Reduced postprandial lipid accumulation in circulating monocytes by olezarsen attenuates pro-inflammatory changes in hypertriglyceridemic patients

Abstract

Aims

The pathophysiological mechanisms linking hypertriglyceridaemia to atherosclerotic cardiovascular disease remain incompletely understood. Accumulating evidence implicates postprandial triglyceride-rich lipoproteins in lipid accumulation and activation of monocytes, thereby exacerbating arterial wall inflammation. Olezarsen lowers fasting and postprandial triglycerides, but its effect on monocyte activation in hypertriglyceridaemia remains unclear.

Methods and results

In this double-blind, randomized, placebo-controlled trial, 28 patients with hypertriglyceridaemia (fasting triglycerides ≥350 mg/dL) received two 80 mg doses of subcutaneous olezarsen or placebo (2:1) at 4-weekly intervals. Peripheral blood mononuclear cells and CD14⁺ monocytes were isolated 4 h after a standardized oral fat load for phenotyping.

Oral fat load induced a marked triglyceride increase, accompanied by a 49.0% [95% CI: 30.2%, 71.0%] (P = 0.0001) increase in lipid droplets per monocyte. Olezarsen reduced postprandial triglyceride area under the curve by 55%, which was associated with a 25.0% decrease [95% CI: 7.9%, 40.5%] (P = 0.0049) in intracellular lipid droplets. Mass cytometry-based profiling of peripheral blood mononuclear cells revealed a decrease of CD11c (P = 0.006), CD11b (P = 0.049), and CD16 (P = 0.039) whilst CD14 remained unchanged. Postprandial monocyte interleukin-1beta gene expression was reduced in olezarsen-treated patients (P = 0.0024). Functionally, monocytes from olezarsen-treated patients exhibited a 14.4% decrease (P = 0.036) in monocyte adhesion in an ex vivo transendothelial migration assay compared to baseline.

Conclusion

Olezarsen reduced postprandial lipid droplet accumulation in monocytes and was associated with decreased expression of activation markers and reduced monocyte adhesion and transendothelial migration of monocytes ex vivo. These findings support an olezarsen-induced reduction in lipid-driven immune-cell activation, an effect associated with a reduced postprandial triglyceride burden.

Graphical Abstract

Graphical Abstract.

Translational perspective.

• Lowering of triglyceride-rich apoB-containing lipoproteins and their remnants with olezarsen reduces postprandial lipid accumulation in monocytes, an effect correlated with altered plasma triglycerides.

• Olezarsen-induced lowering of triglyceride-rich apoB-containing lipoproteins and their cholesterol-containing remnants downregulates CD16, CD11b, and CD11c expression on monocytes, altering their inflammatory phenotype and reducing their transendothelial migratory capacity ex vivo.

Introduction

Hypertriglyceridaemia increases the risk of atherosclerotic cardiovascular disease (ASCVD), independent of low-density lipoprotein-cholesterol (LDL-c) levels.^1–3 Recent genetic and epidemiological studies have provided compelling evidence for a likely causal relationship between triglyceride-rich lipoproteins (TRLs) and their remnants and increased risk of ASCVD.^1,4,5 On a per-particle basis, TRLs have been suggested to have up to a four-fold greater atherogenic impact than LDL particles,⁴ although the latter are more prevalent in the circulation. This heightened atherogenicity has been attributed amongst other factors, to a higher inflammatory potential of TRLs, which may relate to the large load of lipolysis-derived free fatty acids generated from their triglyceride (TG) core and to their capacity for receptor-independent uptake by subendothelial macrophages, eliciting pro-inflammatory skewing and enhanced foam-cell formation.¹ TRLs and their remnants have also been shown to promote activation of the vascular endothelium and the attraction of monocytes to the subendothelial space, contributing to the initiation and progression of atherosclerosis in the arterial wall.^6,7 Less attention has been paid to the potential role of postprandial elevation in TG levels, which has been shown to contribute to increased lipid accumulation within the cytoplasm of circulating monocytes.⁸ Thus, high TRL content may transform monocytes into lipid-laden foamy cells with altered phenotype and pro-inflammatory behaviour,^8,9 linking TG increases to immune cell activation.^8,10–12 In hypertriglyceridaemic patients, such elevations in plasma TGs largely reflect increased exposure to TG-rich apolipoprotein B (apoB)-containing lipoproteins and their cholesterol-containing remnants.

