Differences in Metabolomic Profiles Between Black and White Women and Risk of Coronary Heart Disease: an Observational Study of Women From Four US Cohorts.

BACKGROUND: Racial differences in metabolomic profiles may reflect underlying differences in social determinants of health by self-reported race and may be related to racial disparities in coronary heart disease (CHD) among women in the United States. However, the magnitude of differences in metabolomic profiles between Black and White women in the United States has not been well-described. It also remains unknown whether such differences are related to differences in CHD risk. METHODS: Plasma metabolomic profiles were analyzed using liquid chromatography-tandem mass spectrometry in the WHI-OS (Women's Health Initiative-Observational Study; 138 Black and 696 White women), WHI-HT trials (WHI-Hormone Therapy; 156 Black and 1138 White women), MESA (Multi-Ethnic Study of Atherosclerosis; 114 Black and 219 White women), JHS (Jackson Heart Study; 1465 Black women with 107 incident CHD cases), and NHS (Nurses' Health Study; 2506 White women with 136 incident CHD cases). First, linear regression models were used to estimate associations between self-reported race and 472 metabolites in WHI-OS (discovery); findings were replicated in WHI-HT and validated in MESA. Second, we used elastic net regression to construct a racial difference metabolomic pattern (RDMP) representing differences in the metabolomic patterns between Black and White women in the WHI-OS; the RDMP was validated in the WHI-HT and MESA. Third, using conditional logistic regressions in the WHI (717 CHD cases and 719 matched controls), we examined associations of metabolites with large differences in levels by race and the RDMP with risk of CHD, and the results were replicated in Black women from the JHS and White women from the NHS. RESULTS: Of the 472 tested metabolites, levels of 259 (54.9%) metabolites, mostly lipid metabolites and amino acids, significantly differed between Black and White women in both WHI-OS and WHI-HT after adjusting for baseline characteristics, socioeconomic status, lifestyle factors, baseline health conditions, and medication use (false discovery rate <0.05); similar trends were observed in MESA. The RDMP, composed of 152 metabolites, was identified in the WHI-OS and showed significantly different distributions between Black and White women in the WHI-HT and MESA. Higher RDMP quartiles were associated with an increased risk of incident CHD (odds ratio=1.51 [0.97-2.37] for the highest quartile comparing to the lowest; Ptrend=0.02), independent of self-reported race and known CHD risk factors. In race-stratified analyses, the RDMP-CHD associations were more pronounced in White women. Similar patterns were observed in Black women from the JHS and White women from the NHS. CONCLUSIONS: Metabolomic profiles significantly and substantially differ between Black and White women and may be associated with CHD risk and racial disparities in US women.

Differences in metabolomic profiles between Black and White women in the U.S.: Analyses from two prospective cohorts.

There is growing interest in incorporating metabolomics into public health practice. However, Black women are under-represented in many metabolomics studies. If metabolomic profiles differ between Black and White women, this under-representation may exacerbate existing Black-White health disparities. We therefore aimed to estimate metabolomic differences between Black and White women in the U.S. We leveraged data from two prospective cohorts: the Nurses' Health Study (NHS; n = 2077) and Women's Health Initiative (WHI; n = 2128). The WHI served as the replication cohort. Plasma metabolites (n = 334) were measured via liquid chromatography-tandem mass spectrometry. Observed metabolomic differences were estimated using linear regression and metabolite set enrichment analyses. Residual metabolomic differences in a hypothetical population in which the distributions of 14 risk factors were equalized across racial groups were estimated using inverse odds ratio weighting. In the NHS, Black-White differences were observed for most metabolites (75 metabolites with observed differences ≥ |0.50| standard deviations). Black women had lower average levels than White women for most metabolites (e.g., for N6, N6-dimethlylysine, mean Black-White difference = - 0.98 standard deviations; 95% CI: - 1.11, - 0.84). In metabolite set enrichment analyses, Black women had lower levels of triglycerides, phosphatidylcholines, lysophosphatidylethanolamines, phosphatidylethanolamines, and organoheterocyclic compounds, but higher levels of phosphatidylethanolamine plasmalogens, phosphatidylcholine plasmalogens, cholesteryl esters, and carnitines. In a hypothetical population in which distributions of 14 risk factors were equalized, Black-White metabolomic differences persisted. Most results replicated in the WHI (88% of 272 metabolites available for replication). Substantial differences in metabolomic profiles exist between Black and White women. Future studies should prioritize racial representation.

