Search

Abstract Background Fruit consumption has been associated with a lower cardiovascular disease (CVD) risk but the underlying mechanisms are unclear. We investigated the cross-sectional and prospective associations of fruit consumption with markers of adiposity, blood pressure, lipids, low-grade inflammation, glycaemia, and oxidative stress. Methods The main analyses included 365 534 middle-aged adults from the UK Biobank at baseline, of whom 11 510, and 38 988 were included in the first and second follow-up respectively, free from CVD and cancer at baseline. Fruit consumption frequency at baseline was assessed using a questionnaire. We assessed the cross-sectional and prospective associations of fruit with adiposity (body mass index, waist circumference and %body fat), systolic and diastolic blood pressure, lipids (low-density and high-density lipoproteins, triglycerides and apolipoprotein B), glycaemia (haemoglobin A1c), low-grade inflammation (C-reactive protein) and oxidative stress (gamma-glutamyl-transferase) using linear regression models adjusted for socioeconomic and lifestyle factors. Analyses were repeated in a subset with two to five complete 24-h dietary assessments (n = 26 596) allowing for adjustment for total energy intake. Results Fruit consumption at baseline generally showed weak inverse associations with adiposity and biomarkers at baseline. Most of these relationships did not persist through follow-up, except for inverse associations with diastolic blood pressure, C-reactive protein, gamma-glutamyl transferase and adiposity. However, for most mechanisms, mean levels varied by less than 0.1 standard deviations (SD) between high and low fruit consumption (> 3 vs

Abstract Background The associations of vegetarian diets with risks for site-specific cancers have not been estimated reliably due to the low number of vegetarians in previous studies. Therefore, the Cancer Risk in Vegetarians Consortium was established. The aim is to describe and compare the baseline characteristics between non-vegetarian and vegetarian diet groups and between the collaborating studies. Methods We harmonised individual-level data from 11 prospective cohort studies from Western Europe, North America, South Asia and East Asia. Comparisons of food intakes, sociodemographic and lifestyle factors were made between diet groups and between cohorts using descriptive statistics. Results 2.3 million participants were included; 66% women and 34% men, with mean ages at recruitment of 57 (SD: 7.8) and 57 (8.6) years, respectively. There were 2.1 million meat eaters, 60,903 poultry eaters, 44,780 pescatarians, 81,165 vegetarians, and 14,167 vegans. Food intake differences between the diet groups varied across the cohorts; for example, fruit and vegetable intakes were generally higher in vegetarians than in meat eaters in all the cohorts except in China. BMI was generally lower in vegetarians, particularly vegans, except for the cohorts in India and China. In general, but with some exceptions, vegetarians were also more likely to be highly educated and physically active and less likely to smoke. In the available resurveys, stability of diet groups was high in all the cohorts except in China. Conclusions Food intakes and lifestyle factors of both non-vegetarians and vegetarians varied markedly across the individual cohorts, which may be due to differences in both culture and socioeconomic status, as well as differences in questionnaire design. Therefore, care is needed in the interpretation of the impacts of vegetarian diets on cancer risk.

Abstract Background Circulating total insulin-like growth factor-I (IGF-I) is an established risk factor for prostate cancer. However, only a small proportion of circulating IGF-I is free or readily dissociable from IGF-binding proteins (its bioavailable form), and few studies have investigated the association of circulating free IGF-I with prostate cancer risk. Methods We analyzed data from 767 prostate cancer cases and 767 matched controls nested within the European Prospective Investigation into Cancer and Nutrition cohort, with an average of 14-years (interquartile range = 2.9) follow-up. Matching variables were study center, length of follow-up, age, and time of day and fasting duration at blood collection. Circulating free IGF-I concentration was measured in serum samples collected at recruitment visit (mean age 55 years old; standard deviation = 7.1) using an enzyme-linked immunosorbent assay (ELISA). Conditional logistic regressions were performed to examine the associations of free IGF-I with risk of prostate cancer overall and subdivided by time to diagnosis (≤ 14 and > 14 years), and tumor characteristics. Results Circulating free IGF-I concentrations (in fourths and as a continuous variable) were not associated with prostate cancer risk overall (odds ratio [OR] = 1.00 per 0.1 nmol/L increment, 95% CI: 0.99, 1.02) or by time to diagnosis, or with prostate cancer subtypes, including tumor stage and histological grade. Conclusions Estimated circulating free IGF-I was not associated with prostate cancer risk. Further research may consider other assay methods that estimate bioavailable IGF-I to provide more insight into the well-substantiated association between circulating total IGF-I and subsequent prostate cancer risk.

