The advantages of UK Biobank's open-access strategy for health research.

Ready access to health research studies is becoming more important as researchers, and their funders, seek to maximize the opportunities for scientific innovation and health improvements. Large-scale population-based prospective studies are particularly useful for multidisciplinary research into the causes, treatment and prevention of many different diseases. UK Biobank has been established as an open-access resource for public health research, with the intention of making the data as widely available as possible in an equitable and transparent manner. Access to UK Biobank's unique breadth of phenotypic and genetic data has attracted researchers worldwide from across academia and industry. As a consequence, it has enabled scientists to perform world-leading collaborative research. Moreover, open access to an already deeply characterized cohort has encouraged both public and private sector investment in further enhancements to make UK Biobank an unparalleled resource for public health research and an exemplar for the development of open-access approaches for other studies.

Adherence to the Mediterranean diet and risk of bladder cancer in the EPIC cohort study.

There is growing evidence of the protective role of the Mediterranean diet (MD) on cancer. However, to date no epidemiological study has investigated the influence of the MD on bladder cancer. We evaluated the association between adherence to the MD and risk of urothelial cell bladder cancer (UCC), according to tumor aggressiveness, in the European Prospective Investigation into Cancer and Nutrition (EPIC). The analysis included 477,312 participants, recruited from ten European countries between 1991 and 2000. Information from validated dietary questionnaires was used to develop a relative Mediterranean diet score (rMED), including nine dietary components. Cox regression models were used to assess the effect of the rMED on UCC risk, while adjusting for dietary energy and tobacco smoking of any kind. Stratified analyses were performed by sex, BMI, smoking status, European region and age at diagnosis. During an average follow-up of 11 years, 1,425 participants (70.9% male) were diagnosed with a first primary UCC. There was a negative but non-significant association between a high versus low rMED score and risk of UCC overall (HR: 0.84 [95% CI 0.69, 1.03]) and risk of aggressive (HR: 0.88 [95% CI 0.61, 1.28]) and non-aggressive tumors (HR: 0.78 [95% CI 0.54, 1.14]). Although there was no effect modification in the stratified analyses, there was a significant 34% (p = 0.043) decreased risk of UCC in current smokers with a high rMED score. In EPIC, the MD was not significantly associated with risk of UCC, although we cannot exclude that a MD may reduce risk in current smokers.

Inflammation marker and risk of pancreatic cancer: a nested case-control study within the EPIC cohort.

BACKGROUND: Established risk factors for pancreatic cancer include smoking, long-standing diabetes, high body fatness, and chronic pancreatitis, all of which can be characterised by aspects of inflammatory processes. However, prospective studies investigating the relation between inflammatory markers and pancreatic cancer risk are scarce. METHODS: We conducted a nested case-control study within the European Prospective Investigation into Cancer and Nutrition, measuring prediagnostic blood levels of C-reactive protein (CRP), interleukin-6 (IL-6), and soluble receptors of tumour necrosis factor-α (sTNF-R1, R2) in 455 pancreatic cancer cases and 455 matched controls. Odds ratios (ORs) were estimated using conditional logistic regression models. RESULTS: None of the inflammatory markers were significantly associated with risk of pancreatic cancer overall, although a borderline significant association was observed for higher circulating sTNF-R2 (crude OR=1.52 (95% confidence interval (CI) 0.97-2.39), highest vs lowest quartile). In women, however, higher sTNF-R1 levels were significantly associated with risk of pancreatic cancer (crude OR=1.97 (95% CI 1.02-3.79)). For sTNF-R2, risk associations seemed to be stronger for diabetic individuals and those with a higher BMI. CONCLUSION: Prospectively, CRP and IL-6 do not seem to have a role in our study with respect to risk of pancreatic cancer, whereas sTNF-R1 seemed to be a risk factor in women and sTNF-R2 might be a mediator in the risk relationship between overweight and diabetes with pancreatic cancer. Further large prospective studies are needed to clarify the role of proinflammatory proteins and cytokines in the pathogenesis of exocrine pancreatic cancer.

Alcohol consumption and risk of type 2 diabetes in European men and women: influence of beverage type and body size The EPIC-InterAct study.

