Using machine learning to predict carotid artery symptoms from CT angiography: A radiomics and deep learning approach.

Purpose: To assess radiomics and deep learning (DL) methods in identifying symptomatic Carotid Artery Disease (CAD) from carotid CT angiography (CTA) images. We further compare the performance of these novel methods to the conventional calcium score. Methods: Carotid CT angiography (CTA) images from symptomatic patients (ischaemic stroke/transient ischaemic attack within the last 3 months) and asymptomatic patients were analysed. Carotid arteries were classified into culprit, non-culprit and asymptomatic. The calcium score was assessed using the Agatston method. 93 radiomic features were extracted from regions-of-interest drawn on 14 consecutive CTA slices. For DL, convolutional neural networks (CNNs) with and without transfer learning were trained directly on CTA slices. Predictive performance was assessed over 5-fold cross validated AUC scores. SHAP and GRAD-CAM algorithms were used for explainability. Results: 132 carotid arteries were analysed (41 culprit, 41 non-culprit, and 50 asymptomatic). For asymptomatic vs symptomatic arteries, radiomics attained a mean AUC of 0.96(± 0.02), followed by DL 0.86(± 0.06) and then calcium 0.79(± 0.08). For culprit vs non-culprit arteries, radiomics achieved a mean AUC of 0.75(± 0.09), followed by DL 0.67(± 0.10) and then calcium 0.60(± 0.02). For multi-class classification, the mean AUCs were 0.95(± 0.07), 0.79(± 0.05), and 0.71(± 0.07) for radiomics, DL and calcium, respectively. Explainability revealed consistent patterns in the most important radiomic features. Conclusions: Our study highlights the potential of novel image analysis techniques in extracting quantitative information beyond calcification in the identification of CAD. Though further work is required, the transition of these novel techniques into clinical practice may eventually facilitate better stroke risk stratification.

COVID-19 related myocardial injury is associated with immune dysregulation in symptomatic patients with cardiac MRI abnormalities.

AIMS: While acute cardiovascular complications of COVID-19 are well-described, less is known about longer-term cardiac sequelae. For many individuals with cardiac signs or symptoms arising after COVID-19 infection, the aetiology remains unclear. We examined immune profiles associated with magnetic resonance imaging (MRI) abnormalities in patients with unexplained cardiac injury after COVID-19. METHODS AND RESULTS: Twenty-one participants (mean age 47 [SD 13] years, 71% female) with long COVID (n=17), raised troponin (n=2), or unexplained new-onset heart failure (n=2), who did not have pre-existing heart conditions or recent steroid/immunosuppression treatment were enrolled a mean 346 (SD 191) days after COVID-19 infection in a prospective observational study. Cardiac MRI and blood sampling for deep immunophenotyping using mass cytometry by time of flight and measurement of proteomic inflammatory markers was performed. Nine of 21 (43%) participants had MRI abnormalities (MRI(+)), including non-ischaemic patterns of late gadolinium enhancement and/or visually overt myocardial oedema in 8 people. One patient had mildly impaired biventricular function without fibrosis or oedema, and 2 had severe left ventricular impairment. MRI(+) individuals had higher blood CCL3, CCL7, FGF-23 and CD4 Th2 cells, and lower CD8 T effector memory (TEM) cells, than MRI(-). Cluster analysis revealed lower expression of inhibitory receptors PD1 and TIM3 in CD8 TEM cells from MRI(+) patients than MRI(-) patients, and functional studies of CD8 T αß cells showed higher proportions of cytotoxic granzyme B+ secreting cells upon stimulation. CD8 TEM cells and CCL7 were the strongest predictors of MRI abnormalities in a LASSO regression model (composite AUC 0.96, 95%CI 0.88-1.0). CCL7 was correlated with diffuse myocardial fibrosis/oedema detected by quantitative T1 mapping (r=0.47, p=0.04). CONCLUSION: COVID-19 related cardiac injury in symptomatic patients with non-ischaemic myocarditis-like MRI abnormalities is associated with immune dysregulation, including decreased peripheral CD8 TEM cells and increased CCL7, persisting long after the initial infection.

Low-dose IL-2 enhances the generation of IL-10-producing immunoregulatory B cells.

