Volume 181, Issue 3 p. 319-339
INVITED REVIEW
Open Access

Taming resistant hypertension: The promise of novel pharmacologic approaches and renal denervation

Omar Azzam

Omar Azzam

Dobney Hypertension Centre, Medical School—Royal Perth Hospital Unit, Royal Perth Hospital Medical Research Foundation, The University of Western Australia, Perth, Western Australia, Australia

Department of Nephrology, Royal Perth Hospital, Perth, Western Australia, Australia

Contribution: Writing - original draft (lead), Writing - review & editing (equal)

Search for more papers by this author
Sayeh Heidari Nejad

Sayeh Heidari Nejad

Dobney Hypertension Centre, Medical School—Royal Perth Hospital Unit, Royal Perth Hospital Medical Research Foundation, The University of Western Australia, Perth, Western Australia, Australia

Contribution: Writing - review & editing (supporting)

Search for more papers by this author
Revathy Carnagarin

Revathy Carnagarin

Dobney Hypertension Centre, Medical School—Royal Perth Hospital Unit, Royal Perth Hospital Medical Research Foundation, The University of Western Australia, Perth, Western Australia, Australia

Contribution: Writing - review & editing (supporting)

Search for more papers by this author
Janis M. Nolde

Janis M. Nolde

Dobney Hypertension Centre, Medical School—Royal Perth Hospital Unit, Royal Perth Hospital Medical Research Foundation, The University of Western Australia, Perth, Western Australia, Australia

Contribution: Writing - review & editing (supporting)

Search for more papers by this author
Marcio Galindo-Kiuchi

Marcio Galindo-Kiuchi

Dobney Hypertension Centre, Medical School—Royal Perth Hospital Unit, Royal Perth Hospital Medical Research Foundation, The University of Western Australia, Perth, Western Australia, Australia

Contribution: Writing - review & editing (supporting)

Search for more papers by this author
Markus P. Schlaich

Corresponding Author

Markus P. Schlaich

Dobney Hypertension Centre, Medical School—Royal Perth Hospital Unit, Royal Perth Hospital Medical Research Foundation, The University of Western Australia, Perth, Western Australia, Australia

Department of Nephrology, Royal Perth Hospital, Perth, Western Australia, Australia

Department of Cardiology, Royal Perth Hospital, Perth, Western Australia, Australia

Correspondence

Markus P. Schlaich, Dobney Hypertension Centre, Medical School—Royal Perth Hospital Unit, Royal Perth Hospital Medical Research Foundation, The University of Western Australia, Level 3, MRF Building, Rear 50 Murray St, Perth, WA 6000, Australia.

Email: [email protected]

Contribution: Writing - original draft (equal), Writing - review & editing (equal)

Search for more papers by this author
First published: 15 September 2023

Abstract

Resistant hypertension is associated with an exceedingly high cardiovascular risk and there remains an unmet therapeutic need driven by pathophysiologic pathways unaddressed by guideline-recommended therapy. While spironolactone is widely considered as the preferable fourth-line drug, its broad application is limited by its side effect profile, especially off-target steroid receptor-mediated effects and hyperkalaemia in at-risk subpopulations. Recent landmark trials have reported promising safety and efficacy results for a number of novel compounds targeting relevant pathophysiologic pathways that remain unopposed by contemporary drugs. These include the dual endothelin receptor antagonist, aprocitentan, the aldosterone synthase inhibitor, baxdrostat and the nonsteroidal mineralocorticoid receptor antagonist finerenone. Furthermore, the evidence base for consideration of catheter-based renal denervation as a safe and effective adjunct therapeutic approach across the clinical spectrum of hypertension has been further substantiated. This review will summarise the recently published evidence on novel antihypertensive drugs and renal denervation in the context of resistant hypertension.

Abbreviations

  • AOBP
  • automated office BP
  • CKD
  • chronic kidney disease
  • DBP
  • diastolic blood pressure
  • HTN
  • hypertension
  • MRA
  • mineralocorticoid receptor antagonist
  • RDN
  • renal denervation
  • RHTN
  • resistant hypertension
  • uRDN
  • ultrasound renal denervation
  • SBP
  • systolic blood pressure
  • 1 INTRODUCTION

    Hypertension (HTN) is the single biggest contributor to global burden of disease and mortality, with an estimated 10.4 million deaths attributed to hypertension in 2017 alone (Stanaway et al., 2018). Using diagnostic thresholds of systolic blood pressure (SBP) ≥ 140 mmHg and/or diastolic blood pressure (DBP) readings ≥90 mmHg, the estimated global prevalence of hypertension in the adult population is roughly one third (33.5%) and less than a third of those (31.7%) are controlled when applying a blood pressure (BP) threshold of <140/90 mmHg (Beaney et al., 2020; Nolde et al., 2022). Furthermore, hypertension-mediated organ damage confers incremental cardiovascular disease risk at every level of BP (Unger et al., 2020; Vasan et al., 2022; Whelton et al., 2018; Williams et al., 2018). It is therefore imperative that continued assessment, and initiation and escalation of BP lowering strategies be undertaken in patients with both elevated BP (‘prehypertension’) and mild to severe forms of hypertension.

    Resistant hypertension (RHTN) is defined as uncontrolled BP despite the use of ≥3 antihypertensive drugs of different classes, including a diuretic (usually thiazide/thiazide-like), a long-acting calcium channel (Cav1.2) blocker and a blocker of the renin–angiotensin system (RAS) (angiotensin-converting enzyme [ACE] inhibitor [ACEi] or angiotensin II type 1 [AT1] receptor antagonist), at maximal or maximally tolerated doses. The definition also encompasses patients with BP that is controlled on ≥4 antihypertensive medications, referred to as controlled resistant hypertension (Calhoun et al., 2008; Carey et al., 2018). The reported prevalence of resistant hypertension varies considerably, mainly driven by disparate measurement methods and diagnostic thresholds (Achelrod et al., 2015). A meta-analysis of data from 3.2 million hypertension patients estimates the prevalence of ‘true’ resistant hypertension at roughly 10% (Noubiap et al., 2019). Patients with true resistant hypertension have a greater risk of end-stage renal disease, ischaemic heart disease, heart failure, stroke or death compared with patients with controlled BP (Daugherty et al., 2012; Leung et al., 2022; Sim et al., 2015).

    2 GENERAL APPROACH TO TREATMENT-RESISTANT HYPERTENSION

    When embarking on formulating treatment plans for patients with apparent resistant hypertension, it is important to exclude pseudo-resistance and consequently confirm true resistant hypertension (Calhoun & Grassi, 2017). The causes of pseudo-resistance include measurement artefact, white coat effect, medication non-adherence and clinical inertia (Bhatt et al., 2016; de la Sierra et al., 2011; Jung et al., 2013). Doing so enables identification of patients with an exceedingly high-risk phenotype who may benefit from treatment escalation. Overcoming non-adherence and clinical inertia can be enhanced by prescribing fixed-dose, single-pill combination antihypertensive therapy (Chow et al., 2021; Rose et al., 2007). Utilisation of out-of-office measurements including ambulatory BP monitoring (ABPM) or the less resource-intensive home BP monitoring, mitigates white coat effect and measurement inaccuracies that arise from office readings (de la Sierra et al., 2011; O'Brien et al., 2013). In addition to confirming true resistant hypertension, out-of-office measurements can identify a hypertensive subpopulation at heightened cardiovascular disease and mortality risk, namely masked uncontrolled hypertension or in this context masked resistant hypertension (Banegas et al., 2014; Tientcheu et al., 2015).

    The mechanisms underpinning resistant hypertension are several and often overlapping. It is widely accepted that overactivation of the renin–angiotensin–aldosterone system (RAAS) and the sympathetic nervous system predominate BP regulation in this setting (Acelajado et al., 2019). Elevated sympathetic nervous system activity increases systemic vasoconstriction and, via efferent signalling to the nephron, promotes salt and water retention, renin release, in turn activating the RAAS (Oliva & Bakris, 2014). Aldosterone, a mineralocorticoid receptor (MR/NR3C2) agonist released from the zona glomerulosa of the adrenal cortex, promotes sodium and water retention in the distal nephron and, consequently, volume expansion (Egan & Li, 2014). Beyond initiating and sustaining hypertension, sympathetic nervous system overactivation and a state of aldosterone excess exert deleterious cardiovascular effects that are independent of their BP elevating actions, further emphasising the importance of targeted therapeutic interventions ameliorating these pathophysiologic states (Schlaich et al., 2003). Furthermore, the potent vasoconstrictor, endothelin-1 (ET-1), has been implicated in the pathophysiology of salt-sensitive, low-renin hypertension and in particular resistant hypertension (Pollock & Pollock, 2001). This review will navigate the emerging evidence for both add-on pharmacotherapy and catheter-based interventions in patients with resistant hypertension, a population with an unmet therapeutic need.