OBJECTIVE: To investigate the association between alcohol consumption and type 2 diabetes, and determine whether this is modified by sex, body mass index (BMI) and beverage type. DESIGN: Multicentre prospective case-cohort study. SETTING: Eight countries from the European Prospective Investigation into Cancer and Nutrition cohort. SUBJECTS: A representative baseline sample of 16 154 participants and 12 403 incident cases of type 2 diabetes. INTERVENTIONS: Alcohol consumption assessed using validated dietary questionnaires. MAIN OUTCOME MEASURES: Occurrence of type 2 diabetes based on multiple sources (mainly self-reports), verified against medical information. RESULTS: Amongst men, moderate alcohol consumption was nonsignificantly associated with a lower incidence of diabetes with a hazard ratio (HR) of 0.90 (95% CI: 0.78-1.05) for 6.1-12.0 versus 0.1-6.0 g day(-1) , adjusted for dietary and diabetes risk factors. However, the lowest risk was observed at higher intakes of 24.1-96.0 g day(-1) with an HR of 0.86 (95% CI: 0.75-0.98). Amongst women, moderate alcohol consumption was associated with a lower incidence of diabetes with a hazard ratio of 0.82 (95% CI: 0.72-0.92) for 6.1-12.0 g day(-1) (P interaction gender <0.01). The inverse association between alcohol consumption and diabetes was more pronounced amongst overweight (BMI ≥ 25 kg m(-2) ) than normal-weight men and women (P interaction < 0.05). Adjusting for waist and hip circumference did not alter the results for men, but attenuated the association for women (HR=0.90, 95% CI: 0.79-1.03 for 6.1-12.0 g day(-1) ). Wine consumption for men and fortified wine consumption for women were most strongly associated with a reduced risk of diabetes. CONCLUSIONS: The results of this study show that moderate alcohol consumption is associated with a lower risk of type 2 diabetes amongst women only. However, this risk reduction is in part explained by fat distribution. The relation between alcohol consumption and type 2 diabetes was stronger for overweight than normal-weight women and men.

Concentrations of IGF-I and IGFBP-3 and pancreatic cancer risk in the European Prospective Investigation into Cancer and Nutrition.

BACKGROUND: Insulin-like growth factors (IGFs) and their binding proteins (BPs) regulate cell differentiation, proliferation and apoptosis, and may have a role in the aetiology of various cancers. Information on their role in pancreatic cancer is limited and was examined here in a case-control study nested within the European Prospective Investigation into Cancer and Nutrition. METHODS: Serum concentrations of IGF-I and IGFBP-3 were measured using enzyme-linked immunosorbent assays in 422 cases and 422 controls matched on age, sex, study centre, recruitment date, and time since last meal. Conditional logistic regression was used to compute odds ratios (OR) and 95% confidence intervals (CI) adjusted for confounding variables. RESULTS: Neither circulating levels of IGF-I (OR=1.21, 95% CI 0.75-1.93 for top vs bottom quartile, P-trend 0.301), IGFBP-3 (OR=1.00, 95% CI 0.66-1.51, P-trend 0.79), nor the molar IGF-I/IGFBP-3 ratio, an indicator of free IGF-I level (OR=1.22, 95% CI 0.75-1.97, P-trend 0.27), were statistically significantly associated with the risk of pancreatic cancer. In a cross-classification, however, a high concentration of IGF-I with concurrently low levels of IGFBP-3 was related to an increased risk of pancreatic cancer (OR=1.72, 95% CI 1.05-2.83; P-interaction=0.154). CONCLUSION: On the basis of these results, circulating levels of components of the IGF axis do not appear to be the risk factors for pancreatic cancer. However, on the basis of the results of a subanalysis, it cannot be excluded that a relatively large amount of IGF-1 together with very low levels of IGFBP-3 might still be associated with an increase in pancreatic cancer risk.

Helicobacter pylori infection assessed by ELISA and by immunoblot and noncardia gastric cancer risk in a prospective study: the Eurgast-EPIC project.