Apolipoprotein C-III (apoC-III), an apolipoprotein present on TRLs, has emerged as a critical regulator of lipolytic activity.^13,14 ApoC-III inhibits lipoprotein lipase (LPL) and reduces binding to hepatic lipoprotein receptors, thereby slowing both lipolysis and hepatic removal of TRLs, leading to prolonged retention in the circulation.¹⁴ Recently, Peng and colleagues elucidated the direct link between apoC-III and atherosclerotic plaque development in mice, supporting the preclinical antiatherogenic potential of apoC-III-induced TG lowering.¹²

Olezarsen, an N-acetyl-galactosamine-conjugated (GalNAc-conjugated) antisense oligonucleotide, mediates the degradation of hepatic APOC3 mRNA through RNAse-H1-dependent mechanisms, resulting in reduced apolipoprotein C-III production and subsequent lowering of TG levels.¹³ Recent clinical trials have reported marked reductions in fasting^15–17 as well as postprandial¹⁷ TG levels in hypertriglyceridemic patients. However, the potential effects of olezarsen on lipid-induced inflammatory activation remain to be determined.

This study evaluates how olezarsen-mediated reduction of TG-rich lipoprotein/remnant burden, captured by fasting and postprandial TG responses, reshapes postprandial immune cell signatures, offering insights into the anti-inflammatory potential of apoC-III inhibition in patients with hypertriglyceridaemia.

Methods

A comprehensive description of the methods employed, including oral fat load (OFL) content analysis, immune cell isolation, Nile Red staining for lipid content staining and quantification, detailed mass cytometry (MC) acquisition and analysis, quantitative polymerase chain reaction (qPCR) analysis, the ex vivo transendothelial migration assays and quantifications are provided in the supplementary methods.

Study design

Patient materials used in the present study were derived from a randomized, placebo-controlled, single-centre phase 2b clinical trial involving hypertriglyceridemic (fasting TGs ≥350 mg/dL) patients.¹⁷ Baseline characteristics, exclusion criteria, adverse events, and extensive plasma (lipid) changes, as well as study methods have been reported previously.¹⁷ In short, the baseline visit (visit 2, V2) included blood collection followed by a standardized OFL. The week-7 visit (visit 4, V4) repeated the same protocol three weeks after the second dose of olezarsen or placebo. At both V2 and V4, blood was drawn in the fasting state (t = 0 h) and at 2, 4, and 6 h after the OFL. PBMCs and CD14⁺ monocytes were isolated at V2 and V4 for downstream analyses. Definitions of visits and sampling time points are summarized in Table 1. Participants were subsequently randomized to receive either two doses of 80 mg olezarsen or placebo, following a 2:1 distribution, with a 4-week interval between doses. The study was conducted according to the principles of the Declaration of Helsinki and the study protocol, including amendments, was approved by the local ethics committees. All patients provided written informed consent prior to enrolment. For the present mechanistic substudy, immune-cell phenotyping (lipid droplets, MC, and gene expression) and ex vivo adhesion/migration assays were predefined to be analysed in the postprandial state, where TG-rich lipoprotein exposure and monocyte activation are maximal. Therefore, monocyte data are reported at 4 h post-OFL, and longitudinal changes are expressed as differences between postprandial measurements at V2 (baseline) and V4 (week 7). Postprandial TG area under the curve (AUC, 0–6 h) and its change from V2 to V4 (ΔAUC) were used as integrated measures of postprandial TG exposure. Fasting V4 samples were collected and contributed to the characterization of overall lipid changes in the parent trial, but are not shown for monocyte endpoints in this substudy.

Table 1.

