Differentiation between descending thoracic aortic diseases using machine learning and plasma proteomic signatures.

BACKGROUND: Descending thoracic aortic aneurysms and dissections can go undetected until severe and catastrophic, and few clinical indices exist to screen for aneurysms or predict risk of dissection. METHODS: This study generated a plasma proteomic dataset from 75 patients with descending type B dissection (Type B) and 62 patients with descending thoracic aortic aneurysm (DTAA). Standard statistical approaches were compared to supervised machine learning (ML) algorithms to distinguish Type B from DTAA cases. Quantitatively similar proteins were clustered based on linkage distance from hierarchical clustering and ML models were trained with uncorrelated protein lists across various linkage distances with hyperparameter optimization using fivefold cross validation. Permutation importance (PI) was used for ranking the most important predictor proteins of ML classification between disease states and the proteins among the top 10 PI protein groups were submitted for pathway analysis. RESULTS: Of the 1,549 peptides and 198 proteins used in this study, no peptides and only one protein, hemopexin (HPX), were significantly different at an adjusted p < 0.01 between Type B and DTAA cases. The highest performing model on the training set (Support Vector Classifier) and its corresponding linkage distance (0.5) were used for evaluation of the test set, yielding a precision-recall area under the curve of 0.7 to classify between Type B from DTAA cases. The five proteins with the highest PI scores were immunoglobulin heavy variable 6-1 (IGHV6-1), lecithin-cholesterol acyltransferase (LCAT), coagulation factor 12 (F12), HPX, and immunoglobulin heavy variable 4-4 (IGHV4-4). All proteins from the top 10 most important groups generated the following significantly enriched pathways in the plasma of Type B versus DTAA patients: complement activation, humoral immune response, and blood coagulation. CONCLUSIONS: We conclude that ML may be useful in differentiating the plasma proteome of highly similar disease states that would otherwise not be distinguishable using statistics, and, in such cases, ML may enable prioritizing important proteins for model prediction.

Increased phagocytosis in the presence of enhanced M2-like macrophage responses correlates with increased primary and latent HSV-1 infection.

After HSV-1 infection, macrophages infiltrate early into the cornea, where they play an important role in HSV-1 infection. Macrophages are divided into M1 or M2 groups based on their activation. M1 macrophages are pro-inflammatory, while M2 macrophages are anti-inflammatory. Macrophage phenotypes can shift between M1 or M2 in vitro and in vivo following treatment with specific cytokines. In this study we looked at the effect of M2 macrophages on HSV-1 infectivity using mice either lacking M2 (M2-/-) or overexpressing M2 (M2-OE) macrophages. While presence or absence of M2 macrophages had no effect on eye disease, we found that over expression of M2 macrophages was associated with increased phagocytosis, increased primary virus replication, increased latency, and increased expression of pro- and anti-inflammatory cytokines. In contrast, in mice lacking M2 macrophages following infection phagocytosis, replication, latency, and cytokine expression were similar to wild type mice. Our results suggest that enhanced M2 responses lead to higher phagocytosis, which affected both primary and latent infection but not reactivation.

Association of high-density lipoprotein particle concentration with cardiovascular risk following acute coronary syndrome: A case-cohort analysis of the dal-Outcomes trial.

BACKGROUND: High-density lipoprotein cholesterol (HDL-C) concentration is inversely related to risk of major adverse cardiovascular events (MACE) in epidemiologic studies but is a poorer predictor of MACE in patients with established coronary heart disease. HDL particle concentration (HDLP) has been proposed as a better predictor of risk. We investigated whether HDLP is associated with risk of MACE after acute coronary syndrome (ACS). METHODS: The dal-Outcomes trial compared the CETP inhibitor dalcetrapib with placebo in patients with recent ACS. In a nested case-cohort analysis, total, large, medium, and small HDLPs were measured by nuclear magnetic resonance spectroscopy at baseline (4-12 weeks after ACS) in 476 cases with MACE and 902 controls. Hazard ratios (HRs; case-control) for 1-SD increment of HDLP or HDL-C at baseline were calculated with and without adjustment for demographic, clinical, laboratory, and treatment variables. Similarly, HRs for MACE were calculated for changes in HDLP or HDL-C from baseline to month 3 of assigned treatment. RESULTS: Over median follow-up of 28 months, the risk of MACE was not associated with baseline HDLP (adjusted HR = 0.98, 95% CI = 0.84-1.15, P = .81), any HDLP subclass, or HDL-C. Dalcetrapib increased HDL-C and total, medium, and large HDLP and decreased small HDLP but had no effect on MACE compared with placebo. There were no association of risk of MACE with change in HDLP or HDL-C and no interaction with assigned study treatment. CONCLUSIONS: Neither baseline HDLP nor the change in HDLP on treatment with dalcetrapib or placebo was associated with risk of MACE after ACS.