Abstract Background The associations of adiposity with aggressive prostate cancer risk are unclear. Using two‐sample Mendelian randomization, we assessed the association of metabolically unfavourable adiposity (UFA), favourable adiposity (FA) and for comparison body mass index (BMI), with prostate cancer, including aggressive prostate cancer. Methods We examined the association of these genetically predicted adiposity‐related traits with risk of prostate cancer overall, aggressive and early onset disease using outcome summary statistics from the PRACTICAL consortium (including 15,167 aggressive cases). Results In inverse‐variance weighted models, there was little evidence that genetically predicted one standard deviation higher UFA, FA and BMI were associated with aggressive prostate cancer [OR: 0.85 (95% CI:0.61–1.19), 0.80 (0.53–1.23) and 0.97 (0.88–1.08), respectively]; these associations were largely consistent in sensitivity analyses accounting for horizontal pleiotropy. There was no strong evidence that genetically determined UFA, FA or BMI were associated with overall prostate cancer or early age of onset prostate cancer. Conclusions We did not find differences in the associations of UFA and FA with prostate cancer risk, which suggest that adiposity is unlikely to influence prostate cancer via the metabolic factors assessed; however, these did not cover some aspects related to metabolic health that may link obesity with aggressive prostate cancer, which should be explored in future studies.

Abstract Background The association between protein intake and prostate cancer risk remains unclear. Aims To prospectively investigate the associations of dietary intakes of total protein, protein from different dietary sources, and amino acids with prostate cancer risk and mortality. Methods In 131,425 men from the European Prospective Investigation into Cancer and Nutrition, protein and amino acid intakes were estimated using validated dietary questionnaires. Multivariable‐adjusted Cox regression models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Results During a mean follow‐up of 14.2 years, 6939 men were diagnosed with prostate cancer and 914 died of the disease. Dairy protein was positively associated with overall prostate cancer risk in the three highest fifths compared to the lowest (HRQ3=1.14 (95% CI 1.05–1.23); HRQ4=1.09 (1.01–1.18); HRQ5=1.10 (1.02–1.19)); similar results were observed for yogurt protein (HRQ3=1.14 (1.05–1.24); HRQ4=1.09 (1.01–1.18); HRQ5=1.12 (1.04–1.21)). For egg protein intake and prostate cancer mortality, no association was observed by fifths, but there was suggestive evidence of a positive association in the analysis per standard deviation increment. There was no strong evidence of associations with different tumour subtypes. Discussion Considering the weak associations and many tests, the results must be interpreted with caution. Conclusion This study does not provide strong evidence for an association of intakes of total protein, protein from different dietary sources or amino acids with prostate cancer risk or mortality. However, our results may suggest some weak positive associations, which need to be confirmed in large‐scale, pooled analyses of prospective data.

Abstract Background Recent studies have reported that the associations between dietary carbohydrates and cardiovascular disease (CVD) may depend on the quality, rather than the quantity, of carbohydrates consumed. This study aimed to assess the associations between types and sources of dietary carbohydrates and CVD incidence. A secondary aim was to examine the associations of carbohydrate intakes with triglycerides within lipoprotein subclasses. Methods A total of 110,497 UK Biobank participants with ≥ two (maximum five) 24-h dietary assessments who were free from CVD and diabetes at baseline were included. Multivariable-adjusted Cox regressions were used to estimate risks of incident total CVD (4188 cases), ischaemic heart disease (IHD; 3138) and stroke (1124) by carbohydrate intakes over a median follow-up time of 9.4 years, and the effect of modelled dietary substitutions. The associations of carbohydrate intakes with plasma triglycerides within lipoprotein subclasses as measured by nuclear magnetic resonance (NMR) spectroscopy were examined in 26,095 participants with baseline NMR spectroscopy measurements. Results Total carbohydrate intake was not associated with CVD outcomes. Free sugar intake was positively associated with total CVD (HR; 95% CI per 5% of energy, 1.07;1.03–1.10), IHD (1.06;1.02–1.10), and stroke (1.10;1.04–1.17). Fibre intake was inversely associated with total CVD (HR; 95% CI per 5 g/d, 0.96;0.93–0.99). Modelled isoenergetic substitution of 5% of energy from refined grain starch with wholegrain starch was inversely associated with total CVD (0.94;0.91–0.98) and IHD (0.94;0.90–0.98), and substitution of free sugars with non-free sugars was inversely associated with total CVD (0.95;0.92–0.98) and stroke (0.91;0.86–0.97). Free sugar intake was positively associated with triglycerides within all lipoproteins. Conclusions Higher free sugar intake was associated with higher CVD incidence and higher triglyceride concentrations within all lipoproteins. Higher fibre intake and replacement of refined grain starch and free sugars with wholegrain starch and non-free sugars, respectively, may be protective for incident CVD.