    3 PHARMACOTHERAPY

    Across major guidelines, there is emphasis on ensuring that patients are established on maximally tolerated doses of first-line drugs, including a diuretic (preferably thiazide-like or, in the setting of advanced stage chronic kidney disease [CKD], a loop diuretic). Presently, the steroidal mineralocorticoid receptor antagonist (MRA), spironolactone, is widely considered the preferred fourth-line antihypertensive drug (Unger et al., 2020; Whelton et al., 2018; Williams et al., 2018). Treatment guidelines recommend addition of other antihypertensive drugs/classes when spironolactone is either contraindicated or not tolerated. These include, but are not limited to, eplerenone, amiloride, α1-adrenoceptor antagonists (e.g. doxazosin), β1-adrenoceptor antagonists (e.g. bisoprolol) and centrally acting drugs (e.g. clonidine and moxonidine), all of which have been shown to further lower BP as add-on therapy in the context of resistant hypertension but have limited evidence for an impact on cardiovascular outcomes (Carey et al., 2018; Sheppard et al., 2017; Unger et al., 2020; Williams et al., 2018).

    3.1 Mineralocorticoid receptor antagonists (MRAs)

    Aldosterone excess is strongly implicated in the pathogenesis of hypertension and in particular resistant hypertension (Judd et al., 2014). In a prospective study of 251 patients with resistant hypertension, daytime, night-time and 24-h SBP and DBP were significantly higher in ‘high aldosterone’ compared with ‘normal aldosterone’ status in both males and females, and the effects of aldosterone on ambulatory BP levels were more pronounced with increasing age (Pimenta et al., 2007). Some observational studies have reported a prevalence of primary aldosteronism of up to 20% in patients with apparent resistant hypertension (Florczak et al., 2013). Importantly, a lesser degree of non-classical aldosteronism is also present in a significant proportion of the resistant hypertension population (Gaddam et al., 2008). This is at least in part driven by diuretic-induced activation of the RAAS in a manner similar to that observed with sodium restriction and by aldosterone ‘escape’ associated with RAAS blockade (Schjoedt et al., 2004; Ubaid-Girioli et al., 2009). Therefore, it makes mechanistic sense to supplement triple first-line antihypertensive drug regimens with drugs that target unopposed aldosterone-mediated BP elevating effects, the most widely endorsed being blockade of MRs. Recognition of the proinflammatory and profibrotic effects of aldosterone and the widespread tissue/cell type distribution of MRs through which its deleterious effects are mediated adds further relevance to MR blockade (Kintscher et al., 2022). Indeed, independent of their BP lowering efficacy, MRAs have been shown to favourably modulate hard clinical endpoints in high-risk populations as consistently demonstrated in landmark trials focused on cardiorenal endpoints in several disease states such as diabetic kidney disease and heart failure (Pitt et al., 19992003; Zannad et al., 2011). Furthermore, there is compelling data demonstrating that MRAs are associated with pleotropic effects including renoprotective/antialbuminuric effects, amelioration of endothelial dysfunction, and regression of left ventricular hypertrophy and fibrosis independent of their BP reducing effects (Savoia et al., 2008; Schneider et al., 2017).

    Meta-analyses of randomised and nonrandomised studies consistently show that steroidal MRAs (spironolactone and eplerenone) produce clinically meaningful reductions in both SBP and DBP in patients with resistant hypertension (Dahal et al., 2015; Zhao et al., 2017). PATHWAY-2, a double-blind, placebo-controlled crossover trial compared the BP lowering effects of spironolactone (25–50 mg) with bisoprolol (5–10 mg), doxazosin (modified release 4–8 mg) and placebo. After 12 weeks, randomisation to spironolactone was associated with an 8.7 mmHg placebo-adjusted reduction in home SBP. Compared with bisoprolol and doxazosin, spironolactone reduced mean home SBP by a further 4.48 and 4.03 mmHg, respectively. Spironolactone's BP lowering superiority was most pronounced in patients with lowest renin levels at baseline, a marker of increased sodium retention, therefore strongly implicating aldosteronism in the pathogenesis of resistant hypertension (Williams et al., 2015). More recently, the ReHOT trial compared spironolactone with twice-daily sympatholytic, centrally acting α2 adrenoceptor agonist, clonidine, in true resistant hypertension patients. While spironolactone achieved comparable rates of target 24-h ambulatory BP monitoring control (20.8% vs. 20.5%), it was associated with greater absolute reductions in both 24-h SBP and DBP. The greater absolute reductions in BP and easier posology (once-daily dosing) favour spironolactone over clonidine as a fourth-line drug in resistant hypertension (Kwon et al., 2021). Eplerenone, a second generation, highly selective steroidal MRA, has limited, but nonetheless clinically significant data supporting its utility in the resistant hypertension setting. In an open-label prospective study, eplerenone 50 mg once- or twice-daily lowered clinic BP by 17.6/7.9 mmHg and 24-h mean BP by −12.2/−6.0 mmHg (Calhoun & White, 2008). If eplerenone is used, it should be dosed at 50–200 mg day−1 to achieve meaningful BP reductions (Tam et al., 2017).

    The main limitation with steroidal MRA prescribing in clinical practice is the risk of a dose-dependent class effect, namely hyperkalaemia, primarily through their MR-mediated potassium-sparing actions at the principal cells of the distal nephron (Good, 2007). Following the findings of the landmark RALES trial, there emerged a spike in spironolactone prescribing for heart failure and a concerning abrupt increase in rates of hyperkalaemia in clinical practice (Juurlink et al., 2004). Furthermore, the prevalent concomitant presence of CKD and/or diabetes mellitus amplifies this risk, especially when MRAs are prescribed on top of RAS inhibition (Currie et al., 2016). In the AMBER study, which evaluated the efficacy of the potassium binder patiromer in enabling persistent use of spironolactone in patients with advanced CKD (eGFR 25–45 ml min−1 per 1.73 m2) and resistant hypertension, overall, approximately two in three patients developed at least mild hyperkalaemia and approximately one in four patients had to discontinue spironolactone. Mild hyperkalaemia and discontinuations were both significant even in the patiromer arm (>30% and 7%, respectively) (Agarwal et al., 2019). Avoidance in clinical practice is therefore common and cautious monitoring advised when prescribing MRAs to patients with baseline serum potassium in the upper normal range (>4.5 mmol L−1) and/or eGFR < 45 ml min−1 per 1.73 m2 (Khosla et al., 2009).

    Spironolactone is a first-generation steroidal MRA with structural similarity to progesterone, an endogenous antagonist of the MR (Kagawa et al., 1957). It is associated with various steroidogenic side effects owing to its lack of receptor selectivity and the similarities among the ligand binding domains of sex steroid hormone receptors. Spironolactone can predominantly exert both androgen receptor antagonist and progesterone receptor agonist effects within its MR blockade therapeutic range (Gomez-Sanchez, 2016). A well-recognised and sometimes limiting side effect in males is gynaecomastia, primarily due to its dose-dependent anti-androgenic effect (Corvol et al., 1975). Off-target effects in females are associated with menstrual irregularities, breast enlargement and tenderness (Shaw & White, 2002). Notably, spironolactone can exhibit androgen receptor partial agonist effects in androgen-depleted states and therefore should be used with caution in prostate cancer patients on androgen synthesis inhibitors such as abiraterone (Sundar & Dickinson, 2012).

    Compared with spironolactone, eplerenone has a 500-fold lower affinity for the androgen and progesterone receptors, and therefore a more favourable side effect profile (Garthwaite & McMahon, 2004; Pitt et al., 2003). Although spironolactone is the most studied MRA in resistant hypertension, and data for head-to-head comparisons with spironolactone in resistant hypertension are lacking, eplerenone presents a viable alternative for patients affected by or wishing to avoid spironolactone-related side effects (Manolis et al., 2019). Notwithstanding the greater selectivity and more favourable tolerability profile, its lower affinity for MRs and shorter half-life necessitates twice-daily and higher dosing to effect clinically meaningful and comparable BP reductions to spironolactone (Fagart et al., 2010).

