The Case for Case-Cohort: An Applied Epidemiologist's Guide to Reframing Case-Cohort Studies to Improve Usability and Flexibility

Diagnosis of epilepsy by machine learning of high-performance plasma metabolic fingerprinting

Epilepsy is a chronic neurological disorder that causes a major threat to public health and the burden of disease worldwide. High-performance diagnostic tools for epilepsy need to be developed to improve diagnostic accuracy and efficiency while still missing. Herein, we utilized nanoparticle-enhanced laser desorption/ionization mass spectrometry (NELDI MS) to acquire plasma metabolic fingerprints (PMFs) from epileptic and healthy individuals for timely and accurate screening of epilepsy. The NELDI MS enabled high detection speed (∼30 s per sample), high throughput (up to 384 samples per run), and favorable reproducibility (coefficients of variation <15 %), acquiring high-performed PMFs. We next constructed an epilepsy diagnostic model by machine learning of PMFs, achieving desirable diagnostic capability with the area under the curve (AUC) value of 0.941 for the validation set. Furthermore, four metabolites were identified as a diagnostic biomarker panel for epilepsy, with an AUC value of 0.812-0.860. Our approach provides a high-performed and high-throughput platform for epileptic diagnostics, promoting the development of metabolic diagnostic tools in precision medicine.

Associations of multiple hydroxy-polycyclic aromatic hydrocarbons with serum levels of lipids in the workers from coking and non-ferrous smelting industries

Epidemiological evidence indicates that exposure to polycyclic aromatic hydrocarbons (PAHs) is associated with certain metabolic diseases. However, the relationship between PAHs and serum lipid profiles in exposed subjects remain unknown. Herein, the associations of multiple (8) urinary hydroxylated PAHs (OH-PAHs) in workers of coking (n = 655) and non-ferrous smelting (n = 614) industries with serum lipid levels (marking lipid metabolism) were examined. Multivariable linear regression, Bayesian kernel machine regression, and quantile g-computation were used. Most urinary OH-PAHs were significantly higher (p < 0.001) in coking workers than in non-ferrous smelting workers. In workers of both industries, OH-PAH exposure was associated with elevated levels of serum total cholesterol, total triglyceride, and low-density lipoprotein, as well as reduced high-density lipoprotein levels. Specifically, urinary 4-hydroxyphenanthrene was significantly positively associated with serum total cholesterol, total triglyceride, and low-density lipoprotein levels in non-ferrous smelting workers; however, the completely opposite association of 4-hydroxyphenanthrene with these lipid levels was observed in coking workers. The results of this pioneering examination suggest that exposure to OH-PAHs may contribute to dyslipidemia in coking and non-ferrous smelting workers, and distinct patterns of change were observed. Further prospective studies involving larger sample sizes are needed to further validate the findings.

Prenatal exposure to environmental phenols and fetal growth across pregnancy in the LIFECODES fetal growth study

INTRODUCTION

Environmental phenols are endocrine disrupting chemicals hypothesized to affect early life development. Previous research examining the effects of phenols on fetal growth has focused primarily on associations with measures of size at delivery. Few have included ultrasound measures to examine growth across pregnancy.

OBJECTIVE

Investigate associations between prenatal exposure to phenols and ultrasound and delivery measures of fetal growth.

METHODS

Using the LIFECODES Fetal Growth Study (n = 900), a case-cohort including 248 small-for-gestational-age, 240 large-for-gestational age, and 412 appropriate-for-gestational-age births, we estimated prenatal exposure to 12 phenols using three urine samples collected during pregnancy (median 10, 24, and 35 weeks gestation). We abstracted ultrasound and delivery measures of fetal growth from medical records. We estimated associations between pregnancy-average phenol biomarker concentrations and repeated ultrasound measures of fetal growth using linear mixed effects models and associations with birthweight using linear regression models. We also used logistic regression models to estimate associations with having a small- or large-for-gestational birth.

