Follow-up procedures in EPIC-Germany--data quality aspects. European Prospective Investigation into Cancer and Nutrition

Lipidome changes due to improved dietary fat quality inform cardiometabolic risk reduction and precision nutrition

Current cardiometabolic disease prevention guidelines recommend increasing dietary unsaturated fat intake while reducing saturated fats. Here we use lipidomics data from a randomized controlled dietary intervention trial to construct a multilipid score (MLS), summarizing the effects of replacing saturated fat with unsaturated fat on 45 lipid metabolite concentrations. In the EPIC-Potsdam cohort, a difference in the MLS, reflecting better dietary fat quality, was associated with a significant reduction in the incidence of cardiovascular disease (-32%; 95% confidence interval (95% CI): -21% to -42%) and type 2 diabetes (-26%; 95% CI: -15% to -35%). We built a closely correlated simplified score, reduced MLS (rMLS), and observed that beneficial rMLS changes, suggesting improved dietary fat quality over 10 years, were associated with lower diabetes risk (odds ratio per standard deviation of 0.76; 95% CI: 0.59 to 0.98) in the Nurses' Health Study. Furthermore, in the PREDIMED trial, an olive oil-rich Mediterranean diet intervention primarily reduced diabetes incidence among participants with unfavorable preintervention rMLS levels, suggestive of disturbed lipid metabolism before intervention. Our findings indicate that the effects of dietary fat quality on the lipidome can contribute to a more precise understanding and possible prediction of the health outcomes of specific dietary fat modifications.

Longitudinally changed diet quality scores and their association with type 2 diabetes mellitus and cardiovascular diseases in the EPIC-Potsdam study

Association analyses between longitudinal changes in diet quality scores (DQIs) and cardiometabolic risk remain scarce. Hence, we aimed to investigate how changes in two DQIs are associated with incident type 2 diabetes (T2D), myocardial infarction (MI) and stroke in the EPIC-Potsdam study. Changes in the Mediterranean Pyramid Score (MedPyr) and Healthy Diet Score (HDS) over 7 years from baseline (1994-1998) to follow-up 3 (2001-2005) were investigated in 23,548 middle-aged participants. Adjusted Cox Proportional Hazards Regression models were applied to investigate associations between changes in MedPyr and HDS and chronic disease incidence. More than 60% of the participants increased both DQIs more than 5%. Within a median follow-up time of 5 years 568 cases of T2D, 171 of MI, 189 of stroke were verified. An increased compared to stable MedPyr was associated with lower T2D risk (HR 0.74; 95% CI 0.59-0.92), while a decreased MedPyr was associated with higher stroke risk (HR 1.67; 95% CI 1.02-2.72). A decreased compared to stable HDS was associated with higher stroke risk (HR 1.80; 95% CI 1.02-3.20). The findings contribute further evidence on advantages of changing dietary intake towards a Mediterranean Diet. Although baseline HDS adherence was associated with T2D and stroke risk, longitudinal changes in HDS were only significantly associated with stroke risk.

Biomarker signatures associated with ageing free of major chronic diseases: results from a population-based sample of the EPIC-Potsdam cohort

BACKGROUND

A number of biomarkers denoting various pathophysiological pathways have been implicated in the aetiology and risk of age-related diseases. Hence, the combined impact of multiple biomarkers in relation to ageing free of major chronic diseases, such as cancer, cardiovascular disease and type 2 diabetes, has not been sufficiently explored.

METHODS

We measured concentrations of 13 biomarkers in a random subcohort of 2,500 participants in the European Prospective Investigation into Cancer and Nutrition Potsdam study. Chronic disease-free ageing was defined as reaching the age of 70 years within study follow-up without major chronic diseases, including cardiovascular disease, type 2 diabetes or cancer. Using a novel machine-learning technique, we aimed to identify biomarker clusters and explore their association with chronic disease-free ageing in multivariable-adjusted logistic regression analysis taking socio-demographic, lifestyle and anthropometric factors into account.

RESULTS

Of the participants who reached the age of 70 years, 321 met our criteria for chronic-disease free ageing. Machine learning analysis identified three distinct biomarker clusters, among which a signature characterised by high concentrations of high-density lipoprotein cholesterol, adiponectin and insulin-like growth factor-binding protein 2 and low concentrations of triglycerides was associated with highest odds for ageing free of major chronic diseases. After multivariable adjustment, the association was attenuated by socio-demographic, lifestyle and adiposity indicators, pointing to the relative importance of these factors as determinants of healthy ageing.

CONCLUSION

These data underline the importance of exploring combinations of biomarkers rather than single molecules in understanding complex biological pathways underpinning healthy ageing.

Pre-diagnostic circulating resistin concentrations and mortality among individuals with colorectal cancer: Results from the European Prospective Investigation into Cancer and Nutrition study

Resistin is a protein involved in inflammation and angiogenesis processes and may play a role in the progression of colorectal cancer (CRC). However, it remains unclear whether resistin is associated with increased mortality after CRC diagnosis. We examined pre-diagnostic serum resistin concentrations in relation to CRC-specific and all-cause mortality among 1343 incident CRC cases from the European Prospective Investigation into Cancer and Nutrition cohort. For CRC-specific mortality as the primary outcome, hazard ratios (HRs) and 95% confidence intervals (95% CI) were estimated from competing risk analyses based on cause-specific Cox proportional hazards models and further in sensitivity analyses using Fine-Gray proportional subdistribution hazards models. For all-cause mortality as the secondary outcome, Cox proportional hazards models were used. Subgroup analyses were performed by sex, tumor subsite, tumor stage, body mass index and time to CRC diagnosis. Resistin was measured on a median of 4.8 years before CRC diagnosis. During a median follow-up of 8.2 years, 474 deaths from CRC and 147 deaths from other causes were observed. Resistin concentrations were not associated with CRC-specific mortality (HRQ4vsQ1  = 0.95, 95% CI: 0.73-1.23; Ptrend  = .97; and HRper doubling of resistin concentration  = 1.00; 95% CI: 0.84-1.19; P = .98) or all-cause mortality. Results from competing risk (sensitivity) analysis were similar. No associations were found in any subgroup analyses. These findings suggest no association between pre-diagnostic circulating resistin concentrations and CRC-specific or all-cause mortality among persons with CRC, and the potential insignificance of resistin in CRC progression.

Chronic disease outcome metadata from German observational studies - public availability and FAIR principles

Metadata from epidemiological studies, including chronic disease outcome metadata (CDOM), are important to be findable to allow interpretability and reusability. We propose a comprehensive metadata schema and used it to assess public availability and findability of CDOM from German population-based observational studies participating in the consortium National Research Data Infrastructure for Personal Health Data (NFDI4Health). Additionally, principal investigators from the included studies completed a checklist evaluating consistency with FAIR principles (Findability, Accessibility, Interoperability, Reusability) within their studies. Overall, six of sixteen studies had complete publicly available CDOM. The most frequent CDOM source was scientific publications and the most frequently missing metadata were availability of codes of the International Classification of Diseases, Tenth Revision (ICD-10). Principal investigators' main perceived barriers for consistency with FAIR principles were limited human and financial resources. Our results reveal that CDOM from German population-based studies have incomplete availability and limited findability. There is a need to make CDOM publicly available in searchable platforms or metadata catalogues to improve their FAIRness, which requires human and financial resources.

Retinol and Retinol Binding Protein 4 Levels and Cardiometabolic Disease Risk

BACKGROUND

Despite mechanistic studies linking retinol and RBP4 (retinol binding protein 4) to the pathogenesis of cardiovascular diseases (CVD) and type 2 diabetes (T2D), epidemiological evidence is still conflicting. We investigated whether conflicting results of previous studies may be explained by differences in the association of retinol and RBP4 with cardiometabolic risk across subgroups with distinct sex, hypertension state, liver, or kidney function.

METHODS

We used case-cohorts nested in the EPIC (European Prospective Investigation Into Cancer and Nutrition)-Potsdam cohort (N=27 548) comprising a random sample of participants (n=2500) and all physician-verified cases of incident CVD (n=508, median follow-up time 8.2 years) and T2D (n=820, median follow-up time 6.3 years). We estimated nonlinear and linear multivariable-adjusted associations between the biomarkers and cardiometabolic diseases by restricted cubic splines and Cox regression, respectively, testing potential interactions with hypertension, liver, and kidney function. Additionally, we performed 2-sample Mendelian Randomization analyses in publicly available data.

RESULTS

The association of retinol with cardiometabolic risk was modified by hypertension state (P interaction CVD<0.001; P interaction T2D<0.001). Retinol was associated with lower cardiometabolic risk in participants with treated hypertension (hazard ratio_{per SD} [95% CI]: CVD, 0.71 [0.56-0.90]; T2D, 0.81 [0.70-0.94]) but with higher cardiometabolic risk in normotensive participants (CVD, 1.32 [1.06-1.64]; T2D, 1.15 [0.98-1.36]). Our analyses also indicated a significant interaction between RBP4 and hypertension on CVD risk (P interaction=0.04). Regarding T2D risk, we observed a u-shaped association with RBP4 in women (P nonlinearity=0.01, P effect=0.02) and no statistically significant association in men. The biomarkers' interactions with liver or kidney function were not statistically significant. Hypertension state-specific associations for retinol concentrations with cardiovascular mortality risk were replicated in National Health and Nutrition Examination Survey III.