    After 80 years of remarkable progress in MRA research, the pursuit of an drug in this class with a favourable benefit–risk profile has delivered several potent and highly selective nonsteroidal MRAs of which finerenone has achieved the most advanced stages in development and clinical trials. Finerenone has a higher selectivity and binding affinity for MRs than steroidal MRAs (Bärfacker et al., 2012). Unlike its steroidal counterparts which disproportionately concentrate in the kidney, finerenone has been shown in preclinical models to exhibit a balanced distribution between the heart and kidneys (Kolkhof et al., 2014). Moreover, finerenone has been shown in animal models to confer protection against cardiorenal injury more efficiently than eplerenone at equi-natriuretic doses not associated with BP reduction (Grune et al., 2018; Kolkhof et al., 2014). In the presence of aldosterone, finerenone at equi-natriuretic doses to eplerenone was shown in a mouse model of cardiac fibrosis to be a more potent antagonist of mineralocorticoid-mediated MR coactivator binding and to be more potent at inducing corepressor binding. Uniquely, and in contrast to eplerenone, finerenone exhibited an inverse agonist effect on MR coactivator recruitment/binding in the absence of aldosterone. Consequently, it more efficaciously reduces downstream expression of proinflammatory and profibrotic factors following MR activation (Grune et al., 2018). In the landmark event-driven, placebo-controlled phase 3 trials, FIDELIO-DKD and FIGARO-DKD, randomisation to finerenone was associated with reductions in CKD progression and cardiovascular events in patients with CKD associated with type 2 diabetes (T2D) that were established on RAS blockers (Bakris et al., 2020; Pitt et al., 2021). Consequently, finerenone achieved Food and Drug Administration (FDA) and European Union approval to reduce the risk of sustained eGFR decline, end-stage kidney disease, nonfatal myocardial infarction and hospitalisation for heart failure in this setting (Frampton, 2021).

    A recent post hoc analysis by Agarwal et al. indirectly compared safety and BP lowering efficacy of spironolactone and finerenone in the setting of concomitant resistant hypertension and moderate-to-advanced CKD (eGFR 25–45 ml min−1 per 1.73 m2) across matched populations from the FIDELITY and AMBER trials. Treatment discontinuation due to hyperkalaemia was markedly lower with finerenone (0.3%) compared with spironolactone (7% and 23% in the active and placebo arms of the AMBER trial) (Agarwal et al., 2023). Furthermore, in the overall FIDELITY population, hyperkalaemia leading to permanent treatment discontinuation occurred in just 1.7% of patients receiving finerenone in spite of all patients being established on RAS-blocking drugs and a mean eGFR 57 ml min−1 per 1.73 m2 at baseline (Agarwal, Filippatos, et al., 2022). These findings might be explained by the fact that finerenone has a relatively short half-life and an absence of active metabolites (Agarwal et al., 2021). This contrasts with spironolactone, a prodrug with multiple detectable biologically active urinary metabolites (e.g. canrenone) in humans up to 2–3 weeks after discontinuation (Agarwal et al., 2019). Identified risk factors for hyperkalaemia include higher baseline serum potassium, lower eGFR and increased urine albumin–creatinine ratio at baseline, while co-treatment with sodium/glucose cotransporter 2 (SGLT2) inhibitors was found to be protective (Agarwal, Joseph, et al., 2022). Unlike with traditional MRAs, it appears that significant hyperkalaemia and permanent treatment discontinuation can be effectively mitigated by short treatment interruptions and dose reductions, as was successfully implemented in the FIDELIO-DKD study protocol (Agarwal, Joseph, et al., 2022). Considering its unique molecular pharmacology, finerenone appears capable of imparting clinical endpoint benefits with lesser risk of hyperkalaemia (Pitt et al., 2013).

    In the aforementioned post hoc analysis by Agarwal et al. (2023), finerenone was associated with −7.1 mmHg reduction in unattended office SBP in the resistant hypertension with CKD subpopulation, compared with reductions of approximately −11 mmHg with spironolactone in the matched AMBER study population. Analysis of the overall FIDELIO-DKD population showed that, compared with placebo, finerenone reduced SBP and DBP by −2.71 and −1.03 mmHg, respectively, reductions that are considered clinically modest but potentially explained by modification of background antihypertensive treatment throughout the trial (Ruilope et al., 2022). The most pronounced BP reduction was observed in the highest office SBP quartile of >148.0 mmHg. An analysis adjusting for office SBP changes suggests that the observed cardiorenal endpoint benefits were only in small part attributable to finerenone's BP lowering effect (Ruilope et al., 2022).

    While available data suggest an inferior BP lowering effect with finerenone compared with traditional MRAs, its additive cardioprotective and renoprotective effects and lower risk of hyperkalaemia cannot be dismissed. The totality of evidence thus far warrants direct comparison of safety and efficacy of finerenone with currently recommended steroidal MRAs in the setting of resistant hypertension and concomitant CKD, a cohort with limited treatment options.

    3.2 Aldosterone synthase inhibitors (ASIs)

    To overcome off-target steroidogenic side effects and the counterregulatory renin-dependent increase in aldosterone levels associated with MRAs, targeting synthesis of aldosterone instead of blocking its receptor emerged as a logical alternative approach (Agarwal et al., 2021; MacKenzie et al., 2017). Moreover, reduction in aldosterone synthesis would mitigate the MR-dependent nongenomic effects associated with the counterregulatory increases in aldosterone observed with MR blockade (Funder, 2017; Mihailidou et al., 2019; Pitt et al., 2013). Aldosterone synthase, an enzyme expressed in the zona glomerulosa of the adrenal cortex, catalyses sequential hydroxylations and oxidations in the final three steps of aldosterone synthesis from its precursor, 11-deoxycorticosterone (Bassett et al., 2004) (Figure 1). However, aldosterone synthase (encoded by the CYP11B2 gene) has a 93% sequence similarity to cortisol-producing steroid 11β-hydroxylase (encoded by the CYP110B1 gene), thwarting initial efforts to produce a selective aldosterone synthase inhibitor (MacKenzie et al., 2017; Mornet et al., 1989).

    Details are in the caption following the image
    Aldosterone synthase (encoded by the gene CYP11B2) shares 93% sequence similarity to 11β-hydroxylase (encoded by the gene CYP110B1). Baxdrostat has a high selectivity for aldosterone synthase, effectively blocking synthesis of aldosterone from its precursor, 11-deoxycorticosterone, without affecting cortisol synthesis. Chemical structure of baxdrostat (bottom left).

    This issue became manifest when the first in class, LCI699 (later named osilodrostat), produced non-inferior reductions in office SBP compared with eplerenone in a randomised controlled trial of 524 patients with primary hypertension but was inferior to eplerenone and comparable to placebo in a trial of 155 patients with resistant hypertension (Calhoun et al., 2011; Karns et al., 2013). This might be explained by off-target inhibition of the structurally similar steroid 11β-hydroxylase (CYP11B1), causing an undesired reduction in cortisol synthesis. This results in compensatory enhanced adrenocorticotropic hormone (ACTH) release which in turn stimulates increased production of MR-activating steroid precursor, 11-deoxycorticosterone, therefore offsetting the effects of reduction in aldosterone synthesis (Amar et al., 2010; Oparil & Schmieder, 2015). Initially developed as an aldosterone synthase inhibitor, osilodrostat was repurposed as the first FDA-approved drug for CYP11B1-mediated Cushing’s disease (Duggan, 2020). It became apparent that if an effective aldosterone synthase inhibitor were ever to be developed, this must exhibit extremely high selectivity for inhibition of aldosterone synthase over 11β-hydroxylase (Schumacher et al., 2013). This led to the development of RO6836191/CIN-107, later renamed baxdrostat, a potent second-generation aldosterone synthase inhibitor with more than 100-fold in vitro selectivity for CYP11B2 over CYP11B1 (Bogman et al., 2017) (Figure 1). Recently, examining the interactions of recombinant human CYP11B2 with LCI699 and its fadrozole analogues (S)-fadrozole and (R)-fadrozole using X-ray crystallography provided further valuable clues about the CYP11B2 active site architecture, identifying important differences from cortisol-producing CYP11B1 andproviding valuable insights that guide future design of highly selective aldosterone synthase inhibitors (Brixius-Anderko & Scott, 2021).