Abstract Background Previous studies of primarily Western populations have reported contrasting associations of dairy consumption with certain cancers, including a positive association with prostate cancer and inverse associations with colorectal and premenopausal breast cancers. However, there are limited data from China where cancer rates and levels of dairy consumption differ importantly from those in Western populations. Methods The prospective China Kadoorie Biobank study recruited ~0.5 million adults from ten diverse (five urban, five rural) areas across China during 2004–2008. Consumption frequency of major food groups, including dairy products, was collected at baseline and subsequent resurveys, using a validated interviewer-administered laptop-based food frequency questionnaire. To quantify the linear association of dairy intake and cancer risk and to account for regression dilution bias, the mean usual consumption amount for each baseline group was estimated via combining the consumption level at both baseline and the second resurvey. During a mean follow-up of 10.8 (SD 2.0) years, 29,277 incident cancer cases were recorded among the 510,146 participants who were free of cancer at baseline. Cox regression analyses for incident cancers associated with usual dairy intake were stratified by age-at-risk, sex and region and adjusted for cancer family history, education, income, alcohol intake, smoking, physical activity, soy and fresh fruit intake, and body mass index. Results Overall, 20.4% of participants reported consuming dairy products (mainly milk) regularly (i.e. ≥1 day/week), with the estimated mean consumption of 80.8 g/day among regular consumers and of 37.9 g/day among all participants. There were significant positive associations of dairy consumption with risks of total and certain site-specific cancers, with adjusted HRs per 50 g/day usual consumption being 1.07 (95% CI 1.04–1.10), 1.12 (1.02–1.22), 1.19 (1.01–1.41) and 1.17 (1.07–1.29) for total cancer, liver cancer (n = 3191), female breast cancer (n = 2582) and lymphoma (n=915), respectively. However, the association with lymphoma was not statistically significant after correcting for multiple testing. No significant associations were observed for colorectal cancer (n = 3350, 1.08 [1.00–1.17]) or other site-specific cancers. Conclusion Among Chinese adults who had relatively lower dairy consumption than Western populations, higher dairy intake was associated with higher risks of liver cancer, female breast cancer and, possibly, lymphoma.

Abstract Background The association of adiposity with prostate cancer specific mortality remains unclear. We examined how adiposity relates to fatal prostate cancer and described the cross-sectional associations of commonly used adiposity measurements with adiposity estimated by imaging in UK Biobank. We also conducted a dose-response meta-analysis to integrate the new data with existing prospective evidence. Methods 218,237 men from UK Biobank who were free from cancer at baseline were included. Body mass index (BMI), total body fat percentage (using bioimpedance), waist circumference (WC) and waist-to-hip ratio (WHR) were collected at recruitment. Risk of dying from prostate cancer (primary cause) by the different adiposity measurements was estimated using multivariable-adjusted Cox proportional hazards models. Results from this and other prospective cohort studies were combined in a dose-response meta-analysis. Results In UK Biobank, 661 men died from prostate cancer over a mean follow-up of 11.6 years. In the subsample of participants with magnetic resonance imaging and dual-energy X-ray absorptiometry, BMI, body fat percentage and WC were strongly associated with imaging estimates of total and central adiposity (e.g. visceral fat, trunk fat). The hazard ratios (HR) for prostate cancer death were 1.07 (95% confidence interval = 0.97–1.17) per 5 kg/m2 higher BMI, 1.00 (0.94–1.08) per 5% increase in total body fat percentage, 1.06 (0.99–1.14) per 10 cm increase in WC and 1.07 (1.01–1.14) per 0.05 increase in WHR. Our meta-analyses of prospective studies included 19,633 prostate cancer deaths for BMI, 670 for body fat percentage, 3181 for WC and 1639 for WHR, and the combined HRs for dying from prostate cancer for the increments above were 1.10 (1.07–1.12), 1.03 (0.96–1.11), 1.07 (1.03–1.11), and 1.06 (1.01–1.10), respectively. Conclusion Overall, we found that men with higher total and central adiposity had similarly higher risks of prostate cancer death, which may be biologically driven and/or due to differences in detection. In either case, these findings support the benefit for men of maintaining a healthy body weight.