Immunogenetics of Atherosclerosis-Link between Lipids, Immunity, and Genes.

PURPOSE OF REVIEW: Atherosclerosis is a complex disease process with lipid as a traditional modifiable risk factor and therapeutic target in treating atherosclerotic cardiovascular disease (ACVD). Recent evidence indicates that genetic influence and host immune response also are vital in this process. How these elements interact and modify each other and if immune response may emerge as a novel modifiable target remain poorly understood. RECENT FINDINGS: Numerous preclinical studies have clearly demonstrated that hypercholesterolemia is essential for atherogenesis, but genetic variations and host immune-inflammatory responses can modulate the pro-atherogenic effect of elevated LDL-C. Clinical studies also suggest that a similar paradigm may also be operational in atherogenesis in humans. More importantly each element modifies the biological behavior of the other two elements, forming a triangular relationship among the three. Modulating any one of them will have downstream impact on atherosclerosis. This brief review summarizes the relationship among lipids, genes, and immunity in atherogenesis and presents evidence to show how these elements affect each other. Modulation of immune response, though in its infancy, has a potential to emerge as a novel clinical strategy in treating ACVD.

Infections, atherosclerosis, and coronary heart disease.

Atherosclerosis is a chronic inflammatory disease. Pathophysiological similarities between chronic infections and atherosclerosis triggered interest in a clinical association between these conditions. Various infectious microbes have been linked to atherosclerotic vascular disease in epidemiological studies. However, this association failed to satisfy the Koch's postulates of causation with multiple clinical trials demonstrating inefficacy of anti-infective therapies in mitigating atherosclerotic cardiovascular events. Identification of underlying pathophysiological mechanisms and experience with vaccination against various infectious agents has ushered a new avenue of efforts in the development of an anti-atherosclerotic vaccine. Studies in animal models have identified various innate and adaptive immune pathways in atherosclerosis. In this review, we discuss the patho-biological link between chronic infections and atherosclerosis, evaluate existing evidence of animal and human trials on the association between infections and cardiovascular disease and introduce the concept of an anti-atherosclerotic vaccine.

Role of Interleukin-1 Signaling in a Mouse Model of Kawasaki Disease-Associated Abdominal Aortic Aneurysm.

OBJECTIVE: Kawasaki disease (KD) is the most common cause of acquired cardiac disease in US children. In addition to coronary artery abnormalities and aneurysms, it can be associated with systemic arterial aneurysms. We evaluated the development of systemic arterial dilatation and aneurysms, including abdominal aortic aneurysm (AAA) in the Lactobacillus casei cell-wall extract (LCWE)-induced KD vasculitis mouse model. METHODS AND RESULTS: We discovered that in addition to aortitis, coronary arteritis and myocarditis, the LCWE-induced KD mouse model is also associated with abdominal aorta dilatation and AAA, as well as renal and iliac artery aneurysms. AAA induced in KD mice was exclusively infrarenal, both fusiform and saccular, with intimal proliferation, myofibroblastic proliferation, break in the elastin layer, vascular smooth muscle cell loss, and inflammatory cell accumulation in the media and adventitia. Il1r(-/-), Il1a(-/-), and Il1b(-/-) mice were protected from KD associated AAA. Infiltrating CD11c(+) macrophages produced active caspase-1, and caspase-1 or NLRP3 deficiency inhibited AAA formation. Treatment with interleukin (IL)-1R antagonist (Anakinra), anti-IL-1α, or anti-IL-1ß mAb blocked LCWE-induced AAA formation. CONCLUSIONS: Similar to clinical KD, the LCWE-induced KD vasculitis mouse model can also be accompanied by AAA formation. Both IL-1α and IL-1ß play a key role, and use of an IL-1R blocking agent that inhibits both pathways may be a promising therapeutic target not only for KD coronary arteritis, but also for the other systemic arterial aneurysms including AAA that maybe seen in severe cases of KD. The LCWE-induced vasculitis model may also represent an alternative model for AAA disease.

