Vericiguat for endothelial dysfunction and CRT response in non-responders: the VERCET prospective observational study

Abstract

Background

Endothelial dysfunction (ED) is a negative prognostic factor for patients with heart failure (HF) who are treated with cardiac resynchronization therapy (CRT). ED could lead to CRT non-responders by increasing endothelin-1 (ET-1) expression and inflammatory/oxidative stress markers.

Study hypothesis

Vericiguat is an anti-HF medication that may reduce inflammatory/oxidative stress, ET-1 expression, and improve endothelial function in ED-CRT non-responders, leading to better outcomes.

Methods

We conducted a prospective observational study enrolling ED-CRT non-responders treated with vericiguat (vericiguat users) versus those who did not receive vericiguat (non-vericiguat users). ED was diagnosed by flow-mediated vasodilation (FMD ≤ 7.1%) of the brachial artery. Clinical, echocardiographic, and biochemical assessments (inflammatory markers and ET-1) were performed at baseline and at 12 months. The primary endpoints were normalization of endothelial function (FMD > 7.1%) and CRT response at 1 year. Secondary endpoints included changes in inflammatory/oxidative stress markers, ET-1 levels, and clinical outcomes (HF hospitalizations, cardiac, and all-cause mortality).

Results

A total of 289 ED-CRT non-responders were included, of whom 81 received vericiguat (vericiguat users) and 208 did not receive vericiguat (nonusers). At 1-year follow-up, vericiguat users vs. non-users demonstrated significant improvement in New York Heart Association class, 6-min walking test, and left ventricular remodeling (p < 0.05). Vericiguat users vs. non-users showed a significant reduction in inflammatory/oxidative stress biomarkers, B-type natriuretic peptide, and ET-1 (p < 0.05). At the end of follow-up, a lower rate of vericiguat users versus nonusers was under treatment with loop diuretics, aldosterone blockers, and sodium-glucose-transporter 2 inhibitors (p < 0.05). At follow-up end, a higher rate of vericiguat users vs. nonusers showed normalization of endothelial function (30 (37%) vs. 21 (10.1%), p < 0.05) and restored CRT responders (28 (34.6%) vs. 27 (13%), p < 0.05). At 1-year follow-up, the normalization of endothelial function was predicted by left-ventricle ejection fraction (0.42, CI 95% [0.02–0.72]) and vericiguat (4.35, CI 95% [2.45–7.70]). The CRT responders’ outcome was predicted by vericiguat therapy (2.85, 95% CI [1.67–4.89]).

Conclusions

Our findings demonstrate that vericiguat reduces ED and enhances the response to CRT in prior non-responders by significantly lowering inflammatory/oxidative stress markers and ET-1.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12916-026-04747-6.

Background

Endothelial dysfunction (ED) is a prognostic marker in patients treated with cardiac resynchronization therapy (CRT) [1]. Indeed, despite ED being a recognized hallmark of heart failure (HF) with reduced ejection fraction (HFrEF) [2], accumulating evidence suggests it may be particularly relevant in CRT non-responders, where it could contribute to the lack of therapeutic benefit [3]. ED reduces the response to CRT, and CRT patients with ED show a higher rate of non-response and worse clinical outcomes [2]. In CRT non-responders, ED can be diagnosed by altered serum levels of molecular markers, such as endothelin-1 (ET-1), and by impaired flow-mediated vasodilation (FMD) measured by echocolor Doppler of the brachial artery [2]. ET-1 is a vasoconstrictor produced from the precursor peptide big ET-1 by the action of an endothelin-converting enzyme, and it regulates vascular resistance in response to changes in blood flow and inflammatory stress [3]. ET-1 exerts its effects by binding to two G-protein-coupled receptors, ETA and ETB, which modulate nitric oxide (NO) release and/or promote its degradation—leading to vasoconstriction, cell proliferation, inflammation, and fibrosis [3]. Thus, over-inflammation promotes ET-1 expression, whereas ET receptor antagonism enhances NO bioavailability, improves endothelial function, and reduces inflammation in patients with chronic heart failure (HF) and those undergoing CRT [3]. Notably, patients with ED and CRT (ED-CRT) exhibit higher serum levels of ET-1 and markers of inflammation and oxidative stress, along with lower FMD values [3]. These patients also demonstrate a higher incidence of CRT non-response, HF hospitalizations, and cardiac mortality [3]. Therefore, it may be hypothesized that improving endothelial function could attenuate the serum expression of ET-1 and inflammatory/oxidative stress biomarkers, thereby contributing to more favorable clinical outcomes in CRT non-responder patients with ED. Vericiguat is specifically indicated for HF patients who remain symptomatic despite optimal first-line therapy, particularly following recent episodes of HF decompensation [4, 5]. Mechanistically, vericiguat restores the impaired nitric oxide-soluble guanylate cyclase-cyclic guanosine monophosphate (NO–sGC–cGMP) signaling pathway by increasing intracellular cGMP levels in cardiomyocytes and vascular smooth muscle cells [6]. This leads to improved ED in HF patients by enhancing the NO–sGC–cGMP signaling pathway, which plays a critical role in maintaining endothelial function and mediating other favorable cardiovascular effects [6]. Thus, we hypothesized that compared with conventional maximal anti-HF drug therapy, vericiguat might reduce inflammatory and oxidative stress, decrease serum ET-1 expression, and increase NO bioavailability in ED-CRT non-responders. These effects could enhance FMD values and improve endothelial function, thereby reducing ED and increasing the proportion of CRT responders among vericiguat users compared with non-users. Ultimately, this might result in a lower rate of HF hospitalizations and cardiac deaths in this cohort of patients. Thus, to test this hypothesis in the present study, we evaluated, among ED-CRT non-responders, the prevalence of ED and the rate of CRT responders in patients receiving vericiguat therapy (vericiguat users) compared with those not receiving vericiguat (non-users). As secondary endpoints, we assessed serum inflammatory markers, ET-1 expression, and the incidence of HF hospitalizations, cardiac deaths, and all-cause mortality at 1-year follow-up.