Abstract Background Following a vegetarian diet has become increasingly popular and some evidence suggests that being vegetarian may be associated with a lower risk of cancer overall. However, for specific cancer sites, the evidence is limited. Our aim was to assess the associations of vegetarian and non-vegetarian diets with risks of all cancer, colorectal cancer, postmenopausal breast cancer, and prostate cancer and to explore the role of potential mediators between these associations. Methods We conducted a prospective analysis of 472,377 UK Biobank participants who were free from cancer at recruitment. Participants were categorised into regular meat-eaters (n = 247,571), low meat-eaters (n = 205,385), fish-eaters (n = 10,696), and vegetarians (n = 8685) based on dietary questions completed at recruitment. Multivariable-adjusted Cox regressions were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for all cancer incidence and separate cancer sites across diet groups. Results After an average follow-up of 11.4 years, 54,961 incident cancers were identified, including 5882 colorectal, 7537 postmenopausal breast, and 9501 prostate cancers. Compared with regular meat-eaters, being a low meat-eater, fish-eater, or vegetarian were all associated with a lower risk of all cancer (HR: 0.98, 95% CI: 0.96–1.00; 0.90, 0.84–0.96; 0.86, 0.80–0.93, respectively). Being a low meat-eater was associated with a lower risk of colorectal cancer in comparison to regular meat-eaters (0.91, 0.86–0.96); however, there was heterogeneity in this association by sex (p = 0.007), with an inverse association across diet groups in men, but not in women. Vegetarian postmenopausal women had a lower risk of breast cancer (0.82, 0.68–0.99), which was attenuated and non-significant after adjusting for body mass index (BMI; 0.87, 0.72–1.05); in mediation analyses, BMI was found to possibly mediate the observed association. In men, being a fish-eater or a vegetarian was associated with a lower risk of prostate cancer (0.80, 0.65–0.99 and 0.69, 0.54–0.89, respectively). Conclusion The lower risk of colorectal cancer in low meat-eaters is consistent with previous evidence suggesting an adverse impact of meat intake. The lower risk of postmenopausal breast cancer in vegetarian women may be explained by their lower BMI. It is not clear whether the other differences observed for all cancers and for prostate cancer reflect any causal relationships or are due to other factors such as residual confounding or differences in cancer detection.

Abstract Background There is limited prospective evidence on the association between meat consumption and many common, non-cancerous health outcomes. We examined associations of meat intake with risk of 25 common conditions (other than cancer). Methods We used data from 474,985 middle-aged adults recruited into the UK Biobank study between 2006 and 2010 and followed up until 2017 (mean follow-up 8.0 years) with available information on meat intake at baseline (collected via touchscreen questionnaire), and linked hospital admissions and mortality data. For a large sub-sample (~ 69,000), dietary intakes were re-measured three or more times using an online, 24-h recall questionnaire. Results On average, participants who reported consuming meat regularly (three or more times per week) had more adverse health behaviours and characteristics than participants who consumed meat less regularly, and most of the positive associations observed for meat consumption and health risks were substantially attenuated after adjustment for body mass index (BMI). In multi-variable adjusted (including BMI) Cox regression models corrected for multiple testing, higher consumption of unprocessed red and processed meat combined was associated with higher risks of ischaemic heart disease (hazard ratio (HRs) per 70 g/day higher intake 1.15, 95% confidence intervals (CIs) 1.07–1.23), pneumonia (1.31, 1.18–1.44), diverticular disease (1.19, 1.11–1.28), colon polyps (1.10, 1.06–1.15), and diabetes (1.30, 1.20–1.42); results were similar for unprocessed red meat and processed meat intakes separately. Higher consumption of unprocessed red meat alone was associated with a lower risk of iron deficiency anaemia (IDA: HR per 50 g/day higher intake 0.80, 95% CIs 0.72–0.90). Higher poultry meat intake was associated with higher risks of gastro-oesophageal reflux disease (HR per 30 g/day higher intake 1.17, 95% CIs 1.09–1.26), gastritis and duodenitis (1.12, 1.05–1.18), diverticular disease (1.10, 1.04–1.17), gallbladder disease (1.11, 1.04–1.19), and diabetes (1.14, 1.07–1.21), and a lower IDA risk (0.83, 0.76–0.90). Conclusions Higher unprocessed red meat, processed meat, and poultry meat consumption was associated with higher risks of several common conditions; higher BMI accounted for a substantial proportion of these increased risks suggesting that residual confounding or mediation by adiposity might account for some of these remaining associations. Higher unprocessed red meat and poultry meat consumption was associated with lower IDA risk.