CONCLUSIONS

Our findings suggest a hypertension-dependent relationship between plasma retinol and cardiometabolic risk and complex interactions of RBP4 with sex and hypertension on cardiometabolic risk.

Schulze M. German Diabetes Risk Score for the Determination of the Individual Type 2 Diabetes Risk

BACKGROUND

The German Diabetes Risk Score (GDRS) currently enables prediction of the individual risk of developing type 2 diabetes (T2D) within five years. The aim of this study is to extend the prediction period of the GDRS, including its non-clinical version and its HbA1c extension, to 10 years, and to perform external validation.

METHODS

In data from the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam study (n = 25 393), Cox proportional hazards regression was used to reweight the points that were used to calculate the five-year risk. Two population-based prospective cohorts (EPIC-Heidelberg n = 23 624, GNHIES98 cohort n = 3717) were used for external validation. Discrimination was represented by C-indices, and calibration by calibration plots and the expected-to-observed (E/O) ratio.

RESULTS

Prediction performance in EPIC-Potsdam was very good (C-index for the non-clinical model: 0.834) and was confirmed in EPIC-Heidelberg (0.843) and in the GNHIES98 cohort (0.851). Among persons in the GNHIES98 cohort with a greater than 10% predicted probability of disease, 14.9% developed T2D within 10 years (positive predictive value). The models were very well calibrated in EPIC-Potsdam (E/O ratio for the non-clinical model: 1.08), slightly overestimated the risk in EPIC-Heidelberg (1.34), and predicted T2D very well in the GNHIES98 cohort after recalibration (1.06).

CONCLUSION

The extended GDRS prediction period of 10 years, with a non-clinical version and an HbA1c extension that will soon be available in both German and English, enables the even longer-range, evidence-based identification of high-risk individuals with many different applications, including medical screening.

Dihydroceramide- and ceramide-profiling provides insights into human cardiometabolic disease etiology

Metabolic alterations precede cardiometabolic disease onset. Here we present ceramide- and dihydroceramide-profiling data from a nested case-cohort (type 2 diabetes [T2D, n = 775]; cardiovascular disease [CVD, n = 551]; random subcohort [n = 1137]) in the prospective EPIC-Potsdam study. We apply the novel NetCoupler-algorithm to link a data-driven (dihydro)ceramide network to T2D and CVD risk. Controlling for confounding by other (dihydro)ceramides, ceramides C18:0 and C22:0 and dihydroceramides C20:0 and C22:2 are associated with higher and ceramide C20:0 and dihydroceramide C26:1 with lower T2D risk. Ceramide C16:0 and dihydroceramide C22:2 are associated with higher CVD risk. Genome-wide association studies and Mendelian randomization analyses support a role of ceramide C22:0 in T2D etiology. Our results also suggest that (dh)ceramides partly mediate the putative adverse effect of high red meat consumption and benefits of coffee consumption on T2D risk. Thus, (dihydro)ceramides may play a critical role in linking genetic predisposition and dietary habits to cardiometabolic disease risk.

Endogenous Circulating Sex Hormone Concentrations and Colon Cancer Risk in Postmenopausal Women: A Prospective Study and Meta-Analysis

Background

Observational studies have consistently reported that postmenopausal hormone therapy use is associated with lower colon cancer risk, but epidemiologic studies examining the associations between circulating concentrations of endogenous estrogens and colorectal cancer have reported inconsistent results.

Methods

We investigated the associations between circulating concentrations of estrone, estradiol, free estradiol, testosterone, free testosterone, androstenedione, dehydroepiandrosterone (DHEA), progesterone, and sex hormone-binding globulin (SHBG) with colon cancer risk in a nested case-control study of 1028 postmenopausal European women (512 colon cancer cases, 516 matched controls) who were noncurrent users of exogenous hormones at blood collection. Multivariable conditional logistic regression models were used to compute odds ratios and 95% confidence intervals to evaluate the association between circulating sex hormones and colon cancer risk. We also conducted a dose-response meta-analysis of prospective studies of circulating estrone and estradiol with colorectal, colon, and rectal cancer risk in postmenopausal women. All statistical tests were 2-sided.

Results

In the multivariable model, a nonstatistically significantly positive relationship was found between circulating estrone and colon cancer risk (odds ratio per log2 1-unit increment = 1.17 [95% confidence interval = 1.00 to 1.38]; odds ratioquartile4-quartile1  = 1.33 [95% confidence interval = 0.89 to 1.97], P trend  = .20). Circulating concentrations of estradiol, free estradiol, testosterone, free testosterone, androstenedione, DHEA, progesterone, and SHBG were not associated with colon cancer risk. In the dose-response meta-analysis, no clear evidence of associations were found between circulating estradiol and estrone concentrations with colorectal, colon, and rectal cancer risk.

Conclusion

Our observational and meta-analysis results do not support an association between circulating concentrations of endogenous sex hormones and colon or rectal cancer in postmenopausal women.

A newly developed and externally validated non-clinical score accurately predicts 10-year cardiovascular disease risk in the general adult population

Inclusion of clinical parameters limits the application of most cardiovascular disease (CVD) prediction models to clinical settings. We developed and externally validated a non-clinical CVD risk score with a clinical extension and compared the performance to established CVD risk scores. We derived the scores predicting CVD (non-fatal and fatal myocardial infarction and stroke) in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam cohort (n = 25,992, cases = 683) using competing risk models and externally validated in EPIC-Heidelberg (n = 23,529, cases = 692). Performance was assessed by C-indices, calibration plots, and expected-to-observed ratios and compared to a non-clinical model, the Pooled Cohort Equation, Framingham CVD Risk Scores (FRS), PROCAM scores, and the Systematic Coronary Risk Evaluation (SCORE). Our non-clinical score included age, gender, waist circumference, smoking, hypertension, type 2 diabetes, CVD family history, and dietary parameters. C-indices consistently indicated good discrimination (EPIC-Potsdam 0.786, EPIC-Heidelberg 0.762) comparable to established clinical scores (thereof highest, FRS: EPIC-Potsdam 0.781, EPIC-Heidelberg 0.764). Additional clinical parameters slightly improved discrimination (EPIC-Potsdam 0.796, EPIC-Heidelberg 0.769). Calibration plots indicated very good calibration with minor overestimation in the highest decile of predicted risk. The developed non-clinical 10-year CVD risk score shows comparable discrimination to established clinical scores, allowing assessment of individual CVD risk in physician-independent settings.

Metabolic signatures of greater body size and their associations with risk of colorectal and endometrial cancers in the European Prospective Investigation into Cancer and Nutrition

BACKGROUND

The mechanisms underlying the obesity-cancer relationship are incompletely understood. This study aimed to characterise metabolic signatures of greater body size and to investigate their association with two obesity-related malignancies, endometrial and colorectal cancers, and with weight loss within the context of an intervention study.

METHODS

Targeted mass spectrometry metabolomics data from 4326 participants enrolled in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort and 17 individuals from a single-arm pilot weight loss intervention (Intercept) were used in this analysis. Metabolic signatures of body size were first determined in discovery (N = 3029) and replication (N = 1297) sets among EPIC participants by testing the associations between 129 metabolites and body mass index (BMI), waist circumference (WC), and waist-to-hip ratio (WHR) using linear regression models followed by partial least squares analyses. Conditional logistic regression models assessed the associations between the metabolic signatures with endometrial (N = 635 cases and 648 controls) and colorectal (N = 423 cases and 423 controls) cancer risk using nested case-control studies in EPIC. Pearson correlation between changes in the metabolic signatures and weight loss was tested among Intercept participants.

RESULTS

After adjustment for multiple comparisons, greater BMI, WC, and WHR were associated with higher levels of valine, isoleucine, glutamate, PC aa C38:3, and PC aa C38:4 and with lower levels of asparagine, glutamine, glycine, serine, lysoPC C17:0, lysoPC C18:1, lysoPC C18:2, PC aa C42:0, PC ae C34:3, PC ae C40:5, and PC ae C42:5. The metabolic signature of BMI (OR1-sd 1.50, 95% CI 1.30-1.74), WC (OR1-sd 1.46, 95% CI 1.27-1.69), and WHR (OR1-sd 1.54, 95% CI 1.33-1.79) were each associated with endometrial cancer risk. Risk of colorectal cancer was positively associated with the metabolic signature of WHR (OR1-sd: 1.26, 95% CI 1.07-1.49). In the Intercept study, a positive correlation was observed between weight loss and changes in the metabolic signatures of BMI (r = 0.5, 95% CI 0.06-0.94, p = 0.03), WC (r = 0.5, 95% CI 0.05-0.94, p = 0.03), and WHR (r = 0.6, 95% CI 0.32-0.87, p = 0.01).

CONCLUSIONS

Obesity is associated with a distinct metabolic signature comprising changes in levels of specific amino acids and lipids which is positively associated with both colorectal and endometrial cancer and is potentially reversible following weight loss.