    In a preclinical cynomolgus monkey model, baxdrostat inhibited aldosterone synthesis without affecting the adrenocorticotropic hormone-induced rise in cortisol. Concurrently, Bogman et al. performed a first-in-man study in which they, likewise, demonstrated that complete suppression of aldosterone production was achievable with 10 mg of baxdrostat without affecting cortisol synthesis. Increases in the precursors 11-deoxycorticosterone and 11-deoxycortisol only occurred at a ≥90 mg dose, confirming its high selectivity for aldosterone synthase. It was well tolerated at all doses tested (Bogman et al., 2017). In a phase 1, placebo-controlled trial evaluating safety, pharmacokinetics and pharmacodynamics of baxdrostat in 54 healthy volunteers between the ages of 18 and 55 years (70% male), a dose-dependent reduction of plasma aldosterone occurred with baxdrostat at doses ≥1.5 mg, with reductions sustained at 10 days under both low and normal salt diet conditions. Encouragingly, unlike the 10-fold rise observed with osilodrostat, the rise in 11-deoxycorticosterone was modest (approximately twofold to threefold) and baxdrostat treatment was not associated with any meaningful reduction in cortisol concentrations, confirming its high selectivity for aldosterone synthase. Pharmacokinetic studies showed a mean half-life of 29 h, supporting once-daily dosing (Freeman, Bond, et al., 2023).

    Most recently, the phase 2 BrigHTN trial assessed the safety and efficacy of baxdrostat in the setting of resistant hypertension (Freeman et al., 2023). This was a double-blind, placebo-controlled, parallel-design, dose-ranging study that randomised 275 eligible men and women aged 18 or over with resistant hypertension to baxdrostat 0.5, 1 or 2 mg daily or placebo for 12 weeks. More than 90% of participants were established on an ACEi or AT1 antagonist (angiotensin receptor blocker [ARB]) at baseline. The trial was stopped early by an independent data review committee due to clearcut efficacy determined at a prespecified interim analysis. Dose-dependent reductions in both plasma and urine aldosterone and a compensatory rise in plasma renin activity were observed, confirming effective inhibition of aldosterone synthase. There was a reduction in mean seated SBP of −20.3, −17.5 and −12.1 mmHg with 2, 1 and 0.5 mg baxdrostat respectively, compared with 9.4 mmHg with placebo, resulting in statistically significant placebo-adjusted decreases of −11.0 and −8.1 mmHg for the 2 and 1 mg dose, respectively. Additionally, the baxdrostat 2 mg daily dose achieved the secondary endpoint with a placebo-adjusted reduction in seated DBP of −5.2 mmHg. Approximately 46% of patients in the 2 mg dose arm achieved BP control (SBP < 130 mmHg).

    No serious adverse events were attributed to baxdrostat and there was no observation of adrenal insufficiency. In total six cases of hyperkalaemia occurred and an elevation to ≥6 mmol L−1 occurred in only three patients. This did not recur after withdrawal and re-initiation of the drug and none of the participants discontinued the trial because of hyperkalaemia. It must be noted however that the risk of severe hyperkalaemia was significantly mitigated by absolute exclusion of patients with eGFR <45 ml min−1 per 1.73 m2; the vast majority of participants had eGFR ≥ 60 ml min−1 per 1.73 m2 at baseline. In summary, baxdrostat treatment in patients with resistant hypertension was associated with effective aldosterone suppression, significant dose-dependent reductions in SBP and was overall well tolerated (Freeman et al., 2023).

    The findings from BrigHTN add to the existing body of evidence that a state of aldosterone excess contributes to the pathogenesis of resistant hypertension (Azizi, 2023). Trials that directly compare safety and efficacy of baxdrostat with widely prescribed antihypertensives in resistant hypertension, including MRAs, and inclusion of patients with more advanced stages of CKD are warranted. Relevant phase 3 trials are now underway to substantiate these findings and facilitate the regulatory process for registration in relevant indications once available.

    3.3 Endothelin (ET) receptor antagonists

    Since its discovery in the late 1980s, ET, a potent vasoconstrictor, has attracted much interest as a potential therapeutic target in the treatment of hypertension (Yanagisawa et al., 1988). Its most potent isoform, ET-1, through stimulation of its two main receptors (ETA and ETB), results in different and often opposing effects which under physiological settings regulate vascular tone and BP, but dysregulation of which promotes endothelial dysfunction, cardiorenal injury and systemic hypertension (Feldstein & Romero, 2007). ETA receptor binding and activation promotes vasoconstriction, salt and water retention, inflammation and fibrosis, whereas ETB receptor activation stimulates production of vasodilator substances, such as nitric oxide (NO), promotes sodium excretion and inhibits vascular inflammation and fibrosis (Kohan, 2010). Furthermore, preclinical and clinical evidence shows that ET-1 is a mediator of aldosterone release and ET receptor blockade is associated with reduction in plasma aldosterone levels (Mazzocchi et al., 1996; Rossi et al., 2003). It is well established that ETA receptor blockade lowers BP and confers renoprotection across several progressive renal disease settings (Dhaun et al., 2011; Gagliardini et al., 2011). Nonetheless, the clinical application of ET receptor antagonists has thus far been limited to pulmonary hypertension, where there is unequivocal benefit attributed to vasodilatory and anti-hyperplastic effect in the pulmonary vascular bed (Kuntz et al., 2016).

    The selective ETA antagonist, darusentan, was investigated in the setting of resistant hypertension, yielding mixed efficacy results, mainly influenced by BP measurement methods employed (clinic readings vs. ambulatory BP monitoring). Using ambulatory BP derived estimates, darusentan was associated with a placebo-adjusted reduction in SBP of ~7 mmHg, and a ~5 mmHg more pronounced reduction achieved with the central α2 agonist guanfacine (Bakris et al., 2010; Weber et al., 2009). However, this benefit was offset by fluid retention/oedema reported in up to 28% of participants randomised to darusentan versus ~12% in each of the placebo and guanfacine treatment arms (Bakris et al., 2010). Another ETA-predominant receptor antagonist, avosentan, had been shown to result in significant reductions in albuminuria in diabetic patients (Wenzel et al., 2009). Therefore, its potential renoprotective benefit in the setting of overt diabetic nephropathy was tested in the placebo-controlled trial, ASCEND. Unfortunately, an unacceptable rate of fluid retention and congestive heart failure events in trial participants necessitated premature termination of the study (Mann et al., 2010). Likewise, a trend towards fluid retention was observed in the placebo-controlled SONAR trial of yet another ETA-selective receptor antagonist, atrasentan (Heerspink et al., 2019). Despite employing an ‘enrichment phase’—to enhance detection of treatment benefit while minimising the risk of heart failure—there remained a significant residual signal for fluid retention which offsets the 35% risk reduction in the primary composite renal endpoint (doubling of serum creatinine or end-stage kidney disease) (Heerspink et al., 2019). It therefore became apparent that fluid retention represents the most relevant side effect associated with ETA-selective antagonists.

    In a preclinical rat model selective ETA antagonism, through overactivation of unblocked ETB by endogenous ET-1, was associated with increased plasma arginine vasopressin and aldosterone concentrations, reduced water excretion and increased vascular leakage (Sato et al., 1995). Furthermore, preclinical data suggest that, at high doses, predominant ETA receptor antagonist avosentan may promote sodium and water retention through off-target blockade of renal ETB receptors in proximal renal tubules (Baltatu et al., 2012). A meta-analysis of 4894 patients from 24 randomised placebo-controlled trials reported that the occurrence and severity of fluid retention was influenced by ET receptor selectivity, and by the presence of or absence of disease states that favour the development of fluid retention, namely CKD and heart failure (Wei et al., 2016). These preclinical and clinical findings encouraged research into dual ETA/ETB receptor blockade as a potential alternative to selective ETA receptor blockade.