Methods

Study population

We screened a cohort of CRT non-responders with ED. The CRT non-responders were diagnosed by the absence of the following criteria at 6 months after implantation: left ventricular reverse remodeling (reduction in left ventricular end-systolic volume (LVESV) of ≥ 10% at cardiac echography) and significant change in functional HF class (improvement of the 6-min walk test (6MWT) and Minnesota living with HF scale improvement) [5]. ED was diagnosed in CRT non-responders using brachial artery FMD measured by echocolor Doppler, according to established methods [3]. Interestingly, by applying a cut-off value of 7.1% for FMD, measured via echo-color Doppler of the brachial artery, it is possible to distinguish CRT non-responders with ED (FMD ≤ 7.1%) from those without ED (FMD > 7.1%) [3]. Thus, we selectively screened the CRT non-responders with FMD ≤ 7.1%, which were classified as ED-CRT non-responders [3]. The ED-CRT non-responders study population respected the following inclusion/exclusion criteria:

Inclusion criteria

These are age ≥ 18 years; clinical history of stable chronic HF; NYHA functional class II, III, or IV; left bundle branch block; severe left ventricular ejection fraction reduction (LVEF < 35%); stable sinus rhythm; and diagnosis of CRTd non-responders according to established criteria [5] and diagnosis of ED defined as FMD ≤ 7.1% by brachial artery echocolor.

Exclusion criteria

These are age < 18 or > 75 years, LVEF > 40%, diagnosis of CRTd responders, unstable HF, patients in IV NYHA class, hyperkalemia, systolic hypotension (systolic blood pressure < 90 mmHg), and permanent atrial fibrillation (AF), patients with an estimated glomerular filtration rate (eGFR) < 30 ml per minute per 1.73 m² of the body surface area, and FMD > 7.1% by brachial artery echo color Doppler, the absence of written informed patient consent, and any condition that would make survival for 1 year unlikely. Thus, we finally enrolled a consecutive population of ED-CRT non-responders’ patients. We reported a study flowchart describing patients’ screening, exclusion, inclusion, and final cohort in Supplementary Fig. 1. CRT non-responders in NYHA class II–IV, with a recent hospitalization for HF worsening despite optimal medical therapy (including a beta-blocker, an angiotensin receptor–neprilysin inhibitor [ARNI] or an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, a mineralocorticoid receptor antagonist, and a sodium-glucose cotransporter-2 inhibitor [SGLT2i]), and at high risk of further HF worsening, were eligible to receive vericiguat [4, 5]. The risk of HF worsening was defined according to established criteria: elevated B-type natriuretic peptide or N-terminal pro-B-type natriuretic peptide levels (BNP ≥ 300 ng/L or NT-proBNP ≥ 1000 ng/L in patients in sinus rhythm, BNP ≥ 500 ng/L or NT-proBNP ≥ 1600 ng/L in patients with atrial fibrillation) and either a hospitalization for HF within the previous 6 months or intravenous diuretic therapy within the previous 3 months [4–6]. Vericiguat is a recently introduced anti-HF drug, recommended with a class II level of evidence in the latest American College of Cardiology guidelines on HF therapy [4]. Thus, in our study, all ED-CRT non-responders underwent optimized biventricular pacing in accordance with guideline-recommended protocols, and conduction system pacing was not applied in this population [5]. Patients with persistent or permanent atrial fibrillation were excluded. The study cohort of ED-CRT non-responders who received vericiguat (vericiguat users) consisted of patients enrolled after June 2023, when vericiguat was introduced into clinical practice. A second cohort included ED-CRT non-responders (the enrollment started before June 2023), who did not receive vericiguat (vericiguat non-users). Importantly, the inclusion of both vericiguat-treated and untreated patients reflects real-world prescribing practices rather than a protocol-driven or time-dependent treatment assignment. Patients were not randomized to vericiguat therapy, and not all patients enrolled after June 2023 received vericiguat. Treatment decisions were made by the treating physicians based on individual clinical conditions, contraindications, drug availability, and patient tolerance. For patients receiving vericiguat, the 1-year follow-up was calculated from therapy initiation, whereas for non-vericiguat patients, follow-up began at the corresponding baseline clinical evaluation, ensuring comparable prospective follow-up intervals between groups. Therefore, within our study population, exposure to vericiguat after June 2023 was heterogeneous, and a proportion of patients did not receive this therapy.

Study design

In this prospective, observational, multicenter study conducted from June 2018 to June 2024, we enrolled ED-CRT non-responders divided into two cohorts: (i) ED-CRT non-responders who received vericiguat (vericiguat users) and (ii) ED-CRT non-responders who did not receive vericiguat (non-users). The study was specifically designed to evaluate the effects of vericiguat on endothelial function in CRT non-responders, rather than in the broader HFrEF population, given the particularly poor prognosis and unmet therapeutic needs of this subgroup of patients. Patients were evaluated by clinical and instrumental assessment, as well as telemetric device monitoring, at baseline and at 6- and 12-month follow-up. We assessed the effects of vericiguat therapy on molecular (serum markers of inflammatory burden and ET-1 levels), clinical, and echocardiographic parameters during visits, as well as ED status and CRT response at the end of follow-up. The study was conducted in accordance with the Declaration of Helsinki, and the protocols were approved by the ethics committees of all participating institutions. Written informed consent was obtained from all participants. All patients underwent standardized clinical, hematological, and instrumental evaluations.