BMI and BMI change following incident type 2 diabetes and risk of microvascular and macrovascular complications: the EPIC-Potsdam study

AIMS/HYPOTHESIS

Studies suggest decreased mortality risk among people who are overweight or obese compared with individuals with normal weight in type 2 diabetes (obesity paradox). However, the relationship between body weight or weight change and microvascular vs macrovascular complications of type 2 diabetes remains unresolved. We investigated the association between BMI and BMI change with long-term risk of microvascular and macrovascular complications in type 2 diabetes in a prospective cohort study.

METHODS

We studied participants with incident type 2 diabetes from the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam cohort, who were free of cancer, cardiovascular disease and microvascular disease at diagnosis (n = 1083). Pre-diagnosis BMI and relative annual change between pre- and post-diagnosis BMI were evaluated in multivariable-adjusted Cox models.

RESULTS

There were 85 macrovascular (myocardial infarction and stroke) and 347 microvascular events (kidney disease, neuropathy and retinopathy) over a median follow-up of 10.8 years. Median pre-diagnosis BMI was 29.9 kg/m² (IQR 27.4-33.2), and the median relative annual BMI change was -0.4% (IQR -2.1 to 0.9). Higher pre-diagnosis BMI was positively associated with total microvascular complications (multivariable-adjusted HR per 5 kg/m² [95% CI]: 1.21 [1.07, 1.36], kidney disease 1.39 [1.21, 1.60] and neuropathy 1.12 [0.96, 1.31]) but not with macrovascular complications (HR 1.05 [95% CI 0.81, 1.36]). Analyses according to BMI categories corroborated these findings. Effect modification was not evident by sex, smoking status or age groups. In analyses according to BMI change categories, BMI loss of more than 1% indicated a decreased risk of total microvascular complications (HR 0.62 [95% CI 0.47, 0.80]), kidney disease (HR 0.57 [95% CI 0.40, 0.81]) and neuropathy (HR 0.73 [95% CI 0.52, 1.03]), compared with participants with a stable BMI; no clear association was observed for macrovascular complications (HR 1.04 [95% CI 0.62, 1.74]). The associations between BMI gain compared with stable BMI and diabetes-related vascular complications were less apparent. Associations were consistent across strata of sex, age, pre-diagnosis BMI or medication but appeared to be stronger among never-smokers compared with current or former smokers.

CONCLUSIONS/INTERPRETATION

Among people with incident type 2 diabetes, pre-diagnosis BMI was positively associated with microvascular complications, while a reduced risk was observed with weight loss when compared with stable weight. The relationships with macrovascular disease were less clear.

Association of familial history of diabetes or myocardial infarction and stroke with risk of cardiovascular diseases in four German cohorts

Since family history of diabetes is a very strong risk factor for type 2 diabetes, which is one of the most important risk factors for cardiovascular disease (CVD), it might be also useful to assess the risk for CVD. Therefore, we aimed to investigate the relationship between a familial (parents and siblings) history of diabetes and the risk of incident CVD. Data from four prospective German cohort studies were used: EPIC-Potsdam study (n = 26,054), CARLA study (n = 1,079), SHIP study (n = 3,974), and KORA study (n = 15,777). A multivariable-adjusted Cox regression was performed to estimate associations between familial histories of diabetes, myocardial infarction or stroke and the risk of CVD in each cohort; combined hazard ratios (HR_Meta) were derived by conducting a meta-analysis. The history of diabetes in first-degree relatives was not related to the development of CVD (HR_Meta 0.99; 95% CI 0.88–1.10). Results were similar for the single outcomes myocardial infarction (MI) (HR_Meta 1.07; 95% CI 0.92–1.23) and stroke (HR_Meta 1.00; 95% CI 0.86–1.16). In contrast, parental history of MI and stroke were associated with an increased CVD risk. Our study indicates that diabetes in the family might not be a relevant risk factor for the incidence of CVD. However, the study confirmed the relationship between a parental history of MI or stroke and the onset of CVD.

Predicted basal metabolic rate and cancer risk in the European Prospective Investigation into Cancer and Nutrition

Emerging evidence suggests that a metabolic profile associated with obesity may be a more relevant risk factor for some cancers than adiposity per se. Basal metabolic rate (BMR) is an indicator of overall body metabolism and may be a proxy for the impact of a specific metabolic profile on cancer risk. Therefore, we investigated the association of predicted BMR with incidence of 13 obesity-related cancers in the European Prospective Investigation into Cancer and Nutrition (EPIC). BMR at baseline was calculated using the WHO/FAO/UNU equations and the relationships between BMR and cancer risk were investigated using multivariable Cox proportional hazards regression models. A total of 141,295 men and 317,613 women, with a mean follow-up of 14 years were included in the analysis. Overall, higher BMR was associated with a greater risk for most cancers that have been linked with obesity. However, among normal weight participants, higher BMR was associated with elevated risks of esophageal adenocarcinoma (hazard ratio per 1-standard deviation change in BMR [HR1-SD ]: 2.46; 95% CI 1.20; 5.03) and distal colon cancer (HR1-SD : 1.33; 95% CI 1.001; 1.77) among men and with proximal colon (HR1-SD : 1.16; 95% CI 1.01; 1.35), pancreatic (HR1-SD : 1.37; 95% CI 1.13; 1.66), thyroid (HR1-SD : 1.65; 95% CI 1.33; 2.05), postmenopausal breast (HR1-SD : 1.17; 95% CI 1.11; 1.22) and endometrial (HR1-SD : 1.20; 95% CI 1.03; 1.40) cancers in women. These results indicate that higher BMR may be an indicator of a metabolic phenotype associated with risk of certain cancer types, and may be a useful predictor of cancer risk independent of body fatness.

Vascular injury biomarkers and stroke risk

Objective

Because little is known about associations between biomarkers of vascular injury and stroke risk, we evaluated associations between plasma concentrations of 6 novel biomarkers of vascular injury and stroke risk in a population-based study.

Methods

A case–cohort subset of EPIC-Heidelberg (European Prospective Investigation for Cancer and Nutrition–Heidelberg) including incident stroke cases (n = 335) and a random subcohort (n = 2,418) was selected. Concentrations of intercellular adhesion molecule 3 (ICAM3), soluble E-selectin and P-selectin, soluble thrombomodulin (sTM), thrombopoietin, and glycoprotein IIb/IIIa were measured in baseline plasma samples. Weighted Cox regression analyses were used to assess associations between biomarker levels and stroke risk.

Results

Median follow-up in the subcohort and among cases was 9.8 (range, 0.1–12.5) years and 6.2 (range, 0.01–12.1) years, respectively. ICAM3 levels were associated with increased risk of incident stroke after multivariable adjustment (hazard ratio, highest vs lowest quartile: 1.64 [95% confidence interval, 1.15–2.32]; plinear trend < 0.001). This association was more apparent for ischemic (1.65 [1.12–2.45]; plinear trend < 0.01) than for hemorrhagic stroke (1.29 [0.60–2.78]; plinear trend = 0.3). We further observed a borderline significant trend for a positive association between sTM and overall stroke risk (1.47 [0.99–2.19]; plinear trend = 0.05).

Conclusions

In this population-based study, circulating levels of ICAM3, an adhesion molecule shed by leukocytes, were associated with increased risk of incident stroke. Further mechanistic studies are needed to elucidate the pathophysiology underlying this association.

Classification of evidence

This study provides Class II evidence that plasma levels of ICAM3 are associated with increased stroke risk.

Plasma N-Glycans as Emerging Biomarkers of Cardiometabolic Risk: A Prospective Investigation in the EPIC-Potsdam Cohort Study

OBJECTIVE

Plasma protein N-glycan profiling integrates information on enzymatic protein glycosylation, which is a highly controlled ubiquitous posttranslational modification. Here we investigate the ability of the plasma N-glycome to predict incidence of type 2 diabetes and cardiovascular diseases (CVDs; i.e., myocardial infarction and stroke).

RESEARCH DESIGN AND METHODS

Based on the prospective European Prospective Investigation of Cancer (EPIC)-Potsdam cohort (n = 27,548), we constructed case-cohorts including a random subsample of 2,500 participants and all physician-verified incident cases of type 2 diabetes (n = 820; median follow-up time 6.5 years) and CVD (n = 508; median follow-up time 8.2 years). Information on the relative abundance of 39 N-glycan groups in baseline plasma samples was generated by chromatographic profiling. We selected predictive N-glycans for type 2 diabetes and CVD separately, based on cross-validated machine learning, nonlinear model building, and construction of weighted prediction scores. This workflow for CVD was applied separately in men and women.

RESULTS

The N-glycan-based type 2 diabetes score was strongly predictive for diabetes risk in an internal validation cohort (weighted C-index 0.83, 95% CI 0.78-0.88), and this finding was externally validated in the Finland Cardiovascular Risk Study (FINRISK) cohort. N-glycans were moderately predictive for CVD incidence (weighted C-indices 0.66, 95% CI 0.60-0.72, for men; 0.64, 95% CI 0.55-0.73, for women). Information on the selected N-glycans improved the accuracy of established and clinically applied risk prediction scores for type 2 diabetes and CVD.

CONCLUSIONS

Selected N-glycans improve type 2 diabetes and CVD prediction beyond established risk markers. Plasma protein N-glycan profiling may thus be useful for risk stratification in the context of precisely targeted primary prevention of cardiometabolic diseases.