Definitions of study visits and sampling time points

Term	Definition
Baseline	Pre-treatment visit (visit 2, V2)
Week 7/treatment	Post-treatment visit (visit 4, V4), three weeks after the second dose
Fasting	Blood sample taken before the oral fat load (t = 0 h)
Postprandial	Blood sample taken 4 h after the oral fat load (t = 4 h)
OFL	Oral fat load; standardized high-fat challenge used in both visits
ΔAUC	Change in postprandial triglyceride AUC (0–6 h) between V2 and V4

Mass cytometry

Details of elaborate MC acquisition and analysis can be found in the supplemental methods. In short, cryopreserved peripheral blood mononuclear cells (PBMCs) were thawed, barcoded with a unique combination of three out of seven different anti-CD45 antibodies, and subsequently stained with a panel of 21 preconjugated antibodies according to manufacturer’s instructions (Standard BioTools, South San Francisco, CA, USA). All 21 antibodies were obtained from Standard BioTools and included CD45, CD11b, CD11c, CD40, CD80 (B7-1), CD163, CD19, CD36, CD22, CD38, CD66a, CD24, CD33, HLA-DR, CD279 (PD-1), CD192 (CCR2), CD3, CD7, CD14, CD206, and CD16. Integrin expression (CD11b, CD11c, etc.) was quantified on CD14⁺ monocytes, and median fluorescence intensity values reflect monocyte gates only, not total PBMCs. Analysis was performed using OMIQ software (Milan, Italy). All patients with a single sample containing less than 20 000 CD45 events after cleaning (N = 3 treatment group and n = 1 control group) were excluded from further analysis.

Statistical analysis, sample size, and power considerations

The data are presented as the mean ± SEM. For TGs, the postprandial AUC was calculated using the trapezoidal rule, using plasma samples at t = 0, 2, 4, and 6 h.¹⁷ Correlations were evaluated using Spearman’s rank correlation test. The effects of treatment and time were analysed using one-way analysis of variance with Dunnett’s post hoc test for multiple comparisons.¹⁸ Comparisons between two groups were performed using a Wilcoxon matched-pairs signed-rank test. Outliers were excluded based on the robust regression and outlier removal test (Q = 1%).¹⁹ Data distribution was then assessed using the Shapiro–Wilk test. All statistical analyses were conducted using GraphPad Prism (version 10.4.0). The threshold for statistical significance was set as P ≤ 0.05. The mechanistic substudy was embedded in the randomized, placebo-controlled Phase 2b trial reported previously¹⁷ In that parent trial, the sample size was calculated to provide 80% power (two-sided α = 0.05) to detect a between-group difference of ∼70% in fasting TG levels, resulting in a target of 24 participants and an actual enrolment of 28 patients. The current laboratory analyses therefore share this overall sample size; no additional formal power calculation was performed for individual immune or functional endpoints, which were predefined but considered exploratory and hypothesis-generating.

Results

Olezarsen reduces both fasting and postprandial TG levels

As reported previously,¹⁷ a total of 28 participants, with TG levels >350 mg/dL, completed the study; 19 participants received olezarsen and 9 received a placebo control (2:1 randomization). Visual representation of the study setup can be found in Figure 1A. Baseline clinical and biochemical characteristics, including gluco-metabolic parameters, are summarized in Supplementary material online, Table S1. Fasting baseline TG levels were comparable between the olezarsen and placebo groups, averaging 8.3 [2.3, 14.3] mmol/L and 7.9 [2.0, 13.9] mmol/L (Table 2). The mean age of the participants was 58.6 years, 82.1% male with a mean BMI of 30.9 kg/m². As previously reported,¹⁷ the administration of olezarsen resulted in altered lipid and lipoprotein parameters (Summary in Table 1). Alterations in apoC-III levels were accompanied by reductions in fasting (P < 0.0001) and postprandial (P < 0.0001) TG levels. Postprandial TGs at t = 4 h were reduced by 60.8% [−82.5%, −39.0%] in the olezarsen group compared with a change of 12.2% [−43.7%, 19.4%] in the placebo group, resulting in a 51.6% placebo-controlled reduction in postprandial AUC.¹⁷

Figure 1.

Lipid content analysis in monocytes following oral fat load and olezarsen treatment. (A) Schematic representation of the study setup. (B) Representative images of CD14⁺ monocytes isolated from whole blood, stained with nile red to neutral lipid droplets (cyan) and phospholipids (magenta) using confocal microscopy. (C) The effect of oral fat load on lipid droplets/cell in both olezarsen vs. placebo treatment on postprandial lipid droplets/cell (D) Fasting and 4 h post-oral fat load lipid droplets per monocyte were measured at both baseline (visit 2) and week 7 (visit 4) in the same subjects; treatment or placebo groups were then compared to their own baseline using one-way analysis of variance with Tukey’s post-hoc test. Correlation analysis of plasma triglycerides and lipid droplets/cell was investigated using Spearman’s rho test. ** = P < 0.01; **** = P < 0.0001. n = 8 placebo, n = 18 olezarsen.