Anthropometric evaluations

At baseline and during follow-up, we assessed physical examination findings and vital signs in all ED-CRT non-responders, with subgroup analyses stratified by vericiguat use (vericiguat users vs. non-users). Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared [7]. Clinical evaluations included physical examination, vital signs measurements, and review of adverse events at baseline and follow-up. At each follow-up visit, the NYHA functional class was determined. In addition, patients graded their overall condition using a global self-assessment scale, reporting whether their status was unchanged, had worsened, or improved slightly, moderately, or markedly.

Echocardiographic evaluation

According to the study design and blinded to treatment allocation (vericiguat users vs. non-users), two physicians experienced in transthoracic two-dimensional echocardiography performed baseline and follow-up examinations. Echocardiographic assessments included M-mode recordings, conventional Doppler, and pulsed-wave tissue Doppler imaging (TDI), using a Philips iE33 echocardiography system (Eindhoven, the Netherlands). Images were acquired in parasternal long- and short-axis views, and all measurements were averaged over five consecutive cardiac cycles. The following parameters were assessed: LV end-diastolic diameter (LVEDD), end-diastolic volume (LVEDV), end-systolic diameter (LVESD), end-systolic volume (LVESV), and LVEF calculated using the Simpson’s biplane method [8]. Mitral regurgitation severity was quantified as the ratio of the color Doppler regurgitant jet area to the left atrial systolic area and graded as mild (+), moderate (+ +), moderate to severe (+ + +), or severe (+ + + +) in accordance with established criteria [8].

Laboratory analysis

After an overnight fast, plasma glucose, serum lipids, B-type natriuretic peptide (BNP), and N-terminal pro-BNP (NT-proBNP) were measured in the study cohorts using standard enzymatic assays. To assess inflammatory burden, we determined serum levels of pro-inflammatory cytokines (tumor necrosis factor-α [TNF-α], interleukin-1 [IL-1], and interleukin-6 [IL-6]), systemic inflammatory markers (C-reactive protein [CRP]), and leukocyte and neutrophil counts at baseline and after 12 months of follow-up [9]. Commercially available enzyme-linked immunosorbent assay (ELISA) kits were used to measure TNF-α, IL-1, IL-6, and CRP (TNF-α: Human ELISA Kit KHC3011, Thermo Fisher Scientific; IL-1: Human IL-1α ELISA Kit RAB0269, Sigma-Aldrich; IL-6: Human IL-6 Quantikine ELISA Kit D6050, R&D Systems; CRP: Human ELISA Kit KHA0031, Thermo Fisher Scientific). Blood samples were collected using an ice-cooled system, immediately centrifuged for 10 min at 2500 rpm at 4 °C, and the supernatant serum was isolated and stored at − 80 °C until ELISA analyses were performed. For the evaluation of ED markers in CRT non-responder cohorts, plasma levels of endothelin-1 (ET-1, pmol/L) were measured using a chemiluminescent sandwich immunoassay (CT-proET-1 LIA, B.R.A.H.M.S GmbH, Hennigsdorf/Berlin, Germany), as previously reported [3]. The analytical detection limit for CT-proET-1 was 0.4 pmol/L, with an interlaboratory coefficient of variation < 10% for values > 10 pmol/L. CT-proET-1 was stable in plasma at room temperature for at least 4 h [3, 10].

Echo Doppler measurements of the brachial artery and assessment of endothelial function

Endothelial function was assessed by ultrasound examination of the brachial artery, located above the antecubital fossa, using a 10-MHz linear-array transducer. A blood pressure cuff was placed on the forearm distal to the probe, in accordance with international guidelines [11]. The echographic examinations were performed by experienced physicians in the morning, with patients fasting, resting, and in the supine position for at least 20 min at a room temperature of 22 °C [11]. At rest and during reactive hyperemia, brachial artery peak blood flow velocity was measured using pulsed-wave Doppler, and the internal diameter was measured from the anterior to the posterior wall at the media–intima interface (“M” line) at end diastole. Endothelium-dependent FMD was calculated as the maximum percentage increase in brachial artery diameter during reactive hyperemia induced by the release of an occlusion cuff inflated to 200 mmHg for 5 min on the upper arm distal to the measurement site [11]. Peak flow velocity was recorded at 15 and 30 s after cuff deflation, and arterial diameter changes were continuously monitored for 6 min [11]. Inter- and intra-observer variability for diameter measurements was ≤ 3%. CRT non-responders with FMD ≤ 7.1% were included in the study. As previously reported [4], an FMD value ≤ 7.1% was used as the diagnostic cutoff for ED in CRT non-responder patients.

Study endpoints

At 1 year of follow-up, we evaluated the primary and secondary study endpoints. As the primary study endpoint, we evaluated the rate of CRT non-responders who normalized their endothelial function (without ED, referred to as “NED”) and the rate of CRT responders at 1 year of follow-up. As a secondary study endpoint, we evaluated the serum inflammatory markers, ET-1 expression, and the clinical outcomes: hospitalizations for HF worsening, cardiac deaths, and all-cause deaths at 1 year of follow-up. We assessed endothelial function normalization by brachial artery echocolor Doppler, with FMD measurement at the end of follow-up. Patients with FMD > 7.1% at 1 year were classified as having the “normalization of endothelial function” [3]. CRT responders were defined, 6 months after study enrollment, by the patients showing all of the following criteria: (1) a reduction in LVESV of ≥ 10% on cardiac echocardiography, (2) a significant improvement in the 6-min walk test (6MWT), and (3) a score > 8 on the Minnesota Living with Heart Failure questionnaire. These parameters are consistent with established diagnostic criteria [3–5]. Hospitalizations for HF worsening were defined as hospital admissions due to exacerbation of symptoms or signs of HF requiring treatment intensification, typically with intravenous diuretics, in patients with a prior HF diagnosis and excluding new-onset HF episodes (4). Cardiac deaths were defined as deaths attributable to cardiovascular causes, including sudden cardiac death, within 1 year of follow-up [4].

