Sodium-glucose cotransporter-2 inhibitors (SGLT2i) are currently recommended in clinical guidelines for the treatment of heart failure (HF) over the full spectrum of left ventricular ejection fraction (LVEF).1 Robust evidence supports their clinical benefits, including a consistent reduction in the risk of HF decompensation, hospitalizations, and cardiovascular mortality.2 The beneficial effects of SGLT2i are thought to arise from multiple cardiovascular mechanisms, encompassing both cardiac and renal protective pathways.3

Circulating biomarkers play a pivotal role in elucidating the pathophysiological effects of cardiovascular therapies. Among these, B-type natriuretic peptides are the most extensively validated biomarkers in HF, supporting diagnosis, monitoring, and risk stratification by reflecting myocardial stretch.4 HF pathophysiology involves complex interactions among several maladaptive processes, including inflammation, fibrosis, congestion, and systemic neurohormonal dysregulation. However, the pathophysiology of HF involves a complex interplay of maladaptive biological processes, including inflammation, fibrosis, congestion, and systemic neurohormonal dysregulation. Although SGLT2i have demonstrated broad clinical benefits, their effect on cardiac biomarkers remains incompletely defined. Existing studies often evaluate a limited subset of biomarkers, focus on specific LVEF phenotypes, or report inconsistent findings, particularly with respect to B-type natriuretic peptides and inflammatory markers.5–8 Moreover, there is a notable lack of data addressing the effects of SGLT2i on atrial natriuretic peptides, neurohormonal stress markers, and biomarkers of congestion and fibrosis.

The DAPA-MODA study enrolled stable patients with chronic HF over the entire LVEF spectrum.9 This investigation assessed the impact of initiating dapagliflozin on cardiac remodeling over a 6-month period in patients receiving optimized guideline-directed medical therapy (GDMT) without prior SGLT2i exposure. In the primary analysis, dapagliflozin produced favorable effects on both atrial and ventricular remodeling. Here, we present findings from a prespecified substudy evaluating the effects of dapagliflozin on circulating biomarkers representing 5 key pathophysiological pathways implicated in HF progression.

This study characterizes the pathophysiological profile of a well-defined cohort of patients with established chronic heart failure, using 11 circulating biomarkers that reflect 5 key biological pathways involved in the pathogenesis of HF. It also evaluates their response to the initiation of dapagliflozin. In this population, biomarkers of cardiac stress, inflammation, and neuroendocrine activation emerged as the primary indicators of the presence of disease. Furthermore, dapagliflozin initiation was associated with improvements across several pathophysiological pathways, including reductions in cardiac natriuretic peptides (NT-proBNP and MR-proANP), inflammatory markers within the IL-1 axis (CRP and IL-6), MR-proADM, copeptin, and P1NP. Notably, these effects were most pronounced in patients with elevated baseline levels (figure 5).

Figure 5.

Central illustration. Frequency of elevated biomarkers belonging to the 5 different pathobiological pathways studied in a population of 156 stable ambulatory chronic heart failure patients from the DAPA-MODA trial. The initiation of dapagliflozin was associated with improvements in several pathophysiological systems, predominantly in patients with elevated baseline levels, including both natriuretic peptides (NT-proBNP and MR-proANP), the IL-1 related inflammatory axis (CRP and IL-6), congestion (MR-proADM), neuroendocrine stress (copeptin), and fibrosis (P1NP).

CA-125, cancer antigen 125; CRP, C-reactive protein; GDF-15, growth differentiation factor 15; HF, heart failure; hsTNT, high-sensitivity troponin T; IL-6, interleukin 6; MR-proADM: mid-regional pro adrenomedullin; MR-proANP, mid-regional proatrial natriuretic peptide; NT-proBNP, N-terminal probrain natriuretic peptide; P1NP, procollagen type 1 N-terminal propeptide; ST2, suppression of tumorigenicity 2.

(1.01MB).

We assessed NT-proBNP and MR-proANP levels as indicators of ventricular and atrial cardiac stress, respectively, and hsTnT levels as a marker of myocardial injury. Both natriuretic peptides, NT-proBNP and MR-proANP, showed a significant reduction over the 180-day period in patients with elevated baseline levels. This observation agrees with previous evidence demonstrating that treatment with SGLT2i exerts a beneficial effect in lowering NT-proBNP concentrations in patients with heart failure, irrespective of reduced or preserved ejection fraction.6 Atrial natriuretic peptides have been less extensively studied; however, some small-scale studies have reported a significant reduction following treatment with SGLT2i.5 However, hsTnT levels remained stable throughout the study period, consistent with findings from the DAPA-HF trial, which reported a nonsignificant trend toward attenuation of the increase in hsTnT from baseline to 12 months in patients treated with dapagliflozin compared with placebo.21 It is plausible that SGLT2i may blunt the progressive rise in hsTnT levels in patients with HF, potentially reflecting a deceleration in disease progression.