BACKGROUND: In epidemiological studies, Helicobacter pylori infection is usually detected by enzyme-linked immunosorbent assay (ELISA). However, infection can spontaneously clear from the mucosa during the progression of atrophy and could lead to substantial under-detection of infection and underestimation of its effect on gastric cancer (GC) risk. Antibodies detected by western blot are known to persist longer after the loss of the infection. METHODS: In a nested case-control study from the Eurogast-EPIC cohort, including 88 noncardia GC cases and 338 controls, we assessed the association between noncardia GC and H. pylori infection comparing antibodies detected by western blot (HELICOBLOT2.1) to those detected by ELISA (Pyloriset EIA-GIII(®)). RESULTS: By immunoblot, 82 cases (93.2%) were H. pylori positive, 10 of these cases (11.4%) were negative by ELISA and only 6 cases (6.8%) were negative by both ELISA and immunoblot. Multivariable odds ratio (OR) for noncardia GC comparing immunoglobulin G positive versus negative by ELISA was 6.8 [95% confidence interval (CI) 3.0-15.1], and by immunoblot, the OR was 21.4 (95% CI 7.1-64.4). CONCLUSIONS: Using a western blot assay, nearly all noncardia GC were classified as H. pylori positive and the OR was more than threefold higher than the OR assessed by ELISA, supporting the hypothesis that H. pylori infection is a necessary condition for noncardia GC.

Oral contraceptive use and reproductive factors and risk of ovarian cancer in the European Prospective Investigation into Cancer and Nutrition.

BACKGROUND: It is well established that parity and use of oral contraceptives reduce the risk of ovarian cancer, but the associations with other reproductive variables are less clear. METHODS: We examined the associations of oral contraceptive use and reproductive factors with ovarian cancer risk in the European Prospective Investigation into Cancer and Nutrition. Among 327,396 eligible women, 878 developed ovarian cancer over an average of 9 years. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using Cox proportional hazard models stratified by centre and age, and adjusted for smoking status, body mass index, unilateral ovariectomy, simple hysterectomy, menopausal hormone therapy, and mutually adjusted for age at menarche, age at menopause, number of full-term pregnancies and duration of oral contraceptive use. RESULTS: Women who used oral contraceptives for 10 or more years had a significant 45% (HR, 0.55; 95% CI, 0.41-0.75) lower risk compared with users of 1 year or less (P-trend, <0.01). Compared with nulliparous women, parous women had a 29% (HR, 0.71; 95% CI, 0.59-0.87) lower risk, with an 8% reduction in risk for each additional pregnancy. A high age at menopause was associated with a higher risk of ovarian cancer (>52 vs ≤ 45 years: HR, 1.46; 95% CI, 1.06-1.99; P-trend, 0.02). Age at menarche, age at first full-term pregnancy, incomplete pregnancies and breastfeeding were not associated with risk. CONCLUSION: This study shows a strong protective association of oral contraceptives and parity with ovarian cancer risk, a higher risk with a late age at menopause, and no association with other reproductive factors.

Primary brain tumours and specific serum immunoglobulin E: a case-control study nested in the European Prospective Investigation into Cancer and Nutrition cohort.

BACKGROUND: Case-control studies suggest that patients with allergic diseases have a lower risk of developing glioma but not meningioma or schwannoma. However, those data can be differentially biased. Prospective studies with objective measurements of immunologic biomarkers, like immunoglobulin E (IgE), in blood obtained before cancer diagnosis could help to clarify whether an aetiological association exists. METHODS: The present case-control study nested within the European Prospective Investigation into Cancer and Nutrition (EPIC) measured specific serum IgE as a biomarker for the most common inhalant allergens in 275 glioma, 175 meningioma and 49 schwannoma cases and 963 matched controls using the ImmunoCAP specific IgE test. Subjects with an IgE level ≥0.35 kUA/l (kilo antibody units per litre) were classified as sensitized by allergens. Odds ratios (OR) and 95% confidence intervals (CI) were estimated by adjusted conditional logistic regression models for each tumour subtype. The effect of dose-response relationship was assessed in five increasing IgE level categories to estimate P-values for trend. RESULTS: The risk of glioma was inversely related to allergic sensitization (OR = 0.73; 95% CI 0.51-1.06), especially pronounced in women (OR = 0.53; 95% CI 0.30-0.95). In dose-response analyses, for high-grade glioma, the lowest OR was observed in sera with the highest IgE levels (P for trend = 0.04). No association was seen for meningioma and schwannoma. CONCLUSION: The results, based on serum samples prospectively collected in a cohort study, provide some support for the hypothesis that individuals with allergic sensitization are at reduced risk of glioma and confirm results from previous case-control studies.

Mediterranean dietary pattern and cancer risk in the EPIC cohort.

