Framingham Heart Study Participants Research Articles

The association between platelet reactivity and lipoprotein levels in Framingham Heart Study participants

BackgroundHypertriglyceridemia is an independent risk factor for major adverse cardiovascular events, though the mechanisms linking triglycerides and platelet function with thrombosis, remain elusive. The aim of this study was to assess the association between platelet function and triglyceride levels. MethodsWe included participants from the Framingham Heart Study Third Generation cohort, OMNI, and New Offspring Spouse cohort who attended the third examination cycle (2016–2019). Eligible participants were categorized into four triglyceride subgroups. ResultsThe study comprised a total of 1897 (55.53 %) participants with normal TG levels; 883 (25.85 %) participants with high-normal TGs; 378 (11.07 %) with borderline high TGs; and 258 (7.55 %) participants with hypertriglyceridemia. After adjusting for age, sex, alcohol consumption, aspirin, statin and P2Y12 inhibitors, the levels of ADP-induced platelet aggregation were inversely associated with total cholesterol levels (P < 0.0001). Platelet disaggregation was associated with low-density lipoprotein and high-density lipoprotein cholesterol levels (P < 0.0001). Lastly, in a shear-stress chamber assay mimicking arterial flow velocities, TG levels in the normal-high group were associated with increased levels of collagen-dependent thrombogenicity (β = 24.16, SE = 6.65, P < 0.0001). ConclusionTriglyceride levels are associated with altered platelet activation and aggregation. Furthermore, increased platelet-driven thrombogenicity is directly associated with triglyceride levels after adjusting for medications and other covariates.

Inflammatory biomarkers and MRI visible perivascular spaces: The Framingham Heart Study

We studied the association between inflammatory biomarkers and magnetic resonance imaging (MRI) visible perivascular spaces (PVS) in Framingham Heart Study participants free of stroke and dementia. PVS in the basal ganglia (BG) and centrum semiovale (CSO) were rated with validated methods and categorized based on counts. A mixed score of high PVS burden in neither, one or both regions was also evaluated. We related biomarkers representing various inflammatory mechanisms to PVS burden using multivariable ordinal logistic regression analysis accounting for vascular risk factors and other MRI markers of cerebral small vessel disease. Among 3604 participants (mean age 58±13 years, 47% males), significant associations were observed for intercellular adhesion molecule1, fibrinogen, osteoprotegerin, and P-selectin in relation to BG PVS, P-selectin for CSO PVS, and tumor necrosis factor receptor 2, osteoprotegerin and cluster of differentiation 40 ligand for mixed topography PVS. Therefore, inflammation may have a role in the pathogenesis of cerebral small vessel disease and perivascular drainage dysfunction represented by PVS, with different and shared inflammatory biomarkers depending on PVS topography.

Step Count, Self-reported Physical Activity, and Predicted 5-Year Risk of Atrial Fibrillation: Cross-sectional Analysis.

Physical inactivity is a known risk factor for atrial fibrillation (AF). Wearable devices, such as smartwatches, present an opportunity to investigate the relation between daily step count and AF risk. The objective of this study was to investigate the association between daily step count and the predicted 5-year risk of AF. Participants from the electronic Framingham Heart Study used an Apple smartwatch. Individuals with diagnosed AF were excluded. Daily step count, watch wear time (hours and days), and self-reported physical activity data were collected. Individuals' 5-year risk of AF was estimated, using the Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE)-AF score. The relation between daily step count and predicted 5-year AF risk was examined via linear regression, adjusting for age, sex, and wear time. Secondary analyses examined effect modification by sex and obesity (BMI≥30 kg/m2), as well as the relation between self-reported physical activity and predicted 5-year AF risk. We examined 923 electronic Framingham Heart Study participants (age: mean 53, SD 9 years; female: n=563, 61%) who had a median daily step count of 7227 (IQR 5699-8970). Most participants (n=823, 89.2%) had a <2.5% CHARGE-AF risk. Every 1000 steps were associated with a 0.08% lower CHARGE-AF risk (P<.001). A stronger association was observed in men and individuals with obesity. In contrast, self-reported physical activity was not associated with CHARGE-AF risk. Higher daily step counts were associated with a lower predicted 5-year risk of AF, and this relation was stronger in men and participants with obesity. The utility of a wearable daily step counter for AF risk reduction merits further investigation.

