Gut-heart immuno-metabolic disruption associated with inflammaging and subclinical coronary artery disease in people with HIV on antiretroviral therapy

Abstract

Background

Despite antiretroviral therapy (ART), people with HIV (PWH) face immunological ageing and accelerated comorbidities including coronary artery disease (CAD). Gut mucosal damage, chronic inflammation, and Tryptophan (Trp) catabolism into Kynurenine (Kyn) by indoleamine 2,3-dioxygenase (IDO), are associated with HIV disease progression and atherosclerosis. Thus, we comprehensively assessed the interplay between these perturbations in PWH with subclinical CAD under ART.

Methods

Plasma levels of inflammatory mediators, Trp metabolites, and immune cell subset phenotyping were assessed on blood specimens from PWH and HIV seronegative participants with or without CAD diagnosed by cardiac computed tomography angiography from the Canadian HIV Aging Cohort Study.

Results

Levels of gut damage markers regenerating islet-derived protein 3 alpha (REG3α), intestinal fatty-acid binding protein (I-FABP) and soluble CD14 (sCD14) were increased in PWH, but only I-FABP was associated with CAD. Among inflammatory soluble markers, soluble receptor type II for the tumor necrosis factor (sTNFRII) was elevated in PWH, while increased interferon-gamma (IFN-γ) and interleukin (IL)-17A levels were associated with CAD. Regardless of CAD, PWH were characterized by upregulated IDO/Kyn pathway and increased Trp metabolites Kyn, 3-hydroxykynurenine (3-H-Kyn), anthranilic acid, along with decreased xanthurenic acid (Xan acid). Accordingly, HIV+CAD+ participants exhibited highest Kyn/Trp and 3-H-Kyn/Xan acid ratios. CAD+ participants had increased classical CD14⁺ and decreased non-classical CD16⁺ monocyte frequencies, regardless of HIV status. HIV status was associated with increased frequencies of Ki67⁺ proliferating, CD28⁻CD57⁺ senescent and CCR4⁺CXCR3⁺ CD4 and CD8 T-cells, while T-cell and regulatory T-cells (Treg) atheroprotective CD73-expressing subsets were decreased in PWH.

Conclusions

In PWH, subclinical CAD is associated with a distinctive gut-heart immuno-metabolic feature and inflammaging characterized by elevated plasma markers of gut mucosal damage, increased IFN-γ levels and IDO pathway activity, and altered monocyte and T-cell subsets.

Graphical Abstract

Gut–Heart immuno-metabolic disruption and subclinical Coronary Artery Disease (CAD) in chronic HIV infection. Image was made in part using Servier Medical Art (https://smart.servier.com/).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12979-026-00566-8.

Introduction

Although antiretroviral therapy (ART) successfully achieves durable virologic suppression, people with HIV (PWH) continue to experience persistent chronic low-grade inflammation which typically characterizes immunological ageing and is referred to as “inflammaging” [1–3]. As a result, PWH are at an elevated risk for the early development of non-AIDS comorbidities associated with aging [1–3]. Notably, atherosclerosis and cardiovascular disease (CVD) have become the leading causes of death in PWH [4], with the risk of various CVD manifestations being 1.5 to 2-fold higher in this population compared to seronegative individuals [5]. PWH face an increased risk of both clinical coronary artery disease (CAD), whereby symptoms are present, and subclinical CAD, whereby individuals are asymptomatic but have signs of disease on imaging or clinical tests [6]. Factors driving this elevated risk in PWH are multiple including traditional CVD risk factors such as smoking, hypertension, dyslipidemia, diabetes, insulin resistance, sedentary life-style and substance use, and HIV-specific risk factors such as chronic inflammation and relative toxicities due to long-term administration of ART [7, 8]. Importantly, altered gut mucosal immunity appears very early upon infection and play a pivotal role in HIV pathogenesis and inflammaging, even following ART initiation [2, 3, 9]. Gut mucosal fibrosis in PWH results in gut microbiota dysbiosis and further microbial translocation into the peripheral blood, which in turn, drives chronic immune activation of T-cells and monocytes, resulting in inflammaging and increased pro-inflammatory cytokine levels [2, 3, 10].

CAD is driven by atherosclerosis, characterized by the accumulation of lipids and immune cells in the subendothelial space of the coronary arteries. Importantly, inflammation plays a pivotal role in the formation and growth of atherosclerotic plaques [11, 12]. These plaques progressively obstruct blood flow, compromising the heart’s oxygen supply [11, 12]. Endothelial dysfunction, caused by disruptions in vascular homeostasis, initiates atherosclerosis through the retention and modification of low-density lipoproteins (LDL) in the intima [13]. Modified LDLs are captured by macrophages and vascular smooth muscle cells, promoting foam cells formation [11–13]. The initial recruitment of leukocytes to early atheromatous plaques primarily involve circulating monocytes and resident vascular macrophage [11, 12]. Other immune cells, including T-cells, also contribute to atherosclerotic lesion by releasing inflammatory mediators [11, 12]. These mediators activate pathways that drive lesion progression and intensify vascular inflammation. Key cytokines like interferon-gamma (IFN-γ) and tumor necrosis factor-alpha (TNF-α) play critical roles in atherosclerosis by promoting vascular inflammation, enhancing leukocyte adhesion and migration, and facilitating collagen and extracellular matrix degradation, processes central to both establishment and rupture susceptibility of plaque [14]. Another pro-inflammatory cytokine, Interleukin (IL)−6 enhances the inflammatory response by driving the synthesis of acute-phase reactants, activating endothelial cells, and promoting the proliferation and differentiation of lymphocytes [15, 16]. Elevated IL‑17 (especially IL‑17A) levels are consistently associated with acute coronary syndromes and unstable plaque phenotypes, suggesting their pro‑atherogenic role [17–21].

In ART-treated PWH, chronic immune activation and inflammaging contribute to endothelial dysfunction, abnormal lipid metabolism, hypercoagulation, and fibrosis of coronary vascular tissues, all of which promote the formation of atheromatous plaques and the progression of CAD [22]. HIV proteins (Tat, Nef and gp120) also contribute to the development and progression of atherosclerosis by inducing endothelial dysfunction directly and promoting inflammation and immune activation indirectly [23]. As a result, PWH exhibit elevated arterial inflammation compared to seronegative individuals, correlating with increased levels of pro-inflammatory cytokines like IL-6, IFN-γ and markers of monocyte activation [15, 24, 25]. Additionally, numerous biomarkers of immune activation, inflammation, and thrombosis, including IL-1β, TNF-α, Monocyte chemoattractant protein-1 (MCP-1/CCL2), and Intercellular adhesion molecule-1 (ICAM-1), are consistently elevated in PWH and have been associated with increased risk of CVD and mortality, even in those on ART [8]. Notably, in PWH, IL-17A has been associated with endothelial dysfunction, and visceral adipose tissue mass in obese PWH, thus contributing to the development of CAD [21]. Previous studies have shown an association between levels of plasma markers of myeloid cell activation such as soluble CD14 (sCD14) and soluble CD163 (sCD163), and intermediate monocyte (CD14⁺CD16⁺) counts with atherosclerosis in PWH [26, 27]. In addition to myeloid cells, PWH with subclinical CAD are enriched with activated CD8 T-cells (CD38/HLA-DR), which are potent producers of pro-inflammatory cytokines [28, 29].

An important immuno-metabolic regulatory pathway which is involved in both HIV and CVD pathogenesis is via Tryptophan (Trp) catabolism to Kynurenine (Kyn) by the indoleamine 2,3-dioxygenase 1 (IDO1) [30–32]. Trp is an essential amino acid that is metabolized through two primary endogenous pathways – the Kyn pathway and the serotonin–melatonin pathway – and, in addition, via a third, microbiota-dependent indole pathway. Approximately 90% of Trp is metabolized through the Kyn pathway, in peripheral tissues primarily via the rate-limiting enzymes IDO1 and IDO2, and in the liver via tryptophan 2,3-dioxygenase (TDO) [31, 33]. IDO1 is mainly expressed by antigen-presenting cells such as dendritic cells (DCs) and macrophages, where it plays a key immunoregulatory role [31]. Our team and others have previously shown that Trp catabolism via IDO/Kyn pathway is upregulated in both acute and chronic HIV infections along with imbalance in favor of regulatory T-cells (Treg) over Th17 cells and elevated plasma levels of inflammatory cytokines [9, 34, 35]. Moreover, HIV proteins Tat and Nef, along with IFN-γ, activate IDO1, promoting Trp degradation [31, 34, 36]. Importantly, the plasma Kyn/Trp ratio has been previously linked to carotid artery atherosclerosis and elevated risk of CVD [30, 32, 37–40]. Increased IDO/Kyn pathway was also associated with development of atherosclerosis in a mouse model of HIV infection [41]. Notably, plasma Kynurenic acid/Trp ratio was associated with increased risk of carotid plaques in PWH [42, 43]. However, less is known about other metabolites of the Kyn pathway and their relationship with HIV-induced inflammaging and accelerated CAD in PWH.

The intricate interactions among gut mucosal disruption and immuno-metabolic dysregulation in favor of inflammaging and subclinical CAD in PWH under effective ART remain poorly understood. This study aimed to comprehensively evaluate these interconnected mechanisms in PWH with subclinical CAD to better understand immuno-metabolic features of early disease pathogenesis. This approach has the potential to yield novel insights into the pathophysiology of CAD in this vulnerable population and to identify new biomarkers and therapeutic targets for earlier detection and more effective intervention strategies.

Methods

Study design and population

In this cross-sectional observational study, plasma and peripheral blood mononuclear cells (PBMC) samples from PWH and HIV seronegative control participants were obtained from the multicenter Canadian HIV and Aging Cohort Study (CHACS) cardiovascular imaging sub-study. The design and detailed information of the CHACS and its cardiovascular imaging sub-study have been previously published by our team [6, 44, 45]. Briefly, the inclusion criteria were to be aged 40 years or older or having lived with HIV for 15 years or more, with a 10-year Framingham risk score for cardiovascular disease ranging from 5 to 20%, and to be free of overt cardiovascular disease at baseline. Coronary atherosclerotic plaque volume assessment was done using contrast-enhanced coronary CT angiography for all study participants, by a board-certified cardiothoracic radiologist at the Centre Hospitalier de l’Université de Montréal, QC, Canada. Participants with a measurable coronary artery plaque of any size were considered to have subclinical CAD [6]. In addition, low-attenuation plaque component was also assessed, as defined as a CT plaque density of 30 Hounsfield units or less. Low attenuation is a marker of the plaque risk of rupture [46].

Ethics

This study was conducted in accordance with the Declaration of Helsinki and was authorized by the research ethics board of the Centre Hospitalier de l’Université de Montréal (CHUM) (CE 11.063), MUHC: MP-37–2022-8107 and UQAM: 2022–4663. Written informed consent was obtained from all participants prior to their inclusion in the CHACS study biobanking.

Quantifications of plasma markers of gut mucosal damage, microbial translocation and sTNFRII by ELISA

The intestinal fatty-acid binding protein (I-FABP) and the human regenerating islet-derived protein 3 alpha (REG3α), two markers associated with gut mucosal damage were quantified using the commercially available ELISA kits from Hycult Biotech (Uden, Netherlands) and R&D Systems (Minneapolis, MN, USA), respectively. The soluble form of CD14 (sCD14), a microbial translocation marker, was quantified using a Hycult Biotech ELISA kit. Then, the soluble receptor type II for the tumor necrosis factor (sTNFRII) was also quantified by ELISA (R&D Systems, Minneapolis, MN, USA). All experiments were carried out in duplicate in plasma samples, according to the manufacturer’s instructions.

Quantification of plasma levels of cytokines and chemokines by Luminex

As we reported previously [47], plasma levels of the following inflammatory soluble factors were measured in duplicates using the MILLIPLEX® Human Cytokine/Chemokine/Growth Factor Panel A (MilliporeSigma, Burlington, MA, USA): C-X3-C motif chemokine ligand 1 (CX3CL1), C-X-C motif chemokine ligand 1 (CXCL1), IFN-γ, IL-1β, IL-6, IL-10, Interferon-induced protein 10 (IP-10/CXCL10), MCP-1, TNF-α, IL-17A and IL-17F. The assay was performed according to the manufacturer’s instructions. Samples were run on a MAGPIX instrument (Luminex), and the results were analyzed with Belysa Immunoassay Curve Fitting Software version 1.2 (MilliporeSigma, Burlington, MA, USA).

Tryptophan metabolites quantification

Plasma levels of Trp and its downstream catabolites including kynurenine (Kyn), 3-hydroxykynurenine (3-H-Kyn), kynurenic acid, anthranilic acid, xanthurenic acid (Xan acid), 3-Hydroxyanthranilic acid, picolinic acid and quinolinic acid were quantified by Liquid chromatography in combination with isotope dilution tandem mass spectrometry (LC–MS/MS) method as we previously reported [48].

