Surgical Evaluation and Treatment for Genetic Epilepsies

Epilepsy is a common neurologic disorder, affecting an estimated 50 million people worldwide.¹ Studies have shown that although an estimated two-thirds of people with epilepsy can achieve disease control with medication, the remaining one-third have drug-resistant epilepsy.¹ Recommendations for drug-resistant epilepsy include referral for epilepsy surgery evaluation, which can result in reduced seizure burden or seizure freedom. Epilepsy surgery evaluation varies among individuals, but most individuals initially are evaluated for focal onset of seizures that would be amenable to resection. However, even in the absence of a focus or foci for resection, some individuals may have reduced seizure burden with other surgical treatments, such as vagus nerve stimulator (VNS) implantation, deep brain stimulation, reactive neurologic stimulation, or corpus callosotomy.

Epilepsy surgery evaluation is thorough and can include a video EEG test, MRI of the brain with high cortical resolution, magnetoencephalography, single-photon emission computed tomography (SPECT), PET, stereoelectroencephalography (SEEG), and language and memory testing to help localize seizure onset. With increasing frequency of genetic testing in individuals with drug-resistant epilepsy, the number of identified genetic causes of epilepsy prior to or during epilepsy surgery workups has increased. In addition, genetic testing has become a routine component of an epilepsy surgery workup.² This change has raised questions about the efficacy of surgery in individuals with diagnosed genetic epilepsies.

We review the evidence for epilepsy surgeries among individuals with genetic diagnoses, specifically in people with pathogenic variants in the mTOR pathway genes (collectively known as mTORopathies) or ion channels (channelopathies), epilepsy syndromes with highly localizing findings (sleep-related hypermotor epilepsy [SHE]), and variants leading to gross structural malformations in the brain. There is growing evidence of somatic variants in individuals with focal cortical dysplasia, but that topic is outside of the scope of this article and is reviewed elsewhere. We aim to demonstrate that genetic diagnoses can better inform decision-making for surgical treatment of people with epilepsy and should not be seen as a contraindication to epilepsy surgery.

mTORopathies

The mTOR signaling cascade is a crucial regulator of neuronal growth and proliferation as well as synaptic transmission and neuronal networking.³ Pathogenic mutations in this pathway are associated with focal cortical dysplasias and refractory epilepsy. Perhaps one of the most well-known mTORopathies is tuberous sclerosis, which is caused by autosomal-dominant, heterozygous variants in the TSC1 or TSC2 genes. More than 85% of individuals with tuberous sclerosis will develop epilepsy, with a large portion of these cases being drug-resistant.⁴ Removal of epileptogenic tubers can be a highly effective means of reducing seizure frequency or providing complete seizure control.

In a nationwide analysis of 364 children who underwent surgical resection of tubers, 258 (71%) had achieved seizure freedom by 1-year follow-up.⁵ Early referral is likely beneficial; there was a trend toward higher rates of seizure freedom at the 1- and 4-year follow-up in individuals who underwent surgery at age 10 or younger.⁵ In addition, the underlying genetic variant can offer insight to surgical counseling and surgical outcomes, as individuals with pathogenic variants in TSC1 have been shown to have more favorable outcomes after epilepsy surgery interventions compared with individuals with pathogenic variants in TSC2. One study evaluated almost 200 individuals with tuberous sclerosis and found that 65% of 32 children with pathogenic sequence variants in TSC1 had a favorable response to epilepsy surgery independent of the age at diagnosis, genetic variant, or presence or absence of infantile spasms.⁴ However, even in individuals with TSC2 variants, which are associated with significantly worse epilepsy phenotypes given an earlier age at seizure onset and higher frequency of infantile spasms, almost one-half of the children (54 out of 114) with a pathogenic TSC2 variant had seizure freedom after surgical intervention.⁴ In the same study, 57 individuals underwent VNS implantation rather than resective surgery, 13 of whom had TSC1 variants and 44 with TSC2 variants. Overall, 20 individuals (35%) had greater than 50% reduction in seizures, with similar outcomes regardless of TSC1 or TSC2 variants (31% and 36%, respectively).⁴ These data demonstrate that although resection can be an effective surgical option for many individuals with tuberous sclerosis, even in nonresective candidates, individuals may benefit from VNS implantation.

mTORopathies also may be caused by pathogenic variants in genes such as DEPDC5, NPRL2, and NPRL3, which are components of the GATOR1 complex, a key regulator of the mTOR pathway.⁶ Similar to tuberous sclerosis, variants in this pathway are associated with focal epilepsy syndromes, with more than half of such individuals having drug-resistant epilepsy.⁶ In a recent study of 50 children with GATOR1-associated variants, 27 underwent surgery, all of whom had significant improvement in seizure outcomes, with 92% seizure-free at 6 months.⁶ Although positive findings on imaging were associated with improved seizure control after surgery, 7 of the children from this study had no lesions found on MRI, but, with focal findings on FDG-PET and SEEG, still achieved seizure freedom.⁶ This highlights the importance of a complete surgical workup for individuals with variants in GATOR1-associated epilepsy even in the absence of structural etiology on MRI. A separate study reported that 4 out of 8 children with drug-resistant epilepsy secondary to GATOR1 complex variants achieved seizure freedom after epilepsy surgery.⁷ The same authors conducted a literature review of an additional 30 individuals with GATOR1 variants who underwent epilepsy surgery and reported a seizure freedom rate of 60%.⁷ Between these studies, 47 of 65 (72%) individuals with GATOR1 variants who underwent epilepsy surgery achieved seizure freedom.^6,7 These studies demonstrate that epilepsy surgery can be a highly effective treatment option for individuals with this genetic diagnosis and appropriate individual selection.