Effects of dalcetrapib in patients with a recent acute coronary syndrome.

BACKGROUND: In observational analyses, higher levels of high-density lipoprotein (HDL) cholesterol have been associated with a lower risk of coronary heart disease events. However, whether raising HDL cholesterol levels therapeutically reduces cardiovascular risk remains uncertain. Inhibition of cholesteryl ester transfer protein (CETP) raises HDL cholesterol levels and might therefore improve cardiovascular outcomes. METHODS: We randomly assigned 15,871 patients who had had a recent acute coronary syndrome to receive the CETP inhibitor dalcetrapib, at a dose of 600 mg daily, or placebo, in addition to the best available evidence-based care. The primary efficacy end point was a composite of death from coronary heart disease, nonfatal myocardial infarction, ischemic stroke, unstable angina, or cardiac arrest with resuscitation. RESULTS: At the time of randomization, the mean HDL cholesterol level was 42 mg per deciliter (1.1 mmol per liter), and the mean low-density lipoprotein (LDL) cholesterol level was 76 mg per deciliter (2.0 mmol per liter). Over the course of the trial, HDL cholesterol levels increased from baseline by 4 to 11% in the placebo group and by 31 to 40% in the dalcetrapib group. Dalcetrapib had a minimal effect on LDL cholesterol levels. Patients were followed for a median of 31 months. At a prespecified interim analysis that included 1135 primary end-point events (71% of the projected total number), the independent data and safety monitoring board recommended termination of the trial for futility. As compared with placebo, dalcetrapib did not alter the risk of the primary end point (cumulative event rate, 8.0% and 8.3%, respectively; hazard ratio with dalcetrapib, 1.04; 95% confidence interval, 0.93 to 1.16; P=0.52) and did not have a significant effect on any component of the primary end point or total mortality. The median C-reactive protein level was 0.2 mg per liter higher and the mean systolic blood pressure was 0.6 mm Hg higher with dalcetrapib as compared with placebo (P<0.001 for both comparisons). CONCLUSIONS: In patients who had had a recent acute coronary syndrome, dalcetrapib increased HDL cholesterol levels but did not reduce the risk of recurrent cardiovascular events. (Funded by F. Hoffmann-La Roche; dal-OUTCOMES ClinicalTrials.gov number, NCT00658515.).

Efficacy and safety of a microsomal triglyceride transfer protein inhibitor in patients with homozygous familial hypercholesterolaemia: a single-arm, open-label, phase 3 study.