Abstract MP63: Higher Alternative Healthy Eating Index Scores Including Greater EPA and DHA Consumption Are Associated With Attenuated Platelet Function in the Framingham Heart Study

Introduction: Dietary modification that reflects healthy eating is a notable strategy for CVD prevention on a population level. The cardioprotective role diet plays in hemostasis and thrombosis, particularly in respect to platelet function remains unclear. Hypothesis: Healthier Alternative Healthy Eating Index (AHEI) scores associate with attenuated platelet reactivity in a large-scale population cohort. Methods: Cross-sectional data was obtained from Framingham Heart Study participants (N=3,429). AHEI scores were derived from self-reported Harvard semi-quantitative food frequency questionnaires. Ranging from 0 to 110, higher AHEI scores reflect healthier overall diet quality. Five bioassays measured platelet reactivity traits across several agonists in whole blood and platelet-rich plasma samples. Linear mixed effects models adjusted for age, sex, aspirin use, family relatedness, daily energy intake, and BMI evaluated associations between platelet reactivity and AHEI scores and components. Results: Higher AHEI scores associated with attenuated platelet reactivity among 16 different traits (P&lt;0.01), with notable decreased platelet aggregation to collagen, TRAP6 (PAR1 agonist), ADP, and ristocetin. Of the 11 AHEI components, higher EPA &amp; DHA intake was most strongly associated to decreased platelet function with the greatest number of significant associations to platelet traits (18 traits, P&lt;0.01) and moderate negative effect sizes (Figure 1). Conclusions: We report that higher AHEI scores associate with decreased platelet reactivity. Effects appear strongest for EPA &amp; DHA intake consistent with several small randomized controlled trials. However, we found that a holistic AHEI score reflects a better interpretation of diet as a modulator of platelet function compared to individual components. Our results conclude that the impact of a healthier diet in preventing thrombosis may be partially mediated through a decline in platelet reactivity through multiple independent activation pathways.

Abstract 125: Enlarged Perivascular Spaces In The Centrum Semiovale Are Associated With Incident Late-life Depression: Framingham Heart Study

Introduction: Perivascular spaces visible on brain MRI (PVS) are considered cerebral small vessel disease (CSVD) markers, possibly reflecting impaired brain fluid and metabolite clearance, and could contribute to depression in old age. This study aims to determine the association of PVS with prevalent and incident depression in the community-based Framingham Heart Study (FHS). Methods: FHS participants free of stroke and dementia with available Center for Epidemiological Studies Depression (CES-D) scale and PVS ratings in the basal ganglia (BG) and centrum semiovale (CSO) were included. Depression was defined as CES-D score≥16 or use of antidepressants. PVS were rated on a scale from I-IV in the BG and CSO using a validated method, with grades III-IV representing high PVS burden. A mixed score was created to describe high burden in the BG and/or CSO (0=no high ePVS burden, 1=high burden in the BG, 2=high burden in the CSO, 3=high burden in both regions). Logistic regression analysis was used to assess the association between PVS in the CSO, BG, and mixed regions and prevalent and incident depression. Models adjusted for age, sex, FSH cohort, time interval between CES-D score and MRI measurements, body mass index, smoking status, prevalent hypertension and diabetes. Results: Among 3815 participants included (mean age 57±14, male 47%), incident depression was observed in 19% over the follow up period ranging from 8 to 15 years subsequent to PVS rating. In multivariable adjusted analyses, high burden PVS in the CSO was associated with incident depression (OR 1.75 [1.18, 2.60], p=0.005). In the mixed score, concurrent high burden PVS in both regions (OR 2.10 [1.13, 3.91], p=0.02) and high burden in the CSO only (OR 1.65 [1.05, 2.58], p=0.03) were associated with incident depression. PVS were not associated with prevalent depression. Conclusion: In conclusion, high PVS burden was associated with incident depression in a community-based sample free of stroke and dementia. The association was predominantly driven by high burden of CSO PVS, suggesting that lobar forms of CSVD may play role in late life depression, possibly reflecting disrupted neuroanatomic circuits involved in mood control, and differing from BG PVS.

Multiple Prior Live Births Are Associated With Cardiac Remodeling and Heart Failure Risk in Women

Greater parity has been associated with cardiovascular disease risk. We sought to find whether the effects on cardiac remodeling and heart failure risk are clear. We examined the association of number of live births with echocardiographic measures of cardiac structure and function in participants of the Framingham Heart Study (FHS) using multivariable linear regression. We next examined the association of parity with incident heart failure with preserved (HFpEF) or reduced (HFrEF) ejection fraction using a Fine-Gray subdistribution hazards model in a pooled analysis of n = 12,635 participants in the FHS, the Cardiovascular Health Study, the Multi-Ethnic Study of Atherosclerosis, and Prevention of Renal and Vascular Endstage Disease. Secondary analyses included major cardiovascular disease, myocardia infarction and stroke. Among n = 3931 FHS participants (mean age 48 ± 13 years), higher numbers of live births were associated with worse left ventricular fractional shortening (multivariable β -1.11 (0.31); P = 0.0005 in ≥ 5 live births vs nulliparous women) and worse cardiac mechanics, including global circumferential strain and longitudinal and radial dyssynchrony (P < 0.01 for all comparing ≥ 5 live births vs nulliparity). When examining HF subtypes, women with ≥ 5 live births were at higher risk of developing future HFrEF compared with nulliparous women (HR 1.93, 95% CI 1.19-3.12; P = 0.008); by contrast, a lower risk of HFpEF was observed (HR 0.58, 95% CI 0.37-0.91; P = 0.02). Greater numbers of live births are associated with worse cardiac structure and function. There was no association with overall HF, but a higher number of live births was associated with greater risk for incident HFrEF.

