Re‐PERFUSE: Phase 1b study of AZD3427, a novel relaxin receptor agonist, on renal perfusion in HFrEF patients

Abstract

Aims

Renal impairment frequently coexists with heart failure (HF) and is associated with increased risk of poor clinical outcomes. This highlights the urgent need for therapies targeting both cardiac and renal dysfunction. AZD3427, a long‐acting recombinant fusion protein and relaxin analogue that selectively activates the relaxin family peptide receptor 1 (RXFP1), showed trends of increased stroke volume and estimated glomerular filtration rate (eGFR) in HF patients (NCT04630067). The hypothesis is that AZD3427 may enhance GFR by expanding the functional renal cortex volume and improving renal perfusion.

Methods and results

The Re‐PERFUSE study (NCT06611423) is a Phase 1b, randomised, double‐blind, placebo‐controlled trial aimed at evaluating the effects of AZD3427 on renal perfusion in patients with HF and reduced ejection fraction (HFrEF) with reduced eGFR. Patients with HF, EF ≤ 40% and an eGFR of 30 to 90 mL/min/1.73 m², assessed by the CKD‐EPI 2021, will receive a single dose of subcutaneous AZD3427 (n = 6) or placebo (n = 6). Intravenous dopamine will serve as a positive control for increased renal blood flow. Positron emission tomography (PET) imaging with [¹⁵O]‐labelled water will be used to measure renal cortex blood flow pre‐ and post‐treatment, allowing differentiation between global and focal renal blood flow changes and providing insights into potential nephron recruitment and increased filtration membrane area. Safety and tolerability will be assessed through monitoring of adverse events, clinical laboratory tests and vital signs.

Conclusions

This study, alongside other ongoing AZD3427 studies, aims to evaluate the dual effects of AZD3427 in improving both cardiac and renal function. These insights could guide the development of future therapeutic strategies for managing HF and renal impairment.

Introduction

The cardiorenal syndrome is a multifaceted disorder characterised by the reciprocal deterioration of cardiac and renal functions, where impairment in one organ system accelerates dysfunction in the other, leading to poor outcomes and posing significant challenges to clinical management. ¹ Diminished cardiac output, increased central venous pressure and increased sympathetic nervous system activation result in decreased renal perfusion due to the centralization of circulation and constriction of the afferent arterioles within the nephrons. This impaired renal function leads to fluid retention and electrolyte disturbances, such as hyperkalemia, which subsequently increase cardiac preload and may precipitate arrhythmias. Furthermore, the accumulation of uremic toxins, including ammonia, trimethylamine and indoxyl sulfate, can exert detrimental effects on both the heart and kidneys, exacerbating their dysfunction. ² The complex nature of this syndrome complicates therapeutic strategies, with renal ischaemia and remodelling being contributing factors to diuretic resistance. ³

Relaxin, a naturally occurring hormone, is acknowledged for its vasodilatory, anti‐inflammatory and anti‐fibrotic properties. ⁴ , ⁵ , ⁶ Targeting the relaxin receptor RXFP1 through AZD3427 offers the potential to induce favourable long‐term haemodynamic and structural changes, including enhancing renal perfusion while reducing intrarenal resistance, potentially reversing or halting pathological vascular remodelling. Over the past two decades, attempts to harness relaxin's therapeutic benefits in heart and kidney disease patients have been made ⁷ , ⁸ , ⁹ ; however, the short‐lived nature of these intravenously administered relaxin compounds has limited their chronic therapeutic potentials.

AZD3427 is a recombinant protein, long‐acting relaxin analogue suitable for chronic subcutaneous treatment, facilitating ongoing and long‐term therapeutic effects. Preliminary evidence from a Phase 1b study indicates that AZD3427 can improve eGFR in heart failure, particularly in patients with reduced ejection fraction (HFrEF). However, the mechanisms through which relaxin and AZD3427 elevate eGFR remain obscure. Clarifying whether the increase in eGFR is solely due to a regional rise in renal blood flow (RBF) and consequent increase in intraglomerular pressure, or, as we hypothesise, from the recruitment of previously under‐perfused nephrons that increase the filtration area without raising intraglomerular pressure, is crucial to distinguishing between potentially harmful and clinically beneficial effects (Figure  1 ).