RESULTS

We observed positive associations between 2,4-dichlorophenol, benzophenone-3, and triclosan (TCS) and multiple ultrasound measures of fetal growth. For example, TCS was associated with a 0.09 (95 % CI: 0.01, 0.18) higher estimated fetal weight z-score longitudinally across pregnancy. This effect size corresponds to a 21 g increase in estimated fetal weight at 30 weeks gestation. Associations with delivery measures of growth were attenuated, but TCS remained positively associated with birthweight z-scores (mean difference: 0.13, 95 % CI: 0.02, 0.25). Conversely, methylparaben was associated with higher odds of a small-for-gestational age birth (odds ratio: 1.45, 95 % CI: 1.06, 1.98).

DISCUSSION

We observed associations between some biomarkers of phenol exposure and ultrasound measures of fetal growth, though associations at the time of delivery were attenuated. These findings are consistent with hypotheses that phenols have the potential to affect growth during the prenatal period.

Scientific Committee guidance on appraising and integrating evidence from epidemiological studies for use in EFSA's scientific assessments

EFSA requested its Scientific Committee to prepare a guidance document on appraising and integrating evidence from epidemiological studies for use in EFSA's scientific assessments. The guidance document provides an introduction to epidemiological studies and illustrates the typical biases, which may be present in different epidemiological study designs. It then describes key epidemiological concepts relevant for evidence appraisal. This includes brief explanations for measures of association, exposure assessment, statistical inference, systematic error and effect modification. The guidance then describes the concept of external validity and the principles of appraising epidemiological studies. The customisation of the study appraisal process is explained including tailoring of tools for assessing the risk of bias (RoB). Several examples of appraising experimental and observational studies using a RoB tool are annexed to the document to illustrate the application of the approach. The latter part of this guidance focuses on different steps of evidence integration, first within and then across different streams of evidence. With respect to risk characterisation, the guidance considers how evidence from human epidemiological studies can be used in dose-response modelling with several different options being presented. Finally, the guidance addresses the application of uncertainty factors in risk characterisation when using evidence from human epidemiological studies.

Organophosphate Ester Flame Retardants and Plasticizers in Relation to Fetal Growth in the LIFECODES Fetal Growth Study

BACKGROUND

Organophosphate esters (OPEs), used ubiquitously as flame retardants and plasticizers in consumer products, are suspected of having developmental toxicity.

OBJECTIVES

Our study aimed to estimate associations between prenatal exposure to OPEs and fetal growth, including both ultrasound (head circumference, abdominal circumference, femur length, and estimated fetal weight) and delivery [birth weight z-score, small-for-gestational age (SGA), and large-for-gestational age (LGA)] measures of growth.

METHODS

In the LIFECODES Fetal Growth Study (2008-2018), an enriched case-cohort of 900 babies born at the small and large ends of the growth spectrum, we quantified OPE biomarkers in three urine samples per pregnant participant and abstracted ultrasound and delivery measures of fetal growth from medical records. We estimated associations between pregnancy-averaged log-transformed OPE biomarkers and repeated ultrasound measures of fetal growth using linear mixed-effects models, and delivery measures of fetal growth using linear (birth weight) and logistic (SGA and LGA) regression models.

RESULTS

Most OPE biomarkers were positively associated with at least one ultrasound measure of fetal growth, but associations with delivery measures were largely null. For example, an interquartile range (IQR; 1.31  ng / mL ) increase in bis(2-chloroethyl) phosphate concentration was associated with larger z-scores in head circumference [mean difference (difference): 0.09; 95% confidence interval (CI): 0.01, 0.17], abdominal circumference (difference: 0.10; 95% CI: 0.02, 0.18), femur length (difference: 0.11; 95% CI: 0.03, 0.19), and estimated fetal weight (difference: 0.13; 95% CI: 0.04, 0.22) but not birth weight (difference: 0.04; 95% CI: - 0.08 , 0.17). At delivery, an IQR (1.00  ng / mL ) increase in diphenyl phosphate (DPHP) concentration was associated with an SGA birth (odds ratio: 1.46; 95% CI: 1.10, 1.94).