Bitter- and Umami-Related Genes are Differentially Associated with Food Group Intakes: the Framingham Heart Study

BackgroundAs suboptimal diet quality remains the leading modifiable contributor to chronic disease risk, it is important to better understand the individual-level drivers of food choices. Recently, a genetic component of food choices was proposed based on variants (SNPs) in genes related to taste perception (taste-related SNPs). ObjectivesThis study aimed to determine the cumulative contribution of taste-related SNPs for basic tastes (bitter, sweet, umami, salt, and sour), summarized as “polygenic taste scores,” to food group intakes among adults. MethodsCross-sectional analyses were performed on 6230 Framingham Heart Study participants (mean age ± SD: 50 ± 14 y; 54% female). Polygenic taste scores were derived for tastes with ≥2 related SNPs identified in prior genome-wide association studies, and food group intakes (servings per week [sev/wk]) were tabulated from food frequency questionnaires. Associations were determined via linear mixed-effects models, using false discovery rates and bootstrap resampling to determine statistical significance. ResultsThirty-three taste-related SNPs (9 bitter, 19 sweet, 2 umami, 2 sour, 1 salt) were identified and used to derive polygenic taste scores for bitter, sweet, umami, and sour. Per additional allele for higher bitter perception, whole grain intakes were lower by 0.17 (95% CI: −0.28, −0.06) sev/wk, and for higher umami perception, total and red/orange vegetable intakes were lower by 0.73 (95% CI: −1.12, −0.34) and 0.25 (95% CI: −0.40, −0.10) sev/wk, respectively. Subsequent analyses at the SNP level identified four novel SNP-diet associations—two bitter-related SNPs with whole grains (rs10960174 and rs6782149) and one umami-related SNP with total and red/orange vegetables (rs7691456)—which may have been driving the identified associations. ConclusionsTaste-related genes for bitter and umami were differentially associated with food choices that may impact diet quality. Hence, a benefit could be derived from leveraging knowledge of taste-related genes when developing personalized risk reduction dietary guidance.

Assisted clustering of gene expression data using regulatory data from partially overlapping sets of individuals

As omics measurements profiled on different molecular layers are interconnected, integrative approaches that incorporate the regulatory effect from multi-level omics data are needed. When the multi-level omics data are from the same individuals, gene expression (GE) clusters can be identified using information from regulators like genetic variants and DNA methylation. When the multi-level omics data are from different individuals, the choice of integration approaches is limited. We developed an approach to improve GE clustering from microarray data by integrating regulatory data from different but partially overlapping sets of individuals. We achieve this through (1) decomposing gene expression into the regulated component and the other component that is not regulated by measured factors, (2) optimizing the clustering goodness-of-fit objective function. We do not require the availability of different omics measurements on all individuals. A certain amount of individual overlap between GE data and the regulatory data is adequate for modeling the regulation, thus improving GE clustering. A simulation study shows that the performance of the proposed approach depends on the strength of the GE-regulator relationship, degree of missingness, data dimensionality, sample size, and the number of clusters. Across the various simulation settings, the proposed method shows competitive performance in terms of accuracy compared to the alternative K-means clustering method, especially when the clustering structure is due mostly to the regulated component, rather than the unregulated component. We further validate the approach with an application to 8,902 Framingham Heart Study participants with data on up to 17,873 genes and regulation information of DNA methylation and genotype from different but partially overlapping sets of participants. We identify clustering structures of genes associated with pulmonary function while incorporating the predicted regulation effect from the measured regulators. We further investigate the over-representation of these GE clusters in pathways of other diseases that may be related to lung function and respiratory health. We propose a novel approach for clustering GE with the assistance of regulatory data that allowed for different but partially overlapping sets of individuals to be included in different omics data.

The Association of Protein Biomarkers With Incident Heart Failure With Preserved and Reduced Ejection Fraction.

