This review summarizes the results of randomized controlled trials (RCTs) investigating anti-inflammatory therapies in HF, with a particular focus on the stage of HF at baseline (also in Table 1, Table 2 and Table 3). The serious adverse events and safety profile across these therapeutic trials are also noted in Table 4. The direct impact of these therapies is examined in the context of patient phenotype and disease severity at the time of treatment.

TNF-α is an inflammatory cytokine that is over-expressed in HF and associated with poor prognosis. Early studies investigating anti-TNF-α therapies, particularly etanercept, a recombinant soluble TNF-receptor, demonstrated potential benefits in small pilot trials. Deswal et al. (1999) conducted a double-blind, dose escalation study of etanercept in NYHA Class III HF patients, which showed significantly improved LVEF in etanercept group (+ 3.8 ± 2.0% vs −3.8 ± 2.9%, p = 0.04) compared to placebo. The 6-min walk distance (6MWD), quality-of-life, and ejection fraction (EF) were also improved in the group treated with higher doses (4 and 10 mg/m2) [18]. Similarly, a study by Bozkurt et al. (2001) in NYHA III-IV HF patients demonstrated a dose-dependent improvement (from 5 to 12 mg/m2) in LVEF and in LV end-diastolic and end-systolic volumes (p = 0.01, p = 0.04, p = 0.02, respectively) with etanercept, although the benefits were observed only in NYHA III patients [19].

However, larger scale trials have failed to replicate these benefits. The RENEWAL (Randomized Etanercept Worldwide Evaluation) trial, which combined data from the RECOVER and RENAISSANCE trial, included a patient cohort with more advanced HF staging (46% NYHA IIIa, 25% NYHA IIIb, and 4% NYHA IV). The primary endpoint of death or hospitalization for HF (HHF) was not reduced with etanercept, thereby leading to early trial termination of both trials due to lack of benefit [20]. Similarly, the ATTACH (Anti-TNF-α in Congestive Heart Failure) trial investigating infliximab, a chimeric monoclonal TNF-α antibody, in NYHA III-IV patients highlighted concerns about the safety of TNF-α antagonism in advanced HF. Through the 28 weeks of the trial, the high-dose group (10 mg/kg) experienced significantly higher rates of death or HHF [hazard ratio (HR) 2.84 with 95% confidence interval (CI): 1.01–7.97; p = 0.043] [21]. Although both infliximab groups showed early reduction in CRP and IL-6 within one week and an improvement in LVEF (+ 3.5% with 5 mg/kg and + 2.1% with 10 mg/kg, p = 0.039 for placebo versus both infliximab groups combined), these changes did not persist beyond week 14. By week 28, inflammatory biomarkers and LVEF returned to baseline, which coincided with the upwards trend in adverse events, including mortality and HHF. Given that the last infliximab dose was administered at week 6, this deterioration may reflect a loss of anti-inflammatory effect over time, a rebound inflammatory response following abrupt withdrawal of high-dose immunosuppression, or potential dose-dependent toxicity. Collectively, these large-scale studies suggest that anti-TNF-α therapies may be ineffective or even harmful in advanced HF stages, thus emphasizing the need for clinical phenotyping of HF prior to treatment selection. Although initial studies suggest potential benefits in milder stages (NYHA II–III), the efficacy of TNF-α blockade appears to be lower in more progressed disease, highlighting a narrow therapeutic window in which such interventions may be beneficial.

While these early trials evaluated initial safety of etanercept in HF patients, they were generally small in size, had short duration of follow up, and were not designed to systematically evaluate inflammatory biomarkers and the extent to which these were responsive to therapy. The larger RENEWAL trial was performed in a larger sample of patients and enrolled patients with more advanced HF. While the ATTACH trial had a modest patient size and duration of treatment, it was the only TNF-α inhibitor study that measured inflammatory markers longitudinally. However, the numerical values of baseline CRP levels and magnitude of CRP changes were not reported, limiting the interpretation of infliximab effects. The lack of biomarker data overall in these trials limits our ability to identify the correlation between anti-inflammation and clinical outcomes. This gap underscores the potential importance of integrating therapeutic target-responsive biomarkers in future trials.