Table 2.

Effect olezarsen treatment on TG levels

	Olezarsen (n = 19)	Placebo (n = 9)
Apo-C-III (mg/dL)
Baseline	18.0 [16.6, 27.3]	17.0 [11.5, 24.2]
7 weeks	5.7 [2.9, 7.8]	18.9 [14.5, 27.0]
Mean change from baseline (%)	−74.6% [−80.5, −68.7]	25.2% [−10.3, 60.7]
TG (mmol/L) fast
Baseline	5.9 [4.5, 9.2]	4.5 [4.1, 8.5]
7 weeks	1.6 [1.2, 2.7]	5.4 [2.5, 8.8]
Mean % change (95% CI)	−71.0% [−77.9, −64.2]****	−11.8% [−36.9, 13.5]
TG (mmol/L) postprandial
Baseline	9.9 [6.9, 15.4]	10.4 [4.2,16,7]
7 weeks	4.2 [2.3, 5.3]	9.2 [1.2,17,2]
Mean % change (95% CI)	−57.2% [−78.9,−35.5]****	−11.4% [−42.9,20.1]
AUC (to 6 h post fat load)
Baseline	29.4 [22.4, 36.5]	28.1 [16.3, 40.1]
7 weeks	10.1 [8.0, 14.4]	24.3 [10.9, 37.4]
Mean % change (95% CI)	−61.6% [−70.0, −53.3]****	−11.6% [−34.6, 11.5]

Summary of previously reported data.¹⁷ All data are expressed as mean [95%CI].

iAUC, incremental area under the curve; Apo-CIII, apolipoprotein C-III; TG, triglyceride; AUC, area under the curve.

^**** = P < 0.0001.

Olezarsen reduces lipid accumulation in monocytes

Hyperlipidaemia results in the intracellular accumulation of lipids in lipid droplets within monocytes, transforming them into lipid-laden foamy cells.^8–10,20 Quantification of neutral lipid droplets in monocytes, determined by Nile Red staining, was conducted in patients prior to (V2) and following (V4) olezarsen treatment, both in a fasting state and after the administration of a standardized OFL. The schematics of the study setup are shown in Figure 1A. Representative images of Nile Red-stained monocytes are presented in Figure 1B. Overall, the number of lipid droplets was positively associated with plasma TG levels, (Spearman ρ = 0.667, P < 0.0001; Figure 1D). The percentage of lipid droplet-positive monocytes remained unchanged (see Supplementary material online, Figure S1). A 49.0% [95% CI: 30.2%, 71.0%] (P < 0.0001) increase in the number of lipid droplets per cell was observed 4 h after OFL. Four hours after OFL, olezarsen treatment resulted in a significant reduction of lipid droplets per cell, with a 25.0% decrease [95% CI: 7.9%, 40.5%] (P = 0.0049) compared with untreated post-OFL controls, from 5.7 to 4.1 lipid droplets/monocyte after olezarsen treatment. Monocytes from the placebo-treated group did not show any differences in lipid droplets/monocyte compared with baseline (Figure 1C). These data indicate that olezarsen treatment is associated with reduced postprandial neutral lipid droplet accumulation in monocytes. The change in lipid droplets per cell from V2 to V4 was strongly correlated with the change in postprandial TG area under the curve (ΔAUC) in the olezarsen-treated group (Spearman ρ = 0.699, P = 0.0.001; see Supplementary material online, Figure S2).