    Aprocitentan, an active metabolite of macitentan, is a dual ETA/ETB receptor antagonist that potently inhibits binding of ET-1 to ETA and ETB with an inhibitory potency ratio of 1:16 (i.e. stronger inhibitory effect on ETA than on ETB receptors) (Iglarz et al., 2008). In two rat models of hypertension, spontaneously hypertensive rats and in the deoxycorticosterone acetate (DOCA)-salt model of salt-dependent hypertension, aprocitentan imparted synergistic BP reduction effects when combined with enalapril or valsartan, unlike spironolactone which only had additive effects (Trensz et al., 2019). In contrast to combination with spironolactone, combination of enalapril with aprocitentan did not increase plasma urea or creatinine concentrations in hypertensive animals (Trensz et al., 2019). In the first-in-human study assessing its tolerability, safety, pharmacokinetics and pharmacodynamics, aprocitentan administered up to 600 mg as a single dose and 100 mg once daily (multiple doses) was well tolerated in healthy male and female adults and in healthy elderly subjects. The increase in body weight was in general modest, most pronounced in subjects receiving 100 mg daily. The pharmacokinetic profile was dose proportional and its half-life ~44 h, supporting once-daily dosing. Only minor differences in exposure were observed between males and females or adult and elderly subjects (Sidharta et al., 2019). A 9-day placebo-controlled crossover study assessed its effects on sodium and water homeostasis in 28 healthy normotensive male subjects on high sodium diet that were randomised to either placebo or aprocitentan 10, 25 or 50 mg daily. Aprocitentan was associated with moderate weight gain (<1 kg) over the 9-day period but without oedema (Gueneau de Mussy et al., 2021). Most notably, there was no evidence of a dose-dependent sodium retention with aprocitentan, contrasting with the significant dose-dependent sodium retention observed in a placebo-controlled study of avosentan in 23 healthy male subjects also on a high-salt diet (Gueneau de Mussy et al., 2021; Smolander et al., 2009). Plasma volume expansion in both aprocitentan studies was estimated at ~5% and was dose-independent.

    In a dose-finding, phase 2, placebo- and comparator-controlled study of 490 patients (61% male; mean age 55 years) with essential hypertension, aprocitentan at doses of 10 and 25 mg daily achieved a favourable risk–benefit profile, effecting meaningful placebo-corrected reductions in unattended automated office BP (AOBP) of -7.05/4.93 and 9.90/6.99 mmHg, respectively, with low rates of fluid retention. The reductions were superior to comparator lisinopril 20 mg daily (-4.84/3.81 mmHg) (Verweij et al., 2020). Compared with 10 mg and 25 mg daily, the 50 mg daily dosing was not associated with additional BP reductions but more pronounced reductions in Hb and corresponding increases in plasma volume. Based on these findings, aprocitentan doses of 12.5 and 25 mg were selected for further clinical development.

    The findings from preclinical and early phase human studies forecast promising outcomes from the pivotal randomised parallel-group phase 3 study, PRECISION, which adopted a novel three-part design to assess the short-term BP lowering effects of aprocitentan in patients with resistant hypertension confirmed by unattended automated sitting office BP, and whether these effects are sustained (Danaietash et al., 2022). Following 4 weeks on single-pill standardised triple drug combination therapy and a 4-week placebo run-in period (to exclude pseudo-resistant hypertension and placebo responders), 730 participants (~60% men) were randomised to 4 weeks of aprocitentan 12.5 mg, 25 mg or placebo daily (Part 1); Part 2 was a 32-week single (patient)-blind part, in which all patients received aprocitentan 25 mg; and Part 3 was a 12-week double-blind, randomised and placebo-controlled withdrawal part, in which patients were re-randomised to aprocitentan 25 mg or placebo in a 1:1 ratio (Schlaich et al., 2022). At 4 weeks, aprocitentan 10 and 25 mg daily were associated with statistically significant placebo-corrected reductions in unattended office SBP of −3.8 and −3.7 mmHg, respectively. Its superiority was confirmed by 24-h ambulatory BP monitoring, achieving a placebo-corrected reduction in 24-h SBP of −5.9 mmHg with the 25 mg daily dose (Figure 2A). Of note, treatment with aprocitentan was associated with significant reductions in proteinuria (28% and 31% for the 12.5 m and 25 mg doses, respectively) which were sustained and more pronounced in participants with CKD stages 3 and 4. BP reductions were consistent throughout the 24 h of ambulatory BP monitoring, and importantly, there was a pronounced reduction in night-time SBP, a strong predictor of cardiovascular outcomes (Dolan et al., 2005; Kario et al., 2020) (Figure 2B). Four weeks into the double-blind withdrawal phase (Part 3), there was a significant 5.8 mmHg SBP increase with placebo compared with aprocitentan (secondary endpoint), and this was maintained out to Week 48 (Schlaich et al., 2022) (Figure 2C).

    Details are in the caption following the image
    (a) Aprocitentan 10 and 25 mg daily was associated with greater reductions in unattended office systolic blood pressure (SBP) than placebo, and the reductions were sustained versus placebo at the conclusion of a 4-week withdrawal phase. (b) The blood pressure (BP)-lowering effect was sustained throughout the circadian cycle of BP in both the double-blind and (c) withdrawal double-blind phase when assessed by means of ambulatory blood pressure monitoring. Reproduced with permission from Schlaich et al. (2022).

    Haemoglobin concentrations decreased and estimated plasma volume increased by ~ 10% to 11% with both doses of aprocitentan compared with placebo, an effect that stabilised through Part 2 and reversed upon withdrawal in Part 3. There was a dose-dependent reporting of oedema or fluid retention during Part 1: 9.1%, 18.4% and 2.1% for patients receiving aprocitentan 12.5 mg, 25 mg and placebo, respectively. The rate of oedema or fluid retention in Part 2 (all participants on 25 mg daily) was identical to the aprocitentan 25 mg daily arm in Part 1. This was generally mild to moderate and managed with the addition of a diuretic (Schlaich et al., 2022). Eleven patients experienced hospitalisation for heart failure, 10 of which were on aprocitentan, at least half of which had a history of heart failure with or without CKD at baseline. No signs of hepatotoxicity were observed, consistent with the safety demonstrated in phase 1 single-dose study enrolling patients with moderate hepatic impairment (Fontes et al., 2022).

    It is conceivable that the more favourable effects on salt and water homeostasis of the more balanced ETA and ETB receptor antagonism, as demonstrated in a rat model comparing representative molecules, may at least partly explain the observation of less pronounced fluid retention in aprocitentan treated patients than that which was historically observed with ETA-selective antagonists such avosentan (Vercauteren et al., 2017). The fact that all patients enrolled in PRECISION were on a hydrochlorothiazide (albeit less effective in those with eGFR 15–45 ml min−1 per 1.73 m2) may have also somewhat contributed to a lower rate of fluid retention. It is conceivable that the use of more potent diuretic therapy, particularly in participants with known heart failure or CKD Stages 3 and 4 at baseline, would have resulted in even fewer fluid retention events.

    While fluid retention remains a potential side effect of dual ETA/ETB receptor blockade, this can generally be mitigated by close volume status monitoring (particularly in the first 4 weeks after initiating the drug), initiating treatment with the 12.5 mg daily dose, judicious patient selection and co-prescribing or dose escalation of diuretics (e.g. chlorthalidone, indapamide or furosemide). The current increased prescribing trends of SGLT2 inhibitors for patients with heart failure and CKD may confer additional protection against fluid retention. The findings of this study offer a novel treatment option that targets a pathophysiologic pathway unopposed by first-line antihypertensive s, and a potential alternative to guideline-recommended spironolactone, especially in patients with proteinuric CKD Stages 3 and 4 where hyperkalaemia poses a significant prescribing limitation in day-to-day practice (Clozel, 2022). Future studies directly comparing aprocitentan with spironolactone and long-term evaluation of effects on target-organ damage and cardiorenal endpoints are warranted. Aprocitentan is currently undergoing FDA review for the indication of resistant hypertension.

    3.4 Aminopeptidase A (APA) inhibitors

    Brain RAS hyperactivity has long been implicated in the development and maintenance of hypertension in experimental animal models (Fournie-Zaluski et al., 2004). In the brain, aminopeptidase A cleaves angiotensin II, converting it to angiotensin III, a potent brain effector peptide that promotes an increase in vasopressin release and tonic stimulatory control over BP (Wright et al., 2012). In rats, it was shown that intracerebroventricular administration of a selective aminopeptidase A inhibitor, (3S)-3-amino-4-sulfanyl-butane-1-sulfonic acid (EC33), blocked the pressor response to exogenous angiotensin II (Reaux et al., 1999). As EC33 does not cross the blood–brain barrier, a dose-dependent reduction in BP was not observed with intravenous administration, supporting the concept that the blockade of the formation of brain, but not systemic, Ang III is responsible for the decrease in BP (Reaux et al., 1999). This leads to the development of orally active prodrug of EC33, RB150, which crosses the blood–brain barrier and was shown in a hypertensive DOCA-salt rat model to inhibit brain aminopeptidase A activity, decrease plasma vasopressin levels, increase diuresis, induce a mild natriuresis and significant dose-dependent reduction in BP (Bodineau et al., 2008). Therefore, aminopeptidase A inhibition presented a putative therapeutic target in human hypertension (Wright et al., 2012).