CONCLUSIONS

In a large prospective cohort, gestational OPE exposures were associated with larger fetal size during pregnancy, but associations at delivery were null. DPHP concentrations were associated with heightened risk of an SGA birth. These findings suggest that OPE exposure may affect fetal development. https://doi.org/10.1289/EHP14647.

Associations between use of chemical hair products and epigenetic age: Findings from the Sister Study

Background

Hair products may be a source of harmful chemicals and have been linked to age-related health outcomes. We investigated whether the use of hair products is related to epigenetic age in a sample of Black (both Hispanic and non-Hispanic) and non-Hispanic White women.

Methods

In a subset of 4358 participants aged 35-74 years from the Sister Study, we estimated cross-sectional associations between self-reported use of four chemical hair products (permanent dye, semipermanent dye, straighteners/relaxers, and hair permanents/body waves) in the year before enrollment (2003-2009) and three DNA methylation-based measures of epigenetic age (DunedinPACE, GrimAge age acceleration [GrimAgeAccel], and PhenoAge age acceleration [PhenoAgeAccel]) using survey-weighted multivariable linear regressions. Associations were estimated both overall and by self-identified race and ethnicity, adjusting for chronological age, socioeconomic and lifestyle factors, body mass index, menopausal status, and DNA methylation platform.

Results

Associations between the use of hair products and the three epigenetic age measures were largely null. Use of hair permanents/body waves was modestly associated with higher DunedinPACE among all participants (βever-never = 0.010; 95% confidence interval [CI] = 0.001, 0.019) and with lower PhenoAgeAccel among Black women (βever-never = -1.53; 95% CI = -2.84, -0.21).

Conclusion

In this US-based study, we found little evidence of associations between chemical hair product use and epigenetic age in Black and non-Hispanic White women. Observed associations were modest and largely not supported by dose-response relationships or were inconsistent across epigenetic age measures. Previously observed associations between chemical hair product use and aging-related health outcomes may not be explained by the biological aging pathways captured by DunedinPACE, GrimAgeAccel, or PhenoAgeAccel. Alternative biological pathways are worth investigating in racially diverse samples.

The accuracy of prehospital triage decisions in English trauma networks - a case-cohort study

BACKGROUND

Care for injured patients in England is provided by inclusive regional trauma networks. Ambulance services use triage tools to identify patients with major trauma who would benefit from expedited Major Trauma Centre (MTC) care. However, there has been no investigation of triage performance, despite its role in ensuring effective and efficient MTC care. This study aimed to investigate the accuracy of prehospital major trauma triage in representative English trauma networks.

METHODS

A diagnostic case-cohort study was performed between November 2019 and February 2020 in 4 English regional trauma networks as part of the Major Trauma Triage Study (MATTS). Consecutive patients with acute injury presenting to participating ambulance services were included, together with all reference standard positive cases, and matched to data from the English national major trauma database. The index test was prehospital provider triage decision making, with a positive result defined as patient transport with a pre-alert call to the MTC. The primary reference standard was a consensus definition of serious injury that would benefit from expedited major trauma centre care. Secondary analyses explored different reference standards and compared theoretical triage tool accuracy to real-life triage decisions.

RESULTS

The complete-case case-cohort sample consisted of 2,757 patients, including 959 primary reference standard positive patients. The prevalence of major trauma meeting the primary reference standard definition was 3.1% (n=54/1,722, 95% CI 2.3 - 4.0). Observed prehospital provider triage decisions demonstrated overall sensitivity of 46.7% (n=446/959, 95% CI 43.5-49.9) and specificity of 94.5% (n=1,703/1,798, 95% CI 93.4-95.6) for the primary reference standard. There was a clear trend of decreasing sensitivity and increasing specificity from younger to older age groups. Prehospital provider triage decisions commonly differed from the theoretical triage tool result, with ambulance service clinician judgement resulting in higher specificity.