Circulating Immune Cell Composition and Cancer Risk: A Prospective Study Using Epigenetic Cell Count Measures

Although ample evidence indicates that immune cell homeostasis is an important prognostic outcome determinant in patients with cancer, few studies have examined whether it also determines cancer risk among initially healthy individuals. We performed a case-cohort study including incident cases of breast (n = 207), colorectal (n = 111), lung (n = 70), and prostate (n = 201) cancer as well as a subcohort (n = 465) within the European Prospective Investigation into Cancer and Nutrition-Heidelberg cohort. Relative counts of neutrophils, monocytes, and lymphocyte sublineages were measured by qRT-PCR. HRs and 95% confidence intervals were used to measure the associations between relative counts of immune cell and cancer risks. When relative counts of immune cell types were taken individually, a significant positive association was observed between relative counts of FOXP3⁺ regulatory T cells (Tregs) and lung cancer risk, and significant inverse associations were observed between relative CD8⁺ counts and risks of lung and breast cancer (overall and ER+ subtype). Multivariable models with mutual adjustments across immune markers showed further significant positive associations between higher relative FOXP3⁺ T-cell counts and increased risks of colorectal and breast cancer (overall and ER- subtype). No associations were found between immune cell composition and prostate cancer risk. These results affirm the relevance of elevated FOXP3⁺ Tregs and lower levels of cytotoxic (CD8⁺) T cells as risk factors for tumor development. SIGNIFICANCE: This epidemiologic study supports a role for both regulatory and cytotoxic T cells in determining cancer risk among healthy individuals.See related commentary by Song and Tworoger, p. 1801.

Mediation analysis of the alcohol-postmenopausal breast cancer relationship by sex hormones in the EPIC cohort

Alcohol consumption is associated with higher risk of breast cancer (BC); however, the biological mechanisms underlying this association are not fully elucidated, particularly the extent to which this relationship is mediated by sex hormone levels. Circulating concentrations of estradiol, testosterone, their free fractions and sex-hormone binding globulin (SHBG), were examined in 430 incident BC cases and 645 matched controls among alcohol-consuming postmenopausal women nested within the European Prospective Investigation into Cancer and Nutrition. Mediation analysis was applied to assess whether individual hormone levels mediated the relationship between alcohol intake and BC risk. An alcohol-related hormonal signature, obtained by partial least square (PLS) regression, was evaluated as a potential mediator. Total (TE), natural direct and natural indirect effects (NIE) were estimated. Alcohol intake was positively associated with overall BC risk and specifically with estrogen receptor-positive tumors with respectively TE = 1.17(95%CI: 1.01,1.35) and 1.36(1.08,1.70) for a 1-standard deviation (1-SD) increase of intake. There was no evidence of mediation by sex steroids or SHBG separately except for a weak indirect effect through free estradiol where NIE = 1.03(1.00,1.06). However, an alcohol-related hormonal signature negatively associated with SHBG and positively with estradiol and testosterone was associated with BC risk (odds ratio [OR] = 1.25 [1.07,1.47]) for a 1-SD higher PLS score, and had a statistically significant NIE accounting for a mediated proportion of 24%. There was limited evidence of mediation of the alcohol-BC association by individual sex hormones. However, a hormonal signature, reflecting lower levels of SHBG and higher levels of sex steroids, mediated a substantial proportion of the association.

Dietary Patterns Are Associated with Serum Metabolite Patterns and Their Association Is Influenced by Gut Bacteria among Older German Adults

BACKGROUND

Although dietary intakes and dietary intake patterns (DPs) have been associated with single metabolites, it is unclear whether DPs are also reflected in specific metabolite patterns (MPs). Moreover, the influence of groups of gut bacteria on the relationship between DPs and MPs is underexplored.

OBJECTIVES

We aimed to investigate the association of DPs and serum MPs and also the modifying effect of the gut bacteria compositional patterns (BCPs).

METHODS

This is a cross-sectional investigation among 225 individuals (median age: 63 y; 53% women) from the European Prospective Investigation into Cancer and Nutrition study. Dietary intakes were assessed by three 24-h dietary recalls, gut bacteria composition was quantified by 16S rRNA gene sequencing, and the serum metabolome was profiled by an untargeted approach. We identified DPs and BCPs by the treelet transform analysis. We modeled associations between DPs and 8 previously published MPs and the modifying effect of BCPs by fitting generalized linear models using DataSHIELD R.

RESULTS

We identified 5 DPs and 7 BCPs. The "bread, margarine, and processed meat" and "fruiting vegetables and vegetable oils" DPs were positively associated with the "amino acids" (β = 0.35; 95% CI: 0.02, 0.69; P = 0.03) and "fatty acids" MPs (β = 0.45; 95% CI: 0.16, 0.74; P = 0.01), respectively. The "tea and miscellaneous" was inversely associated with the "amino acids" (β = -0.28; 95% CI: -0.52, -0.05; P = 0.02) and "amino acid derivatives" MPs (β = -0.21; 95% CI: -0.39, -0.02; P = 0.03). One BCP negatively modified the association between the "bread, margarine, and processed meat" DP and the "amino acids" MP (P-interaction = 0.01).

CONCLUSIONS

In older German adults, DPs are reflected in MPs, and the gut bacteria attenuate 1 DP-MP association. These MPs should be explored as biomarkers of these jointly consumed foods while taking into account a potentially modifying role of the gut bacteria.

Biomarkers of Vascular Injury and Type 2 Diabetes: A Prospective Study, Systematic Review and Meta-Analysis

Data on biomarkers of vascular injury and type 2 diabetes (T2D) risk from prospective studies are lacking. We evaluated seven biomarkers of vascular injury in relation to T2D. Additionally, a meta-analysis was performed. From the EPIC–Heidelberg cohort, 2224 participants were followed-up from baseline for 16 (median) years. E-Selectin, P-Selectin, intercellular adhesion molecule 3 (ICAM3), thrombomodulin, thrombopoietin, glycoprotein IIb/IIIa and fibrinogen levels were measured in baseline blood samples. The systematic review and meta-analysis included prospective studies identified through MEDLINE and Web of Science that investigated the association between mentioned biomarkers and T2D. The study population included 55% women, median age was 50 years, and 163 developed T2D. ICAM3 was associated with lower T2D risk (fully adjusted HR_{highest vs. lowest} tertile 0.62 (95% CI: 0.43, 0.91)), but no other studies on ICAM3 were identified. Overall, fifteen studies were included in the systematic review and meta-analysis (6,171 cases). E-Selectin was associated with higher T2D risk HR_{per SD}: 1.34 (95% CI: 1.16, 1.54; I² = 63%, n = 9 studies), while thrombomodulin was associated with lower risk HR_{per SD}: 0.82 (95% CI: 0.71, 0.95; I² = 0%, n = 2 studies). In the EPIC–Heidelberg, ICAM3 was associated with lower T2D risk. The meta-analysis showed a consistent positive association between E-Selectin and T2D. It was also suggestive of an inverse association between thrombomodulin and T2D, although further studies are needed to corroborate this finding.

A cross-sectional investigation into the association between Porphyromonas gingivalis and autoantibodies to citrullinated proteins in a German population

Background

Porphyromonas gingivalis (P.g) is unique among pathogens due to its ability to generate citrullinated proteins in an inflammatory milieu, potentially mediating the loss of immune tolerance, the production of anticitrullinated protein antibodies (ACPAs), and subsequently the development of rheumatoid arthritis (RA). Based on this hypothesis, we set out to investigate whether P.g is linked to ACPAs in a well-characterized German population.

Participants and methods

A total of 600 participants (292 women and 308 men with a mean age of 67 years) of the European Prospective Investigation into Cancer and Nutrition-Potsdam study were selected in 2013, and paired saliva and serum samples were collected. Salivary P.g DNA and serum anticyclic citrullinated peptide (anti-CCP2) levels were quantified by real-time polymerase chain reaction and anti-CCP2 enzyme-linked immunosorbent assay, respectively. In selected participants, additional ACPA fine-specificities were also analysed on a custom-made multiplex peptide array.

Results

Among participants with C-reactive protein greater than 3.0 mg/l, a one-unit increase in P.g DNA was associated with an almost twofold increase in anti-CCP2 levels. Moreover, participants with high P.g DNA had on average approximately 2.8-times higher anti-CCP2 levels when compared with participants with low P.g DNA, (Holm-adjusted p value = 0.01). Furthermore, citrullinated epitopes on α-enolase and vimentin were common ACPA reactivities among participants who also had high P.g DNA and elevated C-reactive protein.

Conclusions

Our study suggests that in specific subgroups of individuals with systemic inflammation, higher salivary P.g DNA is associated with elevated serum ACPA. These data support a role for P.g in the development of anticitrulline immunity.

[Assessing incident cardiovascular and metabolic diseases in epidemiological cohort studies in Germany]

BACKGROUND

Cardiovascular and metabolic diseases are a major cause of mortality and loss of quality of life in Germany. Research into risk factors of these diseases requires large population-based cohort studies. Complete and accurate assessment of the incidence of cardiovascular and metabolic diseases is a key element for valid interpretation of the results from such studies.