    Single oral administration of QGC001 up to 1250 mg in 56 healthy male volunteers did not affect BP and was well tolerated (Balavoine et al., 2014). Renamed firibastat, it was further evaluated in an uncontrolled, open-label, dose-titrating study of 256 ethnically diverse, overweight, either treatment-naïve or treated hypertensive patients, with systolic and diastolic AOBP between 145 and 170 mmHg and below 105 mmHg, respectively. Firibastat lowered systolic AOBP by 9.5 mmHg and diastolic AOBP by 4.2 mmHg (Ferdinand et al., 2019). In a phase 2A proof-of-concept study that enrolled 34 patients (74% male) with uncontrolled hypertension on zero to two medications, randomisation to firibastat treatment was associated with a placebo-corrected reduction in daytime ambulatory SBP of −2.7 mmHg, but the difference between the groups was not statistically significant (Azizi, Courand, et al., 2019). Finally, the results of FRESH, a phase 3 double-blind, placebo-controlled, multicentre, efficacy and safety study of firibastat 500 mg administered twice daily over 12 weeks to evaluate its effect on unattended office BP in patients with uncontrolled hypertension in spite of two drugs (difficult-to-treat hypertension) or three drugs (resistant hypertension) was presented at the American Heart Association (AHA)scientific sessions. There was no significant reduction of BP with firibastat 500 mg twice daily compared with placebo (adjusted difference of +0.03 mmHg P = 0.98). Allergic skin reactions were observed in 5.1% of participants in the firibastat arm (American Heart Association, 2022). In light of the results of FRESH, another phase 3 study of firibastat, REFRESH, was stopped, and development in of firibastat in the cardiovascular arena has now ceased.

    4 INTERVENTIONAL APPROACHES: RENAL DENERVATION (RDN)

    The notion that the sympathetic nervous system plays a role in BP regulation first emerged in the early-mid 20th century with the publications of human experiences with thoracolumbar splanchnicectomy, then endorsed as standard treatment for severe hypertension (Longland & Gibb, 1954; Smithwick & Thompson, 1953). Evidence of sympathetic nervous system overactivation in initiation and maintenance of hypertension, often referred to as neurogenic hypertension, has been consistently documented by various methods including measurement of circulating noradrenaline levels, recording of sympathetic outflow utilising microneurography and measurement of noradrenaline release from sympathetic nerve terminals using isotope dilution methodology (Esler et al., 1984; Schlaich et al., 2004). Accumulated evidence from various experimental animal models points to the renal sympathetic nerves, which lie within and adjacent to the renal arteries, as key players in the pathophysiology of neurogenic hypertension (DiBona, 2002, 2005). Renal sympathetic efferent nerves terminate in the juxtaglomerular apparatus, blood vessels and renal tubules. Stimulation of these nerve fibres promotes renin release (up-regulating RAAS), sodium retention and vasoconstriction of the renal vasculature (Johns et al., 2011). Various forms of renal injury activate afferent sensory renal nerve signalling to sympathetic premotor nuclei in the central nervous system, in turn increasing sympathetic outflow to organs and structures involved in BP regulation, including the heart, kidneys and blood vessels (Sata et al., 2018).

    Characterisation of the anatomy and fundamental role of the renal nerve structures in human hypertension sparked an interest in the concept of ablating the renal nerves, referred to as RDN. This therapeutic approach would, through disruption of kidney–brain afferent and efferent signalling pathways, dampen sympathetic nervous system activation and consequently lower BP. Early work in this space showed impressive BP lowering effects accompanied by reductions in surrogate markers of sympathetic nervous system overactivation (Hering et al., 2013; Schlaich, Sobotka, Krum, Lambert, & Esler, 2009). If definitively found to be successful, it would present the clinician and patient with a therapeutic option that addresses the shortcomings of pharmacotherapy in the setting of hypertension in general and in difficult-to-treat or resistant hypertension in particular. (Schlaich, 2017; Schlaich, Sobotka, Krum, Whitbourn, et al., 2009).

    4.1 First-generation RDN trials: The Symplicity trial programme

    Krum et al. reported the first proof-of-concept case series (SYMPLICITY HTN-1) of 50 patients with resistant hypertension treated with radiofrequency RDN at five Australian and European centres. Bilateral multiple (Unger et al., 2020; Whelton et al., 2018; Williams et al., 2018) radiofrequency ablations of ≤8 W for up to 2 min each were delivered using the mono-electrode Symplicity Flex catheter (initially developed by Ardian, Inc., later acquired by Medtronic, Santa Rosa, CA, USA). After each delivery, the catheter is withdrawn and rotated in an attempt to achieve circumferential ablation of the sympathetic plexus. Significant reductions in office SBP and DBP were evident at 1 month and persisted out to 12 months. Data from a subgroup of patients showed a mean reduction in renal noradrenaline spillover from baseline of 47% (Krum et al., 2009). Twenty-four-month follow-up data for 18 patients showed a mean BP reduction of −32/14 mmHg (Krum et al., 2011). At 36 months, data for 88 participants showed a sustained and significant mean reduction in office BP of −32/14.4 mmHg and a drop of ≥10 mmHg SBP was observed in 93% (Krum et al., 2014). A major limitation was absence of a control group, raising the possibility of significant confounding by regression to the mean, placebo and Hawthorne effects. SYMPLICITY HTN-2 randomised resistant hypertension patients to RDN in addition to continued pharmacological treatment versus pharmacological treatment alone. Office-based BP measurements in the RDN group reduced by −32/12 mmHg from a baseline of 178/96 mmHg, while rather unusually, no significant reduction was observed in the control group (Esler et al., 2010). These findings generated considerable enthusiasm in the scientific community, such that an expert consensus statement published in 2013 considered transluminal RDN a therapeutic strategy for resistant hypertension (Schlaich et al., 2013).

    SYMPLICITY HTN-3, the first and largest-to-date sham-controlled trial randomised 535 patients with resistant hypertension in a 2:1 ratio to RDN or a sham procedure. The mean change in office SBP at 6 months was −14.13 mmHg in the RDN group compared with −11.74 mmHg in the sham-procedure group, for a nonsignificant between-group difference of −2.39 mmHg. A similar lack of superiority was observed for the secondary endpoint of 24-h ABP at 6 months (Bhatt et al., 2014). Similarly, smaller sham-controlled trials employing the same mono-electrode Symplicity Flex catheter found no evidence of significant BP reductions in patients with resistant hypertension (Desch et al., 2015; Mathiassen et al., 2016). The neutral findings of SYMPLICITY HTN-3 startled the RDN and wider hypertension community and had major widespread implications for clinical endorsement and application of RDN in the setting of hypertension such that The Joint UK Societies recommended a moratorium on RDN until the SYMPLICITY HTN-3 outcomes were appropriately scrutinised. An introspective endeavour to unravel the reasons behind the incongruous results between SYMPLICITY HTN trials ensued. Post hoc analyses showed that study design, patient and operator-related factors are combined to effectively nullify any potential clinically significant sham-corrected BP reductions in SYMPLICITY HTN-3. Perhaps the most influential were limited operator experience, suboptimal renal sympathetic nerve ablation, suboptimal patient selection, frequency of medication changes during the trial and likely medication non-adherence (Kandzari et al., 2015).

    In contrast with the SYMPLICITY HTN-3 study, DENERHTN, utilised a more clinically oriented, prospective, open-label randomised controlled trial with blinded endpoint evaluation study design (PROBE). It compared the ambulatory BP lowering efficacy and safety of single-electrode Symplicity RDN added to standardised stepped-care antihypertensive treatment (SSAHT) with SSAHT alone in patients with resistant hypertension. It showed statistically significant placebo-adjusted reductions in mean ambulatory daytime SBP (primary endpoint) of −5.9 mmHg in favour of RDN (Azizi et al., 2015). Importantly, a significant, clinically meaningful reduction in night-time SBP was also observed, an effect strongly associated with reduction in cardiovascular morbidity (Staessen et al., 2001). The percentage of responders, defined as an 24-h BP reduction of >20 mmHg, in the RDN group was more than twice that in controls (44.5% vs. 20.8%) (Gosse et al., 2017). While a major limitation was absence of sham control, unlike SYMPLICITY HTN-3, DENERHTN study strictly controlled for confounding by antihypertensive treatment changes and differential treatment adherence, albeit without confirmatory measurement of plasma and urine drug levels (Azizi et al., 2015). The positive findings from this study reignited interest in RDN as a viable adjunct to pharmacotherapy in the setting of resistant hypertension.