CONCLUSIONS

Prehospital decision making for injured patients in English trauma networks demonstrated high specificity and low sensitivity, consistent with the targets for cost-effective triage defined in previous economic evaluations. Actual triage decisions differed from theoretical triage tool results, with a decreasing sensitivity and increasing specificity from younger to older ages.

Predictors of upstream inflammation and oxidative stress pathways during early pregnancy

BACKGROUND

Inflammation and oxidative stress are critical to pregnancy, but most human study has focused on downstream, non-causal indicators. Oxylipins are lipid mediators of inflammation and oxidative stress that act through many biological pathways. Our aim was to characterize predictors of circulating oxylipin concentrations based on maternal characteristics.

METHODS

Our study was conducted among 901 singleton pregnancies in the LIFECODES Fetal Growth Study, a nested case-cohort with recruitment from 2007 to 2018. We measured a targeted panel of oxylipins in early pregnancy plasma and urine samples from several biosynthetic pathways, defined by the polyunsaturated fatty acid (PUFA) precursor and enzyme group. We evaluated levels across predictors, including characteristics of participants' pregnancy, socioeconomic determinants, and obstetric and medical history.

RESULTS

Current pregnancy and sociodemographic characteristics were the most important predictors of circulating oxylipins concentrations. Plasma oxylipins were lower and urinary oxylipins higher for participants with a later gestational age at sampling (13-23 weeks), higher prepregnancy BMI (obesity class I, II, or III), Black or Hispanic race and ethnicity, and lower socioeconomic status (younger age, lower education, and uninsured). For example, compared to those with normal or underweight prepregnancy BMI, participants with class III prepregnancy obesity had 45-46% lower plasma epoxy-eicosatrienoic acids, the anti-inflammatory oxylipins produced from arachidonic acid (AA) by cytochrome P450, and had 81% higher urinary 15-series F2-isoprostanes, an indicator of oxidative stress produced from non-enzymatic AA oxidation. Similarly, in urine, Black participants had 92% higher prostaglandin E2 metabolite, a pro-inflammatory oxylipin, and 41% higher 5-series F2-isoprostane, an oxidative stress indicator.

CONCLUSIONS

In this large pregnancy study, we found that circulating levels of oxylipins were different for participants of lower socioeconomic status or of a systematically marginalized racial and ethnic groups. Given associations differed along biosynthetic pathways, results provide insight into etiologic links between maternal predictors and inflammation and oxidative stress.

Air pollution and epigenetic aging among Black and White women in the US

BACKGROUND

DNA methylation-based measures of biological aging have been associated with air pollution and may link pollutant exposures to aging-related health outcomes. However, evidence is inconsistent and there is little information for Black women.

OBJECTIVE

We examined associations of ambient particulate matter <2.5 μm and <10 μm in diameter (PM2.5 and PM10) and nitrogen dioxide (NO2) with DNA methylation, including epigenetic aging and individual CpG sites, and evaluated whether associations differ between Black and non-Hispanic White (NHW) women.

METHODS

Validated models were used to estimate annual average outdoor residential exposure to PM2.5, PM10, and NO2 in a sample of self-identified Black (n=633) and NHW (n=3493) women residing in the contiguous US. We used sampling-weighted generalized linear regression to examine the effects of pollutants on six epigenetic aging measures (primary: DunedinPACE, GrimAgeAccel, and PhenoAgeAccel; secondary: Horvath intrinsic epigenetic age acceleration [EAA], Hannum extrinsic EAA, and skin & blood EAA) and epigenome-wide associations for individual CpG sites. Wald tests of nested models with and without interaction terms were used to examine effect measure modification by race/ethnicity.