Metabolomic profiles of chronic distress are associated with cardiovascular disease risk and inflammation-related risk factors.

BACKGROUND: Chronic psychological distress is associated with increased risk of cardiovascular disease (CVD) and investigators have posited inflammatory factors may be centrally involved in these relationships. However, mechanistic evidence and molecular underpinnings of these processes remain unclear, and data are particularly sparse among women. This study examined if a metabolite profile linked with distress was associated with increased CVD risk and inflammation-related risk factors. METHODS: A plasma metabolite-based distress score (MDS) of twenty chronic psychological distress-related metabolites was developed in cross-sectional, 1:1 matched case-control data comprised of 558 women from the Nurses' Health Study (NHS; 279 women with distress, 279 controls). This MDS was then evaluated in two other cohorts: the Women's Health Initiative Observational Cohort (WHI-OS) and the Prevención con Dieta Mediterránea (PREDIMED) trial. We tested the MDS's association with risk of future CVD in each sample and with levels of C-reactive protein (CRP) in the WHI-OS. The WHI-OS subsample included 944 postmenopausal women (472 CHD cases; mean time to event = 5.8 years); the PREDIMED subsample included 980 men and women (224 CVD cases, mean time to event = 3.1 years). RESULTS: In the WHI-OS, a 1-SD increase in the plasma MDS was associated with a 20% increased incident CHD risk (odds ratio [OR] = 1.20, 95% CI: 1.04 - 1.38), adjusting for known CVD risk factors excluding total and HDL cholesterol. This association was attenuated after including total and HDL cholesterol. CRP mediated an average 12.9% (95% CI: 4.9% - 28%, p < 10-15) of the total effect of MDS on CHD risk when adjusting for matching factors. This effect was attenuated after adjusting for known CVD risk factors. Of the metabolites in the MDS, tryptophan and threonine were inversely associated with incident CHD risk in univariate models. In PREDIMED, each one SD increase in the MDS was associated with an OR of 1.19 (95% CI: 1.00 - 1.41) for incident CVD risk, after adjusting all risk factors. Similar associations were observed in men and women. Four metabolites in the MDS were associated with incident CVD risk in PREDIMED in univariate models. Biliverdin and C36:5 phosphatidylcholine (PC) plasmalogen had inverse associations; C16:0 ceramide and C18:0 lysophosphatidylethanolamine(LPE) each had positive associations with CVD risk. CONCLUSIONS: Our study points to molecular alterations that may underlie the association between chronic distress and subsequent risk of cardiovascular disease in adults.

SpiderLearner: An ensemble approach to Gaussian graphical model estimation.

Gaussian graphical models (GGMs) are a popular form of network model in which nodes represent features in multivariate normal data and edges reflect conditional dependencies between these features. GGM estimation is an active area of research. Currently available tools for GGM estimation require investigators to make several choices regarding algorithms, scoring criteria, and tuning parameters. An estimated GGM may be highly sensitive to these choices, and the accuracy of each method can vary based on structural characteristics of the network such as topology, degree distribution, and density. Because these characteristics are a priori unknown, it is not straightforward to establish universal guidelines for choosing a GGM estimation method. We address this problem by introducing SpiderLearner, an ensemble method that constructs a consensus network from multiple estimated GGMs. Given a set of candidate methods, SpiderLearner estimates the optimal convex combination of results from each method using a likelihood-based loss function. K $$ K $$ -fold cross-validation is applied in this process, reducing the risk of overfitting. In simulations, SpiderLearner performs better than or comparably to the best candidate methods according to a variety of metrics, including relative Frobenius norm and out-of-sample likelihood. We apply SpiderLearner to publicly available ovarian cancer gene expression data including 2013 participants from 13 diverse studies, demonstrating our tool's potential to identify biomarkers of complex disease. SpiderLearner is implemented as flexible, extensible, open-source code in the R package ensembleGGM at https://github.com/katehoffshutta/ensembleGGM.

Path-level interpretation of Gaussian graphical models using the pair-path subscore.