Olezarsen attenuates monocyte activation

Previously, postprandial lipid droplets have been associated with altered monocyte phenotype and behaviour.^8,9 To investigate olezarsen-induced postprandial alterations in immune cell composition and activation, a 21-marker MC panel was used to investigate patients’ postprandial PBMCs. Unsupervised dimension reduction of all immune cells revealed different cell populations based on lineage-specific markers (Figure 2A) to discriminate all different immune cell populations including NK-, B-, T-cells, and monocytes (Figure 2B). In total, 15 different clusters (Figure 2C) could be identified, with each cluster expressing a distinct marker profile (Figure 2D). In accordance with previous literature⁸ the cluster that demonstrated the most significant alterations in response to olezarsen treatment was identified as cluster 15, containing CD14⁺ monocytes. In this cluster, median expression of CD16 was reduced (P < 0.039) after olezarsen treatment compared to baseline, whilst monocyte marker CD14 remained unchanged. These alterations indicate a shift from the intermediate monocyte subtype (CD14⁺⁺CD16⁺) towards a higher abundance of the classical (CD14⁺⁺CD16⁻) monocyte subtype after treatment. Compared to baseline, the medians of integrin aX (CD11c) (P < 0.006) and integrin aM (CD11b) (P < 0.049) were reduced after olezarsen treatment. CD16, CD14, CD11b, and CD11c expression remained unchanged in the placebo group (see Supplementary material online, Figure S3). Additionally, a trend towards increased expression of CD163, an anti-inflammatory macrophage marker, was observed (P = 0.0587). Concomitantly, the expression of inhibitory T cell receptor CD279 (PD-1) was significantly elevated by 15.0% [10.8%, 19.2%] (P < 0.05) in all clusters expressing this marker (C1, C3, C6, C7, and C8) (see Supplementary material online, Figure S4).

Figure 2.

Immune cell profiling and phenotypic changes following olezarsen treatment after oral fat load (A) Example marker expression overlayed onto dimension-reduction results of mass cytometry peripheral blood mononuclear cells data. (B) Uniform manifold approximation and projection (UMAP) visualized major immune cell lineages in pooled samples, including natural killer cells, B cells, T cells, and monocytes. (C) Euclidean meta-clustering of UMAPs identified 15 immune cell clusters. (D) Heatmap visualizing relative marker expression for all clusters. (E) Median expression of monocyte-specific markers in the C15-monocyte cluster. Median integrin expression is measured on CD14⁺ monocytes, UMAPs were created using OMIQ. (F) Gene expression of IL1B before (V2) and after (V4) intervention (G) expression of different cell surface adhesion markers. (H) Altered postprandial plasma values before and after treatment. Peripheral blood mononuclear cells were obtained at baseline (visit 2) and week 7 (visit 4), 4 h after the oral fat load, from the same subjects. A Wilcoxon signed-rank test was used to compare two groups, and treatment or placebo compared to baseline were analysed using a one-way analysis of variance with Tukey test. * = P < 0.05, ** = P < 0.01. CYTOFF: n = 9 placebo, n = 18 olezarsen, quantitative polymerase chain reaction: n = 9 placebo, n = 13 olezarsen, Nomics: n = 8–9 placebo, n = 15–17 olezarsen, dependent on detection limits. Abbreviations: IGTAV, Integrin Subunit Alpha V; IGTAM, Integrin Subunit Alpha M; IL-1β, Interleukin-1β.

To further investigate the reduced activation profile, qPCR of isolated CD14⁺ monocytes was performed. A significant reduction in IL1B gene expression (P = 0.0024) following olezarsen treatment was observed, whilst expression remained unchanged in the placebo group (Figure 2F). Additionally, genes encoding surface proteins ITGAM (encoding integrin αM, CD11b) and ITGAV (P = 0.028), (encoding integrin αV) revealed downregulated expression patterns after treatment (Figure 2G), whilst placebo remained unchanged (see Supplementary material online, Figure S5). Analysis of plasma using Nomic Bio nELISA revealed 124 measurable biomarkers of which three significantly changed following olezarsen treatment in the postprandial phase. Specifically, CXCL13 (P = 0.018) and MMP1 (P = 0.041) were upregulated after olezarsen treatment compared to baseline, whilst EMMPRIN was downregulated (P = 0.036). These markers remained unchanged in the placebo group (Figure 2H). 16 compounds were correlated with plasma TGs, including CCL2 (MCP-1), NGAL, MIF, and PCSK9 (see Supplementary material online, Figure S6, Supplementary material online, Table S2)