Statistical methods

All data were collected and analyzed by a qualified statistician. The study population of ED-CRT non-responders was divided into two cohorts: those who used vericiguat (vericiguat users) and those who did not (non-users). A sample size calculation was performed prior to enrollment. Assuming a 20% difference in the primary endpoint (CRT responder rate) between groups, with 80% power and a two-sided alpha of 0.05, a minimum of 74 patients per group was required. To account for potential dropouts and to enhance statistical robustness, we aimed to enroll at least 75 patients in the vericiguat group and 200 in the control group. The safety analyses were performed by comparing the incidence of adverse clinical outcomes and biomarker changes between the two cohorts. Continuous variables were expressed as means ± standard deviations and compared using two-tailed Student’s t-tests. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Predictors of the primary endpoints were assessed using multivariate Cox regression models adjusted for relevant covariates, including age, BMI, sex, smoking status, creatinine levels, diabetes, dyslipidemia, heart rate, systolic blood pressure, and ARNI therapy. A p-value < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS software (version 23.0; SPSS Inc., Chicago, IL, USA).

Results

The clinical characteristics of ED-CRT non-responders, stratified by vericiguat users (n 81) and non-users (n 208), are reported in Table 1 and Supplementary Table 1. At baseline, no significant differences were observed between the two cohorts with respect to clinical characteristics, inflammatory serum markers (Table 1, Supplementary Table 1), serum ET-1 levels, and FMD values (p > 0.05). Similarly, echocardiographic parameters and background anti-HF drug therapy did not differ significantly between groups at baseline (p > 0.05) (Table 1, Supplementary Table 1). The mean baseline FMD values were 3.55 ± 0.15% in vericiguat users and 3.85 ± 0.25% in non-users, with no significant difference (p > 0.05). The mean vericiguat dose in the treatment group was 8.1 ± 1.6 mg/day. Overall, 68% of patients reached the target dose of 10 mg/day, while 21% were maintained on 5 mg/day, and 11% remained at 2.5 mg/day. At 1-year follow-up, vericiguat users showed a significant improvement in NYHA functional class and 6-min walk test (6MWT) performance, along with a significant reduction in LV dimensions and volumes compared with non-users (p < 0.05) (Table 1, Supplementary Table 1). These changes were accompanied by a significant increase in LVEF and a reduction in mitral regurgitation severity (p < 0.05) (Table 1, Supplementary Table 1). Vericiguat users vs. non-users showed a significant reduction in inflammatory/oxidative stress biomarkers, as well as in BNP and Nt-proBNP serum values (Supplementary Table 1). In addition, they exhibited lower serum ET-1 expression (p < 0.05) and higher FMD values (p < 0.05) (Fig. 1). At follow-up end, we noted that the patients with paroxysmal AF represented less than 10% of the population: 6 (7.8%) patients were vericiguat users vs. 19 (9.1%) non-users (p > 0.05), vericiguat was generally well tolerated, and hypotension related was observed in 3 (3.7%) patients of the cohort.

Table 1.

Characteristics of the study population at baseline and1-yearfollow-up, such as vericiguat users (n 81) vs. non-vericiguat users (n 208)