BACKGROUND : Construction of networks from cross-sectional biological data is increasingly common. Many recent methods have been based on Gaussian graphical modeling, and prioritize estimation of conditional pairwise dependencies among nodes in the network. However, challenges remain on how specific paths through the resultant network contribute to overall 'network-level' correlations. For biological applications, understanding these relationships is particularly relevant for parsing structural information contained in complex subnetworks. RESULTS: We propose the pair-path subscore (PPS), a method for interpreting Gaussian graphical models at the level of individual network paths. The scoring is based on the relative importance of such paths in determining the Pearson correlation between their terminal nodes. PPS is validated using human metabolomics data from the Hyperglycemia and adverse pregnancy outcome (HAPO) study, with observations confirming well-documented biological relationships among the metabolites. We also highlight how the PPS can be used in an exploratory fashion to generate new biological hypotheses. Our method is implemented in the R package pps, available at https://github.com/nathan-gill/pps . CONCLUSIONS: The PPS can be used to probe network structure on a finer scale by investigating which paths in a potentially intricate topology contribute most substantially to marginal behavior. Adding PPS to the network analysis toolkit may enable researchers to ask new questions about the relationships among nodes in network data.

A metabolomic analysis of adiposity measures and pre- and postmenopausal breast cancer risk in the Nurses' Health Studies.

BACKGROUND: Adiposity is consistently positively associated with postmenopausal breast cancer and inversely associated with premenopausal breast cancer risk, though the reasons for this difference remain unclear. METHODS: In this nested case-control study of 1649 breast cancer cases and 1649 matched controls from the Nurses' Health Study (NHS) and the NHSII, we selected lipid and polar metabolites correlated with BMI, waist circumference, weight change since age 18, or derived fat mass, and developed a metabolomic score for each measure using LASSO regression. Logistic regression was used to investigate the association between this score and breast cancer risk, adjusted for risk factors and stratified by menopausal status at blood draw and diagnosis. RESULTS: Metabolite scores developed among only premenopausal or postmenopausal women were highly correlated with scores developed in all women (r = 0.93-0.96). Higher metabolomic adiposity scores were generally inversely related to breast cancer risk among premenopausal women. Among postmenopausal women, significant positive trends with risk were observed (e.g., metabolomic waist circumference score OR Q4 vs. Q1 = 1.47, 95% CI = 1.03-2.08, P-trend = 0.01). CONCLUSIONS: Though the same metabolites represented adiposity in pre- and postmenopausal women, breast cancer risk associations differed suggesting that metabolic dysregulation may have a differential association with pre- vs. postmenopausal breast cancer.

Plasma Metabolomic Signature of Early Abuse in Middle-Aged Women.

OBJECTIVE: Metabolomic profiling may provide insights into biological mechanisms underlying the strong epidemiologic links observed between early abuse and cardiometabolic disorders in later life. METHODS: We examined the associations between early abuse and midlife plasma metabolites in two nonoverlapping subsamples from the Nurses' Health Study II, comprising 803 (mean age = 40 years) and 211 women (mean age = 61 years). Liquid chromatography-tandem mass spectrometry assays were used to measure metabolomic profiles, with 283 metabolites consistently measured in both subsamples. Physical and sexual abuse before age 18 years was retrospectively assessed by validated questions integrating type/frequency of abuse. Analyses were conducted in each sample and pooled using meta-analysis, with multiple testing adjustment using the q value approach for controlling the positive false discovery rate. RESULTS: After adjusting for age, race, menopausal status, body size at age 5 years, and childhood socioeconomic indicators, more severe early abuse was consistently associated with five metabolites at midlife (q value < 0.20 in both samples), including lower levels of serotonin and C38:3 phosphatidylethanolamine plasmalogen and higher levels of alanine, proline, and C40:6 phosphatidylethanolamine. Other metabolites potentially associated with early abuse (q value < 0.05 in the meta-analysis) included triglycerides, phosphatidylcholine plasmalogens, bile acids, tyrosine, glutamate, and cotinine. The association between early abuse and midlife metabolomic profiles was partly mediated by adulthood body mass index (32% mediated) and psychosocial distress (13%-26% mediated), but not by other life-style factors. CONCLUSIONS: Early abuse was associated with distinct metabolomic profiles of multiple amino acids and lipids in middle-aged women. Body mass index and psychosocial factors in adulthood may be important intermediates for the observed association.

Associations of depression status with plasma levels of candidate lipid and amino acid metabolites: a meta-analysis of individual data from three independent samples of US postmenopausal women.