We also analyzed 3 inflammatory biomarkers: IL-6, CRP, and GDF-15. Among the IL-1 axis-related markers, both IL-6 and CRP levels decreased over the 30-day and 180-day periods. Interestingly, patients with elevated baseline levels experienced a significant reduction in both biomarkers, whereas those with low baseline levels exhibited a statistically significant increase, although it was not sufficient to exceed the established threshold. The decrease in IL-6 and CRP levels has been previously described in HF patients receiving dapagliflozin,8 and it reveals a significant reduction in the inflammatory state. Regarding GDF-15 levels, we observed a significant increase in patients both below and above the normal reference range. Similarly, data from the EMPEROR program and the EMPIRE-HF Biomarker substudy showed that treatment with empagliflozin was associated with a significant elevation in GDF-15 levels.7,22 Given the concurrent improvements observed in other biomarkers, it seems unlikely that the increase in GDF-15 levels induced by SGLT2i reflects a deterioration in cardiac function. Rather, it has been suggested that GDF-15 may act as a beneficial metabolic regulator, with associations reported between elevated levels and weight loss, improved glucose tolerance, and reduced food intake and appetite.23 Further research is needed to confirm this apparently contradictory elevation of GDF-15, and the underlying mechanisms and clinical implications in the context of SGLT2 inhibition.

CA-125 and MR-proADM were assessed as biomarkers of congestion. While no significant change was observed in CA-125 levels, MR-proADM showed a significant reduction in patients with elevated baseline values, with the majority reaching normal levels by day 180. Unlike in acute HF, where SGLT2i have been associated with a significant reduction in CA125 levels during the early postdischarge period,24 the absence of an effect on CA125 in the present study may reflect the clinical stability of the cohort, characterized by low baseline levels. Similar findings have previously been reported in stable HF populations.25 Interpretation of CA125 may be influenced by the fact that 21% of participants had a history of cancer, although none had an active malignancy in the previous year or a condition affecting life expectancy.

In contrast, the reduction in MR-proADM suggests that this biomarker may reflect changes in vascular permeability rather than a decrease in residual congestion. This biomarker has not been extensively studied in HF patients treated with SGLT2i, but previous studies, such as the EMPIRE-HF trial, reported no significant variation in MR-proADM concentrations.26

We also measured copeptin levels as a biomarker of neurohormonal stress. Although levels appeared to increase at 30 days in the overall population, a significant reduction was observed at 180 days among patients with elevated baseline values. This initial rise is consistent with findings from the EMPIRE-HF trial and the DAPA-SHUTTLE trial at 12 and 4 weeks, respectively, and may reflect enhanced vasopressin release in response to early SGLT2i therapy.26,27 Such a response likely promotes renal water conservation and stabilizes urine volume, thereby counterbalancing the osmotic diuretic effect of glucosuria during the initial phase of treatment.27 However, this effect does not appear to persist over the longer term, and our findings at 180 days suggest a progressive decline in vasopressin demand, indicative of improved neurohormonal status over time.

Lastly, we examined P1NP and ST2 as biomarkers of extracellular matrix remodeling and fibrosis. P1NP levels showed a significant reduction among patients with elevated baseline values. Although this biomarker has not been extensively studied in patients with chronic heart failure (CHF), the EMPEROR trial reported no significant effect of SGLT2i therapy on P1NP levels.28 P1NP is known to reflect collagen synthesis and extracellular matrix remodeling. This is central processes in cardiac fibrosis and myocardial stiffening following injury, which contribute to the pathogenesis and progression of CHF. Our findings may indicate a meaningful improvement in this regard. In contrast, no significant changes in ST2 levels were observed, apart from a slight increase among patients with baseline values within the normal reference range. This may reflect sustained fibrotic activation mediated by the IL-33/ST2 signaling pathway,29,30 which appears unresponsive to SGLT2 inhibition. This finding contrasts with results from a previous small study in which dapagliflozin significantly reduced ST2 levels in HF patients with type 2 diabetes.8