Background There is controversy about associations between total dietary fatty acids, their classes (saturated fatty acids [SFAs], monounsaturated fatty acids, and polyunsaturated fatty acids), and risk of coronary heart disease (CHD). Specifically, the relevance of food sources of SFAs to CHD associations is uncertain. Methods and Results We conducted a case‐cohort study involving 10 529 incident CHD cases and a random subcohort of 16 730 adults selected from a cohort of 385 747 participants in 9 countries of the EPIC (European Prospective Investigation into Cancer and Nutrition) study. We estimated multivariable adjusted country‐specific hazard ratios (HRs) and 95% CIs per 5% of energy intake from dietary fatty acids, with and without isocaloric macronutrient substitutions, using Prentice‐weighted Cox regression models and pooled results using random‐effects meta‐analysis. We found no evidence for associations of the consumption of total or fatty acid classes with CHD, regardless of macronutrient substitutions. In analyses considering food sources, CHD incidence was lower per 1% higher energy intake of SFAs from yogurt (HR, 0.93 [95% CI, 0.88–0.99]), cheese (HR, 0.98 [95% CI, 0.96–1.00]), and fish (HR, 0.87 [95% CI, 0.75–1.00]), but higher for SFAs from red meat (HR, 1.07 [95% CI, 1.02–1.12]) and butter (HR, 1.02 [95% CI, 1.00–1.04]). Conclusions This observational study found no strong associations of total fatty acids, SFAs, monounsaturated fatty acids, and polyunsaturated fatty acids, with incident CHD. By contrast, we found associations of SFAs with CHD in opposite directions dependent on the food source. These findings should be further confirmed, but support public health recommendations to consider food sources alongside the macronutrients they contain, and suggest the importance of the overall food matrix.

Abstract Background Cancer Antigen 125 (CA125) is currently the best available ovarian cancer screening biomarker. However, CA125 has been limited by low sensitivity and specificity in part due to normal variation between individuals. Personal characteristics that influence CA125 could be used to improve its performance as screening biomarker. Methods We developed and validated linear and dichotomous (≥35 U/mL) circulating CA125 prediction models in postmenopausal women without ovarian cancer who participated in one of five large population-based studies: Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO, n = 26,981), European Prospective Investigation into Cancer and Nutrition (EPIC, n = 861), the Nurses’ Health Studies (NHS/NHSII, n = 81), and the New England Case Control Study (NEC, n = 923). The prediction models were developed using stepwise regression in PLCO and validated in EPIC, NHS/NHSII and NEC. Result The linear CA125 prediction model, which included age, race, body mass index (BMI), smoking status and duration, parity, hysterectomy, age at menopause, and duration of hormone therapy (HT), explained 5% of the total variance of CA125. The correlation between measured and predicted CA125 was comparable in PLCO testing dataset (r = 0.18) and external validation datasets (r = 0.14). The dichotomous CA125 prediction model included age, race, BMI, smoking status and duration, hysterectomy, time since menopause, and duration of HT with AUC of 0.64 in PLCO and 0.80 in validation dataset. Conclusions The linear prediction model explained a small portion of the total variability of CA125, suggesting the need to identify novel predictors of CA125. The dichotomous prediction model showed moderate discriminatory performance which validated well in independent dataset. Our dichotomous model could be valuable in identifying healthy women who may have elevated CA125 levels, which may contribute to reducing false positive tests using CA125 as screening biomarker.

Abstract Background Models that accurately predict risk of breast cancer are needed to help younger women make decisions about when to begin screening. Premenopausal concentrations of circulating anti-Müllerian hormone (AMH), a biomarker of ovarian reserve, and testosterone have been positively associated with breast cancer risk in prospective studies. We assessed whether adding AMH and/or testosterone to the Gail model improves its prediction performance for women aged 35–50. Methods In a nested case-control study including ten prospective cohorts (1762 invasive cases/1890 matched controls) with pre-diagnostic serum/plasma samples, we estimated relative risks (RR) for the biomarkers and Gail risk factors using conditional logistic regression and random-effects meta-analysis. Absolute risk models were developed using these RR estimates, attributable risk fractions calculated using the distributions of the risk factors in the cases from the consortium, and population-based incidence and mortality rates. The area under the receiver operating characteristic curve (AUC) was used to compare the discriminatory accuracy of the models with and without biomarkers. Results The AUC for invasive breast cancer including only the Gail risk factor variables was 55.3 (95% CI 53.4, 57.1). The AUC increased moderately with the addition of AMH (AUC 57.6, 95% CI 55.7, 59.5), testosterone (AUC 56.2, 95% CI 54.4, 58.1), or both (AUC 58.1, 95% CI 56.2, 59.9). The largest AUC improvement (4.0) was among women without a family history of breast cancer. Conclusions AMH and testosterone moderately increase the discriminatory accuracy of the Gail model among women aged 35–50. We observed the largest AUC increase for women without a family history of breast cancer, the group that would benefit most from improved risk prediction because early screening is already recommended for women with a family history.