    4.2 Second-generation RDN trials

    In the wake of the disheartening SYMPLICITY HTN-3 trial findings, the SPYRAL HTN trials were designed to overcome identified methodological weaknesses. SPYRAL OFF-MED and SPYRAL ON-MED were two 80-patient randomised, sham-controlled studies that would assess the safety and efficacy of the newly developed multi-electrode radiofrequency RDN delivery system (Medtronic, Galway, Ireland) in both drug-naïve and medication-treated patients. The Symplicity Spyral multi-electrode catheter enables delivery of circumferential radiofrequency ablation treatments in a spiral pattern in the four quadrants of the renal artery and branch vessels between 3 and 8 mm in diameter. This is particularly relevant considering that the sympathetic nerve fibres originate from the abdominal ganglia and run conically towards the distal part of the vessel, and are therefore more greatly affected by RDN assuming a constant depth of penetration (Lauder et al., 2019). This contrasts with the approach in the SYMPLICITY trials, wherein radiofrequency RDN was delivered by single-electrode catheters and was restricted to the main renal artery.

    In SPYRAL HTN-OFF MED, patients with an office SBP of 150–180 mmHg, office DBP ≥ 90 mmHg, mean 24-h ambulatory SBP of 140–170 mmHg and suitable renal artery anatomy were randomly assigned to RDN or sham control. Overall compliance with the requirement to be off antihypertensive medications at baseline and 3 months was approximately 85%, confirmed by urine and plasma analyses. Although not powered for efficacy endpoint, randomisation to RDN was associated with significant between-group BP reductions at 3 months: 24-h SBP −5.0 mmHg, DBP −4.4 mmHg, office SBP −7.7 mmHg, office SBP −10.0 mmHg and office DBP −4.9 mmHg. The positive results were consistent across unadjusted and baseline-adjusted analyses and in the per-protocol analyses when the small number of nonadherent patients was excluded. This study effectively removed confounding by medication adherence and, crucially, proceduralist experience was greater and more uniform than in SYMPLICITY HTN-3 (Townsend et al., 2017). The larger SPYRAL HTN-OFF MED Pivotal study (n = 331) utilised the novel adaptive Bayesian design, leveraging data from the SPYRAL HTN-OFF MED pilot study to both increase study power and decrease the overall number of subjects required for randomisation. At 3 months, RDN versus sham was associated with between-group reductions in 24-h SBP and office SBP of −3.9 and −6.5 mmHg at 3 months, respectively (Figure 3). To minimise procedural variability, the number of proceduralists was restricted to one per trial centre (Böhm et al., 2020).

    Details are in the caption following the image
    The renal denervation (RDN) procedure involves an endovascular catheter-based approach to disrupt renal sympathetic nerves using radiofrequency or ultrasound ablation. Top panel: renal nerve anatomy with the RDN catheter advanced into the main renal artery, Symplicity Spyral multi-electrode catheter (Medtronic, Inc. ©) and Paradise ultrasound catheter (ReCor Medical ©). Bottom panel: between-group difference (RDN vs. sham procedure) of change in 24-h systolic BP in second-generation sham-controlled RDN trials for both multi-electrode radiofrequency (left) and ultrasound (right) ablation systems. Data are mean and 95% confidence intervals (CI). BP, blood pressure; RF, radiofrequency; uRDN, ultrasound renal denervation.

    SPYRAL HTN-ON MED assessed the safety and efficacy of the same multi-electrode RDN system in patients with uncontrolled hypertension who were on one, two or three standard antihypertensive drugs. At 6 months, the change in BP was significantly greater in the RDN group than in the sham-control group for all of office SBP (difference −6.8 mmHg), 24-h SBP (difference −7.4 mmHg), office DBP (difference −3.5 mmHg) and 24-h DBP (difference −4.1 mmHg) (Figure 3). Medication adherence, assessed by means of urine and blood analysis, was ≈60% and similar between groups at all timepoints. Importantly, BP reduction was observed throughout 24 h for the RDN group (Kandzari et al., 2018). Initial findings from the SPYRAL HTN-ON MED Expansion study, also utilising a Bayesian design and considering findings from the SPYRAL ON-MED pilot study (NCT02439775), have been presented at AHA 2022 but are not yet published. Patients from 56 clinical centres worldwide were included in this prospective, randomised, sham-controlled, patient- and assessor-blinded trial. Eligible patients were prescribed one to three antihypertensive medications. Patients were randomised to radiofrequency RDN or sham-control procedure. The primary efficacy endpoint was the baseline-adjusted change in mean 24-h ambulatory SBP at 6 months between groups using a Bayesian trial design and analysis. Drug testing assessed medication adherence. Patients were randomly assigned to undergo RDN (n = 206) or the sham-control procedure (n = 131). While there was no significant difference between groups in the primary efficacy analysis, there was a significant increase in medication intensity among sham-control patients in the trial. RDN was associated with a significant reduction in office SBP compared with sham control at 6 months (adjusted treatment difference: −4.9 mmHg, P = 0.0015) and the win ratio analysis favoured RDN. Adverse safety event rates were rare with one event in 253 assessed patients. Complete analysis and assessment of adherence will be relevant to better understand these findings, as will be review of outcomes after longer term follow-up.

    4.3 Ultrasound-based RDN

    A further technique that had shown efficacy in animal studies and is undergoing ongoing evaluation by randomised clinical trials involves RDN by delivering ultrasound energy. The Paradise endovascular ultrasound system (ReCor Medical, Palo Alto, CA, USA) is designed to deliver a minimum of two sonications of 7 s each, separated longitudinally by 5 mm in the main renal arteries (Versaci et al., 2020).

    RADIANCE-HTN SOLO was a randomised, sham-controlled trial in which drug-naïve patients with combined systolic–diastolic hypertension were randomised 1:1 to endovascular ultrasound RDN (uRDN) or a sham procedure. At 2 months, the baseline-adjusted between-group difference in daytime ambulatory SBP was −6.3 mmHg in favour of uRDN (Figure 3). Likewise, the per-protocol analysis that excluded patients who received antihypertensive medication after randomisation showed a between-group difference of −8.2 mmHg in favour of uRDN. The uRDN group were also significantly more likely to achieve BP control (22% vs. 3%) (Azizi et al., 2018). The RADIANCE-HTN TRIO sham-controlled trial examined the safety and efficacy of endovascular uRDN in patients with confirmed resistant hypertension. At 2 months from randomisation, the adjusted median between-group difference in daytime ambulatory SBP was −4.5 mmHg in favour of uRDN. Importantly, patient adherence to fixed-dose, single-pill combination antihypertensive pharmacotherapy, ascertained by means of urine mass spectrometry, was high and equivalent between both groups (82%) (Azizi et al., 2021). This was the first sham-controlled trial to demonstrate short-term efficacy of uRDN in the setting of resistant hypertension. Patients who had persistent elevation in home BP (≥135/85 mmHg) at 2 months were enrolled in an outcome assessor and patient-blinded stepped-care antihypertensive study phase. At 6 months, there were similar BP reductions in both groups, with fewer additional medications required in the uRDN group (Azizi et al., 2022).