BACKGROUND: Although several studies have investigated the association of the Mediterranean diet with overall mortality or risk of specific cancers, data on overall cancer risk are sparse. METHODS: We examined the association between adherence to Mediterranean dietary pattern and overall cancer risk using data from the European Prospective Investigation Into Cancer and nutrition, a multi-centre prospective cohort study including 142,605 men and 335,873. Adherence to Mediterranean diet was examined using a score (range: 0-9) considering the combined intake of fruits and nuts, vegetables, legumes, cereals, lipids, fish, dairy products, meat products, and alcohol. Association with cancer incidence was assessed through Cox regression modelling, controlling for potential confounders. RESULTS: In all, 9669 incident cancers in men and 21,062 in women were identified. A lower overall cancer risk was found among individuals with greater adherence to Mediterranean diet (hazard ratio=0.96, 95% CI 0.95-0.98) for a two-point increment of the Mediterranean diet score. The apparent inverse association was stronger for smoking-related cancers than for cancers not known to be related to tobacco (P (heterogeneity)=0.008). In all, 4.7% of cancers among men and 2.4% in women would be avoided in this population if study subjects had a greater adherence to Mediterranean dietary pattern. CONCLUSION: Greater adherence to a Mediterranean dietary pattern could reduce overall cancer risk.

Fluid intake and incidence of renal cell carcinoma in UK women.

BACKGROUND: It has been suggested that the apparent protective effect of alcohol intake on renal cell carcinoma may be due to the diluting effect of carcinogens by a high total fluid intake. We assessed the association between intakes of total fluids and of specific beverages on the risk of renal cell carcinoma in a large prospective cohort of UK women. METHODS: Information on beverage consumption was obtained from a questionnaire sent ∼3 years after recruitment into the Million Women Study. Cox proportional hazards models were used to estimate relative risks (RRs) and 95% confidence intervals (CIs) for renal cell carcinoma associated with beverage consumption adjusted for age, region of residence, socioeconomic status, smoking, and body mass index. RESULTS: After an average of 5.2 years of follow-up, 588 cases of renal cell carcinoma were identified among 779,369 women. While alcohol intake was associated with a reduced risk of renal cell carcinoma (RR for ≥ 2 vs <1 drink per day: 0.76; 95% CI: 0.61-0.96; P for trend=0.02), there was no association with total fluid intake (RR for ≥ 12 vs <7 drinks per day: 1.15; 95% CI: 0.91-1.45; P for trend=0.3) or with intakes of specific beverages. CONCLUSIONS: The apparent protective effect of alcohol on the risk of renal cell carcinoma is unlikely to be related to a high fluid intake.

Oral contraceptives, reproductive history and risk of colorectal cancer in the European Prospective Investigation into Cancer and Nutrition.

BACKGROUND: Oral contraceptive use and reproductive factors may initiate long-term changes to the hormonal milieu and thereby, possibly influence colorectal cancer risk. METHODS: We examined the association of hormonal and reproductive factors with risk of colorectal cancer among 337,802 women in the European Prospective Investigation into Cancer and Nutrition, of whom 1878 developed colorectal cancer. RESULTS: After stratification for center and age, and adjustment for body mass index, smoking, diabetes mellitus, physical activity and alcohol consumption, ever use of oral contraceptives was marginally inversely associated with colorectal cancer risk (hazard ratio (HR), 0.92; 95% confidence interval (CI), 0.83-1.02), although this association was stronger among post-menopausal women (HR, 0.84; 95% CI: 0.74-0.95). Duration of oral contraceptive use and reproductive factors, including age at menarche, age at menopause, type of menopause, ever having an abortion, parity, age at first full-term pregnancy and breastfeeding, were not associated with colorectal cancer risk. CONCLUSION: Our findings provide limited support for a potential inverse association between oral contraceptives and colorectal cancer risk.

Serum concentrations of vitamin B12 and folate in British male omnivores, vegetarians and vegans: results from a cross-sectional analysis of the EPIC-Oxford cohort study.