Heart failure with preserved ejection fraction (HFpEF) and heart failure with reduced ejection fraction (HFrEF) are distinct clinical entities, yet there is scant evidence for associations of proteomic signatures with future development of HFpEF versus HFrEF. We evaluated the association of 71 protein biomarkers with incident HFpEF versus HFrEF (left ventricular ejection fraction ≥ versus <50%) among Framingham Heart Study participants using multivariable Cox models. Among 7038 participants (mean age 49 years; 54% women), 5 biomarkers were associated with increased risk of incident HFpEF (false discovery rate q<0.05): NT-proBNP (N-terminal pro-B-type natriuretic peptide; hazard ratio [HR], 2.13; 95% CI, 1.52-2.99; P<0.001), growth differentiation factor-15 (HR, 1.67; 95% CI, 1.32-2.12; P<0.001), adrenomedullin (HR, 1.58; 95% CI, 1.23-2.04; P<0.001), uncarboxylated matrix Gla protein (HR, 1.55; 95% CI 1.23-1.95; P<0.001), and C-reactive protein (HR, 1.46; 95% CI, 1.17-1.83; P=0.001). Fourteen biomarkers were associated with incident HFrEF (multivariable P<0.001, q<0.05 for all). Of these, 3 biomarkers were associated with both HF subtypes (NT-proBNP, growth differentiation factor-15, and C-reactive protein). When compared directly, myeloperoxidase, resistin, and paraoxanase-1 were more strongly associated with HFrEF than HFpEF. We identified 5 protein biomarkers of new-onset HFpEF representing pathways of inflammation, cardiac stress, and vascular stiffness, which partly overlapped with HFrEF. We found 14 biomarkers associated with new-onset HFrEF, with some distinct associations including myeloperoxidase, resistin, and paraoxanase-1. Taken together, these findings provide insights into similarities and differences in the development of HF subtypes. URL: https://clinicaltrials.gov/ct2/show/NCT00005121; Unique identifier: NCT0005121.

Epigenetic signatures of insulin resistance associated with Alzheimer’s Disease and related traits

AbstractBackgroundInsulin resistance (IR) is a major risk factor for Alzheimer’s disease (AD) and is primarily driven by obesity, another AD risk factor. The mechanism by which IR predisposes to AD is unknown. Epigenetic studies may help identify molecular signatures of IR associated with AD, thus contributing to an improved understanding of the biological and regulatory mechanisms of IR in AD.MethodWe conducted an epigenome‐wide association study of IR, quantified using the homeostasis model assessment of IR, in 3,167 Framingham Heart Study (FHS) participants without diabetes. We selected epigenetic markers (outcome) associated with IR at the genome‐wide level (P&lt;10‐7: 0.05/450,000) and evaluated their association with AD, vascular dementia, dementia, and brain volumes (hippocampal, total brain‐TBV, intracranial‐ICV, and lateral ventricular) in up to 3,040 FHS participants. We defined significant associations using P&lt;0.002 (0.05/3/7). All analyses were performed using blood DNA methylation measured using the Illumina HumanMethylation450 BeadChip and linear mixed‐effects models adjusted for age, sex, and batch, while allowing for heterogeneous variance across batches and accounting for familial relatedness. Models including IR or brain volume as a covariate were additionally adjusted for body mass index or ICV respectively.ResultWe confirmed the association of blood DNA methylation with IR at three loci (1p32:cg17901584‐DHCR24, 11q13:cg17058475‐CPT1A, and 21q22:cg06500161‐ABCG1). Blood DNA methylation at cg17058475‐CPT1A was significantly associated with TBV (P=0.0007). The carnitine palmitoyl transferase (CPT) system, crucial for the mitochondrial beta‐oxidation of long‐chain fatty acids, is involved in metabolic syndrome, type 2 diabetes, cardiovascular and neurological diseases including AD. Blood DNA methylation levels at cg17901584‐DHCR24 and cg06500161‐ABCG1 were nominally associated with AD (P=0.03) or vascular dementia (P=0.04) respectively. Both DHCR24 and ABCG1 are involved in cholesterol synthesis or transport. Reduced expression of DHCR24 occurs in AD patient’s temporal cortex while CSF capacity to promote cell cholesterol efflux via ABCG1 is specifically impaired in AD and this impairment correlates with the main AD neurobiochemical markers.ConclusionOur methylation analysis has shed light on genes potentially implicated in both IR and AD. Future work includes association analyses using brain omics from the AD Knowledge Portal.