BACKGROUND: Patients with homozygous familial hypercholesterolaemia respond inadequately to existing drugs. We aimed to assess the efficacy and safety of the microsomal triglyceride transfer protein inhibitor lomitapide in adults with this disease. METHODS: We did a single-arm, open-label, phase 3 study of lomitapide for treatment of patients with homozygous familial hypercholesterolemia. Current lipid lowering therapy was maintained from 6 weeks before baseline through to at least week 26. Lomitapide dose was escalated on the basis of safety and tolerability from 5 mg to a maximum of 60 mg a day. The primary endpoint was mean percent change in levels of LDL cholesterol from baseline to week 26, after which patients remained on lomitapide through to week 78 for safety assessment. Percent change from baseline to week 26 was assessed with a mixed linear model. FINDINGS: 29 men and women with homozygous familial hypercholesterolaemia, aged 18 years or older, were recruited from 11 centres in four countries (USA, Canada, South Africa, and Italy). 23 of 29 enrolled patients completed both the efficacy phase (26 weeks) and the full study (78 weeks). The median dose of lomitapide was 40 mg a day. LDL cholesterol was reduced by 50% (95% CI -62 to -39) from baseline (mean 8·7 mmol/L [SD 2·9]) to week 26 (4·3 mmol/L [2·5]; p<0·0001). Levels of LDL cholesterol were lower than 2·6 mmol/L in eight patients at 26 weeks. Concentrations of LDL cholesterol remained reduced by 44% (95% CI -57 to -31; p<0·0001) at week 56 and 38% (-52 to -24; p<0·0001) at week 78. Gastrointestinal symptoms were the most common adverse event. Four patients had aminotransaminase levels of more than five times the upper limit of normal, which resolved after dose reduction or temporary interruption of lomitapide. No patient permanently discontinued treatment because of liver abnormalities. INTERPRETATION: Our study suggests that treatment with lomitapide could be a valuable drug in the management of homozygous familial hypercholesterolaemia. FUNDING: FDA Office of the Orphan Product Development, Aegerion Pharmaceuticals.

CD8(+)CD25(+) T cells reduce atherosclerosis in apoE(-/-) mice.

BACKGROUND: It is increasingly evident that CD8(+) T cells are involved in atherosclerosis but the specific subtypes have yet to be defined. CD8(+)CD25(+) T cells exert suppressive effects on immune signaling and modulate experimental autoimmune disorders but their role in atherosclerosis remains to be determined. The phenotype and functional role of CD8(+)CD25(+) T cells in experimental atherosclerosis were investigated in this study. METHODS AND RESULTS: CD8(+)CD25(+) T cells were observed in atherosclerotic plaques of apoE(-/-) mice fed hypercholesterolemic diet. Characterization by flow cytometric analysis and functional evaluation using a CFSE-based proliferation assays revealed a suppressive phenotype and function of splenic CD8(+)CD25(+) T cells from apoE(-/-) mice. Depletion of CD8(+)CD25(+) from total CD8(+) T cells rendered higher cytolytic activity of the remaining CD8(+)CD25(-) T cells. Adoptive transfer of CD8(+)CD25(+) T cells into apoE(-/-) mice suppressed the proliferation of splenic CD4(+) T cells and significantly reduced atherosclerosis in recipient mice. CONCLUSIONS: Our study has identified an athero-protective role for CD8(+)CD25(+) T cells in experimental atherosclerosis.

Use of microsomal triglyceride transfer protein inhibitors in patients with homozygous familial hypercholesterolemia: translating clinical trial experience into clinical practice.

Homozygous familial hypercholesterolemia (HoFH) is associated with severe hypercholesterolemia and premature cardiovascular morbidity and mortality. The most frequent cause of HoFH is loss of function mutations in the gene for the low-density lipoprotein receptor, resulting in reduced clearance of low-density lipoprotein (LDL) cholesterol from the circulation. Patients with HoFH have attenuated responsiveness to lipidlowering therapies such as statins, cholesterol absorption inhibition, and bile acid binding resins because of impaired LDL receptor expression. Lomitapide is a novel microsomal triglyceride transfer protein inhibitor that does not depend on the ability to upregulate LDL receptors on the surface of hepatocytes. Lomitapide reduces production of apolipoprotein B-containing lipoproteins, significantly reduces serum levels of LDL cholesterol, and is approved for use in patients with HoFH in the United States and the European Union.

Non-HDL cholesterol vs. apo B for risk of coronary heart disease in healthy individuals: the EPIC-Norfolk prospective population study.

