Blocking SUDEP With a Beta Blocker: Atenolol Reduces Seizure-Related Mortality

Abstract

Atenolol Rescues Premature Mortality in Genetic Mouse Models of Sudden Unexpected Death in Epilepsy

Soh MS, Hu A, Kuanyshbeck A, Mohamed Syazwan ES, Lee HM, McKenzie CE, Phillips AM, Bleakley LE, Semsarian C, Scheffer IE, Berkovic SF, Reid CA. 2025. Epilepsia. Sep 24. doi: 10.1111/epi.18642. Online ahead of print.

Objective: Sudden unexpected death in epilepsy (SUDEP) is the leading cause of premature mortality in epilepsy. Genetic studies have identified that loss-of-function (LOF) KCNH2 variants are enriched in SUDEP patients, suggesting that they may act as a risk factor. KCNH2 encodes the K_V11.1 channel, with LOF pathogenic variants a cause of long-QT syndrome (LQTS), increasing the risk of arrhythmia and sudden cardiac death. Here, we engineered preclinical rodent models that combine epilepsy-causing pathogenic variants with heterozygous Kcnh2 knockout mice to explore the impact of reduced K_V11.1 channel function on mortality. Methods: Both the Gabrg2^R43Q/+ and Hcn1^M294L/+ genetic mouse models of monogenic epilepsy were crossed with Kcnh2^+/− mice. All genotypes were video-recorded post-weaning, and time to death was measured. Additional mice underwent surgery to enable simultaneous electrocorticography and electrocardiography recordings. Atenolol was delivered in drinking water to a subset of mice. Results: Both single mutant Gabrg2^R43Q/+ and Hcn1^M294L/+ mice displayed spontaneous seizures recapitulating the human phenotypes. Single mutant Kcnh2^+/− mice exhibited an LQTS phenotype. Double mutant mice (Gabrg2^R43Q/+/Kcnh2^+/− and Hcn1^M294L/+/Kcnh2^+/−) had both seizure and prolonged QT interval phenotypes that were similar to their respective single mutant mice. Survival analysis revealed that Gabrg2^R43Q/+/Kcnh2^+/− and Hcn1^M294L/+/Kcnh2^+/− mice experienced a disproportionately higher rate of seizure-related death when compared to wild-type and their respective single mutant littermates. Oral administration of the cardiac-selectiveβ-blocker atenolol significantly improved survival in Gabrg2^R43Q/+/Kcnh2^+/−, Hcn1^M294L/+, and Hcn1^M294L/+/Kcnh2^+/− mice. Atenolol attenuated the sympathetic cardiac response to non-terminal seizures. Significance: The data support the premise that LOF KCNH2 variants can contribute to SUDEP risk in a subset of epilepsy patients. Our findings also highlight the potential use of β-blockers as a prevention strategy in SUDEP.

Commentary

Sudden unexpected death in epilepsy (SUDEP) is the leading cause of death in patients with refractory epilepsy. SUDEP is often unwitnessed, making it challenging to understand the mechanistic underpinnings. ¹ It is thought that etiologies for SUDEP are more likely heterogeneous, but that most involve respiratory or cardiac demise triggered by a convulsive seizure. While the large multicenter mortality in epilepsy monitoring units study (MORTEMUS) demonstrated that in most cases, respiratory arrest preceded cardiac arrest, many different cardiac dysrhythmias were noted. ² Mutations in many genes involved in cardiac regulation have been identified in patients who died from SUDEP. ³ Many of these cardiac-related genes are found in both the heart and brain. There are many similarities between another sudden death entity, sudden infant death syndrome (SIDS), including that cardiac and respiratory etiologies have been described in both SIDS and SUDEP. ⁴ Genetic abnormalities that lead to prolonged QT syndrome have been identified in SIDS. ⁵ Similar such polymorphisms have been identified in patients who died from SUDEP. ⁶ Despite this, no specific link between long QT syndrome and SUDEP had been described.

Here, the authors set out to examine whether having long QT syndrome in addition to a genetically driven epilepsy portends increased risk of seizure-related death. ⁷ They did this by first generating novel mouse lines that harbor gene mutations leading to long QT syndrome and gene mutations leading to epilepsy. To do this, they crossed mice heterozygous for a loss-of-function mutation in Kcnh2 (Kcnh2^+/−), which encodes the voltage-gated potassium channel K_v11.1, with mice with 1 of 2 different epilepsy-causing mutations in either Gabrg2, which encodes the gamma₂ subunit of the gamma-aminobutyric acid (GABA)_A receptor, or Hcn1, which encodes the hyperpolarization-activated cyclic nucleotide channel. Both mouse models, Gabrg2^R43Q/+ and Hcn1^M294L/+, respectively, have been shown to display spontaneous seizures.^8,9 From these crosses, they generated mice that had no mutations, had the mutation in just Kcnh2, had a mutation in just Gabgr2 or Hcn1, or had a double mutation of Kcnh2 and Gabrg2 or Hcn1.