RESULTS

Black participants had higher median air pollution exposure than NHW participants. GrimAgeAccel was associated with both PM10 and NO2 among Black participants, (Q4 versus Q1, PM10: β=1.09, 95% CI: 0.16-2.03; NO2: β=1.01, 95% CI 0.08-1.94) but not NHW participants (p-for-heterogeneity: PM10=0.10, NO2=0.20). In Black participants, we also observed a monotonic exposure-response relationship between NO2 and DunedinPACE (Q4 versus Q1, NO2: β=0.029, 95% CI: 0.004-0.055; p-for-trend=0.03), which was not observed in NHW participants (p-for-heterogeneity=0.09). In the EWAS, pollutants were significantly associated with differential methylation at 19 CpG sites in Black women and one in NHW women.

CONCLUSIONS

In a US-wide cohort study, our findings suggest that air pollution is associated with DNA methylation alterations consistent with higher epigenetic aging among Black, but not NHW, women.

Temporal trends and predictors of gestational exposure to organophosphate ester flame retardants and plasticizers

BACKGROUND

Organophosphate esters (OPEs), used as flame retardants and plasticizers, are chemicals of concern for maternal and infant health. Prior studies examining temporal trends and predictors of OPE exposure are primarily limited by small sample sizes.

OBJECTIVES

Characterize temporal trends and predictors of OPE exposure biomarkers.

METHODS

We determined urinary concentrations of eight biomarkers of OPE exposure at three timepoints during pregnancy for participants in the LIFECODES Fetal Growth Study (n = 900), a nested case-cohort recruited between 2007 and 2018. We examined biomarker concentrations, their variability during pregnancy, and temporal trends over the study period. In addition, we identified sociodemographic and pregnancy characteristics associated with biomarker concentrations. Analyses were conducted using both the within-subject pregnancy geometric means and biomarker concentrations measured at individual study visits.

RESULTS

Five OPE biomarkers were detected in at least 60% of the study participants. Biomarkers were not strongly correlated with one another and intraclass correlation coefficients, measuring within-subject variability during pregnancy, ranged from 0.27 to 0.51. Biomarkers exhibited varying temporal trends across study years. For example, bis(1-chloro-2-propyl) phosphate (BCIPP) increased monotonically, whereas bis(1,3-dichloro-2-propyl) phosphate (BDCIPP) and diphenyl phosphate (DPHP), displayed non-monotonic trends with concentrations that peaked between 2011 and 2014. We observed associations between sociodemographic characteristics and OPE biomarkers. In general, concentrations of most OPE biomarkers were higher among participants from racial and ethnic minority populations, participants who were younger, had higher pre-pregnancy body mass index (BMI), and less than a college degree. We observed consistent results using either averaged or visit-specific biomarker concentrations.

SIGNIFICANCE

We observed widespread exposure to several OPEs and OPE biomarkers displayed varying temporal trends in pregnant people from 2007 to 2018. Concentrations of most OPE biomarkers varied according to sociodemographic factors, suggesting higher burdens of exposure among participants with higher pre-pregnancy BMI, those belonging to racial and ethnic minority populations, and lower educational attainment.

Low-density lipoprotein electronegativity and risk of death after acute coronary syndromes: A case-cohort analysis

BACKGROUND AND AIMS

Low-density lipoprotein (LDL)-cholesterol (LDL-C) promotes atherosclerotic cardiovascular disease (ASCVD), with changes in LDL electronegativity modulating its pro-atherogenic/pro-thrombotic effects. Whether such alterations associate with adverse outcomes in patients with acute coronary syndromes (ACS), a patient population at particularly high cardiovascular risk, remains unknown.

METHODS

This is a case-cohort study using data from a subset of 2619 ACS patients prospectively recruited at four university hospitals in Switzerland. Isolated LDL was chromatographically separated into LDL particles with increasing electronegativity (L1-L5), with the L1-L5 ratio serving as a proxy of overall LDL electronegativity. Untargeted lipidomics revealed lipid species enriched in L1 (least) vs. L5 (most electronegative subfraction). Patients were followed at 30 days and 1 year. The mortality endpoint was reviewed by an independent clinical endpoint adjudication committee. Multivariable-adjusted hazard ratios (aHR) were calculated using weighted Cox regression models.