		At baseline			At 1 year of follow-up
Study variables	Vericiguat users n 81		Non-vericiguat users n 208	p-value	Vericiguat users n 81	Non-vericiguat users n 208	p-value
Age, years	70.9 ± 4.7		70.5 ± 6.6	0.335	/	/	/
Male, n (%)	48 (59.3)		108 (51.9)	0.261	/	/	/
BMI > 30 kg/m2, n (%)	7 (8.6)		12 (5.8)	0.376	8 (9.9)	14 (6.7)	0.521
Smokers, n (%)	44 (54.3)		121 (58.2)	0.552	49 (60.5)	127 (61.1)	0.930
Hypertension, n (%)	54 (66.7)		156 (75.0)	0.153	57 (70.4)	163 (78.4)	0.152
Diabetes mellitus, n (%)	56 (69.1)		132 (63.5)	0.364	60 (74.1)	151 (72.6)	0.799
Dyslipidemia, n (%)	38 (46.9)		110 (52.9)	0.362	41 (50.6)	118 (56.7)	0.348
COPD, n (%)	22 (27.2)		55 (26.4)	0.813	23 (28.4)	57 (27.4)	0.866
IDCM, n (%)	49 (60.5)		120 (57.7)	0.664	/	/
I NYHA class, n (%)	/		/	0.634	6 (7.4)	1 (0.5)	0.001*,#,##
II NYHA class, n (%)	30 (37.0)		83 (39.9)		38 (46.9)	60 (28.8)
III NYHA class, n (%)	44 (54.3)		112 (53.8)		34 (42.0)	134 (64.4)
IV NYHA class, n (%)	7 (8.6)		15 (7.2)		1 (1.2)	21 (10.1)
QRS duration, ms	135.4 ± 7.7		136.4 ± 6.6	0.278	142.1 ± 7.7	143.1 ± 6.6	0.310
6MWT	189.48 ± 28.76		192.91 ± 19.57	0.246	254.82 ± 44.92	203.36 ± 26.52	0.008*,#
Medications
Amiodarone, n (%)	18 (22.2)		59 (28.4)	0.289	19 (23.5)	62 (29.8)	0.280
ACE inhibitors, n (%)	31 (38.3)		87 (41.8)	0.571	26 (32.1)	76 (36.5)	0.472
ARS blockers, n (%)	11 (13.6)		41 (19.7)	0.223	9 (11.1)	33 (15.9)	0.303
Sacubitril/valsartan, n (%)	18 (22.2)		57 (27.4)	0.367	23 (28.4)	82 (39.4)	0.080
Beta-blockers
Carvedilol, n (%)	36 (44.4)		103 (49.5)	0.438	39 (48.1)	107 (51.4)	0.615
Bisoprolol, n (%)	27 (33.3)		58 (27.9)	0.361	22 (27.2)	56 (26.9)	0.967
Aspirin, n (%)	30 (37.0)		91 (43.8)	0.299	37 (45.7)	100 (48.1)	0.714
Ticlopidine, n (%)	3 (3.7)		9 (4.3)	0.811	5 (6.2)	14 (6.7)	0.839
Warfarin, n (%)	18 (22.2)		47 (22.6)	0.945	22 (27.2)	59 (28.4)	0.838
NOAC, n (%)	22 (27.2)		49 (23.6)	0.523	23 (28.4)	57 (27.4)	0.866
Calcium antagonist, n (%)	5 (6.2)		15 (7.2)	0.775	7 (8.6)	22 (10.6)	0.623
Ivabradine, n (%)	18 (22.2)		56 (26.9)	0.411	22 (27.2)	71 (34.1)	0.254
Digoxin, n (%)	29 (35.8)		76 (36.5)	0.907	32 (39.5)	80 (38.5)	0.870
Loop diuretics, n (%)	69 (85.2)		185 (88.9)	0.166	67 (82.7)	191 (91.8)	0.001*
Aldosterone blockers, n (%)	46 (56.8)		122 (58.7)	0.773	42 (51.9)	131 (63.0)	0.050*
SGLT2i, n (%)	17 (21)		52 (25)	0.472	18 (22.2)	71 (34.1)	0.050*,##
Statins, n (%)	61 (75.3)		164 (78.8)	0.515	120 (76.9)	331 (79.8)	0.421
Insulin, n (%)	21 (25.9)		61 (29.3)	0.565	24 (29.6)	66 (31.7)	0.729
Metformin, n (%)	48 (59.2)		135 (64.9)	0.371	49 (60.5)	139 (65.1)	0.311
Sulfonylureas, n (%)	20 (24.7)		59 (28.4)	0.674	22 (27.2)	66 (31.7)	0.448
Thiazolidinediones, n (%)	9 (11.1)		26 (12.5)	0.703	11 (13.6)	29 (13.9)	0.776
GLP-1 agonist, n (%)	14 (17.3)		40 (19.2)	0.703	16 (19.8)	45 (21.6)	0.725
DPP-4 inhibitors, n (%)	21 (25.9)		53 (25.5)	0.938	19 (23.5)	50 (24.0)	0.917

BMI body mass index, NYHA New York Heart Association, 6MWT 6-min walking test, LVEF left ventricle ejection fraction, LVEDd left ventricle end-diastolic diameter, LVESd left ventricle end-systolic diameter, LVEDv left ventricle end-diastolic volume, LVESv left ventricle end-systolic volume, LNR lymphocytes/neutrophils ratio, BNP B-type natriuretic peptide, NT-proBNP N-terminal pro-B-type natriuretic peptide, LNR lymphocytes-to-neutrophils ratio, CRP C-reactive protein, IL-6 interleukin 6, TNF-α tumor necrosis factor-alpha, ACE angiotensin-converting enzyme, ARS angiotensin receptor blockers, NOAC new oral anticoagulants, SGLT2i sodium glucose transporter 2 inhibitors. *Statistical significance vs vericiguat users (p < 0.05). #Statistical significance vs. baseline in vericiguat users (p < 0.05). ##Statistical significance vs. baseline in non-vericiguat users (p < 0.05)

Fig. 1.

This figure shows the representation of serum endothelin-1 (ET-1) in the upper part and the flow-mediated vasodilation values of brachial artery (FMD) in lower part, comparing vericiguat users (blue color) vs. non-vericiguat users (red color). For both graphs, the baseline values are reported in the left part (superior and inferior panel), while the follow-up end data are reported in the right part (superior and inferior panel). Endothelin values are expressed with median and interquartile values. *Statistical significance (p < 0.05)

Notably, we compared the level and magnitude of change in ET-1 levels between those who became responders to CRT and those who remained CRT non-responders, as well as those with and without normalization of ED in the “user” group (Supplementary Fig. 2). Regarding the CRT responders vs. CRT non-responders, at baseline, ET-1 levels were comparable across subgroups. However, at follow-up, the restored CRT responders showed a significantly greater reduction in ET-1 levels (median decrease: − 2.3 pg/mL, from ~ 4.0 to ~ 1.7 pg/mL), compared to non-responders (− 0.9 pg/mL, from ~ 3.7 to ~ 2.8 pg/mL) (p < 0.05) (Supplementary Fig. 2).

Regarding the patients with vs. without normalization of ED, thepatients with ED normalization had a more pronounced ET-1 reduction (median decrease: − 2.5 pg/mL, p < 0.05), while those without ED normalization exhibited only a modest change (− 1.0 pg/mL) (p > 0.05) (Supplementary Fig. 2). At the end of follow-up, a smaller proportion of vericiguat users required treatment with loop diuretics, mineralocorticoid receptor antagonists, and SGLT2 inhibitors than non-users (p < 0.05) (Table 1). At 1-year follow-up, compared with baseline, vericiguat users showed significant improvements in NYHA functional class and 6MWT performance, along with reductions in LV volumes, decreased mitral regurgitation severity, and an increase in LVEF (p < 0.05) (Table 1, Supplementary Table 1). These patients also exhibited a significant reduction in cellular and molecular markers of serum inflammation (p < 0.05) (Table 1). In contrast, non-vericiguat users demonstrated worsening of NYHA class and mitral regurgitation severity at 1 year compared with baseline (p < 0.05. Table 1, Supplementary Table 1). Moreover, non-users showed a progressive and significant increase in lymphocyte counts, IL-6, and TNF-α levels, and a higher proportion of patients required SGLT2i therapy at the end of follow-up (p < 0.05) (Table 1).