Recent animal and small clinical studies have suggested depression is related to altered lipid and amino acid profiles. However, this has not been examined in a population-based sample, particularly in women. We identified multiple metabolites associated with depression as potential candidates from prior studies. Cross-sectional data from three independent samples of postmenopausal women were analyzed, including women from the Women's Health Initiative-Observational Study (WHI-OS, n = 926), the WHI-Hormone Trials (WHI-HT; n = 1,325), and the Nurses' Health Study II Mind-Body Study (NHSII-MBS; n = 218). Positive depression status was defined as having any of the following: elevated depressive symptoms, antidepressant use, or depression history. Plasma metabolites were measured using liquid chromatography-tandem mass spectrometry (21 phosphatidylcholines (PCs), 7 lysophosphatidylethanolamines, 5 ceramides, 3 branched chain amino acids, and 9 neurotransmitters). Associations between depression status and metabolites were evaluated using multivariable linear regression; results were pooled by random-effects meta-analysis with multiple testing adjustment using the false discovery rate (FDR). Prevalence rates of positive depression status were 24.4% (WHI-OS), 25.7% (WHI-HT), and 44.7% (NHSII-MBS). After multivariable adjustment, positive depression status was associated with higher levels of glutamate and PC 36 : 1/38 : 3, and lower levels of tryptophan and GABA-to-glutamate and GABA-to-glutamine ratio (FDR-p < 0.05). Positive associations with LPE 18 : 0/18 : 1 and inverse associations with valine and serotonin were also observed, although these associations did not survive FDR adjustment. Associations of positive depression status with several candidate metabolites including PC 36 : 1/38 : 3 and amino acids involved in neurotransmission suggest potential depression-related metabolic alterations in postmenopausal women, with possible implications for later chronic disease.

Gaussian graphical models with applications to omics analyses.

Gaussian graphical models (GGMs) provide a framework for modeling conditional dependencies in multivariate data. In this tutorial, we provide an overview of GGM theory and a demonstration of various GGM tools in R. The mathematical foundations of GGMs are introduced with the goal of enabling the researcher to draw practical conclusions by interpreting model results. Background literature is presented, emphasizing methods recently developed for high-dimensional applications such as genomics, proteomics, or metabolomics. The application of these methods is illustrated using a publicly available dataset of gene expression profiles from 578 participants with ovarian cancer in The Cancer Genome Atlas. Stand-alone code for the demonstration is available as an RMarkdown file at https://github.com/katehoffshutta/ggmTutorial.

New Moms Wellness Study: the randomized controlled trial study protocol for an intervention study to increase fruit and vegetable intake and lower breast cancer risk through weekly counseling and supplemental fruit and vegetable box delivery in breastfeeding women.

BACKGROUND: Laboratory studies indicate that chemicals in fruits and vegetables have anti-carcinogenic and anti-inflammatory activities that can lower breast cancer risk. However, epidemiologic studies of the association between fruit and vegetable intake and breast cancer risk have produced mixed results. Measurement error, confounding, and an emphasis on diet in later adulthood may contribute to weak associations. This paper describes a randomized controlled diet intervention trial in breastfeeding women to examine the effect of high fruit and vegetable intake on breast cancer risk factors, including weight, DNA methylation and inflammatory markers. METHODS: Eligible breastfeeding women who reside within a 35-mile radius of Amherst, MA are enrolled at five to six weeks postpartum and randomly assigned to a Fruit and Vegetable Intervention Arm (target n = 200) or to a USDA MyPlate Control Arm (target n = 200). The Fruit and Vegetable Intervention group receives weekly telephone or video-based counseling to encourage intake of at least eight to ten daily servings of fruits and vegetables and a weekly delivery of a supplemental box of fruits and vegetables for 20 weeks, and less intensive counseling for up to one year. Breastmilk and infant fecal specimens are collected at baseline, 10 and 20 weeks. Anthropometric measurements are obtained at these timepoints and at the 1-year follow-up. The primary outcomes are change in DNA methylation in breast epithelial cells and change in inflammatory markers in breastmilk from randomization to 20 weeks; and change in weight, waist circumference, and fruit and vegetable intake for the period from randomization to 20 weeks and 1 year. DISCUSSION: This 1-year randomized diet intervention trial in breastfeeding women will assess whether intake of at least eight to ten daily servings of fruits and vegetables per day improves biomarkers of breast cancer risk directly in the breast (i.e., DNA methylation and inflammatory markers) and helps women maintain a healthy weight. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04374747. Registered May 5, 2020. https://www. CLINICALTRIALS: gov/ct2/show/NCT04374747 .