Differentially methylated CpG sites between APOE ε4 carriers and non‐carriers in blood and brain from Alzheimer disease cases and controls

AbstractBackgroundThe APOE ε4 allele is the strongest genetic risk factor for late onset Alzheimer disease (AD). Methylation changes have been shown to occur early in AD, however, the influence of APOE genotype on methylation especially in relation to AD remains largely unknown.MethodWe evaluated differential methylation of genome‐wide CpG sites in brain between 178 APOE ε4 carriers and 511 ε4 non‐carriers in the Religious Orders Study and Memory and Aging Project (ROSMAP), in blood between 276 APOE ε4 carriers and 1118 ε4 non‐carriers in the Framingham Heart Study (FHS), and in blood between 114 APOE ε4 carriers and 187 ε4 non‐carriers from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Methylation patterns of the top‐ranked CpG sites were further examined in FHS participants with respect to cognitive and brain imaging traits.ResultThe most differentially methylated CpG sites among the autosomes were observed in the APOE region (chr19:45380000‐45430000) in all datasets. Significantly (P&lt;10‐8) hypomethylated CpG sites in APOE ε4 carriers compared to non‐carriers were observed from the APOC1 gene in the FHS (cg23270113, cg05644480) and ADNI (cg07773593) datasets (P=4.6x10‐9) and from the TOMM40 gene (cg02613937, P=1.3x10‐13) in the ROSMAP dataset. Significantly (P&lt;10‐8) hypermethylated CpG sites in APOE ε4 carriers compared to non‐carriers were observed from the APOE gene in the ROSMAP (cg14123992, cg04406254) and FHS (cg06750524) datasets. The two APOC1 CpG sites were significantly associated (P&lt;0.05) with amyloid‐beta 42 level in plasma and several cognitive tests (Trail A, Paired Associate Learning, Verbal Fluency Test, Boston Naming Test). APOE CpG site cg06750524 was significantly associated with MRI volumetric measures of the cerebrum (P=0.01), temporal cortex (P=0.02), and frontal cortex (P=0.05).ConclusionWe discovered novel CpG methylation differences in blood and brain between APOE ε4 carriers and non‐carriers that are associated with AD‐related traits especially involved in executive function and language. Further studies are required to link the methylation patterns to mechanisms leading to AD.

Systemic Inflammation is Associated with Cardiometabolic Risk Factors and Clinical Outcomes.

Assessing an individual's systemic inflammatory state is vital to understand inflammation's role in cardiometabolic diseases and identify those at the greatest risk of disease. We generated global inflammation scores and investigated their associations with cardiometabolic risk factors and adverse outcomes. Aggregate Inflammation Scores (AIS) and Principal Component Analysis (PCA) scores were generated for 7287 Framingham Heart Study participants using up to 26 inflammation-related proteins, with higher scores reflecting a pro-inflammatory milieu. Multivariable regression and proportional hazards analyses were conducted to investigate the associations of inflammation with cardiometabolic risk factors and outcomes. The primary outcomes for cross-sectional analyses included age, cigarette smoking, fasting lipid and glucose levels, blood pressure, body mass index (BMI), and hypertension, diabetes, and obesity. For prospective analyses, new-onset hypertension, diabetes, obesity, cardiovascular disease and all-cause mortality were investigated. Higher inflammation scores were associated with smoking and older age, higher BMI, systolic blood pressure, lipids, and glucose levels, and with greater odds of hypertension and diabetes after adjusting for age, sex, cohort, and BMI (all p < 0.001). Higher baseline scores were associated with greater odds of new-onset hypertension after adjusting for traditional risk factors (OR [95% CI] per one standard deviation [1-SD] increase, AIS: 1.33 [1.21-1.47], PCA score: 1.26 [1.12-1.42], p < 0.001). The AIS also was associated with new-onset diabetes (1.32 [1.14-1.52], p < 0.001). Proportional hazards analyses revealed greater risk of new-onset cardiovascular disease events and all-cause mortality (HR [95% CI] per 1-SD, AIS: 1.25 [1.14-1.37] and 1.32 [1.23-1.42], PCA score: 1.22 [1.13-1.33] and 1.40 [1.31-1.49], p < 0.001). Global inflammation scores encompassing an array of pro- and anti-inflammatory proteins and pathways may enhance risk assessment for cardiometabolic diseases. The AIS and PCA scores provide further opportunities to investigate the mechanisms of inflammation-related risk of disease.