Olezarsen decreases monocyte adhesion and transendothelial migration

To assess the functional consequences of the changes in gene and protein expression of adhesion molecules and cellular activation markers, we evaluated monocyte adhesion and transendothelial migration ex vivo. For this, we used semi-automated software to discriminate between adhered and transmigrated monocytes based on their morphology (see Supplementary material online, Figure S7). Adhered and transmigrated monocytes (small, round cells and larger, non-round cells, respectively) were visualized using fluorescent phalloidin staining (Figure 3A and B). We observed a significant decrease of 14.4% (P = 0.036) in the number of adhered monocytes, from 192.0 [164.8, 219.2] at baseline to 166.0 [143.6, 188.5] monocytes per field of view after olezarsen treatment (Figure 3C). Spearman correlation analysis revealed that the change in postprandial TG AUC (ΔAUC) was correlated with the change in monocyte adhesion and migration from baseline to week 7 (Spearman ρ = 0.433, P = 0.0214; Supplementary material online, Figure S2). Following olezarsen, monocytes also exhibited reduced migratory progression following treatment, as reflected by a decreased cellular perimeter (Figure 3D and E).

Figure 3.

Lipid-driven functional changes in CD14⁺ monocytes (A) Schematic illustrating monocyte adherence and migration, visualized by brightfield phase contrast and corresponding phalloidin staining. (B) Representative brightfield images of CD14⁺ monocytes adhering to or migrating through activated human aortic endothelial cells ex vivo. (C) Migrating/adhered cells quantified as objects per image at baseline (V2), and olezarsen (V4) and placebo (V4) treatment. (D) Perimeter of manually scored adhered or migrating/migrated monocytes. (E) Perimeter of selected objects at baseline (V2), and after olezarsen (v4) and placebo (V4) treatment. Postprandial (4 h) plasma triglycerides were measured at baseline (visit 2) and week 7 (visit 4) in the same subjects. Wilcoxon signed-rank test was used to compare two groups, and the treatment or placebo compared to baseline was analysed using a one-way analysis of variance with Tukey correction. *=P < 0.05; **=P < 0.01; ***=P < 0.001. n = 8–9 placebo, 15–16 olezarsen.

Discussion

Previous clinical studies have demonstrated that olezarsen, a GalNAc conjugated antisense oligonucleotide targeting apoC-III) effectively reduces both fasting and postprandial TG levels in patients with hypertriglyceridaemia.¹⁷ Here we report several key findings on the impact of olezarsen on postprandial lipid droplets and monocyte activation. In line with previous literature on hypertriglyceridaemia-induced monocyte activation and adhesion,^8–10 neutral lipid droplet accumulation increases markedly during the postprandial phase, with TG levels correlating with the number of intracellular lipid droplets in circulating monocytes. We demonstrate that olezarsen treatment reduces both fasting and postprandial TGs in conjunction with reduced monocyte lipid accumulation. Atherosclerotic risk is mainly driven by the cholesterol cargo of apoB-containing TRL remnants rather than TG itself. In this hypotriglyceridaemic cohort, fasting and postprandial TGs closely track the burden of these cholesterol-containing TRLs and remnants, so the olezarsen-induced reduction in postprandial TGs is interpreted as a reduction in atherogenic TRL/remnant exposure rather than an isolated TG effect. Second, olezarsen downregulates the expression of CD16 and CD11c/CD11b integrins on circulating monocytes isolated after administration of an OFL, which are pivotal for monocyte adhesion and subsequent migration. The decreased integrin expression suggests that olezarsen may effectively hinder the TRL-induced inflammatory activation. Third, this altered integrin expression following olezarsen corresponds to attenuated adhesion and transendothelial migration of postprandial monocytes in patients treated with olezarsen. Collectively, these results demonstrate that the TG-lowering effect of olezarsen is accompanied by reduced lipid-driven monocyte activation and migration ex vivo, highlighting its potential to attenuate monocyte-driven inflammatory processes in the vasculature.

Postprandial TGs comprise both chylomicrons and increased very low-density lipoprotein (VLDL) particles. Whereas chylomicrons are too large for direct cellular uptake, VLDL, VLDL-remnants, and chylomicron-remnants can be readily taken up by monocytes, as evidenced by increased intracellular lipid droplet formation at higher concentrations of these particles.^9,21–23 Olezarsen resulted in a notable reduction in the levels of TRLs, particularly VLDL. Therefore, the observed reduction in monocyte lipid droplets following apoC-III reduction is likely mediated through its effects on smaller TRLs, predominantly VLDL-remnants. ApoC-III on TRLs inhibits LPL-mediated lipolysis and hepatic remnant clearance, whereas apoC-III lowering with olezarsen enhances LPL-dependent TG hydrolysis, promotes remnant and HDL-mediated uptake of remnant-derived lipids, and thereby plausibly underlies the reduced monocyte lipid loading and activation observed in our study.²⁴