OBJECTIVE

Our aim was to identify population-based cohort studies with incidence of cardiovascular and metabolic diseases in Germany and to summarize their methods for assessment and classification of disease endpoints, including myocardial infarction, type 2 diabetes, stroke, heart failure, and arterial hypertension.

METHODS

Within the framework of a workshop, representatives of the ascertained population-based cohort studies in Germany with incidence of cardiovascular or metabolic diseases were invited to present and to systematically provide information on their methods of endpoint identification.

RESULTS

We identified eight studies from different regions in Germany with a total of 100,571 participants, aged 18-83 years at baseline. Self-reporting by study participants is the major source for further inquiries to assess disease endpoints in these studies. Most studies use additional data sources to verify the incidence of diseases, such as documents provided by the treating physician or hospital.

CONCLUSIONS

Our results highlight the central role of self-reporting and the efforts associated with identification and verification of disease endpoints in cohort studies. They also provide a basis for future population-based studies that aim for standardized assessment of the incidence of cardiovascular and metabolic diseases.

Comparison of metabolite networks from four German population-based studies

Background

Metabolite networks are suggested to reflect biological pathways in health and disease. However, it is unknown whether such metabolite networks are reproducible across different populations. Therefore, the current study aimed to investigate similarity of metabolite networks in four German population-based studies.

Methods

One hundred serum metabolites were quantified in European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam (n = 2458), EPIC-Heidelberg (n = 812), KORA (Cooperative Health Research in the Augsburg Region) (n = 3029) and CARLA (Cardiovascular Disease, Living and Ageing in Halle) (n = 1427) with targeted metabolomics. In a cross-sectional analysis, Gaussian graphical models were used to construct similar networks of 100 edges each, based on partial correlations of these metabolites. The four metabolite networks of the top 100 edges were compared based on (i) common features, i.e. number of common edges, Pearson correlation (r) and hamming distance (h); and (ii) meta-analysis of the four networks.

Results

Among the four networks, 57 common edges and 66 common nodes (metabolites) were identified. Pairwise network comparisons showed moderate to high similarity (r = 63–0.96, h = 7–72), among the networks. Meta-analysis of the networks showed that, among the 100 edges and 89 nodes of the meta-analytic network, 57 edges and 66 metabolites were present in all the four networks, 58–76 edges and 75–89 nodes were present in at least three networks, and 63–84 edges and 76–87 edges were present in at least two networks. The meta-analytic network showed clear grouping of 10 sphingolipids, 8 lyso-phosphatidylcholines, 31 acyl-alkyl-phosphatidylcholines, 30 diacyl-phosphatidylcholines, 8 amino acids and 2 acylcarnitines.

Conclusions

We found structural similarity in metabolite networks from four large studies. Using a meta-analytic network, as a new approach for combining metabolite data from different studies, closely related metabolites could be identified, for some of which the biological relationships in metabolic pathways have been previously described. They are candidates for further investigation to explore their potential role in biological processes.

Meal analysis for understanding eating behavior: meal- and participant-specific predictors for the variance in energy and macronutrient intake

BACKGROUND

Meals differ in their nutritional content. This variation has not been fully addressed despite its potential contribution in understanding eating behavior. The aim of this study was to investigate the between-meal and between-individual variance in energy and macronutrient intake as a measure of variation in intake and the meal type-specific relative importance of predictors of these intake variations.

METHODS

Energy and macronutrient intake were derived from three 24 h dietary recalls in an EPIC-Potsdam sub-cohort of 814 German adults. Intra-class correlation was calculated for participants and meal type. Predictors of intake were assessed using meal type-specific multilevel regression models in a structural equation modeling framework at intake and participant levels using the Pratt Index. The importance of the predictor energy misreporting was assessed in sensitivity analyses on 682 participants. 95% confidence intervals were calculated based on 1000 bootstrap samples.

RESULTS

Differences between meal types explain a large proportion of the variation in intake (intra-class correlation: 39% for energy, 25% for carbohydrates, 47% for protein, and 33% for fat). Between-participant variation in intake was much lower, with a maximum of 3% for carbohydrate and fat. Place of meal was the most important intake-level predictor of energy and macronutrient intake (Pratt Index of up to 65%). Week/weekend day was important in the breakfast meal, and prior interval (hours passed since last meal) was important for the afternoon snack and dinner. On the participant level, sex was the most important predictor, with Pratt Index of up to 95 and 59% in the main and in the sensitivity analysis, respectively. Energy misreporting was especially important at the afternoon snack, accounting for up to 69% of the explained variance.

CONCLUSIONS

The meal type explains the highest variation in energy and macronutrient intakes. We identified key predictors of variation in the intake and in the participant levels. These findings suggest that successful dietary modification efforts should focus on improving specific meals.

Nordic diet, Mediterranean diet, and the risk of chronic diseases: the EPIC-Potsdam study

BACKGROUND

The Mediterranean Diet (MedDiet) has been acknowledged as a healthy diet. However, its relation with risk of major chronic diseases in non-Mediterranean countries is inconclusive. The Nordic diet is proposed as an alternative across Northern Europe, although its associations with the risk of chronic diseases remain controversial. We aimed to investigate the association between the Nordic diet and the MedDiet with the risk of chronic disease (type 2 diabetes (T2D), myocardial infarction (MI), stroke, and cancer) in the EPIC-Potsdam cohort.

METHODS

The EPIC-Potsdam cohort recruited 27,548 participants between 1994 and 1998. After exclusion of prevalent cases, we evaluated baseline adherence to a score reflecting the Nordic diet and two MedDiet scores (tMDS, reflecting the traditional MedDiet score, and the MedPyr score, reflecting the MedDiet Pyramid). Cox regression models were applied to examine the association between the diet scores and the incidence of major chronic diseases.

RESULTS

During a follow-up of 10.6 years, 1376 cases of T2D, 312 of MI, 321 of stroke, and 1618 of cancer were identified. The Nordic diet showed a statistically non-significant inverse association with incidence of MI in the overall population and of stroke in men. Adherence to the MedDiet was associated with lower incidence of T2D (HR per 1 SD 0.93, 95% CI 0.88-0.98 for the tMDS score and 0.92, 0.87-0.97 for the MedPyr score). In women, the MedPyr score was also inversely associated with MI. No association was observed for any of the scores with cancer.

CONCLUSIONS

In the EPIC-Potsdam cohort, the Nordic diet showed a possible beneficial effect on MI in the overall population and for stroke in men, while both scores reflecting the MedDiet conferred lower risk of T2D in the overall population and of MI in women.

Participants' decision to enroll in cohort studies with biobanks: quantitative insights from two German studies

AIM

To understand participants' actual motivations to enroll in cohort studies with biobanks.

METHODS

Participants of two such studies currently ongoing in Germany were invited to fill in a questionnaire about their decision to enroll.

RESULTS

From the 623 questionnaires completed, contributing to scientific knowledge appeared as a main motive for enrollment, followed by learning about personal health status and receiving personal medical advice.

CONCLUSION

Motivating participation as a way to contribute to the common good rather than as a way to obtain personal benefits appears to be particularly appropriate to ensure participants' long-term commitment and should therefore be further investigated in this setting.

Red meat consumption and risk of cardiovascular diseases-is increased iron load a possible link?

Background

High iron load and red meat consumption could increase the risk of cardiovascular diseases (CVDs). As red meat is the main source of heme iron, which is in turn a major determinant of increased iron load, adverse cardiometabolic effects of meat consumption could be mediated by increased iron load.

Objective

The object of the study was to assess whether associations between red meat consumption and CVD risk are mediated by iron load in a population-based human study.

Design

We evaluated relations between red meat consumption, iron load (plasma ferritin), and risk of CVD in the prospective EPIC-Heidelberg Study using a case-cohort sample including a random subcohort (n = 2738) and incident cases of myocardial infarction (MI, n = 555), stroke (n = 513), and CVD mortality (n = 381). Following a 4-step mediation analysis, associations between red meat consumption and iron load, red meat consumption and CVD risk, and iron load and CVD risk were assessed by multivariable regression models before finally testing to which degree associations between red meat consumption and CVD risk were attenuated by adjustment for iron status.

Results

Red meat consumption was significantly positively associated with ferritin concentrations and MI risk [HR per 50 g daily intake: 1.18 (95% CI: 1.05, 1.33)], but no significant associations with stroke risk and CVD mortality were observed. While direct associations between ferritin concentrations and MI risk as well as CVD mortality were significant in age- and sex-adjusted Cox regression models, these associations were substantially attenuated and no longer significant after multivariable adjustment for classical CVD risk factors. Strikingly, ferritin concentrations were positively associated with a majority of classical CVD risk factors (age, male sex, alcohol intake, obesity, inflammation, and lower education).

Conclusion

Increased ferritin concentrations may be a marker of an overall unfavorable risk factor profile rather than a mediator of greater CVD risk due to meat consumption.

Derivation and external validation of a clinical version of the German Diabetes Risk Score (GDRS) including measures of HbA1c

OBJECTIVE

The German Diabetes Risk Score (GDRS) is a diabetes prediction model which only includes non-invasively measured risk factors. The aim of this study was to extend the original GDRS by hemoglobin A1c (HbA1c) and validate this clinical GDRS in the nationwide German National Health Interview and Examination Survey 1998 (GNHIES98) cohort.