Flow cytometry analysis

Ex vivo phenotyping of peripheral blood T-cell, dendritic cells (DCs) and monocyte subsets was performed by multi-color flow cytometry on participants’ PBMCs from the CHACS biobank [44, 45]. The LIVE/DEAD fixable Aqua Dead Cell Stain Kit (Invitrogen, Oregon, USA) was used for viability staining. All fluorochrome-conjugated antibodies, used at optimal concentrations in three independent panels, are listed in Supplementary Table 1. For extracellular staining PBMC were incubated with the corresponding antibodies in PBS + 2% fetal bovine serum for 1 h at 4 °C. Following this, the cells were fixed and permeabilized using the Transcription Factor Buffer Set (BD Bioscience, New jersey, USA) as the manufacturer’s protocol, and incubated with the appropriate antibodies in Perm/Wash buffer for 1 h at 4 °C. Acquisition and analysis were performed using a three laser BD LSR Fortessa X-20 cytometer and FlowJo V10.10.0 software (Oregon, USA).

Statistical analysis

All statistical analyses were performed using GraphPad Prism software V10 (Boston, MA, USA). Initially, the variable distribution was evaluated using the Kolmogorov–Smirnov test. To assess significant differences between more than two study groups, the Kruskal–Wallis test was employed. For comparisons of unpaired variables, the non-parametric Mann–Whitney U test was used. Data are presented as medians with interquartile ranges (IQR), and count data are reported as frequencies and percentages. Additionally, the Spearman rank correlation test was applied to evaluate associations between study variables. Only the resulting correlation coefficients statistically significant (p < 0.05) were visualized in the heatmap. To estimate the associations between HIV and CAD status and our study key immuno-metabolic outcomes (Kyn/Trp ratio, I-FABP, and sCD14), we performed multivariate linear regression analyses using R statistical software and the lm() formula, adjusting for potential confounders. First, potential confounders were identified using a priori clinical knowledge. As small sample sizes warranted a parsimonious approach, multivariate models were built only if potential confounders were associated with both outcomes of interest and the exposure. For the three multivariate models, the Change in Estimate (CIE) method was used to identify potential confounders [49, 50]. The considered variables for confounding were based on baseline imbalances between exposure groups: Age, Cigarette packs per year, BMI, LDL, Insulin, Total Cholesterol. Following a manual CIE procedure, variables were discarded if they changed the point of estimate (β) for either exposure variable (HIV and CAD status) by > 10%.

Results

Characteristics of the study populations

This study included a total of 123 participants including 79 PWH and 44 HIV seronegative controls. Demographic and clinical characteristics of the participants are summarized in Table 1. All PWH were on ART with suppressed viral load, and there was no difference between HIV+CAD+ and HIV+CAD- individuals under current protease inhibitor treatment. Participants with CAD had a median age that was 5 years higher than those without CAD. A significantly larger proportion of PWH were male or co-infected by cytomegalovirus (CMV) compared with HIV seronegative controls (82% versus 66%, and 78% versus 30%, respectively). There were more current tobacco smokers among PWH and CAD + participants, while cannabis smoking was elevated only in PWH. Additionally, a higher proportion of PWH were under statin treatment compared to seronegative individuals (25% versus 7%). Accordingly, the lipid profile of PWH showed lower levels of total cholesterol (4.73 versus 5.07 mmol/L), LDL (2.66 versus 3.07 mmol/L), and HDL (1.21 versus 1.35 mmol/L) compared to HIV seronegative controls independent of their CAD status. In contrast, triglyceride and insulin levels were elevated in PWH (1.56 versus 1.24 mmol/L and 77 versus 55.5 pmol/L, respectively). No significant differences were observed between CAD+ and CAD- for their lipid and glucose profiles. HIV+CAD+ participants had a tendency of higher duration since HIV diagnosis (median 23.8 versus 18.35 years, p = 0.08) and since ART initiation (17.59 versus 13.82 years, p = 0.096) compared with HIV+CAD- participants (Table 1). No differences were observed for CD4 and CD8 T-cell counts, CD4/CD8 ratio, and nadir CD4 count between HIV+CAD+ and HIV+CAD- participants (Table 1).

Table 1.

Characteristics of the study participants according to their HIV and subclinical coronary artery disease (CAD) status

	HIV+  N = 79	HIV- N = 44	p-value	CAD+  N = 58	CAD- N = 65	p-value	HIV+CAD+  N = 39	HIV+CAD- N = 40	HIV-CAD+  N = 19	HIV-CAD- N = 25	p-value
Characteristics
Sex [n (%)]			0.048*			0.67*					0.16*
Female	14 (18)	15 (34)		15 (26)	14 (22)		7 (18)	7 (18)	8 (42)	7 (28)
Male	65 (82)	29 (66)		43 (74)	51 (78)		32 (82)	33 (82)	11 (58)	18 (72)
Age at study visit (years) [Median (IQR)]	56.7 (53.4–62.61)	58.75 (51.9–66.1)	0.3#	61.24 (55.26–67.4)	56.2 (51.3–61.31)	0.001#	57.3 (53.6–62.7)	56.2 (52.8–62.41)	64.8 (62–69.5)	55.3 (49.9–60)	0.0007$
HIV status [n (%)]			NA			0.57*					NA
HIV-	NA	NA		19 (33)	25 (38)		NA	NA	NA	NA
HIV+	NA	NA		39 (67)	40 (62)		NA	NA	NA	NA
CAD status [n (%)]			0.57*			NA					NA
CAD-	40 (51)	25 (57)		NA	NA		NA	NA	NA	NA
CAD+	39 (49)	19 (43)		NA	NA		NA	NA	NA	NA
CD4 T-cell count (cells/μL) [Median (IQR)]	597 (432–808)	NA	NA	NA	NA	NA	598 (470–808)	575.5 (369.8–851.8)	NA	NA	0.6#
CD8 T-cell count (cells/μL) [Median (IQR)]	649.3 (514.0–961.2)	NA	NA	NA	NA	NA	646 (530.4–855.6)	658.1 (483.0–1112)	NA	NA	0.88#
CD4:CD8 ratio [Median (IQR)]	0.89 (0.6–1.24)	NA	NA	NA	NA	NA	1 (0.61–1.2)	0.83 (0.57–1.26)	NA	NA	0.41#
Nadir CD4 T-cell count (cells/μL) [Median (IQR)]	180 (100–270)	NA	NA	NA	NA	NA	180 (100–270)	185 (92.5–277.5)	NA	NA	0.96#
Time since HIV diagnosis (years) [Median (IQR)]	20.3 (14.4–26.5)	NA	NA	NA	NA	NA	23.8 (16.2–30.2)	18.35 (13.85–25.68)	NA	NA	0.08#
Time since ART initiation (years) [Median (IQR)]	16.09 (10.6–20.56)	NA	NA	NA	NA	NA	17.59 (10.66–22.32)	13.82 (10–18.56)	NA	NA	0.096#
CMV-IgG status [n (%)]			< 0.0001*			0.52*					< 0.0001*
Negative	12 (15)	24 (55)		16 (28)	20 (31)		5 (13)	7 (18)	11 (58)	13 (52)
Positive	62 (78)	13 (30)		38 (66)	37 (60)		31 (80)	31 (78)	7 (37)	6 (24)
Unknown	5 (7)	7 (15)		4 (6)	8 (9)		3 (7)	2 (4)	1 (5)	6 (24)
Diabetes [n (%)]			0.25*			0.59*					0.33*
No	67 (85)	41 (93)		52 (90)	56 (86)		33 (85)	34 (85)	19 (100)	22 (88)
Yes	12 (15)	3 (7)		6 (10)	9 (14)		6 (15)	6 (15)	0 (0)	3 (12)
Family history of premature CAD [n (%)]			0.81*			0.32*					0.57*
No	34 (43)	22 (50)		23 (40)	33 (51)		13 (33)	21 (53)	10 (53)	12 (48)
Yes	39 (49)	19 (43)		29 (50)	29 (45)		22 (56)	17 (42)	7 (37)	12 (48)
Unknown	6 (8)	3 (7)		6 (10)	3 (4)		4 (11)	2 (5)	2 (10)	1 (4)
Sexual orientation [n (%)]			< 0.0001*			0.84*					0.0006*
Bisexual	4 (5)	0 (0)		2 (3)	2 (3)		2 (5)	2 (5)	0 (0)	0 (0)
Heterosexual	29 (37)	34 (77)		28 (48.5)	35 (54)		12 (31)	17 (43)	16 (84)	18 (72)
Homosexual (MSM)	46 (58)	10 (23)		28 (48.5)	28 (43)		25 (64)	21 (52)	3 (16)	7 (28)
Ethnicity [n (%)]			0.14*			0.03*					0.03*
Caucasian	62 (78)	41 (93)		53 (91)	50 (77)		35 (88)	27 (68)	18 (95)	23 (92)
Asian	1 (1.5)	0 (0)		1 (2)	0 (0)		1 (4)	0 (0)	0 (0)	0 (0)
Black/Caribbean	11 (14)	1 (2)		4 (7)	8 (12)		3 (8)	8 (20)	1 (5)	0 (0)
Black/African	1 (1.5)	0 (0)		0 (0)	1 (2)		0 (0)	1 (2)	0 (0)	0 (0)
Latino	4 (5)	2 (5)		0 (0)	6 (9)		0 (0)	4 (10)	0 (0)	2 (8)
History of tobacco smoking [n (%)]			0.001*			0.002*					0.001*
Current smoker	28 (35)	5 (11)		23 (40)	10 (15)		19 (49)	9 (23)	4 (21)	1 (4)
Ex-smoker	30 (38)	21 (48)		24 (41)	27 (42)		15 (38)	15 (38)	9 (47)	12 (48)
Never smoked	21 (27)	18 (41)		11 (19)	28 (43)		5 (13)	16 (39)	6 (32)	12 (48)
Cigarette packs-years [Median (IQR)]	12 (0–28.1)	1 (0–17.5)	0.02#	14.67 (1.31–26.68)	0.65 (0–16.5)	0.001#	19.8 (4–34.58)	1.43 (0–18.7)	5.1 (0–19.2)	0.1 (0–12)	0.001$
Cannabis use [n (%)]			0.07*			> 0.9999*					0.29*
Active user	19 (24)	4 (9)		11 (19)	12 (18)		10 (26)	9 (23)	1 (5)	3 (12)
Former user	4 (5)	5 (11)		4 (9)	5 (8)		1 (3)	3 (7)	3 (16)	2 (8)
Non-user	56 (71)	35 (80)		43 (72)	48 (74)		28 (71)	28 (70)	15 (79)	20 (80)
Alcohol units per week [Median (IQR)]	1 (0–6)	4 (0–6.75)	0.14#	1.5 (0–7)	30 (0–6)	0.41#	1 (0–7)	1.5 (0–6)	3 (0–7)	4 (1–6.5)	0.43$
Type of alcohol drinker [n (%)]			0.21*			0.006*					0.06*
Do not drink	27 (34)	10 (23)		22 (38)	15 (23)		16 (41)	11 (28)	6 (32)	4 (16)
Ex drinker	4 (5)	0 (0)		2 (3)	2 (3)		2 (5)	2 (4)	0 (0)	0 (0)
Excessive drinker	3 (4)	2 (4)		5 (9)	0 (0)		3 (8)	0 (0)	2 (10)	0 (0)
Social drinker	45 (57)	32 (73)		29 (50)	48 (74)		18 (46)	27 (68)	11 (58)	21 (84)
Illicit drug use ever [n (%)]	35 (44)	10 (23)	0.02*	24 (41)	21 (32)	0.35*	19 (49)	16 (40)	5 (26)	5 (20)	0.09*
Illicit drug use current [n (%)]	24 (30)	4 (9)	0.007*	15 (26)	13 (20)	0.52*	14 (36)	10 (25)	1 (5)	3 (12)	0.03*
Statin use [n (%)]			0.02*			0.31*					0.09*
No	56 (71)	40 (91)		42 (72)	54 (83)		25 (64)	31 (78)	17 (89)	23 (92)
Yes	20 (25)	3 (7)		13 (22)	10 (15)		12 (31)	8 (20)	1 (5.5)	2 (8)
Unknown	3 (4)	1 (2)		3 (6)	1 (2)		2 (5)	1 (2)	1 (5.5)	0 (0)
BMI [n (%)]			0.05*			0.1*					0.07*
Healthy weight (18.5–25)	33 (42)	11 (25)		26 (45)	18 (28)		19 (49)	14 (35)	7 (37)	4 (16)
Underweight (under 18.5)	4 (5)	1 (2)		3 (5)	2 (3)		2 (6)	2 (5)	1 (5)	0 (0)
Overweight (over 25)	42 (53)	30 (68)		29 (50)	43 (66)		18 (46)	24 (60)	11 (58)	19 (76)
Unknown	0 (0)	2 (5)		0 (0)	2 (3)		0 (0)	0 (0)	0 (0)	2 (8)
Systolic blood pressure (mmHg) [Median (IQR)]	124 (112–133)	126 (117–135)	0.87#	122 (110–135.5)	127 (118–133.8)	0.35#	122 (110–135)	127.5 (115–133)	120 (110–137)	126.5 (120–134.8)	0.77$
Framingham score [Median (IQR)]	8 (6–13)	10 (7–14.75)	0.21#	9.5 (7–13.25)	9 (6.-13)	0.39#	8 (6–12)	8.5 (5.25–13.75)	12 (7–15)	9 (6–13)	0.27$
Fasting glucose (mmol/L) [Median (IQR)]	5 (4.5–5.6)	4.9 (4.48–5.4)	0.52#	5 (4.5–5.5)	5 (4.6–5.48)	0.65#	5 (4.55–5.7)	5 (4.53–5.58)	4.85 (4.38–5.33)	5.05 (4.65–5.4)	0.47$
Hb_A1c [Median (IQR)]	0.06 (0.05–5.07)	0.06 (0.05–5.37)	0.47#	0.06 (0.05–5.26)	0.057 (0.05–4.83)	0.31#	0.06 (0.05–5.1)	0.06 (0.05–4.83)	5.2 (0.05–5.42)	0.05 (0.05–5.1)	0.09$
Total Plaque Volume (mm3) [Median (IQR)]	NA	NA	NA	NA	NA	NA	222.6 (75–540.5)	NA	148 (49.70–430.2)	NA	0.26#
Volume of low attenuation plaque (mm3) [Median (IQR)]	NA	NA	NA	NA	NA	NA	82 (17.73–168.9)	NA	43.7 (9–163.7)	NA	0.21#
Total cholesterol (mmol/L) [Median (IQR)]	4.73 (4.04–5.32)	5.07 (4.66–5.47)	0.03#	4.89 (4.26–5.45)	4.92 (4.21–5.36)	0.64#	4.72 (3.98–5.22)	4.79 (4.08–5.34)	5 (4.74–5.47)	5.16 (4.43–5.69)	0.19$
LDL (mmol/L) [Median (IQR)]	2.66 (2.06–3.22)	3.07 (2.68–3.59)	0.009#	2.75 (2.25–3.24)	2.92 (2.33–3.52)	0.38#	2.59 (1.86–3.21)	2.77 (2.09–3.29)	3.05 (2.69–3.59)	3.14 (2.55–3.68)	0.05$
HDL (mmol/L) [Median (IQR)]	1.21 (0.99–1.44)	1.35 (1.14–1.56)	0.04#	1.32 (1.05–1.57)	1.22 (1.01–1.44)	0.24#	1.22 (0.94–1.52)	1.19 (1–1.35)	1.37 (1.2–1.63)	1.3 (1.1–1.51)	0.09$
LDL:HDL ratio [Median (IQR)]	2.07 (1.63–2.9)	2.23 (1.79–3.27)	0.23#	1.94 (1.61–2.94)	2.23 (1.79–2.93)	0.19#	1.88 (1.57–2.94)	2.23 (1.74–2.89)	2.08 (1.7–3.27)	2.29 (1.8–3.48)	0.37$
Triglyceride (mmol/L) [Median (IQR)]	1.56 (1.03–2.14)	1.24 (0.88–2.03)	0.05#	1.53 (0.93–2.09)	1.38 (1.01–2.11)	0.89#	1.9 (0.98–2.65)	1.48 (1.11–2.05)	1.18 (0.87–1.92)	1.27 (0.89–2.17)	0.24$
ApoB (g/L) [Median (IQR)]	0.88 (0.74–1.09)	0.895 (0.66–1.08)	0.64#	0.84 (0.72–1.05)	0.91 (0.74–1.09)	0.36#	0.84 (0.72–1.05)	0.93 (0.76–1.11)	0.83 (0.68–0.97)	0.91 (0.32–1.09)	0.72$
Insulin (pmol/L) [Median (IQR)]	77 (50–153.3)	55.5 (41.5–79.5)	0.005#	69 (43–126.5)	71 (51–128)	0.62#	73 (42 –142)	80 (55–170)	63 (45.25–89.25)	53 (40.5–72.5)	0.02$