RESULTS

Changes in LDL electronegativity were associated with all-cause mortality at 30 days (aHR, 2.13, 95% CI, 1.07-4.23 per 1 SD increment in L1/L5; p=.03) and 1 year (1.84, 1.03-3.29; p=.04), with a notable association with cardiovascular mortality (2.29; 1.21-4.35; p=.01; and 1.88; 1.08-3.28; p=.03). LDL electronegativity superseded several risk factors for the prediction of 1-year death, including LDL-C, and conferred improved discrimination when added to the updated GRACE score (area under the receiver operating characteristic curve 0.74 vs. 0.79, p=.03). Top 10 lipid species enriched in L1 vs. L5 were: cholesterol ester (CE) (18:2), CE (20:4), free fatty acid (FA) (20:4), phosphatidyl-choline (PC) (36:3), PC (34:2), PC (38:5), PC (36:4), PC (34:1), triacylglycerol (TG) (54:3), and PC (38:6) (all p < .001), with CE (18:2), CE (20:4), PC (36:3), PC (34:2), PC (38:5), PC (36:4), TG (54:3), and PC (38:6) independently associating with fatal events during 1-year of follow-up (all p < .05).

CONCLUSIONS

Reductions in LDL electronegativity are linked to alterations of the LDL lipidome, associate with all-cause and cardiovascular mortality beyond established risk factors, and represent a novel risk factor for adverse outcomes in patients with ACS. These associations warrant further validation in independent cohorts.

Methylation-Based Biological Age and Hypertension Prevalence and Incidence

BACKGROUND

Hypertension is common in older individuals and is a major risk factor for cardiovascular disease. Blood DNA methylation profiles have been used to derive metrics of biological age that capture age-related physiological change, disease risk, and mortality. The relationships between hypertension and DNA methylation-based biological age metrics have yet to be carefully described.

METHODS

Among 4419 women enrolled in the prospective Sister Study cohort, DNA methylation data generated from whole blood samples collected at baseline were used to calculate 3 biological age metrics (PhenoAgeAccel, GrimAgeAccel, DunedinPACE). Women were classified as hypertensive at baseline if they had high blood pressure (systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg) or reported current use of antihypertensive medication. New incident cases of hypertension during follow-up were identified via self-report on annual health questionnaires.

RESULTS

All 3 DNA methylation metrics of biological age were positively associated with prevalent hypertension at baseline (per 1-SD increase; PhenoAgeAccel, adjusted odds ratio, 1.16 [95% CI, 1.05-1.28]; GrimAgeAccel, adjusted odds ratio, 1.28 [95% CI, 1.14-1.45]; DunedinPACE, adjusted odds ratio, 1.16 [95% CI, 1.03-1.30]). Among 2610 women who were normotensive at baseline, women with higher biological age were more likely to be diagnosed with incident hypertension (per 1-SD increase; PhenoAgeAccel, adjusted hazard ratio, 1.09 [95% CI, 0.97-1.23]; GrimAgeAccel, adjusted hazard ratio, 1.16 [95% CI, 0.99-1.36]; DunedinPACE, adjusted hazard ratio, 1.16 [95% CI, 1.01-1.33]).

CONCLUSIONS

Methylation-based biological age metrics increase before a hypertension diagnosis and appear to remain elevated in the years after clinical diagnosis and treatment.

Temporal trends and predictors of phthalate, phthalate replacement, and phenol biomarkers in the LIFECODES Fetal Growth Study

BACKGROUND

Exposure to many phthalates and phenols is declining as replacements are introduced. There is little information on temporal trends or predictors of exposure to these newer compounds, such as phthalate replacements, especially among pregnant populations.

OBJECTIVE

Examine temporal trends and predictors of exposure to phthalates, phthalate replacements, and phenols using single- and multi-pollutant approaches.