At the end of follow-up, vericiguat users had a significantly higher rate of patients with normalization of endothelial function (30 [37%] vs. 21 [10.1%], p < 0.05) and CRT responders (28 [34.6%] vs. 27 [13%], p < 0.05) compared with non-users. In addition, FMD values were significantly higher in vericiguat users than in non-users (6.54 ± 0.61 vs. 4.49 ± 0.29, p < 0.01). At the follow-up end, vericiguat users vs. non-users showed a lower rate of HF hospitalizations (11 (13.6%) vs. 51 (24.5%), p < 0.05), cardiac deaths (9 (11.1%) vs. 43 (20.7%), p < 0.05), and all causes of deaths (18 (22.2%) vs. 67 (32.2%), p < 0.05). At 1 year of follow-up, the normalization of endothelial function was predicted by LVEF (0.42, CI 95% [0.02–0.72]) and vericiguat (4.35, CI 95% [2.45–7.70]) (Table 2 Part A). The CRT responders’ outcome was predicted by vericiguat therapy (2.85, CI 95% [1.67–4.89]) (Table 2 Part B). In Fig. 2, we showed the Kaplan curves for the cumulative risk of having normalization of endothelial function and responders at 1 year of follow-up in the study cohorts.

Table 2.

Cox regression analysis for normalization of endothelial function (Part A) and responders (Part B) at 1 year of follow-up in the study population of ED-CRT non-responders (299 patients)

Part A: Normalization of endothelial function
		Univariate analysis			Multivariate analysis
Risk factors	HR	CI 95%	p-value	HR	CI 95%	p-value
Age	0.94	0.90–0.98	0.007*	0.94	0.89–1.01	0.310
BNP	1.00	0.99–1.01	0.611	1.00	0.99–1.01	0.911
LNr	0.66	0.36–1.21	0.180	0.88	0.45–1.74	0.715
NYHA 2	0.96	0.55–1.69	0.893	1.22	0.69–2.16	0.491
LVEF	0.06	0.00–0.79	0.032*	0.42	0.00–0.72	0.029*
6MWT	0.99	0.98–1.00	0.223	0.99	0.98–1.00	0.228
Vericiguat	3.91	2.24–6.80	0.001*	4.35	2.46–7.70	0.001*
Part B: Conversion to CRT responders
		Univariate analysis			Multivariate analysis
Risk factors	HR	CI 95%	p-value	HR	CI 95%	p-value
Age	1.02	0.97–1.06	0.444	1.03	0.99–1.08	0.463
BNP	1.00	0.99–1.01	0.818	0.99	0.99–1.01	0.413
LNr	0.77	0.44–1.35	0.361	0.75	0.39–1.42	0.546
NYHA 2	0.93	0.54–1.61	0.809	1.08	0.62–1.86	0.580
LVEF	0.45	0.04–5.16	0.520	0.20	0.01–3.32	0.768
6MWT	1.00	0.99–1.01	0.838	0.67	0.99–1.01	0.390
Vericiguat	2.72	1.61–4.59	0.001*	2.85	1.67–4.89	0.001*

ED endothelial dysfunction, CRT cardiac resynchronization therapy, BNP B-type natriuretic peptide, LNr lymphocyte-to-neutrophil ratio, NYHA New York Heart Association class 2, LVEF left ventricle ejection fraction, 6MWT 6-min walking text. *Statistical significance (p < 0.05)

Fig. 2.

This figure shows the Kaplan curves for the cumulative risk of having endothelial dysfunction (upper part) and becoming CRT responders (lower part) in the study population of vericiguat users (blue color) vs. non-vericiguat users (red color) at 365 days of follow-up. *Statistical significance (p < 0.05)