    4.4 Long-term durability of RDN

    Management of hypertension is a marathon and requires good long-term BP control in order to reduce overall cardiovascular risk (CV) risk effectively. While short-term BP lowering effects of RDN have been demonstrated, the question as to whether ablated nerves could regenerate or if counterregulatory mechanisms could emerge and consequently whether the effects are sustained beyond early follow-up required clarification. Durability of the BP lowering efficacy of multi-electrode RDN in hypertension patients established on antihypertensive drugs is supported by 36-month follow-up data from SPYRAL HTN-ON MED pilot study which showed significant and sustained sham-adjusted reductions in mean ambulatory SBP (−10.0 mmHg) and DBP (−5.9 mmHg) and, importantly, in night-time SBP (−11.8 mmHg) (Mahfoud et al., 2022). Similarly, 6- and 12-month follow-up data from RADIANCE-HTN SOLO showed that the BP lowering efficacy of uRDN was sustained and, on average, patients required less antihypertensive medications compared with the sham-control group (Azizi et al., 2020; Azizi, Schmieder, et al., 2019). Durability of BP lowering effects was demonstrated at 36-month follow-up of the uRDN arm (Rader et al., 2022). Furthermore, an analysis of data for 1742 patients from the prospective, open-label, Global SYMPLICITY Registry, showed that at 3-year follow-up RDN with the Symplicity mono-electrode Flex catheter was associated with 24-h ambulatory SBP reduction of −8.0 mmHg from baseline (Mahfoud et al., 2019). Most recently, follow-up data on 66 participants across various RDN trials conducted between 2009 and 2014 showed sustained ambulatory SBP and DBP lowering effects out to ≈9 years on less medication and without any signals for adverse renal consequences (Sesa-Ashton et al., 2023).

    Investigating long-term effects of RDN and its ability to maintain lower BP levels will be critical to ultimately appraise the clinical utility of the procedure. Beyond BP lowering, observational data suggest that RDN is associated with an improvement in indices of hypertension-mediated organ damage in patients with resistant hypertension (Kordalis et al., 2018). The impact of RDN-derived BP reductions on cardiovascular endpoints in resistant hypertension has not been directly measured in prospective studies. Notwithstanding, model-based projections suggest that radiofrequency RDN might reduce the relative risk of stroke by 43% and absolute risk of major adverse cardiovascular events from 11.7% in the control group to 8.6% in the RDN group at 3-year follow-up (Schmieder et al., 2023).

    4.5 Safety

    There is an abundance of short- and long-term data supporting the safety of RDN. A theoretical concern associated with catheter-based RDN is new onset renal artery stenosis secondary to vascular injury (Templin et al., 2013). However, systematically undertaken non-invasive renal artery imaging up to 1 year in randomised controlled trials has been reassuring (Azizi et al., 2021; Bhatt et al., 2014; Böhm et al., 2020; Kandzari et al., 2018). Moreover, a large meta-analysis of studies including 5769 patients undergoing radiofrequency RDN estimated an annual incidence of renal artery stenting of 0.2% (Townsend et al., 2020). This is not much different from the incidence of naturally occurring renal artery stenosis in the setting of arterial hypertension (Chrysochou & Kalra, 2009).

    A similar lack of concerning safety signal exists for the risk of both short- or long-term decline in kidney function. A meta-analysis of studies including 2381 patients showed no significant reduction in eGFR at 9.1 months follow-up (Sanders et al., 2017). Data for the largest available cohort of hypertensive patients receiving RDN in a real-world clinical setting (Global SYMPLICITY Registry) showed that the decline in eGFR out to 3 years was within expectation of time-dependent declines observed in the setting of severe hypertension (Mahfoud et al., 2019). Overall, based on a wealth of surveillance data, there is no significant safety signals beyond what would be expected from a transfemoral arterial access procedure, which include access-site vascular complications (haematoma, pseudoaneurysm, fistula and bleeding) and contrast-induced nephropathy, both of which are mitigated if the procedure is performed by highly skilled proceduralists, and with use of adequate periprocedural hydration and minimal or diluted contrast volumes (Bax et al., 2009).

    5 CONCLUSIONS AND FUTURE CONSIDERATIONS

    While spironolactone remains the preferred fourth-line antihypertensive drug, its side effect profile limits its broad application in a population with unmet treatment needs and a heightened risk of morbidity and mortality. The recent publications of two landmark pharmacotherapy trials, BrigHTN and PRECISION, deliver exciting prospects for future therapeutic options in the setting of resistant hypertension and expand insights into its pathophysiology (Azizi, 2023; Touyz & Harrison, 2023). Baxdrostat's high selectivity and mechanism of action overcomes the limitations of its less selective predecessors and of MRAs, respectively. While aprocitentan is associated with a risk of fluid retention, this appears to be less pronounced than with first-generation selective ETA antagonists and, importantly, it does not share spironolactone's risk of hyperkalaemia, therefore offering a potential therapeutic alternative in patients with advanced CKD (Stages 3 and 4). While the nonsteroidal MRA, finerenone, offers relatively modest BP reductions based on indirect comparisons, its unique pharmacology, pleiotropic effects and favourable tolerability profile warrants further exploration in head-to-head trials at doses predicted to effect meaningful BP reductions.

    The safety and efficacy of RDN in the management of hypertension, including resistant hypertension, is supported by evidence from both well-designed sham-controlled trials and large registry data (Kiuchi et al., 2019). When BP targets are not achieved with lifestyle modifications and pharmacotherapy alone, particularly due to intolerance to first-line drugs or spironolactone, RDN may serve as an adjunct therapeutic approach that could deliver durable BP reductions and enable reductions in antihypertensive pill burden (Barbato et al., 2023) (Figure 4). After some early disappointments, the second-generation trials with improved trial designs and newer delivery systems have sparked renewed interest in its potential. The procedure-specific questions that perhaps require further exploration are, (1) the optimal method of RDN delivery (ongoing trials assessing ethanol injection via microneedle: NCT03503773, NCT02910414), with current data indicating excellent safety and effectiveness of radiofrequency and ultrasound systems, and whether (2), additional RDN delivery distal to the main renal artery bifurcation further enhances BP lowering efficacy (Beeftink et al., 2017). Data from preclinical porcine models showed significantly greater reductions in renal noradrenaline spillover when RDN was performed in distal renal artery branches (Henegar et al., 2015; Mahfoud et al., 2015). The appropriate selection of patients that are likely to derive the greatest benefit from this treatment modality needs to be better defined. At this stage, clinical application of RDN remains individualised and best guided by a hypertension specialist with verified expertise and accreditation.

    Details are in the caption following the image
    Current (blue arrows), emerging (red arrows) and potential (green arrow) pharmacological and non-pharmacological approaches in treatment of resistant hypertension. True resistant hypertension is confirmed through ambulatory or home blood pressure measurements, verifying medication adherence and excluding secondary causes of hypertension. (a) Caution if eGFR < 45 ml min−1 per 1.73 m2 or serum potassium >4.5 mmol L−1. (b) Can be used earlier at any step as guideline directed medical therapy in respective indications. MRA, mineralocorticoid receptor antagonist; ETA, endothelin receptor type A; ETB, endothelin receptor type B; T/TL, thiazide/thiazide-like. Patients with *eGFR ≥ 15 ml min−1 per 1.73 m2; **eGFR ≥ 25 ml min−1 per 1.73 m2; ***eGFR ≥ 45 ml min−1 per 1.73 m2.

    5.1 Nomenclature of targets and ligands

    Key protein targets and ligands in this article are hyperlinked to corresponding entries in https://www.guidetopharmacology.org and are permanently archived in the Concise Guide to PHARMACOLOGY 2021/22 (Alexander et al., 2021).

    DECLARATION OF TRANSPARENCY AND SCIENTIFIC RIGOUR

    This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the BJP guidelines for Design and Analysis and as recommended by funding agencies, publishers and other organisations engaged with supporting research.

    AUTHOR CONTRIBUTIONS

    Omar Azzam: Writing—original draft (lead); writing—review and editing (equal). Sayeh Heidari Nejad: Writing—review and editing (supporting). Revathy Carnagarin: Writing—review and editing (supporting). Janis M. Nolde: Writing—review and editing (supporting). Marcio Galindo-Kiuchi: Writing—review and editing (supporting). Markus P. Schlaich: Writing—original draft (equal); writing—review and editing (equal).

    ACKNOWLEDGEMENTS

    Open access publishing facilitated by The University of Western Australia, as part of the Wiley - The University of Western Australia agreement via the Council of Australian University Librarians.

      CONFLICT OF INTEREST STATEMENT

      R.C. is supported by the Australian National Heart Foundation post-doc fellowship. M.G.K. has received consulting fees and/or travel and research support from Medtronic and Abbott. M.P.S. has received consulting fees and/or travel and research support from Medtronic, Abbott, Novartis, Servier, Pfizer and Boehringer-Ingelheim. The other authors have no conflict of interest.

      DATA AVAILABILITY STATEMENT

      N/A-Review.