In this study, the total number of intracellular lipid droplets in monocytes strongly correlated with plasma TG levels, implying direct uptake of TRLs by circulating monocytes. Patients with hypertriglyceridaemia and/or metabolic syndrome are particularly prone to foamy monocyte formation,^8,10 with increasing and prolonged TRL plasma concentrations after ingestion of a high-fat meal.^8,11 Increased neutral lipid load in monocytes is closely associated with increased intracellular pro-inflammatory cytokines like IL-1β¹⁰ and monocyte adhesion marker expression,^8,10 both crucial during initiation and progression of atherosclerosis.^25,26 In this study, olezarsen-induced reduction of postprandial TRLs was associated with a reduction in neutral lipid content and foamy monocyte formation.

Monocyte subsets, defined by varying CD16 and CD14 expression, play distinct roles in the development and progression of atherosclerosis. Consistent with most previous studies^8,27,28 we observed that the median expression of the monocyte subset marker CD16 was reduced in monocytes collected 4 h following OFT in patients treated with olezarsen compared to the median expression before treatment. CD14 levels remained unchanged. This suggests a shift from intermediate (CD14⁺⁺CD16⁺) to classical (CD14⁺⁺CD16⁻) monocytes.

Since intermediate monocytes are recognized as more pro-inflammatory,^29,30 a shift towards classical monocytes following olezarsen may indicate a reduction in the overall inflammatory phenotype of circulating monocytes. Moreover, their proportion is increased in cardiometabolic disorders and is associated with higher cardiovascular risk.^10,31

Following the OFL, CD11b and CD11c integrin expression was significantly reduced on monocytes after olezarsen treatment. CD11b is a surface integrin involved in adhesion and migration and serves as a recognized activation marker on monocytes, with increased expression present during various hyperlipidemic states.^8,9,20,23,25 CD11c is a β2 integrin mediating monocyte adhesion to endothelial cells and subsequent migration into the arterial wall. Murine studies have demonstrated that lipid-loaded monocytes infiltrate early atherosclerotic lesions in a CD11c-dependent manner^20,25; a process accelerated by apoC-III,¹²; whereas genetic deletion of CD11c diminished monocyte arrest.²⁰ This concept is supported by the reduced IL1B gene expression in monocytes collected postprandially, a cytokine which has been shown to fulfil a crucial role in a causal pathway leading to enhanced atherogenesis and an increased cardiovascular risk.^26,32 Previous studies substantiated an increase in in vitro IL-1β production by monocytes following TRL stimulation⁹ as well as elevated postprandial IL-1β levels in individuals with metabolic syndrome,³³ with intermediate monocytes representing the most powerful secretory source amongst monocyte subtypes.^34,35 Altogether, the concomitant reduction of CD11c, CD11b, and IL1B expression implies a diminished monocyte activation state in hypertriglyceridemic patients receiving olezarsen.

Finally, at the functional level, we observe a marked reduction in ex vivo adhesion and transendothelial migration by monocytes isolated after an OFL in patients receiving olezarsen. These effects likely reflect monocyte-intrinsic changes due to reduced TRL-mediated lipid loading rather than altered endothelial activation, as endothelial cells were identically activated in all conditions and monocytes were added for only 30 min. This finding builds upon our previous work and that of others, which demonstrated that hypertriglyceridaemia was associated with increased monocyte adhesion and transmigration.^8,28,36 Our current study extends these observations, specifically for patients with elevated TGs and reduced TRL removal following an OFL. More importantly, a 50% reduction in the area under the curve for postprandial TGs was associated with attenuated monocyte adhesion and transmigration, processes considered early initiating steps in atherogenesis, although these data do not establish a causal relationship.^20,25,37 The correlation between diminished intracellular lipid accumulation and the suppression of adhesive integrins confirmed a mechanistic association, whereby intracellular lipids orchestrate the pro-inflammatory behaviour of circulating monocytes. Whilst our study design cannot fully disentangle the direct effects of apoC-III reduction from the consequences of lowered TRL burden, the strong correlation between plasma TGs and monocyte lipid content supports a lipid-driven mechanism Finally, the present study was not powered or designed to relate these mechanistic readouts directly to clinical events, and the prognostic value of such monocyte signatures will need to be assessed in larger, event-driven cohorts.