Discussion

This study provides the first evidence that treatment with vericiguat is associated with improved endothelial function and a higher rate of restored CRT responders among patients initially classified as ED-CRT non-responders. These favorable effects appear to be associated with reductions in oxidative/inflammatory stress, as well as circulating ET-1 levels, both of which represent key biomarkers of adverse prognosis in CRT non-responders [3]. At 1-year follow-up, vericiguat therapy conferred a 4.4-fold greater likelihood of endothelial function normalization and a threefold greater likelihood of achieving CRT responders. In contrast, a lower baseline LVEF (HR 0.42) was associated with a reduced probability of endothelial function normalization at 1 year. CRT non-responders typically exhibit progressive worsening of endothelial function, driven by increased expression of inflammatory and oxidative stress markers, including ET-1 [1, 3]. Our findings reinforce the importance of endothelial function as a determinant of CRT response and suggest that targeting endothelial dysfunction may be therapeutically relevant [1, 3]. Endothelial dysfunction is linked to impaired NO–sGC–cGMP signaling, thereby amplifying oxidative/inflammatory burden and structural myocardial damage [12]. Vericiguat acts on the impaired NO–sGC–cGMP pathway by stimulating sGC activity, increasing cGMP, and reducing pro-inflammatory and pro-oxidative mediators [12]. These mechanisms likely contribute to the observed improvements in LV reverse remodeling and cardiac function [12]. In our cohort, vericiguat use was associated with improvements in serum markers of ED, significant reductions in ET-1 and inflammatory serum mediators, and functional measures of HF, as evidenced by improved 6MWT distance and NYHA class. Lastly, CRT non-responders evidence elevated ET-1 and markers of inflammatory/oxidative stress, diminished exercise capacity, and progressive worsening of NYHA class [13]. These effects are particularly evident in the ED-CRT non-responder cohort [3]. Preclinical models have demonstrated that vericiguat promotes NO synthesis in cardiomyocytes, thereby stimulating sGC to generate cGMP, which subsequently activates PKG and regulates multiple targets involved in myocyte contraction and cardiac performance [14]. Beyond its effects on the myocardium, cGMP mediates vasodilation and exerts anti-fibrotic and anti-inflammatory actions [14]. In HF, impairment of the NO–sGC–cGMP signaling cascade contributes to ED, enhanced ROS formation, and reduced NO bioavailability [15, 16]. Vericiguat directly stimulates sGC, thereby amplifying NO-dependent cGMP generation and preserving vascular and cardiac function while attenuating remodeling and reducing fibrosis [17]. The improvements in LV geometry, afterload reduction, and contractility observed in our study could support these mechanisms [17]. In our study, vericiguat use was associated with reduced ED and a higher CRT responder rate compared with non-users. Specifically, vericiguat therapy conferred a 4.4-fold greater probability of endothelial function normalization and a threefold greater probability of achieving CRT responders at 1 year. These findings could support the concept of vericiguat as both a cGMP stimulator and an anti-inflammatory/antioxidative therapy in ED-CRT non-responders, ultimately favoring LV reverse remodeling, endothelial function normalization, and improved CRT responsiveness. Importantly, vericiguat use was associated with a lower incidence of HF hospitalizations and all-cause mortality within the ED-CRT non-responders’ cohort. In the VICTORIA trial, comparing the effects of vericiguat (target dose 10 mg once daily) vs. placebo in addition to guideline-based anti-HF medical therapy, authors found that composite endpoint of death from any cause or HF hospitalization was lower in the vericiguat group as compared to the placebo group (HR, 0.90; 95% CI, 0.83 to 0.98; p 0.02) [6]. On the other hand, the VICTORIA trial population consisted of 5050 patients with chronic HF and LVEF < 45%, while our study population was represented by ED-CRT non-responders and so by HF patients with failure to CRT, altered endothelial function, and severe depression of LVEF (LVEF < 35%) [6]. Intriguingly, fewer than 15% of patients in the VICTORIA trial had CRT [6]. Only about 40% of patients were in NYHA class III or IV in the VICTORIA trial, compared with more than 60% in our cohort [6]. This disparity in baseline functional status may partly explain the differences in outcomes observed between the two studies, particularly considering that the primary benefit in VICTORIA was driven by reduced HF hospitalizations rather than the mortality reduction demonstrated in our analysis [6]. However, differences between the study cohorts may explain the outcomes observed in our study compared with those in the VICTORIA trial [6]. Furthermore, beyond endothelial effects, vericiguat was also associated with significant improvements in cardiac structure and function, including increases in LVEF and reductions in LV volumes, consistent with enhanced reverse remodeling. These findings are clinically relevant, as persistent LV dilation and impaired EF are major determinants of CRT non-response and adverse prognosis [5]. These structural and functional changes reinforce the role of CRT combined with vericiguat in enhancing cardiac performance. Given their established prognostic value, EF improvement and LV volume reduction represent key findings of our analysis. However, by improving both endothelial function and myocardial remodeling, vericiguat may therefore act on complementary pathways that support CRT effectiveness. Intriguingly, in the SOCRATES trial, the authors did not observe changes in LVESV or LA volume, or in Nt-proBNP, within a short follow-up period (12 weeks) among patients with chronic HF treated with vericiguat versus placebo [18, 19]. In our study, with a longer follow-up, we found a significant decrease in LVESV despite a significant increase in QRS duration. This suggests that improvements in endothelial function and LV remodeling with vericiguat may occur independently of electrical parameters in CRT non-responders’ cohorts [5, 20]. Notably, reduced cardiac pump efficiency, reflected by a lower baseline LVEF, was associated with a nearly 60% decrease in the likelihood of endothelial function normalization at the follow-up end. Patients with the lowest LVEF had the most severe pump dysfunction both before and after CRT [20]. Nevertheless, CRT non-responders are typically characterized by reduced LVEF and worse clinical prognosis [20]. To date, severely depressed LVEF may contribute to more advanced endothelial dysfunction and worse outcomes in both CRT cohorts, especially non-responders [1–3, 20]. In our study, vericiguat was generally well tolerated, and hypotension-related events were rare (< 5% of the cohort). At follow-up, a lower percentage of vericiguat users were taking diuretics, aldosterone blockers, or SGLT2i than non-users (p < 0.05). This likely reflects clinical improvement and reverse remodeling, rather than drug intolerance. Therefore, the discontinuation of these drugs was not the result of hypotension induced by vericiguat but the probable effect of the LV reverse remodeling with amelioration of endothelial function and clinical status (lower NYHA class and increased 6MWT), accompanied by the best clinical outcomes (CRT response, etc.) in ED-CRT non-responders under vericiguat therapy. Furthermore, our findings indicate that vericiguat use is associated with improvements in endothelial function across multiple levels, including echocardiographic (FMD values), biochemical (ET-1, oxidative/inflammatory stress markers), and clinical (NYHA class, HF hospitalizations) parameters, in CRT non-responders. Finally, by comparing the level and magnitude of change in ET-1 levels between those who became CRT responders and those who remained non-responders, as well as those with and without normalization of ED in the vericiguat-user group, we found that at the follow-up end, the CRT responders showed a significantly greater reduction in ET-1 compared to the non-responders. Conversely, patients with normalized ED had a more pronounced ET-1 reduction than those without ED normalization. Furthermore, these findings suggest that a greater decline in ET-1 levels is associated with both CRT response and reversal of endothelial dysfunction in vericiguat users. These effects further support a mechanistic role of vericiguat in modulating vascular tone and myocardial remodeling in ED-CRT non-responders. This study has several limitations. First, serum samples were obtained from peripheral veins rather than the coronary sinus, as patients had previously undergone CRT implantation and coronary sinus sampling carried a risk of LV lead dislodgement. Second, the follow-up period was limited to 1 year. A longer-term observation is warranted to better assess the effects of vericiguat on survival and sustained outcomes in future research. Third, the relatively small sample size may limit the statistical power and generalizability of our findings regarding the impact of vericiguat on ED and CRT response in non-responders. Fourth, advanced cardiac imaging modalities to evaluate myocardial metabolism and indirect mitochondrial function were not performed, which could have provided additional mechanistic insights. Our findings should be interpreted as exploratory and primarily hypothesis-generating. Indeed, while our study specifically focused on CRT non-responders, future studies adequately powered by extending this analysis to the broader HFrEF population are warranted to determine whether endothelial dysfunction is a general feature of HF or a specific determinant of CRT response. Thus, a key limitation of our work is that we evaluated ED only in CRT non-responders, without including the broader HFrEF population. This restricted design prevents us from determining whether the observed association between vericiguat, endothelial function, and CRT responsiveness is specific to this subgroup or generalizable across all patients with HFrEF. Notably, our study included only patients with left bundle branch block (LBBB), a factor that may influence both CRT response and the overall generalizability of our findings. Conversely, because vericiguat was introduced in June 2023 in clinical practice, we enrolled the study cohorts at two different times: ED-CRT non-responders who received vericiguat (vericiguat users) were enrolled after June 2023, while ED-CRT non-responders who did not receive vericiguat (vericiguat non-users) were enrolled before June 2023. The cohorts were not randomized to vericiguat therapy, and differences in the timing of inclusion in vericiguat therapy could introduce bias into the results. Nevertheless, baseline clinical characteristics and therapeutic management were comparable, and statistical adjustments were applied to minimize potential temporal bias. Our findings should be interpreted within the context of the observational nature and sample size of our study. Given the observational nature of this study, these findings should be interpreted as associative rather than causal. Future investigations with larger and more heterogeneous cohorts are warranted to provide further confirmation. In addition, the use of advanced statistical methodologies, such as propensity score matching, would help validate these results more rigorously, reduce residual confounding, and strengthen the robustness of the observed associations. Future studies with larger cohorts, extended follow-up, and randomized designs are therefore required to confirm and expand upon these findings in ED-CRT non-responders before definitive conclusions can be drawn.