Targeting the IL-1 pathway, a driver of systemic inflammation, has demonstrated potential in HF therapy across early clinical trials. Ikonomidis et al. (2008) investigated anakinra, a recombinant IL-1α receptor antagonist, in patients with rheumatoid arthritis (RA). Anakinra treatment led to a significant decline in IL-6 (p = 0.003) and CRP levels (−80 ± 9% vs −58 ± 6%, p = 0.004) and improvements in vascular and LV function, measured by enhanced coronary flow reserve (+ 29 ± 2% vs + 4 ± 2%, p < 0.001), aortic distensibility (+ 45 ± 3% vs + 2 ± 2%, p < 0.001), and echocardiographic E/Em ratio (−15 ± 1% vs −7 ± 1%, p = 0.005) when compared to placebo [22]. This study demonstrated that anakinra can improve cardiovascular function, providing a basis for exploring IL-1 blockade in HF.

In CANTOS (Canakinumab Anti-inflammatory Thrombosis Outcomes Study), an IL-1β monoclonal antibody canakinumab was investigated in post-myocardial infarction (MI) patients with elevated high-sensitivity CRP (hsCRP). Canakinumab reduced the primary endpoint of stroke, nonfatal MI, or cardiovascular death by 15% in the group treated with 150 mg (p = 0.021) and HHF with a dose-dependent effect (p for trend = 0.025) [23, 40]. At 48 months, treatment with canakinumab led to an hsCRP reduction from baseline that was 26%, 37%, and 41% greater in the 50 mg, 150 mg, and 300 mg group, respectively, compared to placebo group (p < 0.001 for all comparisons of median percent change compared to placebo). Notably, post-hoc analysis indicated that patients who achieved a reduction in hsCRP with canakinumab had better primary endpoint outcomes (HR 0.62; CI: 0.47–0.81) compared to those with sustained hsCRP levels (HR 1.03; CI: 0.81–1.31). While the baseline HF staging was not reported, the HHF incidence rate was higher among treated patients with a history of HF than treated patients without HF at baseline [40]. Moreover, canakinumab did not lead to greater reductions in HHF in patients with a history of HF at baseline versus those with no history of HF, indicating that its benefit is not specifically enhanced in patients with pre-existing HF. Collectively, these findings underscore the potential of IL-1 inhibition in improving HF outcomes for at-risk patients when systemic inflammation, as measured by inflammatory biomarkers, is adequately targeted.

While these studies investigated HF incidence-related endpoints, the D-HART (Diastolic Heart Failure Anakinra Response Trial) pilot study evaluated RA patients with elevated hsCRP and NYHA II-III heart failure with preserved ejection fraction (HFpEF). Patients treated with anakinra saw improved aerobic capacity, measured by peak oxygen consumption (peak VO2), and reduced plasma CRP levels (−6.1 mg/L [−74%], p = 0.006), with reductions correlating with improved peak VO2 (R =  − 0.60, p = 0.002) [24]. The D-HART2 trial showed a significant reduction in CRP (−4.0 mg/L [−65%], p = 0.026) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels (p = 0.022) at 4 weeks in anakinra-treated NYHA II-III HFpEF patients. However, no improvement in peak VO2 or ventilatory efficiency was observed [26]. The advanced disease state (NYHA III predominance) in D-HART2 may account for the differences in cardiopulmonary responses compared to the D-HART study, which had an equal distribution of NYHA II and III patients. Additional factors that may have contributed to these observed outcome differences include: the greater degree of CRP reduction in D-HART and the higher prevalence of severe obesity in D-HART2 (60% of participants with body mass index > 40 kg/m2 compared to 30%), known to limit aerobic capacity independent of cardiac functioning. Despite the absence of pulmonary fitness improvement in D-HART2, anakinra-treated patients still exhibited reductions in NT-proBNP levels and increases in exercise duration, both linked to favorable HF outcomes. This supports an overall benefit of anakinra in the setting of HFpEF.