Abbreviations: HIV Human immunodeficiency virus, CAD Coronary artery disease, CMV-IgG Immunoglobulin G specific for Cytomegalovirus, BMI Body mass index, Hb-A1c Hemoglobin A1c, LDL Low density lipoprotein, HDL High density lipoprotein, NA Not applicable

^*Fisher’s Exact Test

^#Mann–Whitney U test

^$Kruskal–Wallis rank sum test; significant values are presented in bold

Subclinical CAD in PWH was associated with increased levels of plasma markers of gut mucosal damage and microbial translocation

To assess the gut mucosal damage and microbial translocation and their potential association with subclinical CAD in PWH, we quantified plasma levels of several well-established markers, including I-FABP and REG3α, markers of gut mucosal barrier damage, and sCD14 as a marker of microbial translocation [9, 51, 52]. While higher plasma levels of I-FABP, REG3α and sCD14 distinguished PWH from HIV seronegative participants, only the levels of I-FABP, were significantly increased in CAD+ compared to CAD- individuals (Table 2). When participants were grouped according to their HIV and CAD status, the highest levels of I-FABP, REG3α and sCD14 were observed in HIV+CAD+ individuals, with statistically significant differences compared to HIV-CAD- participants (Table 2). Higher plasma levels of I-FABP were observed in CMV+ individuals (Supplementary Table 2). Correlation analysis revealed a negative association between I-FABP and BMI, as well as a positive association of REG3α and sCD14 with total plaque volume and the volume of low attenuation plaque, two major clinical parameters of CAD (Fig. 1). Therefore, elevated levels of plasma markers of gut mucosal dysfunction in ART-treated PWH were associated with subclinical CAD.

Table 2.

Plasma soluble markers of gut damage, systemic inflammation and cytokines

Inflammatory markers according to HIV and CAD status
	HIV+  [Median (IQR)]	HIV- [Median (IQR)]		p-value* (HIV+ vs HIV-)		CAD+  [Median (IQR)]		CAD- [Median (IQR)]		p-value* (CAD+ vs CAD-)
Gut damage markers
I-FABP (pg/mL)	2511 (1548–4873)	1702 (926.9–2215)		0.0002		2372 (1496–5193)		1745 (1190–2840)		0.04
REG3α (pg/mL)	8145 (5649–12,854)	5679 (3819–10,205)		0.01		8412 (5684–12,000)		6540 (4327–12,091)		NS
Inflammation markers
sCD14 (ng/mL)	1536 (1249–1711)	1256 (1050–1535)		0.003		1500 (1152–1689)		1378 (1103–1661)		NS
sTNFRII (pg/mL)	3154 (2380–4471)	2509 (2121–3279)		0.01		3213 (2280–4302)		2647 (2266–3707)		NS
Cytokines
CX3CL1/ Fractalkine (pg/mL)	123.1 (97.12–153)	124.9 (95.81–153.7)		NS		122.8 (96.57–149.8)		127 (97.38–155)		NS
CXCL1 (pg/mL)	38.02 (18.46–66.73)	34.07 (18.01–67.58)		NS		34.65 (21.08–62.26)		36.93 (17.62–77.7)		NS
IFN-γ (pg/mL)	5.14 (3.02–7.72)	4.52 (2.91–6.45)		NS		5.61 (4.02–8.64)		4.31 (2.18–6.5)		0.04
IL-1β (pg/mL)	7.29 (5.54–10.29)	7.00 (5.51- 9.31)		NS		7.37 (4.88–10.6)		7.28 (5.60–9.92)		NS
IL-6 (pg/mL)	1.45 (0.79–2.37)	1.11 (0.55–1.83)		0.09		1.44 (0.73–2.36)		1.25 (0.56–2.09)		NS
IL-10 (pg/mL)	6.69 (4.06–10.01)	6.12 (4.29–10.95)		NS		6.29 (4.33- 8.51)		6.76 (3.98–11.34)		NS
IP-10 (pg/mL)	248.6 (152.9–372.6)	218.7 (153.2–284.5)		NS		227 (166.6–313)		248.6 (145.3–383.8)		NS
MCP-1(pg/mL)	327.1 (223.5–419)	302 (241–342)		NS		311 (231.3–357.4)		306.3 (239.1–418.6)		NS
TNF-α (pg/mL)	37.83 (31.58–53.99)	36.32 (27.47–43.37)		NS		36.37 (29.79–53.57)		37.83 (28.73–47.35)		NS
IL-17A (pg/mL)	3.585 (1.483–5.775)	3.28 (1.73–4.85)		NS		4.41 (2.69–6.05)		2.75 (1.19–4.82)		0.02
IL-17F (pg/mL)	13.88 (8.31–21.27)	14.26 (6.3–23.89)		NS		14.87 (8.78–26.99)		13.16 (6.99–20.95)		NS

Inflammatory markers among 4 study groups
	HIV+CAD+  [Median (IQR)]	HIV+CAD- [Median (IQR)]	HIV-CAD+  [Median (IQR)]	HIV-CAD- [Median (IQR)]	Kruskal- Wallis p-value$	p-value* HIV+CAD+  vs HIV+CAD-	p-value* HIV-CAD+  vs HIV+CAD+	p-value* HIV+CAD+  vs HIV-CAD-	p-value* HIV-CAD+  vs HIV+CAD-	p-value* HIV-CAD- vs HIV+CAD-	p-value* HIV-CAD- vs HIV-CAD+
Gut damage markers
I-FABP (pg/mL)	3653 (2186–7115)	1760 (1235–3002)	1573 (426.1–1939)	1745 (1046–2672)	< 0.0001	0.002	< 0.0001	< 0.0001	0.04	NS	NS
REG3α (pg/mL)	8709 (6236–13,024)	7045 (5326–12,750)	7291 (4365–10,061)	4714 (3608–11,021)	0.06	NS	NS	0.01	NS	0.048	NS
Inflammation markers
sCD14 (ng/mL)	1539 (1189–1711)	1519 (1251–1760)	1282 (1110–1607)	1238 (1013–1383)	0.02	NS	0.06	0.006	NS	0.02	NS
sTNFRII (pg/mL)	3493 (2490– 4563)	2801 (2353–4261)	2559 (2064–3792)	2462 (2134–3223)	0.03	NS	0.05	0.009	NS	0.06	NS
Cytokines
CX3CL1/ Fractalkine (pg/mL)	123 (95.96–146.6)	124.9 (97.38–155.1)	119.9 (95.81–153.7)	133 (93.04–154.1)	NS	NS	NS	NS	NS	NS	NS
CXCL1 (pg/mL)	39.53 (19.58–56.27)	37.25 (17.52–79.39)	34.63 (20.28–74.25)	32.54 (17.22–60.94)	NS	NS	NS	NS	NS	NS	NS
IFN-γ (pg/mL)	6.03 (4.31–8.96)	4.14 (1.90–6.52)	4.32 (1.99–6.25)	4.79 (3.12–7.73)	0.06	0.01	0.06	NS	NS	NS	NS
IL-1β (pg/mL)	7.66 (4.72–10.88)	7.17 (5.57–9.71)	6.86 (4.82–10.12)	7.17 (5.59–9.02)	NS	NS	NS	NS	NS	NS	NS
IL-6 (pg/mL)	1.85 (0.81–2.44)	1.25 (0.75–2.25)	1.01 (0.65–1.93)	1.13 (0.27–1.83)	NS	NS	NS	0.07	NS	NS	NS
IL-10 (pg/mL)	6.29 (4.10–8.21)	6.91 (3.86–10.93)	6.76 (4.72–8.76)	5.84 (4.03–11.83)	NS	NS	NS	NS	NS	NS	NS
IP-10 (pg/mL)	241.6 (175.7–358.1)	251 (145.3–386.9)	218.7 (154.0–268.2)	221.8 (142.7–367)	NS	NS	NS	NS	NS	NS	NS
MCP-1 (pg/mL)	300.4 (220.8–371.1)	376.1 (240.6–443.4)	312.3 (263.3–342.1)	282.3 (237.4–343.8)	NS	NS	NS	NS	NS	NS	NS
TNF-α (pg/mL)	37.69 (31.08–56.42)	37.83 (31.58–48.76)	36.09 (25.99–41.28)	38.15 (27.90–45.08)	NS	NS	NS	NS	NS	NS	NS
IL-17A (pg/mL)	4.64 (2.69–6.26)	2.81 (1.13–4.99)	4.01 (2.24–6.09)	2.33 (1.51–4.72)	NS	0.06	NS	0.04	NS	NS	NS
IL-17F (pg/mL)	16.14 (9.11–23.67)	13.16 (6.99–20.95)	14.4 (5.41–44.06)	13.71 (7.28–20.82)	NS	NS	NS	NS	NS	NS	NS

Abbreviations: HIV Human immunodeficiency virus, CAD Coronary artery disease, I-FABP Intestinal fatty-acid binding protein, REG3α Regenerating islet-derived protein 3 alpha, sCD14 soluble form of CD14, sTNFRII soluble receptor type II for the tumor necrosis factor, CX3CL1 C-X3-C motif chemokine ligand 1, CXCL1 C-X-C motif chemokine ligand 1, IFN-γ Interferon-gamma, IL-1β Interleukin 1 beta, IL-6 Interleukin 6, IL-10 Interleukin 10, IP-10 Interferon γ-induced protein 10, MCP-1 Monocyte chemoattractant protein-1, TNF-α Tumor necrosis factor alpha, IL-17A Interleukin 17A, IL-17F Interleukin 17F

^*Mann–Whitney U test; Significant values are presented in bold; NS: Not significant

^$Kruskal–Wallis rank sum test

Fig. 1.