Mice with no mutations had neither cardiac arrhythmias nor evidence of neural hyperexcitability as evidenced by spike-and-wave discharges (SWDs) or generalized convulsive seizures. Mice with just the Kcnh2 mutation displayed QT-segment prolongation, as did mice harboring either of the epileptogenic mutations and the Kcnh2 mutation. Mice with either of the epileptogenic mutations alone or in conjunction with the Kcnh2 mutation displayed epileptiform spikes, SWDs, and seizures. There was no worsening of the epilepsy phenotype in the double mutants compared to the mice without the Kcnh2 mutation. However, adding the Kcnh2 mutation to either of the epileptogenic mutations increased the rate of seizure-related death.

Given the role of beta-adrenergic blockade in treating arrhythmias associated with long QT syndrome, the authors examined whether treating mice with the cardiac-specific beta blocker, atenolol, could reduce mortality. They found that atenolol could reduce mortality in a dose-dependent manner. Interestingly, mice with seizures due to the Hcn1 mutation alone had a significant mortality rate compared to phenotypically wild-type littermates, as has been seen previously. Atenolol also reduced mortality in mice with the Hcn1 mutation alone, without a specific cardiac abnormality. Mice with just the Gabrg2 mutation did not demonstrate significant mortality.

This study is remarkable for several reasons. First, it very nicely and importantly redemonstrates cardiac phenotypes in Kcnh2^+/− mice, epilepsy phenotypes in Gabr2^R43Q/+ and Hcn1^M294L/+ mice, and the seizure-related death phenotype in Hcn1^M294L/+ mice. Second, they nicely demonstrate that crossing Kcnh2^+/− mice with either Gabr2^R43Q/+ or Hcn1^M294L/+ mice increases mortality without appreciably affecting the epilepsy or excitability phenotype. They demonstrate that a readily available and widely deployed drug, atenolol, is effective in reducing mortality in the double mutant animals. Finally, and perhaps most strikingly, this was also effective in reducing mortality in Hcn1^M294L/+ mice without the Kcnh2 mutation or measurable cardiac abnormalities.

This raises the possibility of prophylactically treating epilepsy patients with atenolol or a similar medication to try to reduce SUDEP risk. While this is incredibly intriguing, more work will be needed to better understand not only which patients are at greatest risk for SUDEP, but also which patients are likely to succumb to SUDEP by which etiology. Ideally, such a prophylactic measure targeting a cardiac mechanism would be deployed in patients with epilepsy who are thought to be at risk of dying from SUDEP due to a cardiac etiology. It is feasible that a subset of patients with epilepsy can be identified who have no known baseline cardiac abnormalities, but who experience particularly profound cardiac changes during and after a seizure. This would be largely dependent on specific cell types involved with the seizure and specific networks engaged and/or regions impacted. Emerging analytical algorithms could be helpful in identifying subtle cardiac effects of seizures that could portend heightened risk for SUDEP. These could also be correlated to mutations in cardiac genes identified through genetic screening.

Now that these mouse models exist, there is the possibility of assessing other physiological measures, such as additional autonomic functions and breathing. It would also be interesting to evaluate the efficacy of atenolol in other animal models of epilepsy with a high SUDEP rate, such as mouse models of Dravet syndrome that generally have mutations in genes encoding sodium channels, Kv1.1 mutant mice with mutations in the Kcna1 gene, and others. Similarly, it may be informative to test this therapeutic option in one of the many mouse models that display seizure-induced respiratory arrest and death. ¹⁰ Of course, there are b₂ receptors in bronchioles, so care should be taken in choosing cardiac (b₁) specific beta blockers so as not to exacerbate respiratory difficulties.

Moving forward, it may also be telling to determine whether seizure-induced cardiac abnormalities are less common in patients already taking beta blockers. Similarly, it would be informative to follow patients prospectively and examine whether SUDEP is less common in those taking beta blockers for a cardiac indication. Such data may already be available in existing epilepsy and/or SUDEP registries. Data would need to be carefully stratified for the underlying indication (eg, dysrhythmia vs hypertension, etc) for the beta blocker.

This study nicely delineates the combined effects on seizure-related mortality of having genetic epilepsy and genetically driven cardiac abnormalities. Perhaps more excitingly, they demonstrate a useful treatment option, albeit in mouse models. While there is still more to learn about the patients that may benefit most from therapy, the fact that a new treatment could be on the horizon is an exciting possibility.