Figure 1.

AZD3427 and potential mechanisms of GFR increase. This figure illustrates two hypothetical scenarios explaining how AZD3427 may increase renal blood flow (RBF) and glomerular filtration rate (GFR). AZD3427, a relaxin analogue, is proposed to enhance RBF through vasodilation. In the first scenario on the left, AZD3427 induces a moderate and global increase in RBF (F) by opening previously underperfused capillaries and recruiting dormant nephrons. This results in reduced intrarenal vascular resistance (R), allowing for increased RBF and filtration without raising—or even with lowering—intraglomerular pressure (P). On PET imaging, this would appear as a diffuse, global increase in RBF (reflected in increase in imaging signal intensity across multiple PET image volume elements, a.k.a. voxels). In contrast, in the second scenario (on the right) a focal and intense increase in RBF is restricted to already perfused nephrons, without recruitment of new capillary beds. In this case, intrarenal resistance remains unchanged, potentially leading to overperfusion of individual nephrons and a potentially harmful increase in intraglomerular pressure. Panel A: Schematic representation of RBF under Scenario 1, showing a widespread but moderate increase in blood flow, and Scenario 2, showing a localised but intense increase in specific renal regions. Panel B: Modelled PET images illustrating changes in RBF under the two scenarios. Panel C: Schematic of glomerular blood flow via afferent (red) and efferent (blue) arterioles. Scenario 1 shows uniform vasodilation across glomeruli, while Scenario 2 depicts selective vasodilation in already perfused glomeruli, with others remaining underperfused. Panel D: Hemodynamic framework illustrating an increase in RBF with or without a decrease in intrarenal resistance. In Scenario 1, RBF increases alongside a reduction in resistance due to recruitment of parallel glomerular arterioles, expanding the filtration surface without elevating glomerular pressure. In Scenario 2, RBF increases without opening additional arterioles—that is, without changing intrarenal resistance—with filtration driven mainly by elevated glomerular pressure.

This study aims to test the hypothesis that AZD3427 increases GFR by enlarging the renal cortex filtration area and enhancing RBF in participants with HFrEF and reduced GFR.

Study design

Re‐PERFUSE is a Phase 1b randomised, double‐blind, placebo‐controlled, single‐dose, single‐centre study to evaluate the renal haemodynamic effects of AZD3427 in participants with HFrEF and reduced eGFR (NCT06611423). Approximately 12 participants will be randomised. Eligible participants will be first randomised equally to IV dopamine diluted in saline (positive control) or saline only placebo, followed by a second randomisation to SC AZD3427 or AZD3427 placebo as shown in Figure 2 . The total study duration (from screening to follow‐up) for each participant will be up to 83 days, depending on the length of the screening period.

Figure 2.

Study design. IV, intravenous; PET, positron emission tomography; R, randomised; SC, subcutaneous.

Objectives

The primary objective of this study is to evaluate the effect of AZD3427 on the volume of perfused renal cortex in participants with HFrEF and reduced eGFR. This will be assessed using quantitative parametric mapping via PET imaging.

Exploratory objectives include evaluating the effect of AZD3427 on total renal perfusion in this population. In addition, changes in plasma and serum biomarkers including renin, aldosterone and copeptin will be evaluated.

Study participants

Eligible participants are adults with a confirmed clinical diagnosis of HFrEF, classified as EF ≤ 40% and NYHA class I to III, for at least 12 months. Participants must have eGFR between 30 and 90 mL/min/1.73 m², assessed using the CKD‐EPI 2021 equation. They must also be on stable standard‐of‐care heart failure medication for at least 4 weeks prior to screening, with no significant dose adjustments or new medications introduced during that time. Female participants must be of non‐childbearing potential. The full list of inclusion and exclusion criteria is presented in Table 1 .

Table 1.