Several previous studies have evaluated these pathophysiological pathways in both acute decompensated HF and chronic HF. The EMPIRE-HF trial investigated the effect of empagliflozin over a 12-week period on a selection of similar biomarkers in patients with stable HF with reduced ejection fraction receiving optimal medical therapy, including NT-proBNP, MR-proANP, MR-proADM, and copeptin, among others.26 In contrast, our study included patients with HF irrespective of LVEF, reflecting the established clinical benefit of SGLT2i throughout the spectrum of LVEF phenotypes. In this population of clinically stable patients receiving optimal HF therapy, dapagliflozin exerted effects on multiple pathophysiological pathways, though its impact was concentrated in a subset of specific biomarkers. These findings provide mechanistic support for the clinical benefits observed with SGLT2 inhibition and suggest which biomarkers may be most useful for monitoring treatment response to this evidence-based therapy.

Our study builds upon insights from previous multibiomarker investigations, such as the EMPIRE-HF trial, and contributes to the growing body of evidence regarding the effects of SGLT2i therapy on distinct pathophysiological pathways in HF. Importantly, it should be emphasized that a simple analysis based solely on absolute biomarker concentrations may lead to misleading or invalid conclusions. Categorization into high and low biomarker levels was prespecified and defined before conducting statistical analyses, based on established reference normal values for each biomarker, and the therapeutic effect of dapagliflozin became evident only when considering baseline overactivation of the relevant pathways. The most pronounced effect was observed in cardiac natriuretic peptides, but meaningful changes were also detected in biomarkers related to inflammation, congestion, neurohormonal stress, and extracellular matrix remodeling.

Overall, all our findings highlight a key observation: the effect of treatment on circulating biomarkers appears to depend on their baseline levels. Patients with elevated baseline values experienced a more pronounced response in biomarkers, whereas those with values within the normal range showed minimal or no changes. This pattern suggests a differential biological response, potentially reflecting a normalization process rather than a uniform pharmacological effect. These results support the relevance of baseline biomarker profiling to better understand individual treatment responses.

In a population of stable chronic HF patients from the DAPA-MODA trial, dapagliflozin initiation was associated with improvements in several pathophysiological systems, including both natriuretic peptides (NT-proBNP and MR-proANP), IL-1 related inflammatory axis (CRP and IL-6), congestion (MR-proADM), neuroendocrine stress (copeptin) and fibrosis (P1NP). These effects were predominantly observed in patients with elevated baseline levels.

This study was sponsored by the Spanish Society of Cardiology and has received funding by an unconditional investigational grant from Astra Zeneca.

The study protocol was approved by the Institutional Ethics Committees (ESR/20/20594), and all participants were required to sign an informed consent prior to enrolment. The investigation was conducted in accordance with the principles outlined in the Declaration of Helsinki, and SAGER guidelines regarding potential sex/gender bias have been followed.

No artificial intelligence was used in the preparation of this article.

Each author contributed significantly to the submitted work as follows: V. Martínez-Pina contributed to data collection and interpretation, as well as drafting and revising the final manuscript. A. Riquelme-Pérez contributed to the statistical analysis and interpretation of the data, and to drafting and revising the final manuscript. A. Bayés Genís, J. Nuñez, I. Cebreiros, H. Morillas, M. Cobo-Marcos, J. M. García-Pinilla, J. F. Rodríguez-Palomares, D. Dobarro, M. A. Restrepo-Córdoba, J. R. González-Juanatey, and J. L. Zamorano critically revised the manuscript for important intellectual content. J. A. Noguera-Velasco and D. A. Pascual-Figal contributed to the conception and design of the work, the analysis and interpretation of data, and critically revised the manuscript for important intellectual content. All authors approved the final version of this manuscript.

D. A. Pascual-Figal has received consultancy and speaker fees from AstraZeneca, Novartis, Roche Diagnostics, Pfizer, Vifor, Rovi, and Bayer. A. Bayés-Genís has participated in advisory boards and/or delivered lectures for Abbott, AstraZeneca, Bayer, Boehringer Ingelheim, Novartis, Roche Diagnostics, and Vifor. H. Morillas has received speaker fees from AstraZeneca. J. Nuñez has received consultancy and speaker fees from AstraZeneca, Alleviant, Amgen, Bayer, Boehringer Ingelheim, CSL Vifor, Daiichi Sankyo, GSK, Lilly, Pfizer, Novartis, Novo Nordisk, and Rovi. M. Cobo reports consultancy and lecture fees from AstraZeneca, Vifor Pharma, Novartis, Rovi, Boehringer Ingelheim, Novo Nordisk, and Bayer. The remaining authors declare no conflicts of interest.