In the REDHART (Recently Decompensated Heart Failure Anakinra Response Trial), NYHA II-III HF patients with recently decompensated HF showed significantly reduced CRP levels (−3.4 mg/L from baseline [−66%], p = 0.011) and improved peak VO₂ from baseline (p = 0.009) with 12 weeks of anakinra treatment. The changes in CRP level correlated with changes in peak VO₂ (R = −0.57, p = 0.001), supporting the use of IL-1 blockade as a systemic anti-inflammatory strategy in inpatient settings for acute decompensation [25].

Overall, these results support the use of IL-1 inhibition as a promising strategy, particularly in early or acute stages of HF. Anakinra has demonstrated improvements in vascular and cardiopulmonary functioning across multiple studies. The CANTOS trial, with its large sample size and extended follow-up (mean of 3.7 years), also provides evidence supporting canakinumab in reducing cardiovascular events in patients with elevated inflammatory risk. However, limitations should be acknowledged. The REDHART, D-HART, and D-HART2 studies were modest in size and length of study and conducted at a single center, which may limit its generalizability across other clinical settings and patient populations. D-HART also only had one male participant and, thus, was not representative of the demographics of the general population. These limitations underscore the need for large, multicenter trials to validate and extend these results.

IL-6, a downstream cytokine of IL-1 signaling, is elevated in HF patients and independently associated with increased risk of mortality and hospitalization [41]. The RESCUE (Trial to Evaluate Reduction in Inflammation in Patients With Advanced Chronic Renal Disease Utilizing Antibody Mediated IL-6 Inhibition) study investigated ziltivekimab, an IL-6 ligand monoclonal antibody, in individuals with chronic kidney disease (CKD), elevated hsCRP, and high cardiovascular risk [27]. After 12 weeks, ziltivekimab produced a significant dose-dependent reduction in the primary outcome of hsCRP (−77% for 7.5 mg, −88% for 15 mg, and −92% for 30 mg, p < 0.0001). Dose-dependent reductions were also observed in atherothrombotic biomarkers, including fibrinogen, serum amyloid A, haptoglobin, secretory phospholipase A2, and lipoprotein(a). These anti-inflammatory effects occurred without changing lipid parameters (i.e., total cholesterol to high-density lipoprotein ratio), supporting a mechanism independent of lipid modulation. A post-hoc analysis also demonstrated a significant reduction in the neutrophil-to-lymphocyte ratio, consistent with suppression of myeloid-driven inflammation [42]. RESCUE did not look into the clinical implications of reducing inflammatory and thrombotic biomarkers. Therefore, building upon these results, the ongoing phase 3 ZEUS (Effect of Ziltivekimab vs Placebo on Cardiovascular Outcomes in Participants with Established Atherosclerotic Cardiovascular Disease) trial evaluates the impact of ziltivekimab on MACEs and renal outcomes in a similar CKD with elevated hsCRP patient population [43].

The RESCUE trial data is also one of the first providing evidence of dose-dependent reduction of atherothrombotic markers via IL-6 inhibition. Although the RESCUE trial had a modest size and duration of therapy, ZEUS will address some of these limitations as it will recruit 6,200 patients and include assessments such as baseline LVEF. The strength of these trials is their focus on patients with elevated baseline inflammation, thereby enriching the study population for individuals more likely to benefit from anti-inflammatory therapy. The targeted enrollment of individuals with CKD and high cardiovascular risk is particularly relevant, as these comorbidities are well-established risk factors for HF, particularly HFpEF. Given the modulatory effects of ziltivekimab on atherogenic inflammatory pathways, if the ZEUS trial also confirms similar clinical benefits, IL-6 blockade may emerge as a potential strategy for individuals with similar at-risk profiles for HF.