RESEARCH DESIGN AND METHODS

Extension of the GDRS was based on the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam study with baseline assessment conducted between 1994 and 1998 (N=27 548, main age range 35-65 years). Cox regression was applied with the original GDRS and HbA1c as independent variables. The extended model was evaluated by discrimination (C-index (95% CI)), calibration (calibration plots and expected to observed (E:O) ratios (95% CI)), and reclassification (net reclassification improvement, NRI (95% CI)). For validation, data from the GNHIES98 cohort with baseline assessment conducted between 1997 and 1999 were used (N=3717, age range 18-79 years). Missing data were handled with multiple imputation.

RESULTS

After 5 years of follow-up 593 incident cases of type 2 diabetes occurred in EPIC-Potsdam and 86 in the GNHIES98 cohort. In EPIC-Potsdam, the C-index for the clinical GDRS was 0.87 (0.81 to 0.92) and the overall NRI was 0.26 (0.21 to 0.30), with a stronger improvement among cases compared with non-cases (NRI_cases: 0.24 (0.19 to 0.28); NRI_non-cases: 0.02 (0.01 to 0.02)). Almost perfect calibration was observed with a slight tendency toward overestimation, which was also reflected by an E:O ratio of 1.07 (0.99 to 1.16). In the GNHIES98 cohort, discrimination was excellent with a C-index of 0.91 (0.88 to 0.94). After recalibration, the calibration plot showed underestimation of diabetes risk in the highest risk group, while the E:O ratio indicated overall perfect calibration (1.02 (0.83 to 1.26)).

CONCLUSIONS

The clinical GDRS provides the opportunity to apply the original GDRS as a first step in risk assessment, which can then be extended in clinical practice with HbA1c whenever it was measured.

Serum metabolites and risk of myocardial infarction and ischemic stroke: a targeted metabolomic approach in two German prospective cohorts

Metabolomic approaches in prospective cohorts may offer a unique snapshot into early metabolic perturbations that are associated with a higher risk of cardiovascular diseases (CVD) in healthy people. We investigated the association of 105 serum metabolites, including acylcarnitines, amino acids, phospholipids and hexose, with risk of myocardial infarction (MI) and ischemic stroke in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam (27,548 adults) and Heidelberg (25,540 adults) cohorts. Using case-cohort designs, we measured metabolites among individuals who were free of CVD and diabetes at blood draw but developed MI (n = 204 and n = 228) or stroke (n = 147 and n = 121) during follow-up (mean, 7.8 and 7.3 years) and among randomly drawn subcohorts (n = 2214 and n = 770). We used Cox regression analysis and combined results using meta-analysis. Independent of classical CVD risk factors, ten metabolites were associated with risk of MI in both cohorts, including sphingomyelins, diacyl-phosphatidylcholines and acyl-alkyl-phosphatidylcholines with pooled relative risks in the range of 1.21–1.40 per one standard deviation increase in metabolite concentrations. The metabolites showed positive correlations with total- and LDL-cholesterol (r ranged from 0.13 to 0.57). When additionally adjusting for total-, LDL- and HDL-cholesterol, triglycerides and C-reactive protein, acyl-alkyl-phosphatidylcholine C36:3 and diacyl-phosphatidylcholines C38:3 and C40:4 remained associated with risk of MI. When added to classical CVD risk models these metabolites further improved CVD prediction (c-statistics increased from 0.8365 to 0.8384 in EPIC-Potsdam and from 0.8344 to 0.8378 in EPIC-Heidelberg). None of the metabolites was consistently associated with stroke risk. Alterations in sphingomyelin and phosphatidylcholine metabolism, and particularly metabolites of the arachidonic acid pathway are independently associated with risk of MI in healthy adults.

Albumin, bilirubin, uric acid and cancer risk: results from a prospective population-based study

BACKGROUND

It has long been proposed that albumin, bilirubin and uric acid may inhibit cancer development due to their anti-oxidative properties. However, there is a lack of population-based studies on blood levels of these molecules and cancer risk.

METHODS

Associations between pre-diagnostic serum albumin, bilirubin and uric acid and the risks of common cancers as well as cancer death in the EPIC-Heidelberg cohort were evaluated by multivariable Cox regression analyses. A case-cohort sample including a random subcohort (n=2739) and all incident cases of breast (n=627), prostate (n=554), colorectal (n=256), and lung cancer (n=195) as well as cancer death (n=761) that occurred between baseline (1994-1998) and 2009 was used.

RESULTS

Albumin levels were inversely associated with breast cancer risk (hazard ratioQuartile 4 vs Quartile 1 (95% CI): 0.71 (0.51, 0.99), Plinear trend=0.004) and overall cancer mortality (HRQ4 vs Q1 (95% CI): 0.64 (0.48, 0.86), Plinear trend<0.001) after multivariable adjustment. Uric acid levels were also inversely associated with breast cancer risk (HRQ4 vs Q1 (95% CI): 0.72 (0.53, 0.99), Plinear trend=0.043) and cancer mortality (HRQ4 vs Q1 (95% CI): 0.75 (0.58, 0.98), Plinear trend=0.09). There were no significant associations between albumin or uric acid and prostate, lung and colorectal cancer. Serum bilirubin was not associated with any cancer end point.

CONCLUSIONS

The present findings indicate that higher levels of albumin and uric acid are related to lower risks of breast cancer and cancer mortality. Further studies are needed to assess whether the observed associations are causal.

Association between chemerin, omentin-1 and risk of heart failure in the population-based EPIC-Potsdam study

The adipokines chemerin and omentin-1 have been suggested to influence cardiovascular function. The study aimed to investigate the longitudinal association between chemerin, omentin-1 concentrations and risk of incident heart failure (HF), respectively. We conducted a case-cohort study nested within the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam cohort (n = 27548) including a randomly drawn subsample and all incident HF cases during a mean follow-up of 8.2 ± 1.5 years. A total of 212 incident HF cases and 2168 individuals free of HF cases were included in the study. After multivariable adjustment for established cardiovascular risk factors chemerin was strongly associated with risk of HF (HR per doubling chemerin: 4.91; 95%-CI: 2.57-9.39; p < 0.0001). Omentin-1 was not significantly related to HF risk in the overall study population. However, the association between omentin-1 and HF risk was modified by prevalent coronary heart disease (CHD), showing that the shape of the association was linear in participants without prevalent CHD (HR doubling omentin-1: 2.11; 95%-CI: 1.36-3.27; p linear = 0.0009) and U-shaped in participants with pre-existing CHD (p non-linear = 0.006). Our study provides first evidence for a strong positive association between chemerin and risk of HF. The association between the adipokine omentin-1 and risk of HF may differ according to pre-existing CHD.

Multiple imputation was a valid approach to estimate absolute risk from a prediction model based on case-cohort data

OBJECTIVE

To compare weighting methods for Cox regression and multiple imputation (MI) in a case-cohort study in the context of risk prediction modeling.

STUDY DESIGN AND SETTING

Based on the European Prospective Investigation into Cancer and Nutrition Potsdam study, we estimated risk scores to predict incident type-2 diabetes using full cohort data and case-cohort data assuming missing information on waist circumference outside the case-cohort (∼90%). Varying weighting approaches and MI were compared with regard to the calculation of relative risks, absolute risks, and predictive abilities including C-index, the net reclassification improvement, and calibration.

RESULTS

The full cohort comprised 21,845 participants, and the case-cohort comprised 2,703 participants. Relative risks were similar across all methods and compatible with full cohort estimates. Absolute risk estimates showed stronger disagreement mainly for Prentice and Self & Prentice weighting. Barlow and Langholz & Jiao weighting methods and MI were in good agreement with full cohort analysis. Predictive abilities were closest to full cohort estimates for MI or for Barlow and Langholz & Jiao weighting.

CONCLUSIONS

MI seems to be a valid method for deriving or extending a risk prediction model from case-cohort data and might be superior for absolute risk calculation when compared to weighted approaches.

Random Survival Forest in practice: a method for modelling complex metabolomics data in time to event analysis

BACKGROUND

The application of metabolomics in prospective cohort studies is statistically challenging. Given the importance of appropriate statistical methods for selection of disease-associated metabolites in highly correlated complex data, we combined random survival forest (RSF) with an automated backward elimination procedure that addresses such issues.

METHODS

Our RSF approach was illustrated with data from the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam study, with concentrations of 127 serum metabolites as exposure variables and time to development of type 2 diabetes mellitus (T2D) as outcome variable. Out of this data set, Cox regression with a stepwise selection method was recently published. Replication of methodical comparison (RSF and Cox regression) was conducted in two independent cohorts. Finally, the R-code for implementing the metabolite selection procedure into the RSF-syntax is provided.

RESULTS

The application of the RSF approach in EPIC-Potsdam resulted in the identification of 16 incident T2D-associated metabolites which slightly improved prediction of T2D when used in addition to traditional T2D risk factors and also when used together with classical biomarkers. The identified metabolites partly agreed with previous findings using Cox regression, though RSF selected a higher number of highly correlated metabolites.