Key inclusion and exclusion criteria

Inclusion criteria
> 20 years‐old at the time of signing the informed consent.
Minimum BMI of 18.0 kg/m2 and a minimum body weight of 50 kg at screening.
Known clinical diagnosis of HF NYHA FC I to III for at least 12 months.
Stable HF standard of care medication, including diuretics (if used), for at least 4 weeks prior to screening. It is also expected that there will be no change in treatment between screening (Visit 1) and Visit 3 in terms of drugs known to affect renal haemodynamics.
LVEF ≤ 40% based on echocardiography conducted within the last 6 months.
eGFR of 30 to 90 mL/min/1.73 m2 (inclusive) assessed by the CKD‐EPI, 2021 (creatinine only) equation at screening.
Female participants must not be pregnant or lactating and must be of non‐childbearing potential, confirmed at screening.

Exclusion criteria
History of any clinically important disease or disorder which, in the opinion of the investigator, may either put the participant at risk because of participation in the study, or influence the results or the participant's ability to participate in the study.
Any clinically important illness, medical/surgical procedure, or trauma within 4 weeks of screening or planned surgical or other procedure before study completion.
Clinically significant, as judged by the investigator, ventricular arrhythmias requiring pharmacological treatment.
Historical or current evidence of a clinically significant disease or disorder including, but not limited to (a) myocardial infarction, stroke, transient ischaemic attack, coronary artery bypass grafting, percutaneous coronary intervention, implantable cardioverter defibrillator implantation and within 12 weeks prior to screening. (b) Primary cardiomyopathy other than dilated, including but not limited to sarcoidosis, amyloidosis, restrictive cardiomyopathy, active myocarditis, constrictive pericarditis, hypertrophic (obstructive) cardiomyopathy and complex congenital heart disease.
Decompensated HF or hospitalization due to any cause < 4 weeks prior to screening.
Severe heart valve disease.
Diagnosis of polycystic kidney disease or anatomical causes of chronic kidney disease. One kidney, renal artery stenosis or glomerulonephritis. Anticipated dialysis or renal transplantation within 1 year. Condition where vasodilatory therapy may be contraindicated for example but not limited to severe aortic stenosis, hypertrophic obstructive cardiomyopathy, restrictive cardiomyopathy. History of active malignancy within 2 years, with the exception of fully excised or treated basal cell carcinoma or ≤2 squamous cell carcinomas of the skin.
If on treatment with other drugs with potential to significantly influence on eGFR (e.g., SGLT2i and ACE‐I), the participant must have been on a stable dose for at least 4 weeks prior to screening. PDE‐5 inhibitors such as sildenafil (e.g., taken for erectile dysfunction or other as‐needed reasons) and NSAIDs such as diclofenac and ibuprofen should not be taken for 2 days before the PET Sessions 2 and 3 (Visit 2 and Visit 3); paracetamol is allowed.
Haemoglobin < 100 g/L at screening. Abnormal vital signs defined as any of the following at screening: sitting SBP > 160 mmHg or sitting DBP > 110 mmHg after a period of at least 5 minutes of rest; sitting SBP < 100 mmHg or sitting DBP < 50 mmHg; resting pulse rate of <50 b.p.m. or >115 b.p.m.
For females only: currently pregnant (confirmed with positive pregnancy test) or breast‐feeding.

ACE‐I, angiotensin‐converting enzyme (ACE) inhibitors; ADAs, antidrug antibodies; BMI, body mass index; CKD EPI, Chronic Kidney Disease Epidemiology Collaboration; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HF, heart failure; HIV, human immunodeficiency virus; LVEF, left ventricular ejection fraction; NSAIDs, nonsteroidal anti‐inflammatory drugs; NYHA FC, New York Heart Association Functional Classification; PDE‐5, phosphodiesterase 5 inhibitors; PET, positron emission tomography; SBP, systolic blood pressure; SGLT2i, sodium‐glucose cotransporter‐2 inhibitors.

Conduct and ethics

This study is being conducted in accordance with the protocol and international guidelines, including the Declaration of Helsinki and the Council for International Organizations of Medical Sciences, and all applicable ethical and regulatory requirements. All participants will provide written informed consent before participating and can withdraw their consent at any time. The study is registered with ClinicalTrials.gov (NCT06611423).

Study assessments

Study procedures and their corresponding timelines are summarised in Figure 2 .

At the screening visit (Visit 1), participant eligibility is confirmed, and baseline assessments are performed. These include the collection of medical history, review of concomitant medications, physical examination, vital signs, electrocardiogram (ECG), haematology, clinical chemistry (including eGFR), urinalysis and other relevant evaluations. On Visit 2 (Day 1) the acute effects of dopamine (positive control) on RBF are assessed through two PET sessions:

• PET Session 1 consists of baseline PET examinations conducted in triplicate to improve the reliability of voxel‐wise estimates of RBF. This is followed by an intravenous infusion of either dopamine diluted in saline or saline alone.

• PET Session 2 involves a single PET examination performed approximately 10 minutes after the infusion of dopamine or saline begins, which continues for around 20 minutes. Following a rest period of at least 5 minutes after PET Session 2, participants receive a single subcutaneous dose of either AZD3427 or AZD3427 placebo before leaving the clinic.

On Visit 3 (Day 8), PET Session 3 is conducted to evaluate the effects of AZD3427 and placebo, eight days after administration of either AZD3427 or placebo. This session includes triplicate PET examinations.

The follow‐up visit (Visit 4, Day 56) occurs approximately five half‐lives of AZD3427 after administration of AZD3427 or placebo. This visit includes follow‐up safety assessments.

PET imaging

PET examinations will be carried out at the Karolinska Institutet (KI), Stockholm, Sweden. The PET measurements of the abdominal area (field of view including kidneys) will be performed with the GE Discovery MI PET‐CT system using a bolus IV injection of ¹⁵O‐radiolabelled water, [¹⁵O]H₂O. The PET data acquisition will be in list mode with a total duration of approximately 10 minutes (single examination) and each subject will undergo seven PET examinations. In preparation for PET measurements, a venous cannula will be inserted in one arm for the administration of [¹⁵O]H₂O, and a second venous cannula will be inserted in the other arm for dopamine/saline placebo injection and blood sampling. All procedures will follow hospital routines. The radioactivity per single injection will not exceed 700 ± 10% MBq/70 kg.

PET image analysis will consist of a number of steps. Firstly, the region of interest (ROI) will be delineated on the CT image for the left and right kidney cortex and the lumen of the descending aorta. Secondly, the ROIs will be applied to the 4D PET image to extract regional time‐activity curves (TAC). The TAC for the aorta will be used as the input function for estimating kinetic parameters. Thirdly, quantitative image analysis will be performed using a one‐tissue kinetic compartment model (1‐TCM). The 1‐TCM fit for voxel TACs within the ROI of the kidney cortex will yield voxel‐wise images of the model parameters: kinetic rate constants K1 (mL/min/g), k2 (1/min) and fractional blood volume, Vb (mL/mL). The K1 parametric image will be used as a voxel‐wise measure of RBF. The repeated baseline and follow‐up images will be utilised by estimating a separate parametric image for each PET measurement and then calculating the average K1 value across the repeated PET measurements (within the imaging session) for each voxel. The change in the K1 parameter between baseline and post‐treatment average images will be calculated for each voxel in the renal cortex.

Study endpoints

The primary study endpoint is the volumetric fraction (%) of the renal cortex with increased RBF from baseline to Day 8, as measured by quantitative parametric mapping of RBF using PET. Exploratory endpoints include change in total RBF from baseline to Day 8, as measured by PET and change in plasma and serum biomarkers including renin, aldosterone and copeptin from baseline to Day 8 and Day 56.

Safety assessments

The safety of AZD3427 in study participants will be ensured by continuous medical monitoring focused on observable and laboratory responses, including serious adverse events/adverse events, vital signs (blood pressure, heart rate, respiratory rate, body temperature and pulse rate), body weight, electrocardiograms, clinical laboratory assessments eGFR, urinalysis, haematology and clinical chemistry), physical examinations and injection‐site reactions.

Statistical analysis

For all pharmacodynamic analyses, the Full Analysis Set population will be used (all participants who are randomised and have received at least one dose of dopamine/saline and one dose of AZD3427/placebo). For safety analysis, the Safety Analysis Set will be used (all participants who have received at least one dose of study intervention). Continuous data will be summarised using descriptive statistics (n, mean, SD, median, minimum and maximum). Categorical data will be summarised as frequency counts and percentages. Baseline is defined as the last available non‐missing measurement before any study intervention. Data will be presented by AZD3427 placebo and AZD3427 regardless of the first study intervention of saline or dopamine. Unless otherwise specified, tests of treatment effects of the second randomised treatment will be conducted at a two‐sided alpha level of 0.1, and the two‐sided 90% confidence interval will be provided.

The primary endpoint, that is, the volumetric fraction (%) of the renal cortex with increased RBF from baseline to Day 8, will be calculated by counting the % fraction of kidney voxels with a positive change in K1. A two‐sample t‐test will be used for comparison of the primary endpoints between active and placebo.

Sample size estimation

In lack of prior literature with the exact voxel‐level primary endpoint of the current study, sample size estimation was based on published ROI‐based RBF imaging data and utilised outcome parameter simulations. In detail, assuming a 30% increase of RBF after AZD3427 and no change to placebo with a total SD (log‐scale) of 0.17 and a two‐sided level of significance of 0.1, a sample size of 12 participants (6 per group) provides at least 70% power. The total SD used in the calculations was comprised of between‐participant (log‐scale of 0.15) and within‐participant (log‐scale of 0.15) variabilities. ¹⁰

Discussion

Declining renal function plays a critical role in the pathophysiology and prognosis of HFrEF, often complicating the optimisation of evidence‐based treatments and worsening outcomes. While available evidence suggests that most guideline‐directed HF therapies are also effective in patients with CKD, there remains a high residual risk. HF patients with CKD experience approximately 2 to 44‐fold increases in hospitalisation and mortality rates compared to those with HF alone. ¹¹ , ¹² Additionally, as HF therapies can directly or indirectly affect renal function and conditions like hyperkalaemia, careful monitoring is required upon initiation. Understanding and managing renal function changes in the context of HFrEF therapies is vital for effective patient care and improved outcomes. ¹³

Relaxin has demonstrated multiple potentially beneficial properties; however, to date, clinical trials in HF and renal disease have not yielded convincingly positive results. ⁷ Currently, there is no relaxin molecule, agonist or mimetic approved for clinical use. A significant limitation of previous relaxin‐based therapies, such as serelaxin, has been their short half‐life necessitating continuous intravenous administration, which is impractical for chronic treatment.

AZD3427 addresses these limitations by fusing a relaxin‐derived peptide to the fragment crystallisable (Fc) portion of human IgG1, resulting in a long‐acting recombinant protein suitable for subcutaneous administration. This design enables sustained therapeutic levels, making it more feasible for chronic use. In a first‐in‐human, Phase 1a/b study (NCT04630067), AZD3427 demonstrated favourable safety and pharmacokinetic profiles, with a terminal half‐life of 13 to 14 days following subcutaneous administration in HF patients. Specifically, in the Phase 1b study, AZD3427 was administered as a once‐weekly subcutaneous injection for 5 weeks to participants with HF. The treatment was well tolerated and resulted in substantial increases in eGFR, accompanied by elevated renin levels, without significant changes in blood pressure. However, a few cases of hypotension were reported. ¹⁴ Due to the small sample size, it is challenging to draw definitive conclusions about the time‐response relationship and the phenotype of responders. Notably, the most significant effects on stroke volume, cardiac output and eGFR were observed 8 days post‐treatment in the cohort with an ejection fraction (EF) of 40% or less.

The increases in plasma renin, an enzyme released by renal juxtaglomerular cells in response to a drop in intraglomerular pressure, coupled with increases in eGFR, a marker of kidney glomerular filtration, suggest that AZD3427 may improve filtration without an increase in intraglomerular pressure. Such improvement in filtration without the potentially deleterious effects of high intracapillary pressure on the glomerulus could be definitively beneficial for HF patients with CKD. In this context, some preclinical and clinical studies suggest that the relaxin‐induced increase in RBF may occur without an accompanying rise in intraglomerular pressure. ¹⁵ , ¹⁶ The proposed mechanism for overall improved renal filtration, without producing a single nephron hyperfiltration, involves the recruitment of previously under‐perfused and under‐functioning nephrons. This recruitment may increase the size of the functional renal cortex and filtration membrane. Such improvements could be caused by a concurrent increase in RBF along with decreased intrarenal resistance. The reduction in resistance may occur as more capillaries and nephrons become available for blood flow (Figure  1 ). Accordingly, this study aims to provide evidence that AZD3427 improves RBF through this novel mechanism.