Dysfunction of interleukin-10 producing regulatory B cells has been associated with the pathogenesis of autoimmune diseases, but whether regulatory B cells can be therapeutically induced in humans is currently unknown. Here we demonstrate that a subset of activated B cells expresses CD25, and the addition of low-dose recombinant IL-2 to in vitro stimulated peripheral blood and splenic human B cells augments IL-10 secretion. Administration of low dose IL-2, aldesleukin, to patients increases IL-10-producing B cells. Single-cell RNA sequencing of circulating immune cells isolated from low dose IL2-treated patients reveals an increase in plasmablast and plasma cell populations that are enriched for a regulatory B cell gene signature. The transcriptional repressor BACH2 is significantly down-regulated in plasma cells from IL-2-treated patients, BACH2 binds to the IL-10 gene promoter, and Bach2 depletion or genetic deficiency increases B cell IL-10, implicating BACH2 suppression as an important mechanism by which IL-2 may promote an immunoregulatory phenotype in B cells.

Low-dose interleukin 2 for the reduction of vascular inflammation in acute coronary syndromes (IVORY): protocol and study rationale for a randomised, double-blind, placebo-controlled, phase II clinical trial.

INTRODUCTION: Inflammation plays a critical role in the pathogenesis of atherosclerosis, the leading cause of ischaemic heart disease (IHD). Studies in preclinical models have demonstrated that an increase in regulatory T cells (Tregs), which have a potent immune modulatory action, led to a regression of atherosclerosis. The Low-dose InterLeukin 2 (IL-2) in patients with stable ischaemic heart disease and Acute Coronary Syndromes (LILACS) study, established the safety of low-dose IL-2 and its biological efficacy in IHD. The IVORY trial is designed to assess the effects of low-dose IL-2 on vascular inflammation in patients with acute coronary syndromes (ACS). METHODS AND ANALYSIS: In this study, we hypothesise that low-dose IL-2 will reduce vascular inflammation in patients presenting with ACS. This is a double-blind, randomised, placebo-controlled, phase II clinical trial. Patients will be recruited across two centres, a district general hospital and a tertiary cardiac centre in Cambridge, UK. Sixty patients with ACS (unstable angina, non-ST elevation myocardial infarction or ST elevation myocardial infarction) with high-sensitivity C reactive protein (hsCRP) levels >2 mg/L will be randomised to receive either 1.5×106 IU of low-dose IL-2 or placebo (1:1). Dosing will commence within 14 days of admission. Dosing will comprise of an induction and a maintenance phase. 2-Deoxy-2-[fluorine-18] fluoro-D-glucose (18F-FDG) positron emission tomography/CT (PET/CT) scans will be performed before and after dosing. The primary endpoint is the change in mean maximum target to background ratios (TBRmax) in the index vessel between baseline and follow-up scans. Changes in circulating T-cell subsets will be measured as secondary endpoints of the study. The safety and tolerability of extended dosing with low-dose IL-2 in patients with ACS will be evaluated throughout the study. ETHICS AND DISSEMINATION: The Health Research Authority and Health and Care Research Wales, UK (19/YH/0171), approved the study. Written informed consent is required to participate in the trial. The results will be reported through peer-reviewed journals and conference presentations. TRIAL REGISTRATION NUMBER: NCT04241601.

Regulatory T-Cell Response to Low-Dose Interleukin-2 in Ischemic Heart Disease.