BACKGROUND/OBJECTIVES: Vegans, and to a lesser extent vegetarians, have low average circulating concentrations of vitamin B12; however, the relation between factors such as age or time on these diets and vitamin B12 concentrations is not clear. The objectives of this study were to investigate differences in serum vitamin B12 and folate concentrations between omnivores, vegetarians and vegans and to ascertain whether vitamin B12 concentrations differed by age and time on the diet. SUBJECTS/METHODS: A cross-sectional analysis involving 689 men (226 omnivores, 231 vegetarians and 232 vegans) from the European Prospective Investigation into Cancer and Nutrition Oxford cohort. RESULTS: Mean serum vitamin B12 was highest among omnivores (281, 95% CI: 270-292 pmol/l), intermediate among vegetarians (182, 95% CI: 175-189 pmol/l) and lowest among vegans (122, 95% CI: 117-127 pmol/l). In all, 52% of vegans, 7% of vegetarians and one omnivore were classified as vitamin B12 deficient (defined as serum vitamin B12 <118 pmol/l). There was no significant association between age or duration of adherence to a vegetarian or a vegan diet and serum vitamin B12. In contrast, folate concentrations were highest among vegans, intermediate among vegetarians and lowest among omnivores, but only two men (both omnivores) were categorized as folate deficient (defined as serum folate <6.3 nmol/l). CONCLUSION: Vegans have lower vitamin B12 concentrations, but higher folate concentrations, than vegetarians and omnivores. Half of the vegans were categorized as vitamin B12 deficient and would be expected to have a higher risk of developing clinical symptoms related to vitamin B12 deficiency.

Reduced muscle power is associated with slower walking velocity and falls in people with Parkinson's disease.

PURPOSE: Muscle strength (force) and power (force x velocity) are reduced in Parkinson's disease (PD). Reduced muscle power is associated with slower walking velocity and falls in the older population, but these associations in people with PD have not previously been investigated. This study investigated the relationships between leg extensor muscle power and strength with walking speed and past falls in people with PD. PARTICIPANTS AND METHODS: Forty people with mild to moderate PD were assessed. Walking velocity was measured over 10 m and the number of falls the participant reported having in the past 12 months was recorded. Leg extensor muscle power and strength were measured using a Keiser leg press machine. RESULTS: Muscle power explained more than half of the variance (R(2) = 0.54) in walking velocity and remained significantly (p < 0.05) associated with walking velocity in models which included Unified Parkinson's Disease Rating Scale (UPDRS) motor scores. Participants with low muscle power were 6 times more likely to report multiple falls in the past year than those with high muscle power (OR = 6.0, 95% CI 1.1 to 33.3), though this association between falls and power was no longer significant in models which included UPDRS motor scores (p = 0.09). CONCLUSION: Muscle power is a significant determinant of walking velocity in PD even after adjusting for UPDRS motor score. Muscle power training warrants investigation in people with PD.

Alcohol consumption patterns, diet and body weight in 10 European countries.

BACKGROUND/OBJECTIVES: Europe has the highest level of alcohol consumption in the world. As drinking patterns are important determinants of the beneficial and harmful effects of alcohol consumption, we investigated alcohol consumption in relation to nutrient intake, place of consumption, education and body weight in a sample of adults from 10 European countries. METHODS: A 24-h dietary recall interview was conducted on 13 025 men and 23 009 women, aged 35-74 years, from 27 centres participating in the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Means and standard errors of alcohol consumption, adjusted for age, were calculated, stratified by gender and centre. RESULTS: In many centres, higher level drinkers (males consuming >24 g of ethanol/day, equivalent to >2 standard drinks and females consuming >12 g of ethanol/day equivalent to >1 standard drink) obtained more energy from fat and protein and less from sugar than did abstainers. The proportion of energy from starch tended to be higher for male and lower for female higher level drinkers than for abstainers. Female higher level drinkers had a lower body mass index than did abstainers, whereas male higher level drinkers generally weighed more. Male higher level drinkers were less educated than abstainers in Mediterranean countries, but were more educated elsewhere. Female higher level drinkers were usually more educated than were abstainers. Outside the home, consumption (both genders) tended to be at friends' homes, particularly among men in Northern and Central Europe, and in bars in Spain. CONCLUSIONS: This study reveals clear geographical differences in drinking habits across Europe, and shows that the characteristics of different alcohol consumption categories also vary.

Cancer incidence in British vegetarians.