Heatmaps of Spearman correlations between study variables, clinical data, and risk factors associated with CAD among HIV+ CAD+ participants. Correlations were calculated using the corrplot() R function. The color scale represents Spearman correlation coefficients ranging from 1 (dark blue, positive correlation) to −1 (dark red, negative correlation). Only correlations with a significant p-value (95% confidence interval) are shown on the matrix. Color intensity and point size indicate the strength of the correlation. The two abscissa axes correspond to clinical data, CAD-related risk factors, gut dysfunction markers, Trp metabolites, and immune cell subset phenotypes

Increased levels of sTNFRII, IFN-γ and IL-17A in PWH with subclinical CAD

Within the inflammatory soluble markers, elevated levels of sTNFRII were observed only in PWH, whereas IFN-γ and IL-17A were elevated only in relation to CAD status (Table 2). Notably, HIV+CAD+ individuals exhibited the highest levels of sTNFRII, IFN-γ, and IL-17A, with their IL-17A levels showed a positive correlation with total plaque volume and volume of low attenuation plaque (Fig. 1). No significant changes were observed in the levels of CX3CL1, CXCL1, IL-1β, IL-6, IP-10, MCP-1, TNF-α, and IL-17F among study groups.

Increased IDO/Kyn pathway activity in HIV+CAD+ individuals

Increased Trp catabolism via IDO/Kyn pathway, as measured by the plasma Kyn/Trp ratio is upregulated during HIV infection and is associated with CAD progression [9, 34, 35, 41, 53]. To confirm and extend these findings, we assessed plasma levels of Trp and its eight main catabolites in our cohort. Overall, as shown in Table 3, mainly HIV status and not CAD was associated with an upregulation of IDO/Kyn pathway. Plasma levels of Kyn, 3-H-Kyn, anthranilic acid, as well as the Kyn/Trp and 3-H-Kyn/Xan acid ratios were significantly higher in PWH compared with HIV seronegative participants. In contrast Xan acid levels were significantly lower in PWH. When analyzed based on CAD alone, only a trend toward a higher 3-H-Kyn/Xan acid ratio in CAD+ versus CAD- participants (p = 0.06). HIV+CAD+ group had the highest levels of Kyn/Trp and 3-H-Kyn/Xan acid ratios and lower levels of Xan acid among four study groups (Table 3). Statin use was also associated with increased 3-H-Kyn/Xan acid ratios (Supplementary Table 2). In the HIV+CAD+ group and in line with previous findings, a positive correlation was observed between plasma Kyn/Trp ratio and IFN-γ levels (Fig. 1) [31]. We also found a positive correlation between 3-H-Kyn, 3-H-Kyn/Kyn ratio, and BMI, as well as between 3-H-Kyn/Xan acid ratio and high-sensitivity C-reactive protein (hs-CRP), a marker of systemic inflammation (Fig. 1).

Table 3.

Plasma circulating tryptophan catabolites in study groups according to participants HIV and subclinical coronary artery disease (CAD) status

Tryptophan catabolites according to HIV and CAD status
	HIV+  [Median (IQR)]		HIV- [Median (IQR)]		p-value* (HIV+ vs HIV-)		CAD+  [Median (IQR)]		CAD- [Median (IQR)]		p-value* (CAD+ vs CAD-)
Trp (µmol/L)	52.25 (44.78–59.33)		55.75 (49.03–60.9)		NS		53.5 (44.9–58.7)		54.1 (47.20–60)		NS
Kyn (µmol/L)	2.28 (1.90–2.82)		2.12 (1.70–2.37)		0.05		2.24 (1.95–2.66)		2.17 (1.79–2.55)		NS
Kyn/Trp ratio	0.045 (0.036–0.054)		0.04 (0.03–0.04)		0.0009		0.044 (0.036–0.052)		0.039 (0.033–0.049)		NS
3-H-Kyn (nmol/L)	37 (29.48–47.65)		32.40 (25.18–39.18)		0.09		36.7 (28.9–47.6)		34.5 (26.8–43.3)		NS
3-H-Kyn/ Kyn ratio	0.016 (0.013–0.02)		0.016 (0.014–0.019)		NS		0.017 (0.014–0.021)		0.016 (0.014–0.019)		NS
Kynurenic acid (nmol/L)	48.05 (37.53–63.8)		48 (39.53–62.03)		NS		48.10 (37.7–64.7)		47.90 (40.7–63)		NS
Kynurenic acid/Trp ratio	0.001 (0.0007–0.0012)		0.0009 (0.0007–0.001)		NS		0.001 (0.0007 – 0.0013)		0.0009 (0.0007 – 0.001)		NS
Anthranilic acid (nmol/L)	15.4 (11.68–21.18)		12.5 (10.6–16.8)		0.02		13.5 (11.3–18.6)		14.4 (11.4–22.1)		NS
Xanthurenic acid (nmol/L)	9.95 (6.05–14.65)		15.45 (10.78–18.93)		0.0006		10.7 (7.3–15.9)		13.5 (6.6–16.9)		NS
3-H-Kyn/ Xan acid ratio	3.48 (2.51–5.85)		2.36 (1.71–3.39)		0.0003		3.16 (2.33–5.74)		2.58 (2.05–4.09)		0.06
3-Hydroxyanthranilic acid (nmol/L)	30.85 (22.1–37.55)		28.25 (22.35–33.23)		NS		30.4 (23–36.7)		29.6 (21.8–34.7)		NS
Picolinic acid (nmol/L)	32.85 (25.35–51.2)		34.05 (28.85–45.83)		NS		35.1 (25.8–48.4)		30.5 (26.2–48.1)		NS
Quinolinic acid (nmol/L)	465 (358.8–596.5)		395.5 (345.3–553.3)		NS		426 (344–591)		424 (365–601)		NS

Tryptophan catabolites among 4 study groups
	HIV+CAD+  [Median (IQR)]	HIV+CAD- [Median (IQR)]		HIV-CAD+  [Median (IQR)]	HIV-CAD- [Median (IQR)]	Kruskal- Wallis p-value$	p-value* HIV+CAD+  vs HIV+CAD-	p-value* HIV-CAD+  vs HIV+CAD+	p-value* HIV+CAD+  vs HIV-CAD-	p-value* HIV-CAD+  vs HIV+CAD-	p-value* HIV-CAD- vs HIV+CAD-	p-value* HIV-CAD- vs HIV-CAD+
Trp (µmol/L)	50.35 (40.08–57.28)	54.35 (47.3–59.63)		56.2 (53.2–59.9)	51 (47.2–63.6)	NS	0.09	0.02	NS	NS	NS	NS
Kyn (µmol/L)	2.26 (1.95–2.77)	2.31 (1.84–2.97)		2.12 (1.93–2.52)	2.11 (1.66–2.29)	NS	NS	NS	NS	NS	0.09	NS
Kyn/Trp ratio	0.048 (0.04–0.06)	0.042 (0.03–0.05)		0.038 (0.03–0.04)	0.03 (0.03–0.04)	0.004	NS	0.002	0.002	NS	NS	NS
3-H-Kyn (nmol/L)	36.8 (30.23–48.63)	37.6 (28.73–45.15)		34.8 (28.10–44.95)	32.2 (24.9–38.8)	NS	NS	NS	0.08	NS	NS	NS
3-H-Kyn/Kyn ratio	0.017 (0.013–0.022)	0.016 (0.014–0.020)		0.016 (0.014–0.020)	0.016 (0.014–0.018)	NS	NS	NS	NS	NS	NS	NS
Kynurenic acid (nmol/L)	49.05 (36.95–66.55)	47.4 (40.93–3.25)		47.8 (41.05–61.65)	48.1 (39.1–63)	NS	NS	NS	NS	NS	NS	NS
Kynurenic acid/Trp ratio	0.001 (0.0007–0.0015)	0.0008 (0.0007–0.001)		0.0009 (0.0007–0.001)	0.001 (0.0008–0.0011)	NS	NS	NS	NS	NS	NS	NS
Anthranilic acid (nmol/L)	15.2 (11.53–20.3)	15.95 (11.75–23.6)		12 (10–15.7)	13.4 (10.9–19.8)	0.07	NS	0.04	NS	0.02	NS	NS
Xanthurenic acid (nmol/L)	8.65 (6.35–14.1)	11.4 (5.15–15.55)		15.3 (8.6–18.4)	15.6 (11.30–19.7)	0.008	NS	0.03	0.004	0.05	0.01	NS
3-H-Kyn/Xan acid ratio	3.99 (2.7–6.35)	2.95 (2.34–5.16)		2.69 (2.28–3.47)	2.14 (1.66–2.79)	0.001	NS	0.05	0.0005	NS	0.002	0.08
3-H-Xyanthranilic acid (nmol/L)	30.65 (19.33–37.88)	30.95 (22.83–37.5)		30.4 (26.65–34.15)	25.4 (16.7–32.3)	NS	NS	NS	NS	NS	NS	NS
Picolinic acid (nmol/L)	34.95 (24.25–55.5)	29.45 (25.83–49.53)		36.6 (28.85–47.25)	31 (28.8–43.8)	NS	NS	NS	NS	NS	NS	NS
Quinolinic acid (nmol/L)	441 (339.5–599.8)	478 (379.5–598.5)		415 (347.5–526.5)	383 (341–621)	NS	NS	NS	NS	NS	NS	NS

Abbreviations: HIV Human immunodeficiency virus, CAD Coronary artery disease, Trp Tryptophan, Kyn Kynurenine, 3-H-Kyn 3-hydroxykynurenine, 3-H-Xyanthranilic acid 3-hydroxyanthranilic acid, Xan acid Xanthurenic acid

^*Mann–Whitney U test; Significant values are presented in bold; NS: Not significant

^$Kruskal–Wallis rank sum test

Peculiar myeloid and lymphoid immune cell features in HIV+CAD+ individuals

The flow cytometry gating strategies used to identify the myeloid cell subsets are shown in Supplementary Figure S1. Regardless of HIV status, classical CD14⁺CD16⁻ monocyte frequencies were enriched in CAD+ participants, while non-classical CD14⁻CD16⁺ monocytes were decreased in these individuals (Table 4). Accordingly, HIV+CAD+ individuals had the highest frequencies of classical and the lowest frequencies of non-classical monocytes among the 4 study groups (Table 4). A trend towards a decrease in intermediate monocytes was observed in PWH versus HIV seronegative individuals (p = 0.05). We further characterized the phenotype of the monocyte subsets by their expression of the activation marker CD163, the adhesion marker CD62L, and the chemokine receptors CCR2 and CX3CR1. We only observed an increased frequency of CCR2⁺ classical monocytes in CAD+ individuals, regardless of the HIV status, with a significant enrichment of CCR2⁺ classical monocytes in HIV+CAD+ versus HIV-CAD- participants (Table 4). No significant differences in the expression of CD163, CD62L, CXCR3, or in the frequency of myeloid DCs (mDCs, CD16⁻CD14⁻CD11c⁺) and plasmacytoid DCs (pDCs, CD16⁻CD14⁻CD123⁺) were observed among the study groups (Table 4).

Table 4.