Methotrexate, a folate antagonist with non-specific anti-inflammatory effects mediated by adenosine, has been studied for its potential as a HF therapy. In the METIS (Methotrexate Therapy Effects in the Physical Capacity of Patients with Ischemic Heart Failure) trial, which investigated chronic ischemic HF patients, there was no improvement in the primary endpoint of 6-min walk test with low-dose methotrexate. However, there was a trend towards improvement in NYHA class after 12 weeks, with 66.7% of the methotrexate group showing improvement compared to 50% of the placebo group (p = 0.2) [29]. The absence of significant findings in METIS could be attributed to the lack of hsCRP reduction in the treatment group, indicating insufficient inflammation control with methotrexate. Additionally, the predominance of patients in advanced HF stages (72% NYHA Class III/IV in the methotrexate group versus 52% in placebo) likely contributed to the limited therapeutic response, as these patients may have progressed beyond a stage where anti-inflammatory therapy could impact cardiac remodeling.

In the CIRT (Cardiovascular Inflammation Reduction Trial), methotrexate was studied in patients at risk for HF, defined as those with a diagnosis of type 2 diabetes mellitus or metabolic syndrome and a prior MI or multivessel coronary artery disease. Methotrexate did not reduce cardiovascular events or HHF in this population. However, with the exception of NYHA IV exclusion, the trial did not report baseline HF classes, limiting interpretation of symptom severity at enrollment. Moreover, the plasma levels of key inflammatory markers, including IL-1β, IL-6, and hsCRP, were not reduced, suggesting either inadequate anti-inflammatory action by methotrexate or limitations in the effects of methotrexate on cardiac-specific inflammatory mediators [28].

In these trials, the failure of methotrexate to reduce inflammatory biomarkers warrants the need to better investigate its potential effects on HF under conditions where it demonstrates sufficient anti-inflammatory effects. Although the METIS trial data was limited by a small sample size and short 12-week treatment duration, it provided preliminary insight into the potential use of methotrexate in the setting of HF. However, its ability to impact physical capacity in patients may have been constrained by its limited treatment window and the lack of dose-dependent evaluation. Furthermore, the METIS trial did not report baseline inflammatory biomarker levels and exclusively focused on ischemic HF, which may not generalize to other HF etiologies. CIRT may also have been limited by its inclusion of patients with only modest elevations in systemic inflammation (median hsCRP levels of 1.6 mg/L), possibly enrolling patients less likely to benefit from anti-inflammatory therapies. The lack of HF phenotype detailing of baseline NYHA class also limits interpretation, raising concerns that these patients were too advanced in their HF staging to experience benefit. Despite these limitations, major strengths of CIRT include its large cohort of 4,786 patients and its extended median follow-up of 2.3 years, offering valuable longitudinal data on cardiovascular outcomes.

Colchicine is an anti-inflammatory alkaloid that disrupts microtubule formation and inhibits nucleotide-binding domain-like receptor protein 3 (NLRP3) inflammasome activation, thereby targeting key inflammatory pathways involved in cardiovascular diseases [44]. The COLCOT (Colchicine Cardiovascular Outcomes Trial) study demonstrated significant reduction in the primary endpoint of cardiovascular death, cardiac arrest, MI, stroke, and urgent angina hospitalization (HR 0.77; CI: 0.61–0.96, p = 0.02) in post-MI patients treated with colchicine. While HHF was comparable across groups, the primary endpoint occurred significantly more frequently in patients with HF at baseline (HR of 1.81; CI: 1.08–3.04, p = 0.03) [30]. This suggests that further stratification by baseline HF status is needed to better understand the potentially different outcomes within the treatment group. Akrami et al. (2021) also found that among patients with acute coronary syndrome (ACS), colchicine significantly reduced major adverse cardiac events (MACE) overall, including ACS, stroke, survival rate, and decompensated HF, with a cumulative incidence of 6.7% in the colchicine group versus 21.7% in placebo group (p = 0.001). The decompensated HF rate itself also trended more favorably for the colchicine group (HR of 1.93; CI: 1.71–2.18), supporting the potential of early colchicine intervention to favorably influence at-risk populations before progression to HF [31].

In a trial by Deftereos et al. (2014) involving patients with stable HF with EF < 40% and average NYHA class of 2.4 ± 0.5 (NYHA IV excluded), colchicine led to significantly greater reductions in left ventricular end-diastolic and end-systolic diameters. Despite overall hsCRP reduction in the colchicine group (mean of −5.1 mg/l, p < 0.001), lack of improvement in the primary endpoint of NYHA class reduction and continuously elevated post-treatment hsCRP level (mean of 5.5 mg/l) suggest an inadequate inflammatory response reduction to drive clinical benefit [32].