CONCLUSIONS

The RSF method appeared to be a promising approach for identification of disease-associated variables in complex data with time to event as outcome. The demonstrated RSF approach provides comparable findings as the generally used Cox regression, but also addresses the problem of multicollinearity and is suitable for high-dimensional data.

The Ratio of Regulatory (FOXP3+) to Total (CD3+) T Cells Determined by Epigenetic Cell Counting and Cardiovascular Disease Risk: A Prospective Case-cohort Study in Non-diabetics

Background

Experimental and clinical evidence indicate that inflammatory processes in atherogenesis and the development of cardiovascular complications are promoted by a loss of regulatory T cell (Treg)-mediated immunological tolerance to plaque antigens. Yet, the association between alterations of systemic Treg frequency and cardiovascular disease incidence remains uncertain.

Methods

A nested case-cohort study was conducted within the European Prospective Investigation into Cancer and Nutrition (EPIC)-Heidelberg, comprising a random subcohort (n = 778) and primary cases of myocardial infarction (MI, n = 276) and ischemic stroke (n = 151). Pre-diagnostic FOXP3 + Treg and total CD3 + T-lymphocyte (tTL) frequencies in blood were measured by epigenetic-based, quantitative real-time PCR-assisted cell counting.

Results

Multivariate, Prentice-weighted Cox regression analyses revealed that lower Treg/tTL ratios were not associated with the risk of either MI (lowest vs. highest sex-specific quartile; hazard ratio: 0.72, 95% confidence interval: 0.46 to 1.13; P_trend = 0.51) or stroke (HR: 0.90, 95% CI: 0.51 to 1.60; P_trend = 0.78). There were no correlations of Treg/tTL ratios with C-reactive protein, HbA1c, and various lipid parameters.

Conclusions

Among middle-aged adults from the general population, imbalances in the relative frequency of Tregs within the total T cell compartment do not confer an increased risk of MI or stroke.

•

We studied if peripheral immune tolerance, as reflected by regulatory (FOXP3+) to total (CD3+) T cells, relates to CVD risk.

•

Epigenetic-based, qPCR assisted cell counting was used to quantify T cell subsets in long-term stored buffy coat samples.

•

Lower Treg-mediated immune tolerance does not confer an increased risk of major CVD events.

Inflammation in the arterial intima plays a central role in atherosclerotic cardiovascular disease and may develop owing to autoimmune-like responses targeted against plaque antigens. While the ratio between regulatory T cells (Tregs) and effector T cells is thought to control such immune response outcomes and tolerance within the T cell compartment, we found no association with incidence of major CVD events. These findings imply that reduced systemic Treg frequencies observed in CVD patients follow rather than precede disease manifestation and that Treg variation within a physiological range may not – as previously reported - constitute a pre-disposing risk factor for CVD.

Participants' Accounts on Their Decision to Join a Cohort Study With an Attached Biobank: A Qualitative Content Analysis Study Within Two German Studies

Reliable participation and sustained retention rates are crucial in longitudinal studies involving human subjects and biomaterials. Understanding the decision to enroll is an essential step to develop adequate strategies promoting long-term participation. Semi-structured interviews were implemented with newly recruited and long-term participants randomly drawn from two ongoing longitudinal studies with a biobank component in Germany. Iterative qualitative content analysis was applied to the transcribed interviews. Participants (n = 31) expressed their decision to enroll or remain in the study as the result of the complex interplay of individual factors, institutional cues, study-related features, and societal dynamics. Different forms of trust were identified as central within the elements used to explain participation and could be compared to Dibben, Morris, and Lean's dynamic model of interpersonal trust. Given these high levels of trust, an investigation of the morality of the trustful relationship at stake between participants and research(ers) is warranted.

Prevalence and Progression of Lower Urinary Tract Symptoms in an Aging Population

OBJECTIVE

To examine the prevalence of lower urinary tract symptoms (LUTS) in males of the general population.

MATERIALS AND METHODS

In our analysis, we included 8627 men, 48-79 years of age, who participated in the fourth follow-up (FUP) of EPIC-Heidelberg (2007-2009) and replied to questions on LUTS. According to the International Prostate Symptom Score questionnaire, men were categorized as having mild (0-7 points), moderate (8-19 points), or severe LUTS (20-35 points). In addition, we examined progression of LUTS among 7821 men, who also participated in FUP 5 (2010-2012).

RESULTS

There were 75.3% of men who reported mild, 22.0% who reported moderate, and 2.7% who reported severe LUTS. The prevalence increased with age. At FUP 4, 5.8% (mild symptoms) to 39.7% (severe LUTS) of participants reported use of any type of benign prostatic hyperplasia or LUTS medication. Nocturia, that is, getting up at night at least twice, was the most common symptom, followed by incomplete emptying of the bladder and urgency. There were 54.8% of men who reported worse LUTS in FUP 5, but 27.1% reported an improvement in symptoms.

CONCLUSION

About a quarter of middle-aged and elderly men reported clinically relevant LUTS. Whereas symptoms in some men actually improve, more than half of men experience worsening of symptoms over a 3-year period in time.

Specific Metabolic Markers Are Associated with Future Waist-Gaining Phenotype in Women

OBJECTIVE

Our study aims to identify metabolic markers associated with either a gain in abdominal (measured by waist circumference) or peripheral (measured by hip circumference) body fat mass.

METHODS

Data of 4 126 weight-gaining adults (18-75 years) from three population-based, prospective German cohort studies (EPIC, KORA, DEGS) were analysed regarding a waist-gaining (WG) or hip-gaining phenotype (HG). The phenotypes were obtained by calculating the differences of annual changes in waist minus hip circumference. The difference was displayed for all cohorts. The highest 10% of this difference were defined as WG whereas the lowest 10% were defined as HG. A total of 121 concordant metabolite measurements were conducted using Biocrates AbsoluteIDQ® kits in EPIC and KORA. Sex-specific associations with metabolite concentration as independent and phenotype as the dependent variable adjusted for confounders were calculated. The Benjamini-Hochberg method was used to correct for multiple testing.

RESULTS

Across studies both sexes gained on average more waist than hip circumference. We could identify 12 metabolites as being associated with the WG (n = 8) or HG (n = 4) in men, but none were significant after correction for multiple testing; 45 metabolites were associated with the WG (n = 41) or HG (n = 4) in women. For WG, n = 21 metabolites remained significant after correction for multiple testing. Respective odds ratios (OR) ranged from 0.66 to 0.73 for tryptophan, the diacyl-phosphatidylcholines (PC) C32:3, C36:0, C38:0, C38:1, C42:2, C42:5, the acyl-alkyl-PCs C32:2, C34:0, C36:0, C36:1, C36:2, C38:0, C38:2, C40:1, C40:2, C40:5, C40:6, 42:2, C42:3 and lyso-PC C17:0.

CONCLUSION

Both weight-gaining men and women showed a clear tendency to gain more abdominal than peripheral fat. Gain of abdominal fat seems to be related to an initial metabolic state reflected by low concentrations of specific metabolites, at least in women. Thus, higher levels of specific PCs may play a protective role in gaining waist circumference.

Insulin-Like Growth Factor 1 and Insulin-Like Growth Factor-Binding Protein 3 in Relation to the Risk of Type 2 Diabetes Mellitus: Results From the EPIC-Potsdam Study

Higher levels of insulin-like growth factor-binding protein 3 (IGFBP-3) might raise the risk of type 2 diabetes mellitus (T2DM) via binding of insulin-like growth factor 1 (IGF-1), an insulin-like hormone that is involved in glucose homeostasis. We investigated serum concentrations of IGF-1 and IGFBP-3 and their molar ratio in relation to T2DM incidence in a nested case-cohort study within the European Prospective Investigation Into Cancer and Nutrition-Potsdam Study. We included a randomly selected subcohort of persons without T2DM at the time of blood sampling (n = 2,269) and 776 individuals with incident T2DM identified between 1994 and 2005. For the highest quartile versus lowest, the multivariable-adjusted hazard rate ratios were 0.91 (95% confidence interval (CI): 0.68, 1.23; P for trend = 0.31) for IGF-1, 1.33 (95% CI: 1.00, 1.76; P for trend = 0.04) for IGFBP-3, and 0.77 (95% CI: 0.57, 1.03; P for trend = 0.03) for IGF-1:IGFBP-3 ratio. IGFBP-3 level remained positively associated with T2DM incidence-and the ratio of IGF-1 to IGFBP-3 was inversely related with T2DM incidence--in models that included adjustment for IGF-1 concentrations (P for trend < 0.05). Therefore, our findings do not confirm an association between total IGF-1 concentrations and risk of T2DM in the general study population, although higher IGFBP-3 levels might raise T2DM risk independent of IGF-1 levels.

Higher plasma levels of lysophosphatidylcholine 18:0 are related to a lower risk of common cancers in a prospective metabolomics study

BACKGROUND

First metabolomics studies have indicated that metabolic fingerprints from accessible tissues might be useful to better understand the etiological links between metabolism and cancer. However, there is still a lack of prospective metabolomics studies on pre-diagnostic metabolic alterations and cancer risk.