Circulating myeloid cell subsets and their surface molecules expression

Myeloid cell subsets according to HIV and CAD status
	HIV+  [Median (%) (IQR)]	HIV- [Median (%) (IQR)]		p-value* (HIV+ vs HIV-)		CAD+  [Median (%) (IQR)]		CAD- [Median (%) (IQR)]		p-value* (CAD+ vs CAD-)
Classical monocytes	82.89 (69.72–90.26)	82.91 (72.34–88.17)		NS		85.84 (75.9–91.08)		81.60 (65.71–88.13)		0.02
Non-classical monocytes	9.97 (4.73–24.63)	10.46 (6.89–15.01)		NS		9.45 (4.2–14.86)		11.73 (7.07–26.73)		0.03
Intermediate monocytes	3.71 (1.63–6.72)	5.11 (2.98–6.95)		0.05		4.35 (1.86–6.93)		3.94 (1.99–6.67)		NS
Classical monocytes CCR2+	99 (98–99.4)	98.8 (98.18–99.4)		NS		99.2 (98.55–99.4)		98.7 (97.8–99.3)		0.008
Classical monocytes CCR2+ MFI	31,114 (27,410–33,921)	31,831 (29,570–33,923)		NS		31,945 (29,377–34,414)		30,341 (27,361–33,698)		NS
Classical monocytes CX3CR1+	48.12 (29.47–62.75)	49.85 (35.59–63.04)		NS		46.71 (30.95–60.74)		50.27 (35.42–65.51)		NS
Classical monocytes CD163+	49.98 (33.43–71)	48.16 (35.24–73.22)		NS		51.51 (34.87–70.92)		47.22 (32.30–72)		NS
Classical monocytes CD62L+	6.61 (2.1–14.14)	8.32 (3.13–12.12)		NS		8.07 (2.915–13.68)		6.44 (2.04–12.99)		NS
Non-classical monocytes CCR2+	28.01 (16.53–41.1)	23.13 (13.19–39.51)		NS		27.56 (16.4–39.02)		26.57 (13.21–41.4)		NS
Non-classical monocytes CCR2+ MFI	4049 (2044–6093)	3947 (1855–5912)		NS		3777 (2099–5825)		4117 (1677–6203)		NS
Non-classical monocytes CX3CR1+	66.90 (54–76.4)	67.89 (55.55–80.53)		NS		66.8 (55.2–77.15)		67.9 (52.9–80.29)		NS
Non-classical monocytes CD163+	57.8 (44.3–76.97)	61.93 (40.36–77.07)		NS		57.59 (42.17–76.24)		60.65 (43.32–77.22)		NS
Non-classical monocytes CD62L+	9.55 (3.82–14.2)	9.34 (5.33–14.82)		NS		9.55 (4.83–13.93)		9.25 (3.12–14.5)		NS
Intermediate monocytes CCR2+	98.1 (95.9–99.1)	97.4 (95.93–98.83)		NS		98.3 (96.78–99.1)		97.5 (95.19–98.85)		0.06
Intermediate monocytes CCR2+ MFI	31,263 (26,134–34,167)	30,269 (27,941–33,216)		NS		31,414 (27,975–33,962)		29,838 (25,550–33,236)		NS
Intermediate monocytes CX3CR1+	83.49 (69.04–91.34)	85 (73.37–91.46)		NS		84.59 (69.21–91.15)		83.87 (69.8–92.47)		NS
Intermediate monocytes CD163+	80 (62.5–93.3)	73.81 (61.95–96.98)		NS		79.40 (58.8–95.14)		79.45 (64.15–94.35)		NS
Intermediate monocytes CD62L+	20.58 (7.49–30.99)	20.23 (12.73–30.95)		NS		20.43 (9.19–31)		20.31 (8.55–30.96)		NS
mDCs CD16−CD14−CD11c+	0.031 (0.01–0.06)	0.025 (0.014–0.048)		NS		0.029 (0.01–0.05)		0.033 (0.02–0.07)		NS
pDCs CD16−CD14−CD123+	0.005 (0.003–0.009)	0.006 (0.003–0.01)		NS		0.005 (0.004–0.01)		0.005 (0.003–0.01)		NS

Myeloid cell subsets among 4 study groups
	HIV+CAD+  [Median (%) (IQR)]	HIV+CAD- [Median (%) (IQR)]	HIV-CAD+  [Median (%) (IQR)]	HIV-CAD- [Median (%) (IQR)]	Kruskal- Wallis p-value$	p-value* HIV+CAD+  vs HIV+CAD-	p-value* HIV-CAD+  vs HIV+CAD+	p-value* HIV+CAD+  vs HIV-CAD-	p-value* HIV-CAD+  vs HIV+CAD-	p-value* HIV-CAD- vs HIV+CAD-	p-value* HIV-CAD- vs HIV-CAD+
Classical monocytes	86.27 (76.12–93.07)	75.65 (63.11–88.41)	83.47 (72.34–89.37)	82.32 (69.61–87.84)	NS	0.02	NS	NS	NS	NS	NS
Non-classical monocytes	8.04 (3.57–14.79)	14.72 (6.82–31.89)	10.63 (5.19–19.19)	10.42 (7.07–14.27)	0.06	0.009	NS	NS	NS	NS	NS
Intermediate monocytes	3.78 (1.83–7.12)	3.32 (1.42–6.45)	5.16 (2.06–6.95)	4.95 (3.03–8.05)	NS	NS	NS	NS	NS	0.04	NS
Classical monocytes CCR2+	99.2 (98.6–99.4)	98.85 (97.8–99.38)	99.3 (98.4–99.5)	98.30 (97.88–99.28)	0.06	0.09	NS	0.03	0.08	NS	0.04
Classical monocytes CCR2+ MFI	31,793 (27,410–33,436)	30,450 (27,361–34,601)	33,037 (29,855–35,068)	30,196 (27,516–33,096)	NS	NS	0.09	NS	NS	NS	0.07
Classical monocytes CX3CR1+	47.73 (29.47–61.68)	49.6 (29.94–66.73)	44.51 (35.11–60.47)	50.9 (36.84–64.86)	NS	NS	NS	NS	NS	NS	NS
Classical monocytes CD163+	56.16 (36.15–75.4)	45.55 (25.48–66.93)	45.44 (27.93–65.26)	53.95 (40.02–89.32)	NS	NS	NS	NS	NS	NS	NS
Classical monocytes CD62L+	8.07 (2.76–16.55)	5.18 (1.09–13.76)	7.39 (2.87–11.22)	8.45 (3.51–12.62)	NS	NS	NS	NS	NS	NS	NS
Non-classical monocytes CCR2+	27.01 (17.16–35.4)	31.1 (15.05–46.38)	30.24 (15.27–42.72)	20.62 (10.91–34.53)	NS	NS	NS	NS	NS	NS	NS
Non-classical monocytes CCR2+ MFI	3754 (2223–5018)	4284 (1585–7361)	4644 (1868–6395)	3671 (1762–4617)	NS	NS	NS	NS	NS	NS	NS
Non-classical monocytes CX3CR1+	66.8 (55.4–76.4)	67.9 (50.74–78)	67.64 (53.75–79.44)	69.15 (56.2–85.75)	NS	NS	NS	NS	NS	NS	NS
Non-classical monocytes CD163+	57.59 (45.52–75.49)	58.6 (39.43–79.6)	59.03 (37.76–77.37)	62.34 (44.77–76.98)	NS	NS	NS	NS	NS	NS	NS
Non-classical monocytes CD62L+	10.05 (4.64–15.05)	8.72 (2.36–13.77)	8.84 (4.62–11.2)	9.99 (5.51–16.6)	NS	NS	NS	NS	NS	NS	NS
Intermediate monocytes CCR2+	98.3 (97.2–99.1)	97.8 (94.25–99.08)	98.4 (95.93–99.34)	97.32 (95.85–98.43)	NS	NS	NS	0.05	NS	NS	NS
Intermediate monocytes CCR2+ MFI	31,263 (26,764–33,516)	31,045 (24,570–36,815)	32,138 (29,783–35,316)	29,167 (27,345–31,207)	NS	NS	NS	NS	NS	NS	0.02
Intermediate monocytes CX3CR1+	82.52 (69.04–91.34)	83.65 (68.1–93.46)	85.00 (72.68–89.59)	84.75 (73.19–92.47)	NS	NS	NS	NS	NS	NS	NS
Intermediate monocytes CD163+	80 (59.2–94.58)	79.45 (62.83–92)	73.3 (55.2–96.08)	78.36 (65.23–97.35)	NS	NS	NS	NS	NS	NS	NS
Intermediate monocytes CD62L+	21.9 (8.56–33.13)	19.18 (5.89–30.81)	16.1 (10.61–30.95)	22.22 (13.5–33.92)	NS	NS	NS	NS	NS	NS	NS
mDCs CD16−CD14−CD11c+	0.03 (0.01–0.05)	0.03 (0.02–0.07)	0.03 (0.01–0.06)	0.025 (0.02–0.04)	NS	NS	NS	NS	NS	NS	NS
pDCs CD16−CD14−CD123+	0.005 (0.004–0.009)	0.005 (0.002–0.01)	0.006 (0.003–0.02)	0.006 (0.003–0.01)	NS	NS	NS	NS	NS	NS	NS

Abbreviations: HIV Human immune deficiency virus, CAD Coronary artery disease, MFI Median fluorescence intensity, mDCs myeloid dendritic cells, pDCs plasmacytoid dendritic cells

^*Mann–Whitney U test; Significant values are presented in bold; NS: Not significant

^$Kruskal–Wallis rank sum test

The flow cytometry gating strategies for identifying the T-cell subsets are presented in Supplementary Figures S2 and S3. In the T-cell compartment, CAD was associated with an increased frequency of central memory (CM) CD4 and CD8 T-cells, which contrasted with the effect of HIV, where CM CD4 T-cells were decreased (Table 5). Furthermore, HIV status was associated with increased frequencies of both CD4 and CD8 effector memory (EM) cells regardless of the CAD status. Within PWH, CAD+ individuals exhibited a higher frequency of CM CD4 and CD8 T-cells compared to HIV+CAD- individuals, while frequencies of EM CD4 and CD8 T-cells were higher in HIV+CAD+ participants compared to HIV-CAD- individuals (Table 5). In PWH, irrespective of CAD status, we observed increased frequencies of proliferative CD4 and CD8 T-cells (CD45RA⁻Ki67⁺), activated HLA-DR⁺ CD4 T-cells, and senescent CD4 and CD8 T-cells (CD28⁻CD57⁺). These elevations remained higher in HIV+CAD+ participants compared to HIV-CAD- individuals (Table 5). The frequencies of co-expressing CCR4⁺CXCR3⁺ CD4 and CD8 T-cells, as well as Tregs, were higher in PWH compared to HIV seronegative individuals (Table 5). This feature was also observed in HIV+CAD+ participants versus HIV-CAD- individuals. Lastly, the frequencies of CD4, CD8 and Treg cells expressing the atheroprotective ectonucleotidase CD73 were reduced in PWH and in HIV+CAD+ participants compared to HIV-CAD- individuals, whereas immunosuppressive CD39⁺ Tregs were increased in PWH (Table 5). Although there was a trend toward Th1 enrichment (p = 0.05) and reduced Th2 frequencies (p = 0.09) based on the HIV status, no significant differences were found in frequencies of total Tregs, Th17, or Th1Th17 cells. CMV status was associated with increased frequencies of proliferative Ki67⁺ T-cells, their differentiating toward EM phenotype, and diminished expression of CD73 (Supplementary Table 2). When analyzing the correlation results, the frequency of Th1Th17 cells in the HIV+CAD+ group showed a positive correlation with the volume of low-attenuation plaque. In conclusion, our findings reveal profound alterations in circulating immune cell profiles characterised by elevated frequencies and activation of classical monocytes, along with increased frequencies of activated and senescent atheroprotective CD4 and CD8 T-cell subsets in HIV+CAD+ individuals.

Table 5.