The COLICA (Colchicine in Acute Heart Failure) trial observed that colchicine significantly reduced inflammatory markers CRP (−70.8% in colchicine group vs −51.1% in placebo; ratio of change of 0.60, p < 0.01) and IL-6 (−49.9% in colchicine group vs −30.1% in placebo; ratio of change of 0.72, p = 0.019) in patients with acute HF (NYHA Class II-IV). However, there was no clinical benefit and no between-group differences in NT-proBNP levels, NYHA class, or HHF and/or mortality risks [33]. The inclusion of predominantly late-stage NYHA Class III and IV patients and subsequent results underscore the potentially diminishing clinical impact of colchicine in later HF stages compared to the more optimal results observed in trials examining earlier stages of the disease. Moreover, COLICA did not select patients with elevated baseline inflammatory markers, which may explain the lack of clinical benefit despite the significantly reduced hsCRP and IL-6.

These trials investigating colchicine in cardiovascular disease and HF have several strengths and limitations. One of the main strengths of COLCOT is its large sample size of 4,745 patients. However, baseline hsCRP was only measured in a small subset of 207 patients, limiting insight into the degree of immune modulation and how this correlated to clinical outcomes. Both the COLICA trial and the study led by Akrami et. al (2021) were limited by small sample size, though the latter revealed improved clinical endpoints. COLICA also had high therapy discontinuation rates in both groups (~ 25% of participants) and an older study population (median age 75 years) compared to other trials, which may limit generalizability. The trial by Akrami et. al also did not measure inflammatory biomarker levels, thus data on the correlation between anti-inflammation and MACE reduction remains unknown. Overall, these limitations highlight the importance of selecting patients most likely to benefit from anti-inflammatory therapy and highlight the need for future studies on colchicine to stratify by HF phenotype and disease stage.

Myeloperoxidase (MPO) is a leukocyte-derived enzyme that links reactive oxygen species to inflammation, making it a potential target for cardiovascular disease. In the SATELLITE trial, mitiperstat, an MPO inhibitor, was evaluated in patients with NYHA class II-IV symptoms, specifically HFpEF or heart failure with mildly reduced ejection fraction (HFmrEF). While the SATELLITE trial was terminated prematurely due to the COVID-19 pandemic, it demonstrated a significant reduction in its primary endpoint of MPO activity by 75% (p = 0.001). Additionally, there was a trend towards improvement in quality-of-life, depicted by the + 6.283 point least squares mean difference (CI: 0.461–13.027, p = 0.067) in the Kansas City Cardiomyopathy Questionnaire (KCCQ) summary score [34].

The ENDEAVOR trial also further studied the effects of mitiperstat in symptomatic NYHA Class II-IV HFpEF or HFmrEF patients. Although the full report of ENDEAVOR has yet to be published at the time of this review, the preliminary results, presented at the AHA Scientific Sessions 2024, revealed no improvements in the primary outcomes of baseline KCCQ scores and 6MWD at 16 weeks among mitiperstat-treated patients (2.5 mg and 5 mg groups combined) compared to placebo [36]. Secondary outcomes, including KCCQ and 6MWD at 24 and 48 weeks, MACE, HF hospitalizations, NT-proBNP, inflammatory markers (hsCRP, IL-6), and echocardiogram parameters, also showed no differences among groups [36]. The lack of anti-inflammatory effects could also explain the absence of clinical benefits in these participants.

In both the SATELLITE and ENDEAVOR trials, the absence of baseline NYHA Class distribution limits assessment of whether disease stage influenced treatment response. Additionally, SATELLITE enrolled fewer participants than anticipated, resulting in exploratory rather than inferential statistical analyses and did not measure inflammatory markers besides MPO activity. The ENDEAVOR trial also did not recruit patients with elevated systemic inflammation at baseline, as indicated by hsCRP or increased MPO-specific activity, which could potentially account for the lack of therapeutic benefit observed with mitiperstat in this patient population. Nonetheless, given that these trials represent some of the first investigations into MPO inhibition in HF, its possible role as a target for future HF therapeutics remains an open question under investigation.