BACKGROUND: There is an ongoing debate about the performance of non-high-density lipoprotein cholesterol (non-HDL-C) compared with apolipoprotein B (apo B) in the prediction of coronary heart disease (CHD) risk. Therefore, we compared the associations between non-HDL-C and apo B in regard to CHD among apparently healthy Western European individuals. DESIGN: In the European Prospective Investigation into Cancer and Nutrition (EPIC)-Norfolk prospective population study, 25,639 men and women aged 45-79 years were followed for 11·4 ± 2·8 years. Those with diabetes or prevalent CHD at baseline were excluded. A total of 2066 (12·1%) participants developed CHD during 195 692 person-years follow-up. RESULTS: The multivariable-adjusted hazard ratio [HR] of future CHD per one standard deviation increase was 1·22 [95% confidence interval (CI): 1·17-1·27] for LDL-C, 1·26 (95% CI 1·20-1·31) for non-HDL-C and 1·19 (95% CI 1·14-1·24) for apo B, respectively. The multivariable-adjusted HR of future CHD in the highest quartile LDL-C was 1·67 (95% CI: 1·47-1·91). For non-HDL-C and apo B, these respective HRs were 1·87 (95% CI: 1·62-2·15) and 1·56 (95% CI: 1·36-1·78). Kaplan-Meier survival analyses showed that there was incremental and comparable increase in risk of CHD with increasing quartiles of both non-HDL-C and apo B. CONCLUSIONS: In this prospective study, non-HDL-C and apo B were comparable in their ability to predict risk of future CHD.

Marked acceleration of atherosclerosis after Lactobacillus casei-induced coronary arteritis in a mouse model of Kawasaki disease.

OBJECTIVE: The purpose of this study was to investigate whether Lactobacillus casei cell wall extract-induced Kawasaki disease (KD) accelerates atherosclerosis in hypercholesterolemic mice. Method and Results- Apolipoprotein E knockout or low-density lipoprotein receptor knockout mice were injected with Lactobacillus casei cell wall extract (KD mice) or PBS, fed high-fat diet for 8 weeks, and atherosclerotic lesions in aortic sinuses, arch (AC), and whole aorta were assessed. KD mice had larger, more complex aortic lesions with abundant collagen, and both extracellular and intracellular lipid and foam cells, compared with lesions in control mice despite similar cholesterol levels. Both apolipoprotein E knockout KD and low-density lipoprotein receptor knockout KD mice showed dramatic acceleration in atherosclerosis versus controls, with increases in en face aortic atherosclerosis and plaque size in both the aortic sinuses and AC plaques. Accelerated atherosclerosis was associated with increased circulating interleukin-12p40, interferon-γ, tumor necrosis factor-α, and increased macrophage, dendritic cell, and T-cell recruitment in lesions. Furthermore, daily injections of the interleukin-1Ra, which inhibits Lactobacillus casei cell wall extract-induced KD vasculitis, prevented the acceleration of atherosclerosis. CONCLUSIONS: Our results suggest an important pathophysiologic link between coronary arteritis/vasculitis in the KD mouse model and subsequent atherosclerotic acceleration, supporting the concept that a similar relation may also be present in KD patients. These results also suggest that KD in childhood may predispose to accelerated and early atherosclerosis as adults.

Impaired tolerance to the autoantigen LL-37 in acute coronary syndrome.

Background: LL-37 is the only member of the cathelicidin family of antimicrobial peptides in humans and is an autoantigen in several autoimmune diseases and in acute coronary syndrome (ACS). In this report, we profiled the specific T cell response to the autoimmune self-antigen LL-37 and investigated the factors modulating the response in peripheral blood mononuclear cells (PBMCs) of healthy subjects and ACS patients. Methods and results: The activation induced marker (AIM) assay demonstrated differential T cell profiles characterized by the persistence of CD134 and CD137, markers that impair tolerance and promote immune effector and memory response, in ACS compared to Controls. Specifically, CD8+CD69+CD137+ T cells were significantly increased by LL-37 stimulation in ACS PBMCs. T effector cell response to LL-37 were either HLA dependent or independent as determined by blocking with monoclonal antibody to either Class-I HLA or Class-II HLA. Blocking of immune checkpoints PD-1 and CTLA-4 demonstrated the control of self-reactive T cell response to LL-37 was modulated predominantly by CTLA-4. Platelets from healthy controls down-modulated CD8+CD69+CD137+ T cell response to LL-37 in autologous PBMCs. CD8+CD69+CD137+ T cell AIM profile negatively correlated with platelet count in ACS patients. Conclusions: Our report demonstrates that the immune response to the autoantigen LL-37 in ACS patients is characterized specifically by CD8+CD69+CD137+ T cell AIM profile with persistent T cell activation and the generation of immunologic memory. The results provide potentially novel insight into mechanistic pathways of antigen-specific immune signaling in ACS.