T cell subset characterization

T cell subset characterization according to HIV and CAD status
	HIV+  [Median (%) (IQR)]	HIV- [Median (%) (IQR)]		p-value* (HIV+ vs HIV-)		CAD+  [Median (%) (IQR)]		CAD- [Median (%) (IQR)]		p-value* (CAD+ vs CAD-)
T cell memory subsets
Naïve CD4 T cells	19.85 (12.03–31.03)	22.00 (12.6–32.23)		NS		20.7 (12.3–30.5)		19.8 (12.2–33.8)		NS
Naïve CD8 T cells	13.2 (7.22–26)	17.75 (12.20–34.55)		0.03		14 (7.18–25.35)		17.2 (9.83–34.9)		NS
Central memory CD4 T cells	31.55 (25.33–36.98)	35.25 (30.45–39.6)		0.009		33.5 (29.55–39.45)		32 (24.2–36.6)		0.04
Central memory CD8 T cells	5.73 (2.78–12.05)	5.63 (3.01–10.13)		NS		6.33 (3.995–14.1)		4.51 (2.34–7.9)		0.01
Effector memory CD4 T cells	2.09 (0.63–6)	0.83 (0.25–2.01)		0.0003		1.58 (0.55–3.98)		1.47 (0.36–3.88)		NS
Effector memory CD8 T cells	17.6 (10.25–26.48)	11.90 (7.04–18.78)		0.003		16.2 (9.62–22.85)		14.3 (8–25.1)		NS
Transitional memory CD4 T cells	33.4 (23.7–42.95)	29.45 (22.78–39.13)		NS		32.3 (23.55–40.45)		32.5 (22.7–43.4)		NS
Transitional memory CD8 T cells	25.2 (16.18–42.68)	26.60 (19.28–35.5)		NS		27.7 (19.5–42.25)		23.9 (15.5–40.2)		NS
Terminally differentiated CD4 T cells	0.12 (0.029–0.483)	0.062 (0.013–0.16)		0.03		0.073 (0.02–0.29)		0.099 (0.02–0.4)		NS
Terminally differentiated CD8 T cells	6.99 (2.99–13.45)	5.81 (3.95–11.93)		NS		6.49 (2.995–13.00)		6.91 (3.47–13.6)		NS
CD4 T cell subsets
Th1	11.4 (7.67–15.4)	10.03 (5.79–12.63)		0.05		11.7 (8.06–14.25)		9.10 (5.87–14.68)		NS
Th2	15.5 (11.70–22.30)	18.7 (14.95–23.78)		0.09		18 (14.60–22.2)		17.15 (12.65–25.88)		NS
Th17	6.4 (3.8–8.35)	5.94 (4.06–7.51)		NS		5.97 (3.78–8.395)		6.28 (4.02–8.08)		NS
Th1/Th17	2.31 (1.46–3.6)	2.11 (1.17–3.46)		NS		2.41 (1.54–3.66)		1.99 (1.23–3.35)		NS
Tregs	2.92 (2.23–4.56)	2.75 (2.21–3.83)		NS		3.06 (2.23–4.24)		2.84 (2.22–4.41)		NS
Th17/Tregs ratio	1.86 (1.27–2.71)	1.99 (1.19–3.06)		NS		1.95 (1.40–2.65)		1.84 (1.18–3.06)		NS
T cell activation markers
CD4+HLA-DR+	8.47 (5.45–13.73)	6.32 (3.83–10.05)		0.02		7 (4.32–11.1)		7.73 (4.6–13.3)		NS
CD8+HLA-DR+	12.3 (6.198–17.83)	7.52 (4.195–16.1)		NS		9.71 (5.53–18.55)		10.7 (4.82–16.4)		NS
CD4+CD38+	43.55 (30–50.33)	40.95 (31.4–54.45)		NS		40 (33.4–50.7)		44.2 (29.9–52.3)		NS
CD8+CD38+	11.3 (6.62–20.05)	11.6 (7.37–18.08)		NS		10.9 (6.71–18.25)		12.5 (7.49–20.4)		NS
CD4+HLA-DR+CD38+	3 (1.47–5.34)	2.21 (1.15–4.78)		NS		2.28 (1.39–4.84)		2.95 (1.43–5.29)		NS
CD8+HLA-DR+CD38+	2.76 (1.46–4.67)	1.95 (1.05–3.62)		0.08		2.51 (1.64–4.44)		1.92 (1.15–4.06)		NS
CD4+CD45RA−Ki67+	0.89 (0.59–1.58)	0.63 (0.45–0.95)		0.001		0.77 (0.52–1.4)		0.74 (0.49–1.36)		NS
CD8+CD45RA−Ki67+	0.83 (0.51–1.46)	0.595 (0.34–1.05)		0.008		0.79 (0.48–1.32)		0.74 (0.35–1.21)		NS
T cell exhaustion
CD4+CTLA4+	5.45 (3.13–7.84)	4.7 (3.09–8.43)		NS		5.41 (3.13–8.44)		4.44 (3.1–7.62)		NS
CD8+CTLA4+	1.71 (0.69–3.72)	1.71 (0.83–4.71)		NS		1.93 (0.75–4.18)		1.7 (0.64–4.04)		NS
T cell senescence
CD4+CD28−CD57+	1.15 (0.14–5.41)	0.21 (0.03–1.38)		0.002		0.65 (0.09–3.02)		0.63 (0.07–4.43)		NS
CD8+CD28−CD57+	21.7 (11.68–33.28)	16.1 (8.76–24.2)		0.05		20 (10.06–30.05)		20 (9.61–28)		NS
Migration markers
CD4+CD45RA−CXCR3+	23.9 (15.4–31.6)	17.05 (11.65–22.58)		0.0002		22.3 (15.1–27.4)		18.7 (13.15–30.93)		NS
CD8+CD45RA−CXCR3+	20.3 (12.6–29.8)	13.35 (7.27–18.75)		0.0005		18.4 (11.65–30.7)		15.1 (9.71–23.48)		NS
CD4+CD45RA−CCR4+	34.2 (27.7–43)	31.6 (26.1–39.58)		NS		33.8 (27.5–40.4)		33.25 (26.25–42.08)		NS
CD8+CD45RA−CCR4+	15.7 (9.72–29.4)	14.85 (9.06–22.53)		NS		16.7 (12–25.15)		13.6 (8.77–23.15)		0.055
CD4+CD45RA−CCR6+	14.6 (9.97–20)	12.3 (8.92–16.43)		NS		13.9 (8.85–18.65)		12.75 (9.95–18.65)		NS
CD8+CD45RA−CCR6+	11.5 (8.48–15)	10.7 (8.09–14.6)		NS		12.1 (9.21–14.95)		9.88 (6.97–14.25)		0.06
CD4+CD45RA−CCR4+CXCR3+	9.63 (6.42–15)	6.14 (3.47–8.55)		< 0.0001		8.61 (5.84–11.1)		7.76 (4.295–13.55)		NS
CD8+CD45RA−CCR4+CXCR3+	4.48 (2.06–7.41)	3.24 (1.42–5.77)		0.046		4.64 (2.57–8.93)		3.21 (1.67–6.13)		0.04
Tregs CD45RA−CCR4+CXCR3+	11.7 (8.41–18)	8.40 (4.9–12)		0.0002		11 (7.45–14.1)		10.3 (5.48–17.78)		NS
Ectonucleotidases
CD4+CD73+	8.22 (5.08–12)	12.25 (9.77–16.13)		< 0.0001		9.06 (6.96–14.10)		10.55 (5.73–14.2)		NS
CD8+CD73+	36.9 (24–46.4)	49.45 (31.28–60.13)		0.004		35.7 (26.7–50)		41.35 (27.45–57.05)		NS
CD4+CD39+	6.95 (3.51–8.84)	4.98 (2.27–6.51)		0.02		5.91 (3.53–8.33)		5.99 (3.14–7.88)		NS
CD8+CD39+	1.1 (0.4–2.03)	1.2 (0.66–2.77)		NS		1.21 (0.45–2.13)		1.08 (0.52–2.15)		NS
CD4+CD39+CD73+	0.26 (0.12–0.37)	0.23 (0.11–0.35)		NS		0.21 (0.13–0.35)		0.27 (0.11–0.38)		NS
CD8+CD39+CD73+	0.23 (0.09–0.41)	0.29 (0.13–0.53)		NS		0.27 (0.11–0.44)		0.23 (0.097–0.46)		NS
Tregs CD39+	53.3 (35.00–67.6)	45.9 (26.8–58.83)		0.049		49.6 (35.75–61.05)		52.30 (24.95–64.03)		NS
Tregs CD73+	5.91 (3.76–9.24)	7.86 (6.02–12.45)		0.005		6.58 (3.97–9.73)		7.44 (4.54–10.7)		NS

T cell subset characterization among 4 study groups
	HIV+CAD+  [Median (%) (IQR)]	HIV+CAD- [Median (%) (IQR)]	HIV-CAD+  [Median (%) (IQR)]	HIV-CAD- [Median (%) (IQR)]	Kruskal- Wallis p-value$	p-value* HIV+CAD+  vs HIV+CAD-	p-value* HIV-CAD+  vs HIV+CAD+	p-value* HIV+CAD +  vs HIV-CAD-	p-value* HIV-CAD+  vs HIV+CAD-	p-value* HIV-CAD- vs HIV+CAD-	p-value* HIV-CAD- vs HIV-CAD+
T cell memory subsets
Naïve CD4 T cells	20.7 (12.8–30)	19.3 (11.4–33.8)	20.95 (11.9–32.08)	23.1 (15.4–33.6)	NS	NS	NS	NS	NS	NS	NS
Naïve CD8 T cells	13.3 (6.19–25.2)	13.1 (9.1–30.2)	14.75 (11.35–28.38)	26.1 (12.73–36.43)	0.07	NS	NS	0.009	NS	0.06	NS
Central memory CD4 T cells	32.2 (29.4–38.4)	29.2 (22.5–35.3)	37.2 (32.35–40.3)	34.40 (30.15–37.55)	0.008	0.04	NS	NS	0.001	0.03	NS
Central memory CD8 T cells	6.5 (3.97–14.2)	4.78 (2.16–7.23)	5.995 (4.53–10.94)	3.86 (2.36–10.18)	NS	0.04	NS	NS	0.09	NS	NS
Effector memory CD4 T cells	1.91 (0.57–4.33)	2.11 (0.91–7.69)	1.27 (0.41–2.46)	0.57 (0.18–2.02)	0.002	NS	NS	0.002	0.08	NS	0.0003
Effector memory CD8 T cells	16.9 (10.6–24.1)	18.9 (9.95–27.9)	13.65 (8.62–22.55)	11.25 (5.18–16.68)	0.01	NS	NS	0.008	NS	0.005	NS
Transitional memory CD4 T cells	33.4 (23.1–41.7)	33.4 (23.9–46)	30.35 (23.78–39.78)	28.95 (20.1–38.7)	NS	NS	NS	NS	NS	NS	NS
Transitional memory CD8 T cells	29.5 (16.4–42.6)	23.7 (15.5–43)	26.6 (21.78–41.83)	25.85 (13.70–33.45)	NS	NS	NS	NS	NS	NS	NS
Terminally differentiated CD4 T cells	0.097 (0.022–0.550)	0.160 (0.030–0.460)	0.051 (0.017–0.135)	0.069 (0.008–0.205)	NS	NS	NS	NS	0.06	NS	NS
Terminally differentiated CD8 T cells	7.05 (2.4–13.2)	6.91 (3.46–13.6)	5.72 (4.22–11.78)	6.46 (3.82–13.15)	NS	NS	NS	NS	NS	NS	NS
CD4 T cell subsets
Th1	11.7 (8.57–14.8)	9.63 (6.15–15.58)	11.4 (5.24–13.43)	8.53 (5.72–12.08)	NS	NS	NS	0.03	NS	NS	NS
Th2	17.9 (12.1–22.1)	16.2 (11.88–26.9)	19.8 (16.35–24.03)	18.7 (15.03–23.65)	NS	NS	NS	NS	NS	NS	NS
Th17	6.1 (3.71–8.44)	6.56 (4.07–8.26)	5.97 (4.07–8.47)	4.98 (3.75–7.13)	NS	NS	NS	NS	NS	NS	NS
Th1/Th17	2.41 (1.52–3.65)	2.14 (1.42–3.55)	2.44 (1.48–3.82)	1.86 (0.89–2.94)	NS	NS	NS	NS	NS	NS	NS
Tregs	3.57 (2.23–4.63)	2.86 (2.19–4.53)	2.96 (2.09–3.3)	2.71 (2.24–4.32)	NS	NS	NS	NS	NS	NS	NS
Th17/Tregs ratio	1.86 (1.23–2.54)	1.89 (1.35–3.10)	2.28 (1.46–4.14)	1.83 (0.93–3.05)	NS	NS	0.06	NS	NS	NS	NS
T cell activation markers
CD4+HLA-DR+	7.49 (4.34–13.6)	8.48 (6.17–15.7)	6.16 (3.57- 9.41)	7.68 (3.98–10.15)	0.08	NS	NS	NS	0.007	0.09	NS
CD8+HLA-DR+	12 (5.6–18.9)	12.4 (7.06–16.7)	9.54 (4.62–16.7)	7.3 (3.97–16.3)	NS	NS	NS	NS	NS	NS	NS
CD4+CD38+	42.1 (33.9–50.1)	43.8 (24.5–52.3)	36.4 (29.40–55.15)	47.1 (32.58–54.88)	NS	NS	NS	NS	NS	NS	NS
CD8+CD38+	11.6 (7.61–19.9)	11 (5.93–20.7)	9.6 (5.53–14.18)	13.7 (9.93–19.7)	NS	NS	NS	NS	NS	NS	0.07
CD4+HLA-DR+CD38+	2.57 (1.43–5.95)	3.01 (1.52–5.29)	1.97 (1.24–4.53)	2.44 (1.08–5.54)	NS	NS	NS	NS	NS	NS	NS
CD8+HLA-DR+CD38+	2.8 (1.69–4.44)	2.76 (1.31–5.24)	2.25 (0.89–4.60)	1.69 (1.08–3.59)	NS	NS	NS	0.08	NS	NS	NS
CD4+CD45RA-Ki67+	0.88 (0.49–1.56)	0.92 (0.69–1.89)	0.76 (0.54–1.05)	0.54 (0.42–0.85)	0.005	NS	NS	0.009	0.09	0.0005	NS
CD8+CD45RA-Ki67+	0.85 (0.51–1.57)	0.82 (0.50–1.32)	0.7 (0.44–1.18)	0.5 (0.28–0.98)	0.04	NS	NS	0.01	NS	0.02	NS
T cell exhaustion
CD4+CTLA4+	5.55 (3.84–8.84)	4.38 (3.10–7.49)	4.48 (2.63–8.24)	4.76 (3.12–8.82)	NS	NS	NS	NS	NS	NS	NS
CD8+CTLA4+	1.95 (0.74–3.76)	1.59 (0.59–3.53)	1.33 (0.998–5.21)	2.26 (0.82–4.14)	NS	NS	NS	NS	NS	NS	NS
T cell senescence
CD4+CD28−CD57+	0.97 (0.13–3.83)	1.18 (0.14–5.87)	0.42 (0.03–1.55)	0.13 (0.02–1.72)	0.02	NS	NS	0.02	0.04	0.005	NS
CD8+CD28−CD57+	20 (10.3–32)	22.4 (12.8–33.9)	18.65 (9.75–25.2)	13.3 (8.03–24.08)	NS	NS	NS	0.06	NS	0.05	NS
Migration markers
CD4+CD45RA−CXCR3+	23.6 (15.4–30.9)	24.55 (15.4–35.85)	19.75 (12.43–23.43)	15.25 (10.38–20.43)	0.004	NS	0.02	0.002	0.03	0.001	NS
CD8+CD45RA−CXCR3+	20.3 (13.3–30.8)	19.95 (12.45–27.95)	13.35 (8.35–23.33)	13.35 (6.42–17.4)	0.004	NS	0.07	0.001	NS	0.004	NS
CD4 + CD45RA−CCR4+	33.8 (27–43)	33.25 (26.20–44.58)	31.6 (25.60–38.23)	30.95 (23.20–39.4)	NS	NS	NS	NS	NS	NS	NS
CD8+CD45RA−CCR4+	15.3 (11.4–25.7)	13.85 (8.30–29.4)	18.35 (11.20–25.75)	13.15 (8.87–21.03)	NS	NS	NS	0.07	NS	NS	NS
CD4+CD45RA−CCR6+	13.9 (8.13–18.2)	14.75 (10.05–20)	13.6 (8.92–19.23)	12.15 (7.80–15.38)	NS	NS	NS	NS	NS	NS	NS
CD8+CD45RA−CCR6+	11.7 (9.2–15)	10.9 (6.97–15.23)	13.45 (10.21–15)	8.81 (6.99–12.68)	0.08	NS	NS	0.04	NS	NS	0.01
CD4+CD45RA−CCR4+CXCR3+	9.17 (6.69–12.6)	10.07 (6.22–15.85)	6.14 (4.69- 9.55)	5.98 (2.66–7.95)	< 0.0001	NS	0.006	0.0001	0.006	0.0003	NS
CD8+CD45RA−CCR4+CXCR3+	4.48 (2.81–8.59)	4.38 (1.94–7.33)	5.18 (2.42–10.4)	2.62 (0.92–4.65)	0.01	NS	NS	0.002	NS	0.01	0.02
Tregs CD45RA−CCR4+CXCR3+	11.3 (8.41–15.8)	12.5 (7.18–21.33)	8.41 (5.63–12.4)	6.56 (4.47–11.55)	0.003	NS	0.07	0.004	0.02	0.001	NS
Ectonucleotidases
CD4+CD73+	8.1 (6.29–10.9)	9.11 (4.28–13.33)	13.15 (9.51–15.68)	11.85 (9.98–16.18)	0.001	NS	0.005	0.0003	0.02	0.005	NS
CD8+CD73+	32.5 (24–44.3)	39.65 (23.6–48.05)	40 (30.83–56.23)	54.6 (32.4–65.55)	0.02	NS	NS	0.007	NS	0.008	NS
CD4+CD39+	7.08 (3.41–9.35)	6.58 (3.51–8.75)	4.73 (3.36–6.05)	5.07 (1.22–6.83)	NS	NS	0.06	0.07	NS	NS	NS
CD8+CD39+	1.21 (0.4–2.18)	1 (0.39–1.97)	1.24 (0.48–2.3)	1.16 (0.7–2.84)	NS	NS	NS	NS	NS	NS	NS
CD4+CD39+CD73+	0.2 (0.13–0.35)	0.27 (0.10–0.38)	0.23 (0.11–0.32)	0.23 (0.11–0.36)	NS	NS	NS	NS	NS	NS	NS
CD8+CD39+CD73+	0.29 (0.09–0.45)	0.22 (0.08–0.4)	0.26 (0.11–0.46)	0.39 (0.15–0.58)	NS	NS	NS	NS	NS	NS	NS
Tregs CD39+	50.5 (35–64.9)	54.6 (33.18–67.68)	47.8 (35.18–57.95)	45.25 (16.15–59.28)	NS	NS	NS	NS	NS	0.06	NS
Tregs CD73+	6.39 (3.76–8.27)	5.85 (3.77–10.57)	7.53 (6.1–12.55)	8.13 (5.98–12.05)	0.04	NS	0.059	0.01	NS	0.058	NS