Conclusions

Vericiguat use was associated with lower circulating ET-1 levels, improved FMD, and restored endothelial function in ED-CRT non-responders. In addition, vericiguat was associated with improvements in clinical status (as measured by 6MWT performance and NYHA class), cardiac geometry, and systolic function (increase in LVEF). These favorable associations were observed alongside a higher rate of patients with improved endothelial function and CRT responders. Furthermore, vericiguat use was associated with a lower rate of HF hospitalizations and mortality within the ED-CRT non-responder cohort.

Abbreviations

6MWT

Six-minute walk test

ACE

Angiotensin-converting enzyme

AF

Atrial fibrillation

ARNI

Angiotensin receptor–neprilysin inhibitor

ARS

Angiotensin receptor blockers

BMI

Body mass index

BNP

B-type natriuretic peptide

COPD

Chronic obstructive pulmonary disease

CRP

C-reactive protein

CRT

Cardiac resynchronization therapy

CRTd

Cardiac resynchronization therapy with defibrillator

ED

Endothelial dysfunction

ED-CRT

Endothelial dysfunction in cardiac resynchronization therapy patients

EF

Ejection fraction

ET-1

Endothelin-1

FMD

Flow-mediated dilation

HF

Heart failure

HFrEF

Heart failure with reduced ejection fraction

IL-1

Interleukin-1

IL-6

Interleukin-6

LBBB

Left bundle branch block

LV

Left ventricle

LVEDV

Left ventricular end-diastolic volume

LVESV

Left ventricular end-systolic volume

LVEF

Left ventricular ejection fraction

LNR

Lymphocyte-to-neutrophil ratio

MRA

Mineralocorticoid receptor antagonist

NO

Nitric oxide

NOAC

Non-vitamin K oral anticoagulant

NT-proBNP

N-terminal pro-B-type natriuretic peptide

NYHA

New York Heart Association

PKG

Protein kinase G

ROS

Reactive oxygen species

Sgc

Soluble guanylate cyclase

SGLT2i

Sodium-glucose cotransporter 2 inhibitor

TNF-α

Tumor necrosis factor alpha

Authors’ contributions

CS conceptualized the study and drafted the manuscript. CF, LR, FN, CC, NDO, and CT contributed to study methodology, supervision, and data interpretation. LVM, JR, DC, MV, CLM, NM, VG, AL, MA, LM, and MLB were involved in data acquisition, investigation, and laboratory analyses. FCS, DC, and RM critically revised the manuscript for important intellectual content. All authors read and approved the final manuscript.

Funding

The authors received no specific funding for this study.

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki. The study protocol was approved by the ethics committees of all participating institutions (no. 1203). Written informed consent to participate in the study was obtained from all patients prior to enrollment.

Consent for publication

All authors have reviewed and approved the final version of the manuscript and consent to its publication.

Competing interests

The authors declare no competing interests.