Abstract Background Few published breast cancer (BC) risk prediction models consider the heterogeneity of predictor variables between estrogen-receptor positive (ER+) and negative (ER-) tumors. Using data from two large cohorts, we examined whether modeling this heterogeneity could improve prediction. Methods We built two models, for ER+ (ModelER+) and ER- tumors (ModelER-), respectively, in 281,330 women (51% postmenopausal at recruitment) from the European Prospective Investigation into Cancer and Nutrition cohort. Discrimination (C-statistic) and calibration (the agreement between predicted and observed tumor risks) were assessed both internally and externally in 82,319 postmenopausal women from the Women’s Health Initiative study. We performed decision curve analysis to compare ModelER+ and the Gail model (ModelGail) regarding their applicability in risk assessment for chemoprevention. Results Parity, number of full-term pregnancies, age at first full-term pregnancy and body height were only associated with ER+ tumors. Menopausal status, age at menarche and at menopause, hormone replacement therapy, postmenopausal body mass index, and alcohol intake were homogeneously associated with ER+ and ER- tumors. Internal validation yielded a C-statistic of 0.64 for ModelER+ and 0.59 for ModelER-. External validation reduced the C-statistic of ModelER+ (0.59) and ModelGail (0.57). In external evaluation of calibration, ModelER+ outperformed the ModelGail: the former led to a 9% overestimation of the risk of ER+ tumors, while the latter yielded a 22% underestimation of the overall BC risk. Compared with the treat-all strategy, ModelER+ produced equal or higher net benefits irrespective of the benefit-to-harm ratio of chemoprevention, while ModelGail did not produce higher net benefits unless the benefit-to-harm ratio was below 50. The clinical applicability, i.e. the area defined by the net benefit curve and the treat-all and treat-none strategies, was 12.7 × 10− 6 for ModelER+ and 3.0 × 10− 6 for ModelGail. Conclusions Modeling heterogeneous epidemiological risk factors might yield little improvement in BC risk prediction. Nevertheless, a model specifically predictive of ER+ tumor risk could be more applicable than an omnibus model in risk assessment for chemoprevention.

Abstract Background Accumulating evidence suggests that individual circulating saturated fatty acids (SFAs) are heterogeneous in their associations with cardio-metabolic diseases, but evidence about associations of SFAs with metabolic markers of different pathogenic pathways is limited. We aimed to examine the associations between plasma phospholipid SFAs and the metabolic markers of lipid, hepatic, glycaemic and inflammation pathways. Methods We measured nine individual plasma phospholipid SFAs and derived three SFA groups (odd-chain: C15:0 + C17:0, even-chain: C14:0 + C16:0 + C18:0, and very-long-chain: C20:0 + C22:0 + C23:0 + C24:0) in individuals from the subcohort of the European Prospective Investigation into Cancer and Nutrition (EPIC)-InterAct case-cohort study across eight European countries. Using linear regression in 15,919 subcohort members, adjusted for potential confounders and corrected for multiple testing, we examined cross-sectional associations of SFAs with 13 metabolic markers. Multiplicative interactions of the three SFA groups with pre-specified factors, including body mass index (BMI) and alcohol consumption, were tested. Results Higher levels of odd-chain SFA group were associated with lower levels of major lipids (total cholesterol (TC), triglycerides, apolipoprotein A-1 (ApoA1), apolipoprotein B (ApoB)) and hepatic markers (alanine transaminase (ALT), aspartate transaminase (AST), gamma-glutamyl transferase (GGT)). Higher even-chain SFA group levels were associated with higher levels of low-density lipoprotein cholesterol (LDL-C), TC/high-density lipoprotein cholesterol (HDL-C) ratio, triglycerides, ApoB, ApoB/A1 ratio, ALT, AST, GGT and CRP, and lower levels of HDL-C and ApoA1. Very-long-chain SFA group levels showed inverse associations with triglycerides, ApoA1 and GGT, and positive associations with TC, LDL-C, TC/HDL-C, ApoB and ApoB/A1. Associations were generally stronger at higher levels of BMI or alcohol consumption. Conclusions Subtypes of SFAs are associated in a differential way with metabolic markers of lipid metabolism, liver function and chronic inflammation, suggesting that odd-chain SFAs are associated with lower metabolic risk and even-chain SFAs with adverse metabolic risk, whereas mixed findings were obtained for very-long-chain SFAs. The clinical and biochemical implications of these findings may vary by adiposity and alcohol intake.