Bayesian variable selection for high dimensional predictors and self-reported outcomes.

BACKGROUND: The onset of silent diseases such as type 2 diabetes is often registered through self-report in large prospective cohorts. Self-reported outcomes are cost-effective; however, they are subject to error. Diagnosis of silent events may also occur through the use of imperfect laboratory-based diagnostic tests. In this paper, we describe an approach for variable selection in high dimensional datasets for settings in which the outcome is observed with error. METHODS: We adapt the spike and slab Bayesian Variable Selection approach in the context of error-prone, self-reported outcomes. The performance of the proposed approach is studied through simulation studies. An illustrative application is included using data from the Women's Health Initiative SNP Health Association Resource, which includes extensive genotypic (>900,000 SNPs) and phenotypic data on 9,873 African American and Hispanic American women. RESULTS: Simulation studies show improved sensitivity of our proposed method when compared to a naive approach that ignores error in the self-reported outcomes. Application of the proposed method resulted in discovery of several single nucleotide polymorphisms (SNPs) that are associated with risk of type 2 diabetes in a dataset of 9,873 African American and Hispanic participants in the Women's Health Initiative. There was little overlap among the top ranking SNPs associated with type 2 diabetes risk between the racial groups, adding support to previous observations in the literature of disease associated genetic loci that are often not generalizable across race/ethnicity populations. The adapted Bayesian variable selection algorithm is implemented in R. The source code for the simulations are available in the Supplement. CONCLUSIONS: Variable selection accuracy is reduced when the outcome is ascertained by error-prone self-reports. For this setting, our proposed algorithm has improved variable selection performance when compared to approaches that neglect to account for the error-prone nature of self-reports.

Metabolic Predictors of Incident Coronary Heart Disease in Women.

BACKGROUND: Although metabolomic profiling offers promise for the prediction of coronary heart disease (CHD), and metabolic risk factors are more strongly associated with CHD in women than men, limited data are available for women. METHODS: We applied a liquid chromatography-tandem mass spectrometry metabolomics platform to measure 371 metabolites in a discovery set of postmenopausal women (472 incident CHD cases, 472 controls) with validation in an independent set of postmenopausal women (312 incident CHD cases, 315 controls). RESULTS: Eight metabolites, primarily oxidized lipids, were significantly dysregulated in cases after the adjustment for matching and CHD risk factors in both the discovery and validation data sets. One oxidized phospholipid, C34:2 hydroxy-phosphatidylcholine, remained associated with CHD after further adjustment for other validated metabolites. Subjects with C34:2 hydroxy-phosphatidylcholine levels in the highest quartile had a 4.7-fold increase in CHD odds in comparison with the lowest quartile; C34:2 hydroxy-phosphatidylcholine also significantly improved the area under the curve (P<0.01) for CHD. The C34:2 hydroxy-phosphatidylcholine findings were replicated in a third replication data set of 980 men and women (230 cardiovascular events) with a stronger association observed in women. CONCLUSIONS: These data replicate known metabolite predictors, identify novel markers, and support the relationship between lipid oxidation and subsequent CHD.

Estimating the receiver operating characteristic curve in matched case control studies.

The matched case-control design is frequently used in the study of complex disorders and can result in significant gains in efficiency, especially in the context of measuring biomarkers; however, risk prediction in this setting is not straightforward. We propose an inverse-probability weighting approach to estimate the predictive ability associated with a set of covariates. In particular, we propose an algorithm for estimating the summary index, area under the curve corresponding to the Receiver Operating Characteristic curve associated with a set of pre-defined covariates for predicting a binary outcome. By combining data from the parent cohort with that generated in a matched case control study, we describe methods for estimation of the population parameters of interest and the corresponding area under the curve. We evaluate the bias associated with the proposed methods in simulations by considering a range of parameter settings. We illustrate the methods in two data applications: (1) a prospective cohort study of cardiovascular disease in women, the Women's Health Study, and (2) a matched case-control study nested within the Nurses' Health Study aimed at risk prediction of invasive breast cancer.

An Empirical Dietary Inflammatory Pattern Score Is Associated with Circulating Inflammatory Biomarkers in a Multi-Ethnic Population of Postmenopausal Women in the United States.