Limitations

As previously discussed,¹⁷ limitations of our trial include the relatively small sample size of the trial, which comprised mostly male participants (23/28), limiting the generalizability of the findings to broader populations. Second, to assess lipid uptake in monocytes, we measured the amount of lipid droplets, but not the size of individual droplets. However, previous work on VLDL-induced lipid droplets in monocytes reveals that total lipid accumulation, rather than enlargement of individual droplets, is more closely associated with inflammatory activation in monocytes.³⁸ In addition, the Nile Red staining in this study quantifies neutral lipid droplets vs. phospholipids but does not identify specific lipid species, and detailed plasma and cellular lipidomics will be required in future work to define which lipid classes drive monocyte lipid. Third, olezarsen is a recently developed therapeutic, and whilst it has demonstrated strong TG-lowering effects, its impact on atherosclerotic plaque burden in humans remains to be established. Preclinical apoC-III antisense studies in Ldlr^−/− apoC-III transgenic mice have shown reductions in plasma TGs, foamy monocytes, CD11c expression, and atherosclerotic lesion size, accompanied by fewer CD11c⁺ cells within the plaque,¹² findings that are consistent with the monocyte changes observed here. Large-scale Phase 3 trials, including studies using coronary computed tomography angiography to assess changes in non-calcified plaque volume with olezarsen, are currently underway.³⁹

Detailed postprandial lipoprotein subfraction profiling (e.g. separation of VLDL- and chylomicron-derived remnants) was not performed in this mechanistic substudy, as available sample volume and study design were focused on monocyte phenotyping and functional assays. As a result, TGs were used as a surrogate for TRL/remnant burden, and future studies with comprehensive lipoprotein phenotyping will be required to define which specific cholesterol-containing remnant subfractions most strongly drive monocyte activation.

Conclusion

In conclusion, the marked reduction in exposure to TG-rich apoB-containing lipoproteins and their cholesterol-containing remnants during olezarsen treatment is accompanied by reduced inflammatory activation as well as an attenuated adhesive and migratory response of circulating monocytes collected during the postprandial phase. These findings imply that potent apoC-III inhibition may partly exert antiatherogenic effects by reducing cellular inflammation in hypertriglyceridemic patients. Our study provides novel insights into the potential mechanisms by which apoC-III inhibition could modulate ASCVD-risk beyond its direct effects on lipid metabolism.

Data availability

All data are available from the corresponding author upon reasonable request.

Supplementary material

Supplementary material is available at European Heart Journal Open online.

Author contributions

Merel C. Peletier (Conceptualization, Data curation, Formal analysis, Visualization, Project administration, Writing—original draft), Jordan M. Kraaijenhof (Conceptualization, Data curation, Formal analysis, Visualization, Project administration, Writing-original draft), Miranda Versloot (Investigation, Data curation, Writing—review & editing), Hester Koppejan (Data curation, Formal Analysis, Writing—review & editing), Kim E. Dzobo (Investigation, Data curation, Writing—review & editing), Veronica J. Alexander (Writing—review & editing), Sotirios Tsimikas (Writing—review & editing), Erik S. G. Stroes (Conceptualization, Funding acquisition, Supervision, Writing—review & editing), and Jeffrey Kroon (Conceptualization, Funding acquisition, Supervision, Writing—review & editing)

Ethical approval

The study was conducted according to the principles of the Declaration of Helsinki and the study protocol, including amendments, were approved by the local ethics committee. All patients provided written informed consent prior to enrolment.

Funding

The study was funded by a collaborative study grant from IONIS Pharmaceuticals. M.C.P. is supported by the Dutch Heart Foundation (03-004-2021-T045). J.M.K. is partly funded by the Klinkerpad foundation and Novo Nordisk. S.T. is supported by NHLBI grants R01 HL159156 and HL170224. J.K. was supported by the Dutch Heart Foundation [Senior Scientist Dekker grant (03-004-2021-T045)], the European Union (ERC, ENDOMET-STEER, 101076407), and by an NWO-Vidi grant (09150172310053). Views and opinions expressed are, however, those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency.