Impact of prolonged utilization of neprilysin inhibition on the cognitive function of heart failure patients

Introduction

One of the most important pharmacological therapies in patients with chronic heart failure and reduced ejection fraction (HFrEF) is to administer agents that modulate the renin–angiotensin–aldosterone system (RAAS), such as angiotensin-converting enzyme inhibitors (ACE-Is) or angiotensin receptor blockers (ARBs) and mineralocorticoid antagonists (MRAs). These drugs limit the effects of hormones that cause the body to constrict blood vessels as well as to retain sodium. On the other hand, vasoactive substances such as bradykinin and adrenomedullin, counteract vasoconstriction by promoting vasodilation of blood vessels. Neprilysin, formerly called neutral endopeptidase (NEP), is a circulating protease that is responsible for the breakdown of these vasoactive peptides including natriuretic peptides (NPs), neuropeptides (e.g. substance P, enkephalins) and the amyloid-β (Aβ) peptides.^1,2 Due to these degradation properties, NEP has emerged as a pharmaceutical target of interest in the management of cardiovascular disease (CVD) and Alzheimer’s disease (AD). It has been hypothesized that inhibiting NEP may elevate levels of these beneficial peptides and thus counteract systemic overactivity which generally leads to heart failure (HF) symptoms.³

In CVD, the use of a neprilysin inhibitor (NEPi) is aimed at increasing and prolonging the beneficial effects of NPs and vasoactive peptides. The use of a NEPi alone, such as candoxatrilat, proved to be ineffective for the treatment of hypertension due to an aldosterone-independent increase in angiotensin II.⁴ Ultimately, a NEPi was combined either with an ACE-I (e.g. omapatrilat) or an ARB (LCZ696) to overcome this problem.^5,6 For the treatment of hypertension, both omapatrilat and LCZ696 (sacubitril-valsartan) were respectively found to be superior to enalapril (ACE-I) and valsartan (ARB).^7,8 On the other hand, according to the PARADIGM-HF trial, chronic administration of LCZ696 was found to be superior when compared with enalapril in patients with chronic HFrEF by improving outcomes such as a risk reduction in first hospitalization for HF and cardiovascular mortality as well as a favorable biomarker response.^6,9 These promising results hastened the United States Food and Drug Administration approval of LCZ696 for the treatment of HFrEF. However, the role of NEP as a practical pharmaceutical target is being questioned, since the populations suffering from CVD and AD are overlapping.³

Discussion

While cardiologists are excited about getting the new drug LCZ696 in their toolbox, some neurologists are asking question about the possibility of LCZ696 increasing Aβ levels and thus the risk of AD, if taken for long periods. AD is a devastating neurodegenerative disorder that leads to behavioral, cognitive and memory deficits. The pathology of AD involves accumulation of extracellular Aβ containing plaques and intracellular neurofibrillary tau tangles in the brain. Generally, in AD and cerebral amyloid angiopathy (CAA), the Aβ peptide gets deposited in the brain long before the onset of the clinical symptoms. In healthy people, this deposition is prevented by the degradation of Aβ peptide by several proteases, such as NEP, ACE, endothelin-converting enzyme (ECE-1) and the insulin degrading enzyme. Out of all these, NEP is the most effective protease.¹⁰

The role of NEP in AD can be supported by the development of AD-like disease in NEP-deficient mice and the development of AD-like lesions in the brain after the intracerebral infusion of NEPi, such as thiorphan or phosphoramidon.^11–13 Additionally, the brain of patients with AD shows the lower expression of NEP.^14,15 From these supporting evidences, it can be anticipated that the chronic administration of NEPi may compromise Aβ peptide degradation in the brain, and may thus accelerate AD and CAA progression in patients with high risk for AD based on genetic factors¹⁶ or vascular factors.¹⁷ The probable reason behind this deleterious effect in at-risk patients could be the ability of NEPi to cross the blood–brain barrier (BBB), since intracerebral infusion of the NEPi provoked AD lesions in animal models.^12,13 However, it is difficult to predict NEPi permeability across the BBB especially in AD patients as they have altered BBB permeability.¹⁸

As previously discussed, accumulation of Aβ occurs decades before the onset of AD symptoms and patients may already have BBB dysfunction at the time of presentation, thus it is difficult to assess the BBB permeability in these patients. Similarly, the dysfunction of BBB also occurs in CAA and other neuropathologies.¹⁹ McMurray and colleagues²⁰ reported that when LCZ696 was administered in cynomolgus monkeys, levels of Aβ_1-40 and Aβ_1-42 were found to be increased in cerebrospinal fluid (CSF) but not in the brain. On the other hand, when LCZ696 was administered in healthy human volunteers for 2 weeks, no changes were noted in the CSF levels of Aβ_1-40 and Aβ_1-42. This finding in humans could support the cerebral safety of LCZ696 administration over a short span of time.

