collagen metabolism in vivo in rat heart. Int. J. Biochem. Cell Biol. 1997;29:129–134. [PubMed]
8. Shivkumar K, et al. Magnesium deficiency enhances oxidative stress and collagen synthesis in vivo in the aorta of rats. Int. J. Biochem. Cell Biol. 1997;29:1273–1278. [PubMed]
9. Hans C, et al. Magnesium deficiency increases oxidative stress in rats. Indian J. Exp. Biol. 2002;40:1275–1279. [PubMed]
10. Malpuech-Brugere C, et al. Accelerate thymus involution in magnesium-deficient rats is related to enhanced apoptosis and sensitivity to oxidative stress. Br. J. Nutr. 1999;81:405–411. [PubMed]
11. Song Y, et al. Magnesium intake, C-reactive protein, and the prevalence of metabolic syndrome in middle-aged and older U.S. women. Diabetes Care. 2005;28:1438–1444. [PubMed]
12. King D, et al. Dietary magnesium and C-reactive protein levels. J. Am. Coll. Nutr. 2005;24:166–171. [PubMed]
13. Siemes C, et al. C-reactive protein levels, variation in the C-reactive protein gene, and cancer risk: the Rotterdam Study. J. Clin. Oncol. 2006;24:5216–5222. [PubMed]
14. Littman A, et al. Chlamydia pneumoniae and lung cancer: epidemiologic evidence. Cancer Epidemiol. Biomarkers Prev. 2005;14:773–778. [PubMed]
15. Littman A, et al. Prior lung disease and risk of lung cancer in a large prospective study. Cancer Causes Control. 2004;15:819–827. [PubMed]
16. Brown D, et al. Asthma and risk of death from lung cancer: NHANES II Mortality Study. J. Asthma. 2005;42:597–600. [PubMed]
17. Daniels C, et al. Does interstitial lung disease predispose to lung cancer? Curr. Opin. Pulm. Med. 2005;11:431–437. [PubMed]
18. Mahabir S, et al. Dietary zinc, copper and selenium and risk of lung cancer. Int. J. Cancer. 2006;120:1108–1115. [PubMed]
19. Hudmon K, et al. Identifying and recruiting healthy control subjects from a managed care organization: a methodology for molecular epidemiologic case-control studies of cancer. Cancer Epidemiol. Biomarkers Prev. 1997;6:565–571. [PubMed]
20. Block G, et al. Comparisons of two dietary questionnaires validated against multiple dietary records collected during a 1-year period. J. Am. Diet. Assoc. 1992;92:686–693. [PubMed]
21. Block G, et al. Revision of dietary analysis software for the health habits and history questionnaire. Am. J. Epidemiol. 1994;139:1190–1196. [PubMed]
22. Wei Q, et al. Repair of tobacco carcinogen-induced DNA adducts and lung cancer risk: a molecular epidemiologic study. J. Natl Cancer Inst. 2000;92:1764–1772. [PubMed]
23. McCuchan J, et al. Enhancement of the infectivity of simian virus 40 deoxyribonucleic acid with diethylaminoethyl-dextran. J. Natl Cancer Inst. 1968;41:351–356. [PubMed]
24. Willett W, et al. Total energy intake: implications for epidemiologic analyses. Am. J. Epidemiol. 1986;124:17–27. [PubMed]
25. Nomura A, et al. Body mass index as a predictor of cancer in men. J. Natl Cancer Inst. 1985;74:319–323. [PubMed]
26. Kabat G, et al. Body mass index and lung cancer risk. Am. J. Epidemiol. 1992;135:769. [PubMed]
27. Knekt P, et al. Leanness and lung-cancer risk. Int. J. Cancer. 1991;49:208–213. [PubMed]
28. Rubin H. Magnesium: the missing element in molecular views of cell proliferation control. Bioessays. 2005;27:311–320. [PubMed]
29. Britton J, et al. Dietary magnesium, lung function, wheezing, and airway hyperreactivity in a random adult population sample. Lancet. 1994;344:357–362. [PubMed]
30. Yang M, et al. Effects of the ADH3, CYP2E1, and GSTP1 genetic polymorphisms on their expressions in Caucasian lung tissue. Lung Cancer. 2002;38:15–21. [PubMed]
31. Petrault I, et al. Changes in gene expression in rat thymocytes identified by cDNA array support the occurrence of oxidative stress in early magnesium deficiency. Biochim. Biophys. Acta. 2002;1586:92–98. [PubMed]
32. Larsson S, et al. Magnesium intake in relation to risk of colorectal cancer in women. JAMA. 2005;293:83–89. [PubMed]
33. Folsom A, et al. Magnesium intake and reduced risk of colon cancer in a prospective study of women. Am. J. Epidemiol. 2005;163:232–235. [PubMed]
34. Leone N, et al. Zinc, copper, and magnesium and risk of all-cause, cancer, and cardiovascular mortality. Epidemiology. 2006;17:308–314. [PubMed]
35. Fain S, et al. Early emphysematous changes in asymptomatic smokers: detection with 3He MR imaging. Radiology. 2006;239:875–883. [PubMed]
36. Barnes P, et al. Chronic obstructive pulmonary disease: molecular and cellular mechanisms. Eur. Respir. J. 2003;22:672–688. [PubMed]
37. Schabath M, et al. Opposing effects of emphysema, hay fever, and selected genetic variants on lung cancer risk. Am. J. Epidemiol. 2005;161:412–422. [PubMed]
38. Mazur A, et al. Magnesium and the inflammatory response: potential physiological implications. Arch. Biochem. Biophys. 2006 in press.
39. Ervin R, et al. Dietary Intake of Selected Minerals for the United States Population: 1999–2000. Advanced Data from Vital and Health Statistics; No. 341. 2004. Hayattsville, MD: National Center for Health Statistics. [PubMed]
40. Mahabir S, et al. Calorie intake misreporting by diet record and food frequency questionnaire compared to doubly labeled water among postmenopausal women. Eur. J. Clin. Nutr. 2006;60:561–565. [PubMed]
41. Subar A, et al. Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study. Am. J. Epidemiol. 2003;158:1–13. [PubMed]
42. Schatzkin A, et al. A comparison of a food frequency questionnaire with a 24-hour recall for use in an epidemiological cohort study: results from the biomarker-based Observing Protein and Energy Nutrition (OPEN) study. Int. J. Epidemiol. 2003;32:1054–1062. [PubMed]
43. Bingham S, et al. Are imprecise methods obscuring a relation between fat and breast cancer? Lancet. 2003;362:212–214. [PubMed]
44. Bingham S, et al. Comparison of dietary assessment methods in nutritional epidemiology: weighed records v. 24 h recalls, food frequency questionnaires and estimated-diet records. Br. J. Nutr. 1994;72:619–643. [PubMed]
45. Athas W, et al. Development and field-test validation of an assay for DNA repair in circulating human lymphocytes. Cancer Res. 1991;51:5786–5793. [PubMed]
46. Wei Q, et al. DNA repair and aging in basal cell carcinoma: a molecular epidemiology study. Proc. Natl Acad. Sci. USA. 1993;90:1614–1618. [PubMed]
47. Qiao Y, et al. Rapid assessment of repair of ultraviolet DNA damage with a modified host-cell reactivation assay using a luciferase reporter gene and correlation with polymorphisms of DNA repair genes in normal human lymphocytes. Mutat. Res. 2002;30:165–174. [PubMed]

Articles from Carcinogenesis are provided here courtesy of Oxford University Press