Abstract Background The relationship between body size and prostate cancer risk, and in particular risk by tumour characteristics, is not clear because most studies have not differentiated between high-grade or advanced stage tumours, but rather have assessed risk with a combined category of aggressive disease. We investigated the association of height and adiposity with incidence of and death from prostate cancer in 141,896 men in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. Methods Multivariable-adjusted Cox proportional hazards models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs). After an average of 13.9 years of follow-up, there were 7024 incident prostate cancers and 934 prostate cancer deaths. Results Height was not associated with total prostate cancer risk. Subgroup analyses showed heterogeneity in the association with height by tumour grade (P heterogeneity = 0.002), with a positive association with risk for high-grade but not low-intermediate-grade disease (HR for high-grade disease tallest versus shortest fifth of height, 1.54; 95% CI, 1.18–2.03). Greater height was also associated with a higher risk for prostate cancer death (HR = 1.43, 1.14–1.80). Body mass index (BMI) was significantly inversely associated with total prostate cancer, but there was evidence of heterogeneity by tumour grade (P heterogeneity = 0.01; HR = 0.89, 0.79–0.99 for low-intermediate grade and HR = 1.32, 1.01–1.72 for high-grade prostate cancer) and stage (P heterogeneity = 0.01; HR = 0.86, 0.75–0.99 for localised stage and HR = 1.11, 0.92–1.33 for advanced stage). BMI was positively associated with prostate cancer death (HR = 1.35, 1.09–1.68). The results for waist circumference were generally similar to those for BMI, but the associations were slightly stronger for high-grade (HR = 1.43, 1.07–1.92) and fatal prostate cancer (HR = 1.55, 1.23–1.96). Conclusions The findings from this large prospective study show that men who are taller and who have greater adiposity have an elevated risk of high-grade prostate cancer and prostate cancer death.

Abstract Background Little is known about how pre-diagnostic metabolites in blood relate to risk of prostate cancer. We aimed to investigate the prospective association between plasma metabolite concentrations and risk of prostate cancer overall, and by time to diagnosis and tumour characteristics, and risk of death from prostate cancer. Methods In a case-control study nested in the European Prospective Investigation into Cancer and Nutrition, pre-diagnostic plasma concentrations of 122 metabolites (including acylcarnitines, amino acids, biogenic amines, glycerophospholipids, hexose and sphingolipids) were measured using targeted mass spectrometry (AbsoluteIDQ p180 Kit) and compared between 1077 prostate cancer cases and 1077 matched controls. Risk of prostate cancer associated with metabolite concentrations was estimated by multi-variable conditional logistic regression, and multiple testing was accounted for by using a false discovery rate controlling procedure. Results Seven metabolite concentrations, i.e. acylcarnitine C18:1, amino acids citrulline and trans-4-hydroxyproline, glycerophospholipids PC aa C28:1, PC ae C30:0 and PC ae C30:2, and sphingolipid SM (OH) C14:1, were associated with prostate cancer (p

To detect modest associations of dietary intake with disease risk, observational studies need to be large and control for moderate measurement errors. The reproducibility of dietary intakes of macronutrients, food groups and dietary patterns (vegetarian and Mediterranean) was assessed in adults in the UK Biobank study on up to five occasions using a web-based 24-h dietary assessment (n 211 050), and using short FFQ recorded at baseline (n 502 655) and after 4 years (n 20 346). When the means of two 24-h assessments were used, the intra-class correlation coefficients (ICC) for macronutrients varied from 0·63 for alcohol to 0·36 for polyunsaturated fat. The ICC for food groups also varied from 0·68 for fruit to 0·18 for fish. The ICC for the FFQ varied from 0·66 for meat and fruit to 0·48 for bread and cereals. The reproducibility was higher for vegetarian status (κ > 0·80) than for the Mediterranean dietary pattern (ICC = 0·45). Overall, the reproducibility of pairs of 24-h dietary assessments and single FFQ used in the UK Biobank were comparable with results of previous prospective studies using conventional methods. Analyses of diet–disease relationships need to correct for both measurement error and within-person variability in dietary intake in order to reliably assess any such associations with disease in the UK Biobank.