The renal effects of relaxin are also supported by data from other clinical studies. Serelaxin, a recombinant form of human relaxin, has been shown to increase eGFR after acute administration in healthy participants, chronic HF patients and acute HF patients, as well as after chronic administration to diffuse systemic scleroderma patients. ¹⁷ , ¹⁸ , ¹⁹ Acute IV administration of serelaxin has also been shown to increase RBF in healthy volunteers and chronic HF patients. ²⁰ Recently, healthy participants treated with recombinant relaxin showed improvements in kidney perfusion by increased effective renal plasma flow, filtration fraction and renal vascular resistance. ¹⁵ In patients with HF and renal impairment (baseline eGFR < 60 mL/min/1.73 m²), renal dysfunction was associated with higher cardiovascular and all‐cause mortality after 6 months in placebo‐treated participants, but not in participants acutely treated with serelaxin, suggesting a potential beneficial effect in this patient population. ²¹

The present trial is designed to test the hypothesis that AZD3427 increases eGFR by expanding the size of the renal cortex filtration area and increasing RBF in participants with HF and reduced eGFR (30 to 90 mL/min/1.73 m²). To test this hypothesis, AZD3427‐induced changes in renal cortex perfusion will be evaluated using PET imaging with ¹⁵O‐labelled water, aiming to distinguish between increases in RBF resulting from global changes in renal cortex blood flow and those associated with only focal changes in RBF. The former may be indicative of renal capillary recruitment and improved perfusion of non‐perfused or under‐perfused nephrons, which could be beneficial in HF patients with renal impairment and as discussed above.

Dopamine will be administered to participants in this study as a positive control (known to increase RBF). Namely, dopamine at a low dose is a potent renal vasodilator known to increase RBF in humans but does not exert a significant effect on systemic arterial blood pressure and heart rate. ²² Dopamine/saline placebo will be infused IV via infusion pump for approximately 20 minutes at a constant rate of 2 to 4 μg/kg/min (after PET Session 1, and prior to and during PET Session 2 examinations). The short half‐life of dopamine (approximately 2 min) means that it should be sufficiently cleared from circulation before AZD3427 or placebo administration.

A limitation of the present study is that GFR is not being measured using gold‐standard methods (e.g., iohexol clearance), as the primary aim is to assess renal blood flow using PET. eGFR is included only as an exploratory endpoint due to the small sample size and its inherent variability. Additionally, the relatively broad inclusion criteria in terms of eGFR and NYHA class—necessary to facilitate recruitment in this single‐site setting—may have introduced heterogeneity, potentially resulting in differences in the magnitude of the treatment effect between patients with more versus less advanced cardiorenal disease.

Conclusions

This study aims to evaluate the effect of AZD3427 on RBF, focusing on distinguishing between increases due to global changes in RBF and those associated with only focal changes. Making this distinction is crucial for understanding the mechanism behind AZD3427‐induced increases in eGFR and the potential for long‐term clinical benefit. Specifically, global changes may indicate the recruitment of nephrons and renal capillaries, leading to improved filtration by enlarging the surface area of the filtration membrane, rather than through focal increases and hyperfiltration in nephrons.

Funding

This work was supported by AstraZeneca.

Conflict of interest

MU, MQH, KC, DP, ZC, AJ, MS and MC are employees and stockholders of AstraZeneca. LHL's institution receives consulting fees for the design of the present trial. Unrelated to the present manuscript, LHL's institution receives grants, consulting and honoraria from: Alleviant, Amgen, AstraZeneca, Bayer, Biopeutics, Boehringer Ingelheim, Edwards, Novartis, Novo Nordisk, Owkin and Pharmacosmos.

Ufnal, M. , Quintana‐Hayashi, M. P. , Connolly, K. , Pettersen, D. , Cselényi, Z. , Jucaite, A. , Schou, M. , Varrone, A. , Chan, M. , and Lund, L. H. (2025) Re‐PERFUSE: Phase 1b study of AZD3427, a novel relaxin receptor agonist, on renal perfusion in HFrEF patients. ESC Heart Failure, 12: 4495–4502. 10.1002/ehf2.15412.