The gut microbiome is associated with markers of cerebral small vessel disease

AbstractBackgroundCerebral small vessel disease (cSVD) is a major cause of cognitive impairment and dementia in the elderly. Dysregulation of the gut microbiome has been linked to cSVD, stroke, and dementia in human and animal studies. However, whether dysbiosis is associated with markers of cSVD is unclear. We aimed to assess the cross‐sectional association between the gut microbiome and MRI markers of cSVD and executive function in the Framingham Heart Study (FHS).MethodWe included FHS participants from the Third‐generation cohort without a history of stroke or dementia. We estimated the gut microbiota composition using 16S rRNA gene sequencing. MRI markers of cSVD included white matter hyperintensities (WMH) and peak width of skeletonized mean diffusivity (PSMD). Executive function was assessed as the difference between the Trail Making Test (TMT) part B and A. We used multivariable regression analysis to evaluate the association between the gut microbiota and WMH, PSMD, and TMT, adjusting for age, age2, sex, BMI, and education. Additionally, we used differential abundance analysis to explore extremes of cSVD markers dichotomized by their distribution with the upper quintile representing the least healthy group. Finally, we used PICRUSt (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States) to gain insight into the functional profiles of the underlying microbial communities.ResultOur sample included 972 participants (age: 55 ± 10 years, 56% female). Using multivariable models, we found that greater abundance of genera Pseudobutyrivibrio and Ruminococcus were associated with lower (healthier) WMH and PSMD (p‐values &lt; 0.001), as well as better executive function (p‐values &lt; 0.01). Furthermore, the differential analysis indicated a lower abundance of the genus Prevotella in the group of participants with the highest (least healthy) PSMD and WHM (p‐values &lt; 0.05). Finally, functional analyses using PICRUSt implicated various KEGG pathways, including carbohydrate metabolism, glycosaminoglycan, and fatty acid biosynthesis previously associated with cognitive performance.ConclusionOur study suggests abundance of certain gut bacteria in genus Pseudobutyrivibrio, Ruminococcus, and Prevotella, that produce butyric acid and other metabolites used by the human brain, are associated with healthier brain vasculature and better cognitive function in middle‐aged healthy adults.

Blood Pressure Responses During Exercise: Physiological Correlates and Clinical Implications.

Abnormal blood pressure (BP) responses to exercise can predict adverse cardiovascular outcomes, but their optimal measurement and definitions are poorly understood. We combined frequently sampled BP during cardiopulmonary exercise testing with vascular stiffness assessment to parse cardiac and vascular components of exercise BP. Cardiopulmonary exercise testing with BP measured every two minutes and resting vascular tonometry were performed in 2858 Framingham Heart Study participants. Linear regression was used to analyze sex-specific exercise BP patterns as a function of arterial stiffness (carotid-femoral pulse wave velocity) and cardiac-peripheral performance (defined by peak O2 pulse). Our sample was balanced by sex (52% women) with mean age 54±9 years and 47% with hypertension. We observed variability in carotid-femoral pulse wave velocity and peak O2 pulse across individuals with clinically defined exercise hypertension (peak systolic BP [SBP] in men ≥210 mm Hg; in women ≥190 mm Hg). Despite similar resting SBP and cardiometabolic profiles, individuals with higher peak O2 pulse displayed higher peak SBP (P≤0.017) alongside higher fitness levels (P<0.001), suggesting that high peak exercise SBP in the context of high peak O2 pulse may in fact be favorable. Although both higher (favorable) O2 pulse and higher (adverse) arterial stiffness were associated with greater peak SBP (P<0.0001 for both), the magnitude of association of carotid-femoral pulse wave velocity with peak SBP was higher in women (sex-carotid-femoral pulse wave velocity interaction P<0.0001). In sex-specific models, exercise SBP measures accounting for workload (eg, SBP during unloaded exercise, SBP at 75 watts, and SBP/workload slope) were directly associated with the adverse features of greater arterial stiffness and lower peak O2 pulse. Higher peak exercise SBP reflects a complex trade-off between arterial stiffness and cardiac-peripheral performance that differs by sex. Studies of BP responses to exercise accounting for vascular and cardiac physiology may illuminate mechanisms of hypertension and clarify clinical interpretation of exercise BP.

Elucidating the genetic architecture of DNA methylation to identify promising molecular mechanisms of disease