BACKGROUND: Atherosclerosis is a chronic inflammatory disease of the artery wall. Regulatory T cells (Tregs) limit inflammation and promote tissue healing. Low doses of interleukin (IL)-2 have the potential to increase Tregs, but its use is contraindicated for patients with ischemic heart disease. METHODS: In this randomized, double-blind, placebo-controlled, dose-escalation trial, we tested low-dose subcutaneous aldesleukin (recombinant IL-2), given once daily for 5 consecutive days. In study part A, the primary end point was safety, and patients with stable ischemic heart disease were randomly assigned to receive placebo or to one of five dose groups (range, 0.3 to 3.0 × 106 IU daily). In study part B, patients with acute non-ST elevation myocardial infarction or unstable angina were randomly assigned to receive placebo or to one of two dose groups (1.5 and 2.5 × 106 IU daily). The coprimary end points were safety and the dose required to increase circulating Tregs by 75%. Single-cell RNA-sequencing of circulating immune cells was used to provide a mechanistic assessment of the effects of aldesleukin. RESULTS: Forty-four patients were randomly assigned to either study part A (n=26) or part B (n=18). In total, 3 patients withdrew before dosing, 27 received active treatment, and 14 received placebo. The majority of adverse events were mild. Two serious adverse events occurred, with one occurring after drug administration. In parts A and B, there was a dose-dependent increase in Tregs. In part B, the estimated dose to achieve a 75% increase in Tregs was 1.46 × 106 IU (95% confidence interval, 1.06 to 1.87). Single-cell RNA-sequencing demonstrated the engagement of distinct pathways and cell­cell interactions. CONCLUSIONS: In this phase 1b/2a study, low-dose IL-2 expanded Tregs without adverse events of major concern. Larger trials are needed to confirm the safety and to further evaluate the efficacy of low-dose IL-2 as an anti-inflammatory therapy for patients with ischemic heart disease. (Funded by the Medical Research Council, the British Heart Foundation, and others; ClinicalTrials.gov number, NCT03113773)

Innate Lymphoid Cells Promote Recovery of Ventricular Function After Myocardial Infarction.

BACKGROUND: Innate lymphoid cells type 2 (ILC2s) play critical homeostatic functions in peripheral tissues. ILC2s reside in perivascular niches and limit atherosclerosis development. OBJECTIVES: ILC2s also reside in the pericardium but their role in postischemic injury is unknown. METHODS: We examined the role of ILC2 in a mouse model of myocardial infarction (MI), and compared mice with or without genetic deletion of ILC2. We determined infarct size using histology and heart function using echocardiography. We assessed cardiac ILC2 using flow cytometry and RNA sequencing. Based on these data, we devised a therapeutic strategy to activate ILC2 in mice with acute MI, using exogenous interleukin (IL)-2. We also assessed the ability of low-dose IL-2 to activate ILC2 in a double-blind randomized clinical trial of patients with acute coronary syndromes (ACS). RESULTS: We found that ILC2 levels were increased in pericardial adipose tissue after experimental MI, and genetic ablation of ILC2 impeded the recovery of heart function. RNA sequencing revealed distinct transcript signatures in ILC2, and pointed to IL-2 axis as a major upstream regulator. Treatment of T-cell-deficient mice with IL-2 (to activate ILC2) significantly improved the recovery of heart function post-MI. Administration of low-dose IL-2 to patients with ACS led to activation of circulating ILC2, with significant increase in circulating IL-5, a prototypic ILC2-derived cytokine. CONCLUSIONS: ILC2s promote cardiac healing and improve the recovery of heart function after MI in mice. Activation of ILC2 using low-dose IL-2 could be a novel therapeutic strategy to promote a reparative response after MI.

Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events.

Radiomics, quantitative feature extraction from radiological images, can improve disease diagnosis and prognostication. However, radiomic features are susceptible to image acquisition and segmentation variability. Ideally, only features robust to these variations would be incorporated into predictive models, for good generalisability. We extracted 93 radiomic features from carotid artery computed tomography angiograms of 41 patients with cerebrovascular events. We tested feature robustness to region-of-interest perturbations, image pre-processing settings and quantisation methods using both single- and multi-slice approaches. We assessed the ability of the most robust features to identify culprit and non-culprit arteries using several machine learning algorithms and report the average area under the curve (AUC) from five-fold cross validation. Multi-slice features were superior to single for producing robust radiomic features (67 vs. 61). The optimal image quantisation method used bin widths of 25 or 30. Incorporating our top 10 non-redundant robust radiomics features into ElasticNet achieved an AUC of 0.73 and accuracy of 69% (compared to carotid calcification alone [AUC: 0.44, accuracy: 46%]). Our results provide key information for introducing carotid CT radiomics into clinical practice. If validated prospectively, our robust carotid radiomic set could improve stroke prediction and target therapies to those at highest risk.

Atherosclerosis imaging using PET: Insights and applications.