Differentiation between Descending Thoracic Aortic Diseases using Machine Learning and Plasma Proteomic Signatures.

Background: Descending thoracic aortic aneurysms and dissections can go undetected until severe and catastrophic, and few clinical indices exist to screen for aneurysms or predict risk of dissection. Methods: This study generated a plasma proteomic dataset from 75 patients with descending type B dissection (Type B) and 62 patients with descending thoracic aortic aneurysm (DTAA). Standard statistical approaches were compared to supervised machine learning (ML) algorithms to distinguish Type B from DTAA cases. Quantitatively similar proteins were clustered based on linkage distance from hierarchical clustering and ML models were trained with uncorrelated protein lists across various linkage distances with hyperparameter optimization using 5-fold cross validation. Permutation importance (PI) was used for ranking the most important predictor proteins of ML classification between disease states and the proteins among the top 10 PI protein groups were submitted for pathway analysis. Results: Of the 1,549 peptides and 198 proteins used in this study, no peptides and only one protein, hemopexin (HPX), were significantly different at an adjusted p-value <0.01 between Type B and DTAA cases. The highest performing model on the training set (Support Vector Classifier) and its corresponding linkage distance (0.5) were used for evaluation of the test set, yielding a precision-recall area under the curve of 0.7 to classify between Type B from DTAA cases. The five proteins with the highest PI scores were immunoglobulin heavy variable 6-1 (IGHV6-1), lecithin-cholesterol acyltransferase (LCAT), coagulation factor 12 (F12), HPX, and immunoglobulin heavy variable 4-4 (IGHV4-4). All proteins from the top 10 most important correlated groups generated the following significantly enriched pathways in the plasma of Type B versus DTAA patients: complement activation, humoral immune response, and blood coagulation. Conclusions: We conclude that ML may be useful in differentiating the plasma proteome of highly similar disease states that would otherwise not be distinguishable using statistics, and, in such cases, ML may enable prioritizing important proteins for model prediction.

Functional Hypothalamic Amenorrhea and Preclinical Cardiovascular Disease.

CONTEXT: Endothelial dysfunction is a preclinical cardiovascular disease (CVD) marker. Due to various neuroendocrine aberrations, functional hypothalamic amenorrhea (FHA) may be a sex-specific risk factor for CVD in young women. OBJECTIVE: To investigate endothelial function in women with FHA, compared with eumenorrheic controls and recently menopausal women. METHODS: We performed a cross-sectional analysis among women with FHA (n = 30), eumenorrheic controls (n = 29), and recently menopausal women (n = 30). FHA was defined as amenorrhea ≥3 consecutive months, estradiol <50 pg/mL, follicle-stimulating hormone (FSH) < 10 mIU/mL, and luteinizing hormone (LH) < 10 mIU/mL, excluding other etiologies. Participants were recruited through obstetrics and gynecology referrals, social media advertising, and review of electronic health records. Preclinical CVD was measured using EndoPAT 2000 to calculate reactive hyperemic index (RHI). RHI ≤1.67 indicates endothelial dysfunction. RESULTS: Mean estradiol levels in women with FHA, as compared with eumenorrheic controls and recently menopausal women, were 29.0 ± 18.1, 46.4 ± 15.7, and 10.9 ± 14.4 pg/mL (P < .0001), respectively. Women with FHA had lower insulin (P = .0095) and higher cortisol (P = .0004) compared with controls. RHI was significantly lower in women with FHA compared with eumenorrheic controls and recently menopausal women (1.8 ± 0.5 vs 2.2 ± 0.5 vs 2.2 ± 0.6, respectively; P = .008), and 35% of women with FHA had RHI ≤1.67, consistent with endothelial dysfunction. CONCLUSION: These results demonstrate endothelial dysfunction in 1 out of 3 young women with FHA. FHA may be a contributor to preclinical CVD, and it is not explained by hypoestrogenemia alone.