BACKGROUND: Few prospective studies have examined cancer incidence among vegetarians. METHODS: We studied 61,566 British men and women, comprising 32,403 meat eaters, 8562 non-meat eaters who did eat fish ('fish eaters') and 20,601 vegetarians. After an average follow-up of 12.2 years, there were 3350 incident cancers of which 2204 were among meat eaters, 317 among fish eaters and 829 among vegetarians. Relative risks (RRs) were estimated by Cox regression, stratified by sex and recruitment protocol and adjusted for age, smoking, alcohol, body mass index, physical activity level and, for women only, parity and oral contraceptive use. RESULTS: There was significant heterogeneity in cancer risk between groups for the following four cancer sites: stomach cancer, RRs (compared with meat eaters) of 0.29 (95% CI: 0.07-1.20) in fish eaters and 0.36 (0.16-0.78) in vegetarians, P for heterogeneity=0.007; ovarian cancer, RRs of 0.37 (0.18-0.77) in fish eaters and 0.69 (0.45-1.07) in vegetarians, P for heterogeneity=0.007; bladder cancer, RRs of 0.81 (0.36-1.81) in fish eaters and 0.47 (0.25-0.89) in vegetarians, P for heterogeneity=0.05; and cancers of the lymphatic and haematopoietic tissues, RRs of 0.85 (0.56-1.29) in fish eaters and 0.55 (0.39-0.78) in vegetarians, P for heterogeneity=0.002. The RRs for all malignant neoplasms were 0.82 (0.73-0.93) in fish eaters and 0.88 (0.81-0.96) in vegetarians (P for heterogeneity=0.001). CONCLUSION: The incidence of some cancers may be lower in fish eaters and vegetarians than in meat eaters.

Plasma phyto-oestrogens and prostate cancer in the European Prospective Investigation into Cancer and Nutrition.

We examined plasma concentrations of phyto-oestrogens in relation to risk for subsequent prostate cancer in a case-control study nested in the European Prospective Investigation into Cancer and Nutrition. Concentrations of isoflavones genistein, daidzein and equol, and that of lignans enterolactone and enterodiol, were measured in plasma samples for 950 prostate cancer cases and 1042 matched control participants. Relative risks (RRs) for prostate cancer in relation to plasma concentrations of these phyto-oestrogens were estimated by conditional logistic regression. Higher plasma concentrations of genistein were associated with lower risk of prostate cancer: RR among men in the highest vs the lowest fifth, 0.71 (95% confidence interval (CI) 0.53-0.96, P trend=0.03). After adjustment for potential confounders this RR was 0.74 (95% CI 0.54-1.00, P trend=0.05). No statistically significant associations were observed for circulating concentrations of daidzein, equol, enterolactone or enterodiol in relation to overall risk for prostate cancer. There was no evidence of heterogeneity in these results by age at blood collection or country of recruitment, nor by cancer stage or grade. These results suggest that higher concentrations of circulating genistein may reduce the risk of prostate cancer but do not support an association with plasma lignans.

Animal foods, protein, calcium and prostate cancer risk: the European Prospective Investigation into Cancer and Nutrition.

We examined consumption of animal foods, protein and calcium in relation to risk of prostate cancer among 142 251 men in the European Prospective Investigation into Cancer and Nutrition. Associations were examined using Cox regression, stratified by recruitment centre and adjusted for height, weight, education, marital status and energy intake. After an average of 8.7 years of follow-up, there were 2727 incident cases of prostate cancer, of which 1131 were known to be localised and 541 advanced-stage disease. A high intake of dairy protein was associated with an increased risk, with a hazard ratio for the top versus the bottom fifth of intake of 1.22 (95% confidence interval (CI): 1.07-1.41, P(trend)=0.02). After calibration to allow for measurement error, we estimated that a 35-g day(-1) increase in consumption of dairy protein was associated with an increase in the risk of prostate cancer of 32% (95% CI: 1-72%, P(trend)=0.04). Calcium from dairy products was also positively associated with risk, but not calcium from other foods. The results support the hypothesis that a high intake of protein or calcium from dairy products may increase the risk for prostate cancer.

Multi-factor dimensionality reduction applied to a large prospective investigation on gene-gene and gene-environment interactions.