BACKGROUND: The empirical dietary inflammatory pattern (EDIP) score has been associated with concentrations of circulating inflammatory biomarkers in European Americans. OBJECTIVE: We used the EDIP score, a weighted sum of 18 food groups that characterizes dietary inflammatory potential based on circulating concentrations of inflammatory biomarkers, to test the hypothesis that a pro-inflammatory dietary pattern is associated with inflammatory biomarker concentrations in a US multi-ethnic population. METHODS: In this cross-sectional study, we calculated EDIP scores using baseline food frequency questionnaire data from 31,472 women, aged 50-79 y, in the Women's Health Initiative observational study and clinical trials. Circulating biomarkers outcomes at baseline were: C-reactive protein (CRP), interleukin-6 (IL-6), tumor necrosis factor (TNF)-α, TNF receptor (TNFR) 1 and 2, and adiponectin. We used multivariable-adjusted linear regression analyses to estimate absolute concentrations and relative differences in biomarker concentrations, overall and in subgroups of race/ethnicity and BMI (body mass index) categories. RESULTS: Independent of energy intake, BMI, physical activity, and other potential confounding variables, higher EDIP scores were significantly associated with higher (lower for adiponectin) absolute concentrations of all 6 biomarkers. On the relative scale, the percentage of difference in the concentration of biomarkers, among women in the highest compared to the lowest EDIP quintile, was: CRP, +13% (P-trend < 0.0001); IL-6, +15% (P-trend < 0.0001); TNF-α, +7% (P-trend = 0.0007); TNFR1, +4% (P-trend = 0.0009); TNFR2, +5% (P-trend < 0.0001); and adiponectin, -13% (P-trend <0.0001). These associations differed by racial/ethnic groups and by BMI categories. Whereas the absolute biomarker concentrations were lower among European-American women and among normal-weight women, the associations with diet were stronger than among women of African-American or Hispanic/Latino origin and among overweight and obese women. CONCLUSIONS: Findings demonstrate the successful replication of an empirical hypothesis-oriented a posteriori dietary pattern score in a multi-ethnic population of postmenopausal women, with subgroup differences by race/ethnicity and body weight. Future research needs to apply the score in non-US populations.

White blood cell DNA methylation and risk of breast cancer in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO).

BACKGROUND: Several studies have suggested that global DNA methylation in circulating white blood cells (WBC) is associated with breast cancer risk. METHODS: To address conflicting results and concerns that the findings for WBC DNA methylation in some prior studies may reflect disease effects, we evaluated the relationship between global levels of WBC DNA methylation in white blood cells and breast cancer risk in a case-control study nested within the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (PLCO) cohort. A total of 428 invasive breast cancer cases and 419 controls, frequency matched on age at entry (55-59, 60-64, 65-69, ≥70 years), year of entry (on/before September 30, 1997, on/after October 1, 1997) and period of DNA extraction (previously extracted, newly extracted) were included. The ratio of 5-methyl-2' deoxycytidine [5-mdC] to 2'-deoxyguanine [dG], assuming [dG] = [5-mdC] + [2'-deoxycytidine [dC]] (%5-mdC), was determined by liquid chromatography-electrospray ionization-tandem mass spectrometry, an especially accurate method for assessing total genomic DNA methylation. RESULTS: Odds ratio (OR) estimates and 95% confidence intervals (CI) for breast cancer risk adjusted for age at entry, year of entry, and period of DNA extraction, were 1.0 (referent), 0.89 (95% CI, 0.6-1.3), 0.88 (95% CI, 0.6-1.3), and 0.84 (95% CI, 0.6-1.2) for women in the highest compared to lowest quartile levels of %5md-C (p for trend = .39). Effects did not meaningfully vary by time elapsed from WBC collection to diagnosis. DISCUSSION: These results do not support the hypothesis that global DNA hypomethylation in WBC DNA is associated with increased breast cancer risk prior to the appearance of clinical disease.

Study design for non-recurring, time-to-event outcomes in the presence of error-prone diagnostic tests or self-reports.