Abbreviations: Tregs regulatory T-cells, HIV Human immunodeficiency virus, CAD Coronary artery disease

^*Mann–Whitney U test; Significant values are presented in bold; NS: Not significant

^$Kruskal–Wallis rank sum test

Multivariate analysis reveals distinct immuno-metabolic signatures associated with HIV infection and subclinical CAD

Multivariate linear regression analyses were used to adjust for the impact of potential confounders on the associations between HIV and CAD status and key immuno-metabolic outcomes including Kyn/Trp ratio, I-FABP, and sCD14, (Table 6). Based on the Change in Estimate (CIE) method, Model 1 (Kyn/Trp ratio outcome) was adjusted for: Age, Cigarette packs per year, BMI, LDL, Insulin, Total Cholesterol; Model 2 (I-FABP outcome) was adjusted for: BMI, Insulin; and, Model 3 (sCD14 outcome) was adjusted for: Age, Cigarette packs per year, LDL, Insulin. In the Model 1 (Kyn/Trp ratio outcome), the CIE method included every potential confounder to the adjusted model; therefore, the crude and adjusted β were identical. HIV status appeared as a significant predictor for the Kyn/Trp ratio outcome, sharing the effect with Age and Total Cholesterol variables, as all three variables showed a p-value < 0.05. For the second and third models, adjustment for identified confounders slightly decreased the crude point of the exposures estimates yet retained their statistical significance in the models. In the Model 2 (I-FABP outcome), both HIV and CAD status were significant predictors of the I-FABP outcome, with HIV status decreasing from 1574.55 to 1563.57 (a 0.70% decrease; p-value < 0.05) and CAD status decreasing from 1243.23 to 1198.99 (a 3.56%; p-value < 0.05). In this model, BMI emerged as a significant confounder. Despite the shared effect, independent and significant associations for HIV, CAD and BMI remained on the outcome. In the third model, both HIV status (crude β of 315.32 to adjusted β of 300.28) and CAD status (crude β of 34.52 to an adjusted β of 34.05) showed slight decreases in point estimates from the crude to adjusted analysis although significance remained, with a p-value for HIV status < 0.05 in the adjusted model. No potential confounder appeared as a significant predictor for this model.

Table 6.

Multivariate linear regression analysis for associations between HIV and CAD status and primary study outcomes

Outcome model and Exposure	Crude β (CI)	Adjusted β (CI)	Adj.p
Model 1: Kyn/Trp ratio outcome
HIV status	1.04 × 10–2 [3.9 × 10–3, 1.68 × 10–2]	1.04 × 10–2 [3.9 × 10–3, 1.68 × 10–2]	0.002
CAD status	5.71 × 10–6 [-6.06 × 10–3, 6.07 × 10–3]	5.71 × 10–6 [-6.06 × 10–3, 6.07 × 10–3]	0.99
Significant confounders
Age	4.48 × 10–4 [5.34 × 10–5, 8.43 × 10–4]	4.48 × 10–4 [5.34 × 10–5, 8.43 × 10–4]	0.03
Total Cholesterol	-5.70 × 10–3 [-1.12 × 10–2, -2.22 × 10–4]	-5.70 × 10–3 [-1.12 × 10–2, -2.22 × 10–4]	0.04
Model 2: I-FABP outcome
HIV status	1574.55 [634.39, 2514.72]	1563.57 [655.94, 2471.21]	0.0009
CAD status	1243.23 [371.34, 2115.13]	1198.99 [395.06, 2002.93]	0.004
Significant confounders
BMI	-127.29 [-216.07, -38.51]	-120.19 [-208.32, -32.05]	0.008
Model 3: sCD14 outcome
HIV status	315.32 [106.13, 524.51]	300.28 [100.87, 499.69]	0.004
CAD status	34.52 [-158.88, 227.93]	34.05 [-157.90, 226.00]	0.73
Significant confounders
None	None	None	None

Model 1 adjusted for: Age, Cigarette packs per year, BMI, LDL, Insulin, Total Cholesterol

Model 2 adjusted for: BMI, Insulin

Model 3 adjusted for: Age, Cigarette packs per year, LDL, Insulin

Significant p-value (p < 0.05) are presented in bold, β: point of estimate, CI: 95% Confidence Interval, Adj.p: p-value from the adjusted model. Significance level for confounders was p < 0.05

Discussion

The primary goal of this study was to perform a comprehensive analysis to better understand immuno-metabolic features linked to inflammaging and disruptions in the gut-heart axis during the early stages of CAD pathophysiology in ART-treated PWH, diagnosed by coronary CT angiography. Overall, our results indicate that in PWH, subclinical CAD is associated with a peculiar gut-heart immuno-metabolic features characterized by (i) elevated levels of markers of gut mucosal damage and microbial translocation, along with (ii) increased plasma IFN-γ and sTNFRII levels, (iii) upregulated IDO/Kyn pathway activity, (iv) increased frequencies of classical monocytes and their activation, and (v) increased T-cell activation, senescence, and homing capacity, accompanied by a depletion of cardioprotective T-cell subsets. Importantly, most of these observed changes were primarily associated with HIV status.

Plasma levels of markers such as I-FABP, indicative of enterocyte turnover, and REG3α, reflecting intestinal barrier damage, serve as direct indicators of ongoing gut mucosal damage, and are increased during chronic HIV infection [9, 51, 54]. Additionally, sCD14, a marker of monocyte activation in response to lipopolysaccharide (LPS), provides an indirect measure of microbial translocation [54, 55]. Among PWH on ART, microbial translocation has been associated with hypercoagulability and an enhanced risk of thrombosis, which can potentially lead to cardiovascular events [56]. Our results showed an increase in plasma levels of the aforementioned markers in PWH versus HIV seronegative individuals, while only I-FABP was also elevated in relation to CAD status, consistent with a previous study showing its association with CAD [57, 58]. In multivariate analyses, BMI emerged as a significant confounder of I-FABP levels, while both HIV and CAD status maintained their independent association with I-FABP levels. This may reflect the influence of body composition on the intestinal epithelial barrier integrity [59]. Moreover, an inverse association between BMI and I-FABP was observed in our study, consistent with previous findings in PWH [60]. In addition, elevated plasma levels of sCD14 and REG3α were positively correlated with the total plaque volume and the volume of low attenuation plaque. Notably, HIV+ CAD+ participants had the highest levels of I-FABP, REG3α and sCD14 among all study groups. Importantly, sCD14 is recognized as an independent predictor of CAD risk, CVD-related outcomes, and mortality in Black populations [61, 62]. In our study, however, the majority of both PWH participants and HIV seronegative individuals were Caucasian. Overall, these findings support prior evidence suggesting that, despite effective ART, gut barrier dysfunction persists in PWH and is associated with presence of premature CAD and an elevated risk of CVD in these individuals.

In addition to gut-related alterations, elevated plasma levels of inflammatory markers have been linked to coronary atherosclerosis in PWH [8]. Imaging studies using PET/CT have demonstrated increased arterial inflammation in this population, which correlates with circulating inflammatory markers [63, 64]. Among several plasma inflammatory mediators analyzed in our study, elevated plasma levels of sTNFRII, a predictor of coronary heart disease [65], were observed in PWH regardless of CAD status, and with the highest concentrations observed in CAD+HIV+ individuals. These findings suggest the involvement of chronic TNF-α pathway activation in the early stages of CAD in PWH. Notably, plasma levels of two other well-known pro-inflammatory cytokines implicated in atherosclerosis, IFN-γ and IL-17A, were elevated in CAD+ individuals regardless of HIV status, with further increase observed in HIV+CAD+ participants. As mentioned, IFN-γ is involved in various steps of vascular inflammation, atherosclerotic plaque formation and rupture [14]. Importantly, we observed that the levels of IFN-γ were associated with REG3α and sCD14 suggesting its link with gut damage and microbial translocation. Regarding IL-17A, we observed a positive correlation with both total plaque volume and low-attenuation plaque volume, specifically in HIV+CAD+ participants, consistent with its role as a pro-inflammatory cytokine involved in atherogenesis [21]. Our results also align with a previous study showing that, in a subset of CAD+ patients, coronary artery-infiltrating T-cells simultaneously produce both IL-17 and IFN-γ, which act synergistically to amplify pro-inflammatory responses in vascular smooth muscle cells [66]. Furthermore, elevated IL-17A levels have been reported in the plasma of hypertensive PWH [18] and are increased in mouse models of hypertension and atherosclerosis [17]. Moreover, in PWH, IL-17A levels have been associated with flow-mediated dilation, a key indicator of endothelial dysfunction [20].