It is becoming increasingly evident that single-locus effects cannot explain complex multifactorial human diseases like cancer. We applied the multi-factor dimensionality reduction (MDR) method to a large cohort study on gene-environment and gene-gene interactions. The study (case-control nested in the EPIC cohort) was established to investigate molecular changes and genetic susceptibility in relation to air pollution and environmental tobacco smoke (ETS) in non-smokers. We have analyzed 757 controls and 409 cases with bladder cancer (n=124), lung cancer (n=116) and myeloid leukemia (n=169). Thirty-six gene variants (DNA repair and metabolic genes) and three environmental exposure variables (measures of air pollution and ETS at home and at work) were analyzed. Interactions were assessed by prediction error percentage and cross-validation consistency (CVC) frequency. For lung cancer, the best model was given by a significant gene-environment association between the base excision repair (BER) XRCC1-Arg399Gln polymorphism, the double-strand break repair (DSBR) BRCA2-Asn372His polymorphism and the exposure variable 'distance from heavy traffic road', an indirect and robust indicator of air pollution (mean prediction error of 26%, P<0.001, mean CVC of 6.60, P=0.02). For bladder cancer, we found a significant 4-loci association between the BER APE1-Asp148Glu polymorphism, the DSBR RAD52-3'-untranslated region (3'-UTR) polymorphism and the metabolic gene polymorphisms COMT-Val158Met and MTHFR-677C>T (mean prediction error of 22%, P<0.001, mean CVC consistency of 7.40, P<0.037). For leukemia, a 3-loci model including RAD52-2259C>T, MnSOD-Ala9Val and CYP1A1-Ile462Val had a minimum prediction error of 31% (P<0.001) and a maximum CVC of 4.40 (P=0.086). The MDR method seems promising, because it provides a limited number of statistically stable interactions; however, the biological interpretation remains to be understood.

Diet, serum insulin-like growth factor-I and IGF-binding protein-3 in European women.

OBJECTIVE: The aim of this study was to examine the relationship of diet with serum insulin-like growth factor-I (IGF-I) and IGF-binding protein-3 in women. DESIGN: Cross-sectional study. SETTING AND SUBJECTS: The population are 2109 women who were control subjects in a case-control study of breast cancer nested in the European Prospective Investigation into Cancer and Nutrition. Control subjects were randomly chosen among risk sets consisting of female cohort members alive and free of cancer (except non-melanoma skin cancer) at the time of diagnosis of the index case. Matching criteria were age at enrolment, follow-up time, time of the day of blood collection and study centre. Diet was measured through validated questionnaires. Serum hormone concentrations were measured by enzyme-linked immunosorbent assays. The relationship between serum IGF-I, IGFBP-3, and intake of nutrients and foods was explored by linear regression in models adjusted for energy intake, age, body mass index, smoking, physical activity, centre and laboratory batch. RESULTS: Serum IGF-I levels were positively related to protein intake (P(trend)<0.001), but not related to energy, fat or carbohydrate intake. Positive relationships were observed with the intake of milk (P(trend)=0.007), calcium (P(trend)<0.001), magnesium (P(trend)=0.003), phosphorus (P(trend)<0.001), potassium (P(trend)=0.002), vitamin B6 (P(trend)=0.03), vitamin B2 (P(trend)=0.001) and inverse relationships with vegetables (P(trend)=0.02) and beta-carotene (P(trend)=0.02). IGFBP-3 was not related with most of the nutrients and foods in this study. CONCLUSIONS: In this population, circulating IGF-I is modestly related with the intake of protein and minerals, and with milk and cheese, while IGFBP-3 does not appear to be related with diet.

Relationship of alcohol intake and sex steroid concentrations in blood in pre- and post-menopausal women: the European Prospective Investigation into Cancer and Nutrition.

OBJECTIVE: Women with a moderate intake of alcohol have higher concentrations of sex steroids in serum, and higher risk of developing breast cancer, compared to non-drinkers. In the present study, we investigate the relationships between alcohol consumption and serum levels of sex steroids and sex-hormone binding globulin (SHBG) in 790 pre- and 1,291 post-menopausal women, who were part of the European Prospective Investigation into Cancer and Nutrition (EPIC). METHODS: Serum levels of testosterone (T), androstenedione (Delta4), dehydroepiandrosterone sulphate (DHEAS), estrone (E1), estradiol (E2) and SHBG were measured by direct immunoassays. Free T (fT) and free E2 (fE2) were calculated according to mass action laws. Current alcohol intake exposure to alcohol was assessed from dietary questionnaires. RESULTS: Pre-menopausal women who consumed more than 25 g/day of alcohol had about 30% higher DHEAS, T and fT, 20% higher Delta4 and about 40% higher E1, concentrations compared to women who were non-consumers. E2, fE2 and SHBG concentrations showed no association with current alcohol intake. In post-menopausal women, DHEAS, fT, T, Delta4, and E1 concentrations were between 10% and 20% higher in women who consumed more than 25 g/day of alcohol compared to non-consumers. E2 or fE2 were not associated with alcohol intake at all. SHBG levels were about 15% lower in alcohol consumers compared to non-consumers. CONCLUSION: This study supports the hypothesis of an influence of alcohol intake on sex hormone concentrations in blood.