Recently, Cannon and colleagues²¹ analyzed reports of cognition and memory related adverse effects in the PARADIGM-HF trial. Similarly, they also analyzed cognition related adverse events in three other trials in HFrEF patients which reported such events in similar fashion, such as Val-HeFT, ATMOSPHERE and CORONA-HF. They reported the annual rate of dementia-related adverse effects as ~1 per 100 patient-years in the ATMOSPHERE and PARADIGM-HF trials, while in the older patients of the CORONA-HF trial, the rate was found to be ~70% higher. Additionally, they found that the patients with such adverse effects were significantly older than the ones without such events. The ATMOSPHERE, CORONA-HF and PARADIGM-HF clinical trials showed similar age-adjusted rates, but the Val-HeFT trial showed higher age-adjusted rates for dementia-related adverse events. Furthermore, it is difficult to put these results in the context as none of the HF clinical trials have published the dementia-related adverse effects. They also reported that the concern about the effect of NEPi on Aβ peptides in brain is just a theoretical aspect and has no clinical and experimental evidence. Additionally, data from human genetics do not support the link between NEP and cognitive impairment (AD).^22–24

Moreover, the cognitive decline in HF patients may not be the result of only AD, but could also be due to vascular abnormalities and decreased cardiac function.^21,25–27 Furthermore, there is a higher prevalence of CVD, such as coronary artery disease, atrial fibrillation and cerebrovascular accident (CVA) in patients showing dementia-related adverse effects. Additionally, some evidence suggested that the effective treatment for CVD, especially with ARBs might have a favorable impact on cognitive function.^25–27 Another possible and important contributor to the cognitive impairment in HFrEF patients is the unplanned hospital admission due to decompensation as such episodes of critical illness are usually associated with cognitive impairment over a period of time.²⁸ It has been suggested that the long-term use of LCZ696 may have a beneficial effect on cognitive function in patients with HFrEF as it improves the cardiovascular function and reduces the hospital admission rate in such patients.²¹ This favorable effect on cognitive function could be due to the ARB component (valsartan) of LCZ696, since vascular diseases like hypertension are risk factors for the progression of AD and ARB limits the impact of this risk factor.^29–31

Most chronic heart failure (CHF) patients may not express cognitive impairment before dying, since such patients are usually elderly and have a limited lifespan. This time factor is critical and it makes it difficult to analyze the potential effects of a NEPi on cognitive function in such patients. However, the 5-year survival has increased in CHF patients treated optimally using guideline-directed medical therapy and it ranges 40–50%.^32–34 On the contrary, patients who showed the maximum survival benefit from LCZ696 did not have advanced CHF and thus those patients are expected to have a longer lifespan which may further increase their exposure to LCZ696. Therefore, it is critical to evaluate the effects of a NEPi on cognition in clinical trials.

The PARADIGM-HF trial excluded patients with AD and did not formally evaluate cognitive function.⁶ However, the trial included the reports of dementia-related adverse effects. There were no differences reported between the treatment groups as far as dementia-related adverse effects were concerned and it was confirmed that the cognitive evaluation by serial cognitive tests will be conducted in the ongoing PARAGON-HF trial in patients with heart failure and preserved ejection fraction (HFpEF).²⁰ This cognitive evaluation is critical as it was not a part of the original PARADIGM-HF trial. The PARAGON-HF trial will assess the efficacy and safety of LCZ696 compared with valsartan in HFpEF patients.³⁵ However, it is difficult to say if the duration of this trial would be enough to evaluate cognitive function, since the cognitive impairment takes years to develop. Therefore, it is advisable to follow patients for a long duration even after the completion of a trial to assess the cerebral effects of LCZ696.³

Conclusion

While LCZ696 has been found to be superior to enalapril alone in the treatment of chronic HFrEF, it also increases the levels of vasoactive peptides, neuropeptides and Aβ peptides through a NEPi effect. Theoretically, this increase in Aβ peptides can potentially increase the risk of AD in patients taking LCZ696. Various clinical trials have suggested possible dementia-related adverse events in patients taking LCZ696, however the link between the use of a NEPi and cognitive impairment is yet to be proved in a clinical trial. Cognitive decline in patients with HF can be due to a variety of causes that form the basis of HF and dementia including vascular disease and advanced age. Another contributor to cognitive impairment in this population could be due to recurrent episodes of critical illness from decompensated HF resulting in hospitalization. On the contrary, there is some evidence that LCZ696 can even improve cognition by improving cardiovascular function and reducing hospitalizations. As AD takes years to develop, a clinical trial with longer follow-up period is critical to assess the cerebral effects of LCZ696. The PARAGON-HF trial will include evaluation of cognition by serial tests, but it is unclear whether the length of trial will be enough to suggest a definitive relationship. The population with HF is rising due to increased survival resulting from the advancements in therapy or in other words we can say that the population at risk for cognitive impairment is rising. Therefore, it is necessary to address this possible adverse effect of a NEPi on cognitive function in future trials.