DNA methylation commonly occurs at cytosine-phosphate-guanine sites (CpGs) that can serve as biomarkers for many diseases. We analyzed whole genome sequencing data to identify DNA methylation quantitative trait loci (mQTLs) in 4126 Framingham Heart Study participants. Our mQTL mapping identified 94,362,817 cis-mQTLvariant-CpG pairs (for 210,156 unique autosomal CpGs) at P < 1e−7 and 33,572,145 trans-mQTL variant-CpG pairs (for 213,606 unique autosomal CpGs) at P < 1e−14. Using cis-mQTL variants for 1258 CpGs associated with seven cardiovascular disease (CVD) risk factors, we found 104 unique CpGs that colocalized with at least one CVD trait. For example, cg11554650 (PPP1R18) colocalized with type 2 diabetes, and was driven by a single nucleotide polymorphism (rs2516396). We performed Mendelian randomization (MR) analysis and demonstrated 58 putatively causal relations of CVD risk factor-associated CpGs to one or more risk factors (e.g., cg05337441 [APOB] with LDL; MR P = 1.2e−99, and 17 causal associations with coronary artery disease (e.g. cg08129017 [SREBF1] with coronary artery disease; MR P = 5e−13). We also showed that three CpGs, e.g., cg14893161 (PM20D1), are putatively causally associated with COVID-19 severity. To assist in future analyses of the role of DNA methylation in disease pathogenesis, we have posted a comprehensive summary data set in the National Heart, Lung, and Blood Institute’s BioData Catalyst.

Abstract 13538: Associations Between Circulating Liver Enzymes and Platelet Reactivity

Introduction: The liver plays an essential role in thrombopoiesis and coagulation, and circulating liver enzymes are associated with altered platelet counts. However, the role of liver enzymes in the modulation of platelet reactivity remains unclear. Methods: A total of 3,429 Framingham Heart Study participants were included in this study. Platelet reactivity to 8 agonists, including adenosine diphosphate (ADP), thromboxane A 2 mimetic (U46619) and arachidonic acid (AA), was measured using light transmission aggregometry (LTA), Multiplate impedance aggregometry and Optimul aggregometry. Platelet counts and platelet-red blood cell aggregates were enumerated using flow cytometry. Levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were also measured. Associations of platelet reactivity phenotypes with liver circulating liver enzyme levels analyzed using linear mixed-effects models and adjusted for sex, age, relatedness, smoking, aspirin-, and statin-use. To correct for multiple testing, a Bonferroni-correction threshold of P&lt;0.0011 was used to account for multiple comparisons. Results: Higher ALT levels were associated with platelet disaggregation in LTA to ADP (β=0.056, SE=0.02, P=3.54E-04), and with decreased platelet reactivity in response to U46619 in Optimul (EC 50 : β=0.064, SE=0.019, P=7.65E-04). Interestingly, higher AST levels were associated with decreased platelet counts (β=-0.082, SE=0.019, P=1.04E-05), platelet reactivity to U46619 (β=-0.069, SE=0.018, P=1.04E-04) and AA (β=-0.047, SE=0.013, P=3.12E-04), and circulating platelet-red blood cell aggregates (β=-0.062, SE=0.018, P=9.54E-04). Conclusion: Our findings suggest that ALT levels could alter primary platelet aggregation. AST may indirectly modulate platelet counts. In addition, lower AST levels may contribute to platelet hyperresponsiveness. Lastly, ALT and AST levels may have antagonistic effects on the formation of platelet and red blood cell aggregates, which may indirectly influence platelet margination within vessels. Additional studies are underway to examine whether liver fat, a marker of fibrosis associated with cardiovascular disease, is also associated with altered platelet function.

Abstract 10538: Metabolite Profiling Highlights Links Between Healthy Diet and Greater Cardiorespiratory Fitness in the Community

Introduction: Cardiorespiratory fitness (CRF) is a powerful predictor of health outcomes, but its determinants are incompletely understood at the population level. Prior studies have suggested associations between dietary components and CRF, but they have generally been limited by small sample sizes, reliance on estimated CRF, variable dietary instruments, and lack of molecular interrogation to support shared biology. Hypothesis: Shared metabolite associations across dietary quality and CRF will prioritize metabolites potentially responsible for linking dietary quality with improved CRF in the community. Methods: Framingham Heart Study (FHS) participants underwent maximum effort cardiopulmonary exercise tests for quantification of CRF by peak oxygen uptake (VO 2 ) and completed food frequency questionnaires at a routine study visit. Dietary quality was assessed by the Mediterranean-style Diet Score (MDS), and fasting levels of 201 metabolites were quantified. Results: In 2380 FHS participants (age 54 ± 9 years, 54% women, BMI 28 ± 5 kg/m 2 ), a 1-SD higher MDS was associated with 4.4% (3.5% - 5.3%, p &lt; 0.0001) greater peak VO 2 in linear models adjusted for age, sex, cardiovascular risk factors, and physical activity. In a representative subsample that had metabolite profiling (N = 1154), 80 metabolites were statistically significantly associated with MDS, 57 with peak VO 2 , and 27 with both MDS and peak VO 2 in multivariable linear models (FDR &lt; 5% for all, Figure). These metabolites included dimethylguanidino valeric acid, medium chain acylcarnitines, and C16 ceramide, which are markers of insulin resistance and increased risk for cardiovascular disease. Metabolites with less established functions include C38:7 plasmalogens, which correlate with fish intake and may have anti-inflammatory effects. Conclusions: Healthy diet is associated with greater CRF in the community and may be partially explained by favorable metabolic health.