METHODS

Associations between pre-diagnostic levels of 120 circulating metabolites (acylcarnitines, amino acids, biogenic amines, phosphatidylcholines, sphingolipids, and hexoses) and the risks of breast, prostate, and colorectal cancer were evaluated by Cox regression analyses using data of a prospective case-cohort study including 835 incident cancer cases.

RESULTS

The median follow-up duration was 8.3 years among non-cases and 6.5 years among incident cases of cancer. Higher levels of lysophosphatidylcholines (lysoPCs), and especially lysoPC a C18:0, were consistently related to lower risks of breast, prostate, and colorectal cancer, independent of background factors. In contrast, higher levels of phosphatidylcholine PC ae C30:0 were associated with increased cancer risk. There was no heterogeneity in the observed associations by lag time between blood draw and cancer diagnosis.

CONCLUSION

Changes in blood lipid composition precede the diagnosis of common malignancies by several years. Considering the consistency of the present results across three cancer types the observed alterations point to a global metabolic shift in phosphatidylcholine metabolism that may drive tumorigenesis.

Quality Control Measures over 30 Years in a Multicenter Clinical Study: Results from the Diabetes Control and Complications Trial / Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study

Implementation of multicenter and/or longitudinal studies requires an effective quality assurance program to identify trends, data inconsistencies and process variability of results over time. The Diabetes Control and Complications Trial (DCCT) and the follow-up Epidemiology of Diabetes Interventions and Complications (EDIC) study represent over 30 years of data collection among a cohort of participants across 27 clinical centers. The quality assurance plan is overseen by the Data Coordinating Center and is implemented across the clinical centers and central reading units. Each central unit incorporates specific DCCT/EDIC quality monitoring activities into their routine quality assurance plan. The results are reviewed by a data quality assurance committee whose function is to identify variances in quality that may impact study results from the central units as well as within and across clinical centers, and to recommend implementation of corrective procedures when necessary. Over the 30-year period, changes to the methods, equipment, or clinical procedures have been required to keep procedures current and ensure continued collection of scientifically valid and clinically relevant results. Pilot testing to compare historic processes with contemporary alternatives is performed and comparability is validated prior to incorporation of new procedures into the study. Details of the quality assurance plan across and within the clinical and central reading units are described, and quality outcomes for core measures analyzed by the central reading units (e.g. biochemical samples, fundus photographs, ECGs) are presented.

Reproductive factors and risk of mortality in the European Prospective Investigation into Cancer and Nutrition; a cohort study

BACKGROUND

Reproductive events are associated with important physiologic changes, yet little is known about how reproductive factors influence long-term health in women. Our objective was to assess the relation of reproductive characteristics with all-cause and cause-specific mortality risk.

METHODS

The analysis was performed within the European Investigation into Cancer and Nutrition prospective cohort study, which enrolled >500,000 women and men from 1992 to 2000, who were residing in a given town/geographic area in 10 European countries. The current analysis included 322,972 eligible women aged 25-70 years with 99 % complete follow-up for vital status. We assessed reproductive characteristics reported at the study baseline including parity, age at the first birth, breastfeeding, infertility, oral contraceptive use, age at menarche and menopause, total ovulatory years, and history of oophorectomy/hysterectomy. Hazard ratios (HRs) and 95 % confidence intervals (CIs) for mortality were determined using Cox proportional hazards regression models adjusted for menopausal status, body mass index, physical activity, education level, and smoking status/intensity and duration.

RESULTS

During a mean follow-up of 12.9 years, 14,383 deaths occurred. The HR (95 % CI) for risk of all-cause mortality was lower in parous versus nulliparous women (0.80; 0.76-0.84), in women who had ever versus never breastfed (0.92; 0.87-0.97), in ever versus never users of oral contraceptives (among non-smokers; 0.90; 0.86-0.95), and in women reporting a later age at menarche (≥15 years versus <12; 0.90; 0.85-0.96; P for trend = 0.038).

CONCLUSIONS

Childbirth, breastfeeding, oral contraceptive use, and a later age at menarche were associated with better health outcomes. These findings may contribute to the development of improved strategies to promote better long-term health in women.

Cross-Sectional Association Between Number of Teeth and High-Sensitivity C-Reactive Protein Among Middle-Aged Germans

BACKGROUND

The association between number of teeth and low-grade systemic inflammation deserves consideration within the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam since the association between number of teeth and myocardial infarction has been established.

METHODS

Two subsamples (n = 2,439 and 728) were randomly selected from EPIC-Potsdam. Participants provided information on number of natural teeth, anthropometry, lifestyle factors, and illness-related factors. High-sensitivity C-reactive protein (hsCRP) was measured from serum. Adjusted means of hsCRP across five categories of numbers of teeth in each subsample and in the combined sample were determined, and linear trends were checked. Non-linear associations were investigated with restricted cubic spline (RCS) regression.

RESULTS

In the first subsample, the full multivariable-adjusted model showed that participants with 28 to 32, 24 to 27, 18 to 23, 1 to 17, and 0 teeth had mean hsCRP values of 1.32, 1.39, 1.54, 1.38, and 1.48 mg/L, respectively; in the second subsample, mean hsCRP values were 1.64, 1.67, 1.73, 1.47, and 1.87 mg/L; combined hsCRP values were 1.49, 1.53, 1.64, 1.44, and 1.65 mg/L. No linear trend was observed in these models, and RCS regression showed no non-linear association.

CONCLUSION

This study shows that number of teeth has a weak association with hsCRP, if any, thereby excluding this marker of low-grade systemic inflammation as a possible explanation for the association between number of teeth and myocardial infarction.

Treg-Mediated Immune Tolerance and the Risk of Solid Cancers: Findings From EPIC-Heidelberg

BACKGROUND

Laboratory-based, mechanistic, and prognosis studies suggest that a shift from antitumor immunity towards tumor-immune tolerance plays a major role in carcinogenesis. However, prospective epidemiological studies on the consequences of differing immune tolerance levels prior to clinical manifestation are missing.

METHODS

A case-cohort study embedded in EPIC-Heidelberg was conducted comprising incident cases of breast (n = 399), colorectal (n = 185), lung (n = 149), and prostate (n = 378) cancer, which occurred during 6.6 years of follow-up, and a subcohort (n = 807). Foxp3+ regulatory T-lymphocytes and CD3+ T-lymphocytes were measured by quantitative polymerase chain reaction-based DNA methylation analysis in prediagnostic leukocyte samples. Hazard ratios (HRs) for associations of cancer risk with the ratio of both parameters, the "cellular ratio of immune tolerance" (ImmunoCRIT), were estimated using Cox regression models. All statistical tests were two-sided.

RESULTS

ImmunoCRIT values were positively associated with the risk of lung (highest vs lowest tertile, HR = 1.98, 95% confidence interval = 1.06 to 3.69, P trend = .0263) and colorectal cancer (HR = 1.59, 95% CI = 0.99 to 2.54, P trend = .0069) after multivariable adjustment, but not with prostate cancer risk. Regarding breast cancer significant heterogeneity by estrogen receptor (ER) status was observed (P heterogeneity = .02), and the ImmunoCRIT was associated with the risk of ER-negative breast cancer (HR = 3.34, 95% CI = 1.52 to 7.35, P trend ≤ .001), but not ER-positive breast cancer.

CONCLUSION

The present study indicates that increased peripheral immune tolerance may be an independent risk factor for lung, colorectal, and ER-negative breast cancer, whereas its role on the development of prostate and ER-positive breast tumors remains uncertain.

[Mortality follow-up of the German Health Interview and Examination Survey for Adults (DEGS) : methods and first results]

OBJECTIVES

Within the framework of the German Health Interview and Examination Survey for Adults (DEGS), the Robert Koch Institute (RKI) conducted a nationwide mortality follow-up study. As there is no national mortality register in Germany, mortality and causes of death were investigated individually and under observance of state-specific data protection conditions.

METHODS

The German Health Interview and Examination Survey 1998 (GNHIES98) provided the database including 7,124 participants aged 18-79 years. A total of 6,979 participants of GNHIES98 (98 %) who consented to be re-contacted were invited between October 2008 and October 2011 to also participate in the first data collection wave of DEGS (DEGS1). In this context, the vital status and the causes of death for deceased participants were assessed. Age- and sex-specific probabilities of survival and death rates were calculated and grouped by main causes of death according to ICD-10 groups.

RESULTS

A total of 671 individuals (285 women, 386 men) died between the two survey contacts. For all deceased persons the date of death and for 539 (80.3 %) the causes of death could be determined. With a median follow-up time of 12.0 years, 8,0742.5 person years were available for survival analysis. The crude overall death rate amounted to 8.3 per 1,000 persons-years (women: 7.2; men: 9.5). Among 539 persons with available information on causes of death, 209 (38.8 %) were attributable to cardiovascular diseases, 188 (34.9 %) to cancer, 135 (25.0 %) to other causes, and seven (1.3 %) could not be unambiguously assigned.

CONCLUSIONS

A mortality follow-up was successfully integrated in the longitudinal component of DEGS as part of the national health monitoring at the RKI. Death rates and cause-specific mortality in relation to highly prevalent chronic diseases and risk factors provide essential information for assessing the potential of prevention and quality of care among adults in Germany. This requires a regular and complete conduction of mortality follow-ups.