Sequentially administered, laboratory-based diagnostic tests or self-reported questionnaires are often used to determine the occurrence of a silent event. In this paper, we consider issues relevant in design of studies aimed at estimating the association of one or more covariates with a non-recurring, time-to-event outcome that is observed using a repeatedly administered, error-prone diagnostic procedure. The problem is motivated by the Women's Health Initiative, in which diabetes incidence among the approximately 160,000 women is obtained from annually collected self-reported data. For settings of imperfect diagnostic tests or self-reports with known sensitivity and specificity, we evaluate the effects of various factors on resulting power and sample size calculations and compare the relative efficiency of different study designs. The methods illustrated in this paper are readily implemented using our freely available R software package icensmis, which is available at the Comprehensive R Archive Network website. An important special case is that when diagnostic procedures are perfect, they result in interval-censored, time-to-event outcomes. The proposed methods are applicable for the design of studies in which a time-to-event outcome is interval censored. Copyright © 2016 John Wiley & Sons, Ltd.

Impact of incident diabetes on atherosclerotic cardiovascular disease according to statin use history among postmenopausal women.

To compare impact of incident diabetes on atherosclerotic cardiovascular disease (ASCVD) risk among postmenopausal women according to statin use. Prospective data from 120,499 postmenopausal women without prevalent diabetes or cardiovascular disease at baseline from the Women's Health Initiative were used. Incident diabetes was self-reported annually and defined as treatment with pills or injectable medication for diabetes. Current statin use was determined at enrollment and years 1, 3, 6, 9 and 13.5 in the three clinical trial arms, and at baseline, year 3, and 13.5 for the observational study. The primary outcome was incident ASCVD events, self-reported annually and adjudicated by blinded local and central physicians. Incident diabetes and statin use status were fitted as time-varying covariates in Cox regression models to assess ASCVD risk during an average follow-up of 13.6 years. For those not on statins at the time of diabetes diagnosis, there was a 42 % increased risk of ASCVD [hazard ratio (HR) 1.42, 95 % CI 1.28-1.58] among women with incident diabetes versus those without diabetes. Among women on statins, there was a 39 % increased risk of ASCVD (HR 1.39, 95 % CI 1.12-1.74) in women with incident diabetes versus those without diabetes. The increased ASCVD risk due to diabetes was similar between women before or after initiating statins (P = 0.89). Whether diabetes was diagnosed before or after statin use did not alter the increased risk of ASCVD associated with diabetes. Mitigating the increased incidence of diabetes in statin users could increase the ASCVD benefit-to-risk ratio of statins.

Marginal structural models for the estimation of the risk of Diabetes Mellitus in the presence of elevated depressive symptoms and antidepressant medication use in the Women's Health Initiative observational and clinical trial cohorts.

BACKGROUND: We evaluate the combined effect of the presence of elevated depressive symptoms and antidepressant medication use with respect to risk of type 2 diabetes among approximately 120,000 women enrolled in the Women's Health Initiative (WHI), and compare several different statistical models appropriate for causal inference in non-randomized settings. METHODS: Data were analyzed for 52,326 women in the Women's Health Initiative Clinical Trials (CT) Cohort and 68,169 women in the Observational Study (OS) Cohort after exclusions. We included follow-up to 2005, resulting in a median duration of 7.6 years of follow up after enrollment. Results from three multivariable Cox models were compared to those from marginal structural models that included time varying measures of antidepressant medication use, presence of elevated depressive symptoms and BMI, while adjusting for potential confounders including age, ethnicity, education, minutes of recreational physical activity per week, total energy intake, hormone therapy use, family history of diabetes and smoking status. RESULTS: Our results are consistent with previous studies examining the relationship of antidepressant medication use and risk of type 2 diabetes. All models showed a significant increase in diabetes risk for those taking antidepressants. The Cox Proportional Hazards models using baseline covariates showed the lowest increase in risk , with hazard ratios of 1.19 (95 % CI 1.06 - 1.35) and 1.14 (95 % CI 1.01 - 1.30) in the OS and CT, respectively. Hazard ratios from marginal structural models comparing antidepressant users to non-users were 1.35 (95 % CI 1.21 - 1.51) and 1.27 (95 % CI 1.13 - 1.43) in the WHI OS and CT, respectively - however, differences among estimates from traditional Cox models and marginal structural models were not statistically significant in both cohorts. One explanation suggests that time-dependent confounding was not a substantial factor in these data, however other explanations exist. Unadjusted Cox Proportional Hazards models showed that women with elevated depressive symptoms had a significant increase in diabetes risk that remained after adjustment for confounders. However, this association missed the threshold for statistical significance in propensity score adjusted and marginal structural models. CONCLUSIONS: Results from the multiple approaches provide further evidence of an increase in risk of type 2 diabetes for those on antidepressants.