Non-specific immunomodulation therapies (IMT), such as intravenous immunoglobulins (IVIG), influence the concentrations of cytokines and its modulators. Gullestad et. al (2001) studied IVIG in HF patients with LVEF < 40% and NYHA class II-III (NYHA 2.6 ± 0.1) symptoms. IVIG significantly induced an anti-inflammatory state compared to baseline, evidenced by a 57% increase in IL-1Ra (p < 0.001), 65% increase in IL-10 (p < 0.001), and a 37% reduction in TNF-α/soluble TNF receptors ratio (p < 0.01). Interestingly, the IVIG group also had a significant increase in LVEF (p < 0.01) and 24% reduction in N-terminal pro-atrial natriuretic peptide (NT-pro-ANP) levels (p < 0.001) from baseline, while the placebo group had no differences [37]. However, within the IVIG group, participants with markedly reduced LVEF and longer duration of HF symptoms did not experience LVEF improvement, possibly representing a subgroup of chronic HF with irreversible myocardial damage due to long-standing disease. These findings suggest that IVIG effectively induces an anti-inflammatory state and potentially improves hemodynamic parameters in selected HF patient populations, particularly those without advanced, irreversible cardiac remodeling.

In the ACCLAIM trial, Celacade, a device-based IMT that treats autologous blood ex vivo with oxidative stress and UV light before reinjection, was evaluated in HF patients (predominantly NYHA Class II-III with only 4% NYHA IV). Compared to placebo, the IMT group saw a significant improvement in quality of life, based on the −6.9 score change (p = 0.04) in the Minnesota Living with Heart Failure Questionnaire (MLHFQ). Although the primary endpoint of death or cardiovascular hospitalization did not differ significantly between groups, stratification by NYHA class revealed that patients with NYHA II symptoms who received IMT experienced significantly better primary endpoint outcomes than their placebo counterpart (HR 0.61; CI: 0.46–0.80; p = 0.0003) [38].

The CORTAHF (Effect of Short-Term Prednisone Therapy on CRP Change in Emergency Department Patients with Acute Heart Failure and Elevated Inflammatory Markers) trial showed that 7 days of prednisone therapy significantly reduced hsCRP (adjusted geometric mean ratio of 0.30 in prednisone group vs 0.40 in placebo, p = 0.0498) in acute HF patients with elevated hsCRP at baseline. Prednisone also resulted in a decreased 90-day risk of worsening HF, HF readmission, or death (10.4%) compared to usual care (30.8%) (HR of 0.31; CI: 0.11–0.86, p = 0.016). Quality of life, as measured by the EuroQol 5-Dimension 5-Level analogue scale, also improved more in the prednisone group (least squares mean difference 2.57, 95% CI 0.12–5.01 points, p = 0.040) [39]. In CORTAHF, the patients had the following baseline NYHA class II-IV composition: II (16%), III (73%), and IV (11%). However, since the results were not stratified by NYHA class, it remains unclear whether disease severity influenced the observed benefits. As a pilot study and one of the first to investigate corticosteroids in HF, these findings warrant further investigation in larger trials to better define patient selection, long-term safety, and the potential role of steroids in HF.

Together, these exploratory studies highlight the potential benefit of non-specific IMT, particularly in patients with earlier-stage HF or with acute HF, where anti-inflammatory treatments may offer improvements in both hemodynamic parameters and quality of life. However, several limitations do exist within these trials. The study led by Gullestad et. al (2001) and CORTAHF had small number of participants. The investigators assessing outcomes in CORTAHF were also not blinded. However, the risk of bias introduction was mitigated by the use of objective endpoints, such as hsCRP and readmission rates. The ACCLAIM trial, despite its large sample size, failed to meet its primary endpoint, possibly due to inadequate inflammation suppression, as depicted by unchanged hsCRP levels. However, its promising results in NYHA II participants, as well as findings from IVIG and corticosteroid studies, underscore the need for larger trials with more precise HF patient stratification and selection.