METHODS

We analyzed data from 900 singleton pregnancies in the LIFECODES Fetal Growth Study, a nested case-cohort with recruitment from 2007 to 2018. We measured and averaged concentrations of 12 phthalate metabolites, four phthalate replacement metabolites, and 12 phenols in urine at three timepoints during pregnancy. We visualized and analyzed temporal trends and predictors of biomarker concentrations. To examine chemical mixtures, we derived clusters of individuals with shared exposure profiles using a finite mixture model and examined temporal trends and predictors of cluster assignment.

RESULTS

Exposure to phthalates and most phenols declined across the study period, while exposure to phthalate replacements (i.e., di(isononyl) cyclohexane-1,2-dicarboxylic acid, diisononyl ester [DINCH] and di-2-ethylhexyl terephthalate [DEHTP]) and bisphenol S (BPS) increased. For example, the sum of DEHTP biomarkers increased multiple orders of magnitude, with an average concentration of 0.92 ng/mL from 2007 to 2008 and 61.9 ng/mL in 2017-2018. Biomarkers of most chemical exposures varied across sociodemographic characteristics, with the highest concentrations observed in non-Hispanic Black or Hispanic participants relative to non-Hispanic White participants. We identified five clusters with shared exposure profiles and observed temporal trends in cluster membership. For example, at the end of the study period, a cluster characterized by high exposure to phthalate replacements was the most prevalent.

SIGNIFICANCE

In a large and well-characterized pregnancy cohort, we observed exposure to phthalate replacements and BPS increased over time while exposure to phthalates and other phenols decreased. Our results highlight the changing nature of exposure to consumer product chemical mixtures.

Vitamin D concentrations and breast cancer incidence among Black/African American and non-Black Hispanic/Latina women

BACKGROUND

Vitamin D may protect against breast cancer. Although Black/African American women and Hispanic/Latina women have lower circulating vitamin D levels than non-Hispanic White women, few studies have examined the association between vitamin D and breast cancer within these racial/ethnic groups.

METHODS

The vitamin D-breast cancer association was evaluated using a case-cohort sample of self-identified Black/African American and non-Black Hispanic/Latina women participating in the US-wide Sister Study cohort. Circulating 25-hydroxyvitamin D (25(OH)D) and 24,25-dihydroxyvitamin D (24,25(OH)2D) were measured using liquid chromatography-tandem mass spectrometry in blood samples collected at the baseline from 415 women (290 Black/African American women and 125 non-Black Hispanic/Latina women) who developed breast cancer. These were compared to concentrations in 1545 women (1084 Black/African American women and 461 Hispanic/Latina women) randomly selected from the cohort. Multivariable-adjusted Cox regression was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).

RESULTS

Over a mean follow-up of 9.2 years, women with circulating 25(OH)D concentrations above the clinical cut point for deficiency (20.0 ng/mL) had lower breast cancer rates than women with concentrations ≤ 20 ng/mL (HR, 0.79; 95% CI, 0.61-1.02). The inverse association was strongest among Hispanic/Latina women (HR, 0.52; 95% CI, 0.29-0.93), with a weaker association observed among Black/African American women (HR, 0.89; 95% CI, 0.68-1.18; P for heterogeneity = 0.13). There were no clear differences by menopausal status, follow-up time, estrogen receptor status, or invasiveness. Neither 24,25(OH)₂ D nor the 24,25(OH)₂ D to 25(OH)D ratio were independently associated with breast cancer risk.

CONCLUSIONS

This prospective study supports the hypothesis that vitamin D may be protective against breast cancer incidence in women, including non-Black Hispanic/Latina and Black/African American women.; LAY SUMMARY: Vitamin D may protect against breast cancer. Although women of color have lower average vitamin D levels than non-Hispanic White women, few studies have considered the role of race/ethnicity. In a sample of self-identified Black/African American and Hispanic/Latina women, we observed that vitamin D concentrations measured in blood were inversely associated with breast cancer, particularly among Latinas. These findings indicate that vitamin D may protect women against breast cancer, including those in racial/ethnic groups with low average circulating levels.