Metabolomics may help to elucidate mechanisms underlying diet-disease relationships and identify novel risk factors for disease. To inform the design and interpretation of such research, evidence on diet-metabolite associations and cross-assay comparisons is needed. We aimed to compare nuclear magnetic resonance (NMR) metabolite profiles between meat-eaters, fish-eaters, vegetarians and vegans, and to compare NMR measurements to those from mass spectrometry (MS), clinical chemistry and capillary gas-liquid chromatography (GC). We quantified 207 serum NMR metabolite measures in 286 male participants of the European Prospective Investigation into Cancer and Nutrition (EPIC)-Oxford cohort. Using univariate and multivariate analyses, we found that metabolite profiles varied by diet group, especially for vegans; the main differences compared to meat-eaters were lower levels of docosahexaenoic acid, total n-3 and saturated fatty acids, cholesterol and triglycerides in very-low-density lipoproteins, various lipid factions in high-density lipoprotein, sphingomyelins, tyrosine and creatinine, and higher levels of linoleic acid, total n-6, polyunsaturated fatty acids and alanine. Levels in fish-eaters and vegetarians differed by metabolite measure. Concentrations of 13 metabolites measured using both NMR and MS, clinical chemistry or GC were mostly similar. In summary, vegans’ metabolite profiles were markedly different to those of men consuming animal products. The studied metabolomics platforms are complementary, with limited overlap between metabolite classes.

UK Biobank is an open access prospective cohort of 500 000 men and women. Information on the frequency of consumption of main foods was collected at recruitment with a touchscreen questionnaire; prior to examining the associations between diet and disease, it is essential to evaluate the performance of the dietary touchscreen questionnaire. The objectives of the present paper are to: describe the repeatability of the touchscreen questionnaire in participants (n 20 348) who repeated the assessment centre visit approximately 4 years after recruitment, and compare the dietary touchscreen variables with mean intakes from participants (n 140 080) who completed at least one of the four web-based 24-h dietary assessments post-recruitment. For fish and meat items, 90 % or more of participants reported the same or adjacent category of intake at the repeat assessment visit; for vegetables and fruit, and for a derived partial fibre score (in fifths), 70 % or more of participants were classified into the same or adjacent category of intake (κweighted > 0·50 for all). Participants were also categorised based on their responses to the dietary touchscreen questionnaire at recruitment, and within each category the group mean intake of the same food group or nutrient from participants who had completed at least one web-based 24-h dietary assessment was calculated. The comparison showed that the dietary touchscreen variables, available on the full cohort, reliably rank participants according to intakes of the main food groups.

Background Non-Hodgkin’s lymphomas are a heterogeneous group of neoplasms arising from the lymphopoietic system including a wide range of subtypes of either B-cell or T-cell lymphomas. The few established risk factors for the development of these neoplasms include viral infections and immunological abnormalities, but their etiology remains largely unknown. Evidence suggests that certain medical conditions may be linked, through immunosuppression, to the risk of non-Hodgkin’s lymphoma. Multiple myeloma is a neoplasm of plasma cells that accounts for approximately 15% of lymphopoietic cancers. Increases in the incidence of non-Hodgkin’s lymphoma and multiple myeloma in the past implicate environmental factors as potential causal agents.Design and Methods In the European Prospective Investigation into Cancer and Nutrition (EPIC), 1,213 histologically confirmed incident cases of non-Hodgkin’s lymphoma and multiple myeloma (594 men; 619 women) were identified during a follow-up of 8.5 years. Cox proportional hazard models were used to explore the association between self-reported diabetes, diagnosed after 30 years of age, and the risk of non-Hodgkin’s lymphoma overall and multiple myeloma and various lymphoma subtypes.Results We found no association between a personal history of diabetes and the risk of non-Hodgkin’s lymphoma overall in men (HR: 1.28, 95% CI: 0.89–1.84), in women (HR: 0.71, 95% CI: 0.41– 1.24), or in men and women combined (HR: 1.09, 95% CI: 0.80–1.47). Among the B-non-Hodgkin’s lymphoma subtypes, we observed a statistically significant increased risk of B-cell chronic lymphocytic leukemia (HR: 2.0, 95% CI: 1.04–3.86) in men, but not in women (HR: 1.07, 95% CI: 0.33–3.43).Conclusions This prospective study did not provide evidence for a role of self-reported diabetes in the etiology of non-Hodgkin’s lymphoma overall or multiple myeloma. We found an increased risk of B-cell chronic lymphocytic leukemia among men with diabetes, but not among women. We hypothesize that diabetes may not play a causal role in the etiology of B-cell chronic lymphocytic leukemia, though the underlying pathogenic mechanisms of both disorders may include shared genetic, host and/or environmental susceptibility factors.