Association of Symptoms After COVID-19 Vaccination With Anti–SARS-CoV-2 Antibody Response in the Framingham Heart Study

This cohort study examines the association of self-reported postvaccination symptoms with anti–SARS-CoV-2 antibody response among Framingham Heart Study participants contributing to the Collaborative Cohort of Cohorts for COVID-19 Research study.

Lighter sleep is associated with higher enlarged perivascular spaces burden in middle-aged and elderly individuals

BackgroundWhile healthy sleep is suggested to promote glymphatic clearance in the brain, poorer sleep may be associated with higher enlarged perivascular spaces (ePVS) burden, potentially representing impaired perivascular drainage. This study aims to evaluate the association between ePVS burden and polysomnographic sleep characteristics in a large community-based sample. Methods552 dementia and stroke-free Framingham Heart Study participants (age: 58.6 ± 8.9 years; 50.4% men) underwent a full-night in-home polysomnography. Three years later on average, participants underwent a brain MRI. ePVS were rated in the basal ganglia and centrum semiovale, and dichotomized as low burden (<20 counts, grades 1 and 2) or high burden (>20 counts, grades 3 and 4). Logistic regression analyses relating sleep variables to subsequent ePVS burden were used, adjusted for age, sex, time interval between polysomnography and MRI, ApoE ε4 allele carrier status, hypertension, and smoking. ResultsLonger N1 sleep and shorter N3 sleep duration were associated with higher ePVS burden in the centrum semiovale. When stratifying these associations by subpopulations, longer N1 sleep duration with ePVS burden was observed especially in older individuals and hypertensive participants. Associations between ePVS burden and other sleep characteristics such as total sleep time and REM sleep duration varied according to ApoE ε4 allele carrier status. ConclusionsLighter sleep, as characterized by longer N1 sleep and shorter slow-wave sleep, is associated with higher ePVS burden. These findings suggest that sleep architecture may be involved in glymphatic clearance and cerebral small vessel disease, which could be an important biological link between sleep and dementia risk.

Response of circulating metabolites to an oral glucose challenge and risk of cardiovascular disease and mortality in the community

BackgroundNew biomarkers to identify cardiovascular disease (CVD) risk earlier in its course are needed to enable targeted approaches for primordial prevention. We evaluated whether intraindividual changes in blood metabolites in response to an oral glucose tolerance test (OGTT) may provide incremental information regarding the risk of future CVD and mortality in the community.MethodsAn OGTT (75 g glucose) was administered to a subsample of Framingham Heart Study participants free from diabetes (n = 361). Profiling of 211 plasma metabolites was performed from blood samples drawn before and 2 h after OGTT. The log2(post/pre) metabolite levels (Δmetabolites) were related to incident CVD and mortality in Cox regression models adjusted for age, sex, baseline metabolite level, systolic blood pressure, hypertension treatment, body mass index, smoking, and total/high-density lipoprotein cholesterol. Select metabolites were related to subclinical cardiometabolic phenotypes using Spearman correlations adjusted for age, sex, and fasting metabolite level.ResultsOur sample included 42% women, with a mean age of 56 ± 9 years and a body mass index of 30.2 ± 5.3 kg/m2. The pre- to post-OGTT changes (Δmetabolite) were non-zero for 168 metabolites (at FDR ≤ 5%). A total of 132 CVD events and 144 deaths occurred during median follow-up of 24.9 years. In Cox models adjusted for clinical risk factors, four Δmetabolites were associated with incident CVD (higher glutamate and deoxycholate, lower inosine and lysophosphatidylcholine 18:2) and six Δmetabolites (higher hydroxyphenylacetate, triacylglycerol 56:5, alpha-ketogluturate, and lower phosphatidylcholine 32:0, glucuronate, N-monomethyl-arginine) were associated with death (P < 0.05). Notably, baseline metabolite levels were not associated with either outcome in models excluding Δmetabolites. The Δmetabolites exhibited varying cross-sectional correlation with subclinical risk factors such as visceral adiposity, insulin resistance, and vascular stiffness, but overall relations were modest. Significant Δmetabolites included those with established roles in cardiometabolic disease (e.g., glutamate, alpha-ketoglutarate) and metabolites with less defined roles (e.g., glucuronate, lipid species).ConclusionsDynamic changes in metabolite levels with an OGTT are associated with incident CVD and mortality and have potential relevance for identifying CVD risk earlier in its development and for discovering new potential therapeutic targets.