PET imaging is able to harness biological processes to characterise high-risk features of atherosclerotic plaque prone to rupture. Current radiotracers are able to track inflammation, microcalcification, hypoxia, and neoangiogenesis within vulnerable plaque. 18 F-fluorodeoxyglucose (18 F-FDG) is the most commonly used radiotracer in vascular studies and is employed as a surrogate marker of plaque inflammation. Increasingly, 18 F-FDG and other PET tracers are also being used to provide imaging endpoints in cardiovascular interventional trials. The evolution of novel PET radiotracers, imaging protocols, and hybrid scanners are likely to enable more efficient and accurate characterisation of high-risk plaque. This review explores the role of PET imaging in atherosclerosis with a focus on PET tracers utilised in clinical research and the applications of PET imaging to cardiovascular drug development.

Integrated cardiovascular assessment of atherosclerosis using PET/MRI.

Atherosclerosis is a systemic inflammatory disease typified by the development of lipid-rich atheroma (plaques), the rupture of which are a major cause of myocardial infarction and stroke. Anatomical evaluation of the plaque considering only the degree of luminal stenosis overlooks features associated with vulnerable plaques, such as high-risk morphological features or pathophysiology, and hence risks missing vulnerable or ruptured non-stenotic plaques. Consequently, there has been interest in identifying these markers of vulnerability using either MRI for morphology, or positron emission tomography (PET) for physiological processes involved in atherogenesis. The advent of hybrid PET/MRI scanners offers the potential to combine the strengths of PET and MRI to allow comprehensive assessment of the atherosclerotic plaque. This review will discuss the principles and technical aspects of hybrid PET/MRI assessment of atherosclerosis, and consider how combining the complementary modalities of PET and MRI has already furthered our understanding of atherogenesis, advanced drug development, and how it may hold potential for clinical application.

Imaging unstable plaque.

Recent advances in imaging technology have enabled us to utilise a range of diagnostic approaches to better characterise high-risk atherosclerotic plaque. The aim of this article is to review current and emerging techniques used to detect and quantify unstable plaque in the context of large and small arterial systems and will focus on both invasive and non-invasive imaging techniques. While the diagnosis of clinically relevant atherosclerosis still relies heavily on anatomical assessment of arterial luminal stenosis, evolving multimodal cross-sectional imaging techniques that encompass novel molecular probes can provide added information with regard to plaque composition and overall disease burden. Novel molecular probes currently being developed to track precursors of plaque rupture such as inflammation, micro-calcification, hypoxia and neoangiogenesis are likely to have translational applications beyond diagnostics and have the potential to play a part in quantifying early responses to therapeutic interventions and more accurate cardiovascular risk stratification.

Choline Kinase Alpha as an Androgen Receptor Chaperone and Prostate Cancer Therapeutic Target.

BACKGROUND: The androgen receptor (AR) is a major drug target in prostate cancer (PCa). We profiled the AR-regulated kinome to identify clinically relevant and druggable effectors of AR signaling. METHODS: Using genome-wide approaches, we interrogated all AR regulated kinases. Among these, choline kinase alpha (CHKA) expression was evaluated in benign (n = 195), prostatic intraepithelial neoplasia (PIN) (n = 153) and prostate cancer (PCa) lesions (n = 359). We interrogated how CHKA regulates AR signaling using biochemical assays and investigated androgen regulation of CHKA expression in men with PCa, both untreated (n = 20) and treated with an androgen biosynthesis inhibitor degarelix (n = 27). We studied the effect of CHKA inhibition on the PCa transcriptome using RNA sequencing and tested the effect of CHKA inhibition on cell growth, clonogenic survival and invasion. Tumor xenografts (n = 6 per group) were generated in mice using genetically engineered prostate cancer cells with inducible CHKA knockdown. Data were analyzed with χ(2) tests, Cox regression analysis, and Kaplan-Meier methods. All statistical tests were two-sided. RESULTS: CHKA expression was shown to be androgen regulated in cell lines, xenografts, and human tissue (log fold change from 6.75 to 6.59, P = .002) and was positively associated with tumor stage. CHKA binds directly to the ligand-binding domain (LBD) of AR, enhancing its stability. As such, CHKA is the first kinase identified as an AR chaperone. Inhibition of CHKA repressed the AR transcriptional program including pathways enriched for regulation of protein folding, decreased AR protein levels, and inhibited the growth of PCa cell lines, human PCa explants, and tumor xenografts. CONCLUSIONS: CHKA can act as an AR chaperone, providing, to our knowledge, the first evidence for kinases as molecular chaperones, making CHKA both a marker of tumor progression and a potential therapeutic target for PCa.