We also observed significant changes in plasma levels of Trp and its IDO1-induced downstream metabolites, such as increased Kyn, 3-H-Kyn, and anthranilic acid levels, as well as Kyn/Trp and 3-H-Kyn/Xan acid ratios in PWH compared to HIV seronegative controls, irrespective of CAD status. Importantly, HIV+CAD+ participants had the highest Kyn/Trp and 3-H-Kyn/Xan acid ratios which are well-established markers of CVD and CAD risk [30, 32, 37–40, 67]. Multivariate analysis identified age and total cholesterol as significant confounders of the Kyn/Trp ratio, suggesting that the IDO/Kyn pathway may be influenced by aging and total cholesterol levels, in line with previous studies [68, 69]. The significant upregulation of the IDO/Kyn pathway is consistent with elevated IFN-γ levels and a positive correlation between IFN-γ and the Kyn/Trp ratio in our HIV+CAD+ participants, supporting the role of IFN-γ as a key inducer of IDO1 expression and downstream pathway activation [31]. Accordingly, the negative correlation between Trp and sCD14 levels observed in HIV+CAD+ is consistent with a link between IDO1 upregulation and monocyte activation [55]. The upregulation of IDO/Kyn pathway might also be related to gut dysbiosis since Trp is an essential dietary amino acid and its catabolism is also depends on gut microbiota composition [31]. We previously reported CAD-associated gut microbiota dysbiosis in PWH within the CHACS cohort [70]. When analyzed based on the CAD status, only a trend toward increased 3-H-Kyn/Xan acid ratio was observed, which is in line with a previous report showing that elevated plasma levels of 3-H-Kyn and 3-H-Kyn/Xan acid were linked to increased mortality in patients with CAD and heart failure [30]. In line with our results, elevated plasma levels of Kyn, anthranilic acid, and 3-H-Kyn are associated with oxidative stress, vascular endothelial dysfunction, metabolic syndrome and CVD-related deaths [67, 71]. A previous study found that elevated plasma kynurenic acid/Trp ratios were linked to increased carotid plaque risk, with reduced kynurenic acid levels observed only in viremic PWH but not in aviremic participants [42]. Accordingly, in our study, aviremic PWH on effective ART showed no significant differences in kynurenic acid levels or kynurenic acid/Trp ratios compared to HIV seronegative controls. Interestingly, we observed a marked reduction in plasma xanthurenic acid levels among PWH, similar to another recent study [72]. Importantly, plasma xanthurenic acid has been shown to be inversely correlated to CVD risk and CAD mortality [30, 71]. Thus, our data suggest that ART-treated PWH in our cohort represent with metabolic reprogramming by enhanced IDO activity toward pro-atherogenic metabolites (Kyn, 3-H-Kyn, anthranilic acid) with a depletion of Xan acid, a metabolite that is associated in other cohorts with reduced cardiovascular mortality. This imbalance may potentiate early stages of CAD progression in PWH by diminishing innate metabolic protection and amplifying inflammatory vascular injury.

Considering that IDO1 is predominantly expressed by myeloid cells [31], which are key contributors to atherosclerosis [73, 74], we next analyzed circulating myeloid subsets. Our data revealed a shift in monocyte subset distribution, characterized by increased frequencies of classical monocytes (CD14⁺CD16^-), along with reduced frequencies of non-classical monocytes (CD14^-CD16⁺) in CAD+ individuals regardless of HIV status, and in HIV+CAD+ participants. In humans, classical monocytes (the predominant subset of circulating monocytes), are key inflammatory mediators and constitute the major monocyte population detected within atherosclerotic plaques [75]. Importantly, classical monocytes are known to independently predict future cardiovascular risk and progression of CAD in humans [76, 77]. Additionally, higher frequencies of classical monocytes have been linked to an increased risk of death and early clinical worsening in patients with stroke [78, 79].

The increases in classical monocyte frequencies observed in HIV+CAD+ participants is in line with a similar trend observed in another recent CHACS cohort study, which also reported a trend in higher frequencies of classical monocytes in HIV+CAD+ individuals versus HIV+CAD-, and lower frequencies of intermediate monocytes in PWH versus HIV seronegative individuals [80]. However, in contrast to our findings, that study found increased classical monocyte frequencies based on the HIV status [80]. This discrepancy may be explained by differences in the CHACS cohort enrollment dates at which biological specimens were collected, as well as by minor variations in the flow cytometry panels used in two studies. In our study, we also observed a CAD-associated increase in the frequency of CCR2⁺ classical monocytes, which were also enriched in HIV+CAD+ individuals versus HIV-CAD- participants. Previous experimental murine studies have demonstrated that CCR2 expression on circulating monocytes plays a critical role in atherosclerotic processes [81, 82]. In humans, elevated CCR2 expression on monocytes has been linked to arterial wall inflammation in patients at increased risk of cardiovascular disease [83]. Notably, within atherosclerotic lesions, CCL2 becomes anchored to the plasma membrane of endothelial cells, where it interacts with CCR2 on circulating monocytes, promoting their adhesion to the endothelium and subsequent transmigration into the subendothelial space [84]. Altogether, in HIV+CAD+ individuals, increased CCR2⁺ classical monocytes and upregulated IDO/Kyn pathway, together with elevated plasma sCD14 levels, reflect chronic monocyte activation, possibly due to gut mucosal damage and increased microbial translocation in this group.

Regarding T-cells, independent of the CAD status, PWH exhibited elevated frequencies of activated (HLADR⁺) CD4 T-cells as well as proliferating (Ki67⁺) and senescent (CD28⁻CD57⁺) CD4 and CD8 T-cells. These findings align with previous studies showing that chronic immune activation and inflammation persist in PWH, despite effective ART [85, 86]. Persistent immune activation is known to be associated to accelerated T-cell turnover and differentiation, which may favor the accumulation of senescent T-cell populations with reduced effector capacity [3]. We observed a distinct memory CD4 and CD8 T-cell profile based on HIV and CAD status, with HIV infection associated with increased EM T-cells, while CAD was rather linked to a CM phenotype. Notably, HIV+CAD+ individuals exhibited higher frequencies of CM T-cells compared to HIV+CAD_⁻ participants, potentially reflecting an intermediate or transitional state in which chronic HIV-associated immune activation sustains memory T-cell differentiation and atherosclerotic immune priming. Increased IFN-γ levels observed in HIV+CAD+ participants might be explained by the increasing trend observed in Th1 cell frequencies in PWH. However, no difference has been observed for Th17 cells among our study groups. Notably, contrasting results exist regarding the Th17 in the context of PWH with CAD, since one study showed that higher frequencies of Th17 cells was significantly associated with an increased risk for incident CVD [87], while a recent study from our group using samples from the same cohort demonstrated that ART-treated PWH with subclinical coronary plaques showed lower Th17 frequencies than those without plaques [80]. Additionally, we observed an increased frequency of CXCR3⁺ CD4 and CD8 T-cells in PWH, in line with previous studies reporting elevated CXCR3 expression in the context of chronic HIV infection [88]. Also, PWH compared to HIV seronegative individuals had increased frequency of memory CD4, CD8, and Treg cells co-expressing T-cell heart homing markers CCR4 and CXCR3 (CD45RA⁻CCR4⁺CXCR3⁺). It has been shown that T-cells can be detected within atherosclerotic plaques and are recruited into the heart through mechanisms involving the expression of chemokine receptors CCR4 and CXCR3 along with the interaction between the tyrosine-protein kinase c-Met and hepatocyte growth factor (HGF) expressed by the heart [89, 90]. We further observed alterations in purinergic metabolic pathway, mediated by the ectonucleotidases CD39 and CD73 on T-cells. CD39 hydrolyzes pro-inflammatory extracellular ATP into ADP and AMP, which are subsequently converted into immunosuppressive adenosine by CD73, another ectonucleotidase [91, 92]. The purinergic pathway is also well-recognized for its protective role against cardiovascular diseases [93, 94]. In our study, atheroprotective CD73⁺ subsets within CD4, CD8 and Tregs were reduced in PWH, whereas immunosuppressive CD39⁺ Tregs were increased. Importantly, HIV+CAD+ participants had the lowest T-cell expression of CD73 among all study groups. This imbalance in purinergic pathway aligns with our previous studies demonstrating that increased CD39⁺ Treg frequencies are associated with suboptimal immune responses and HIV disease progression [95–97], and, confirm our previous observations in the CHACS cohort [98].

This study has several limitations. First, the observational and cross-sectional study design restricts the ability to draw causal inferences between immuno-metabolic alterations observed in PWH and atherosclerotic plaque development in these individuals, and limits our study outcomes to be hypothesis-generating Therefore, further longitudinal studies are needed to assess how the peculiar immuno-metabolic profile observed in HIV+CAD+ individuals may be associated with atheroma plaque development over time. Second, our study groups included relatively small sample size which did not allow us to apply a False Discovery Rate (FDR) correction for multiple comparisons. Third, our cohort had a higher proportion of males than females in both the PWH and uninfected control groups, which may limit generalizability across sexes. Notably, sex-related differences have been reported in both HIV- and CAD-related pathophysiology [99], with a higher CAD risk observed among women compared with men living with HIV [100]. Accordingly, we observed reduced atheroprotective CD73⁺ T-cell subsets in male participants. Lastly, a large proportion of PWH in our study were co-infected with CMV, a factor known to influence systemic immune activation and inflammaging. As such, we observed higher I-FABP levels and more advanced T-cell differentiation and senescence in CMV+ participants, along with their reduced atheroprotective CD73⁺ T-cell subsets.

Conclusion

In conclusion, our findings provide new insights into a peculiar immuno-metabolic feature associated with disrupted gut-heart axis in early stages of CAD among ART-treated PWH, which are mainly associated with HIV status. Our findings suggest a peculiar model of inflammaging in PWH that gut mucosal damage and microbial translocation may lead to chronic immune activation, marked by elevated pro-inflammatory IFN-γ and TNF-α signaling, and an upregulated IDO/Kyn metabolic pathway. Additionally, CAD in PWH is associated with a distinct peripheral immune cell phenotype, characterized by increased frequency and activation of classical monocytes, along with profound alterations in CD4 and CD8 T-cell subsets, including features of chronic activation, senescence, heart-homing potential, and reduced expression of cardioprotective purinergic markers. Altogether, these results suggest that despite ART, inflammaging is a feature in PWH potentially contributing to the development of CAD. Further studies are needed to explore whether novel therapeutic approaches targeting these pathways could help prevent and reduce the risk of cardiovascular disease in PWH during early phases of CAD development.

Authors’ contributions

Conceptualization: CTC, MAJ; Data curation: AKDM, KFB, RSMB, MD, MMP, CCL; Data analysis, interpretation and validation: AKDM, KFB, MD, JGB, PA, IK, CTC, MAJ; Funding acquisition: CTC, MAJ, MD, CT; Methodology: MD, IK, MAJ; Resources: MD, CT, ME, CCL, SM, CTC; Supervision: CTC, MAJ; Validation: MD, IK, CTC, MAJ; Writing-original draft: AKDM, MAJ; All authors contributed to the refinement of the study and reviewed, revised, and approved the final version of the manuscript.

Funding

This research was funded by Canadian Institutes of Health Research (CIHR) grant #177334 to CTC, MAJ, MD, CT, ME, SM; the CIHR Pan-Canadian Network for HIV and STBBI Clinical Trials Research (CTN PT043) to CTC, MD, CT, MAJ; NIH (grant #R01AG054324) to CT, MD, ME, PA; and in part, by the Réseau SIDA et maladies infectieuses du Fonds de recherche du Québec-Santé (FRQ-S) to MAJ. RSMB was supported by FRQ-S and CIHR postdoctoral fellowships. MD and CTC are recipients of FRQ-S Senior Chercheur Boursier Clinicien career award. CT holds the Pfizer Chair in Clinical and Translational Research on HIV. MAJ holds the tier 2 CIHR Canada Research Chair in Immuno-Virology. The funders had no role in the design of this study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Data availability

Data from this manuscript is available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki and was authorized by the research ethics board of the Centre Hospitalier de l’Université de Montréal (CHUM) (CE 11.063), MUHC: MP-37–2022-8107 and UQAM: 2022–4663. Written informed consent was obtained from all participants prior to their inclusion in the CHACS study biobanking.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.