Channelopathies

Whereas surgery is clearly beneficial in the presence of pathogenic variants resulting in focal cortical dysplasia, the benefit of surgery in individuals with channelopathies attributable to SCN1A mutations is more variable, with clinical phenotype playing a larger role in determining the likelihood of success of epilepsy surgery. Children with pathogenic variants in SCN1A have a wide range of phenotypes, ranging from the milder GEFS+ (genetic epilepsy with febrile seizures plus) to the severe Dravet syndrome, which is frequently drug-resistant. One case series evaluating 8 children with SCN1A pathogenic variants reported improvement in seizures for 5 of the 8 children who underwent epilepsy surgery.⁸ Those who responded well to resective epilepsy surgery had focal seizures with additional structural abnormalities, such as hippocampal sclerosis. The other 3 individuals in the case series who had a Dravet syndrome phenotype did not respond well to surgical resection of lesions found on MRI.⁸ These findings suggest that although a channelopathy such as SCN1A should not preclude children from undergoing an epilepsy surgery evaluation, the best outcomes will be seen in children who have a focal semiology and focal structural abnormality.

Although many individuals with channelopathies will not be candidates for a focal resection, there remains benefit for alternative surgical approaches. In a case study of 20 children with drug-resistant epilepsy and pathogenic variants in various genes, including SCN1A, SLC35A2, and MECP2, 2 (10%) achieved seizure freedom after VNS implantation, and 11 (55%) had improvement in their seizure burden.⁹ In this same study, 3 out of 4 children with Dravet syndrome secondary to SCN1A variants had a greater than 50% reduction in seizure frequency, suggesting that this may be a viable surgical alternative for children with drug-resistant epilepsy.⁹ A separate case study reported significant improvement in seizure frequency after corpus callosotomy for an individual with a KCNQ2 mutation.¹⁰ Although resection and seizure freedom are rare in channelopathies, these palliative surgical options can improve seizure burden and quality of life.

Epilepsy Syndromes with Highly Localizing Findings

Inherited epilepsy syndromes in which a focal seizure phenotype is present may be caused by multiple separate genes. One such disorder is SHE, previously known as autosomal-dominant sleep-related hypermotor epilepsy or autosomal-dominant nocturnal frontal lobe epilepsy. This syndrome, which is characterized by childhood onset of focal seizures from the frontal lobe that occur during sleep, can be associated with sequence variants in genes such as CHRNA4, KCNT1, and DEPDC5, among others.¹¹ Because up to 30% of individuals with this condition have seizures that are refractory to medications,¹² these individuals make ideal candidates for surgical evaluation. In a 2007 study¹² evaluating 21 individuals with drug-resistant SHE, 16 (76%) achieved seizure freedom, and the remaining 5 had significant improvement in seizure burden after resective surgery. A more recent study¹³ demonstrated Engel Class 1 outcomes (freedom from disabling seizures) in 104 of 127 individuals (82%), with 86 individuals (68%) having complete seizure freedom. Notably, these studies lack data regarding underlying genetic causes for the epilepsy syndrome. About 30% of SHE cases are thought to have a genetic cause. A recent study¹⁴ of 103 patients with SHE found pathogenic variants in 9 (8.7%) of the patients. Of the 9 patients, 4 had pathogenic mutations in DEPDC5 and 1 had a pathogenic mutation in NPRL2, both of which are genes involved in the mTOR pathway. Only one patient had a mutation in KCNT1. Overall, the genes associated with this disorder have widely varying mechanisms; KCNT1 encodes a potassium channel, whereas DEPDC5 is part of the GATOR1 complex. Future research stratifying response to epilepsy surgery by specific variant may help better predict outcomes of epilepsy surgery for this population and provide guidance on which individuals to refer for epilepsy surgery evaluation.

Cortical Malformations

There have been documented cases of success with surgical resection of gross cortical malformations such as those seen with Sturge-Weber syndrome and Sotos syndrome. Sturge-Weber syndrome is associated with pathogenic variants in somatic mutations in the GNAQ gene, resulting in (among other symptoms) leptomeningeal angiomas and resultant seizures early in infancy. A review of 90 children who underwent various epilepsy surgeries for treatment of epilepsy (including focal resections and hemispherectomies) resulted in seizure freedom in more than 80% of the cases.¹⁵

Sotos syndrome is an autosomal dominant inherited disease associated with pathogenic mutations in NSD1, which results in cerebral overgrowth and seizures in half of these individuals. The majority of individuals with epilepsy secondary to Sotos syndrome achieve seizure freedom with medications. However, a recent case report of an individual with Sotos syndrome and drug-resistant epilepsy described seizure freedom achieved after temporal lobe resection.¹⁶

Conclusion

Genetic testing is becoming a standard tool in the evaluation and management of epilepsy. Although individuals with genetic diagnoses are underrepresented in epilepsy surgery cohorts, there is increasing evidence to support that a genetic diagnosis is not a contraindication to epilepsy surgery, but rather is an important piece of information to guide the epilepsy surgery workup. As demonstrated in the Table, genetic results can be used to guide which surgical treatment would be most beneficial for the individual, whether it be curative surgical resection or palliative surgical treatment with VNS implantation, corpus callosotomy, hemispherectomy, or additional surgical techniques. Providers must be familiar with potential surgical outcomes in the presence of various variants, so that individuals can be directed toward the most effective treatment option; for example, although many individuals with Dravet syndrome may not benefit from resective surgery, emerging data are demonstrating that palliative VNS placement may be an effective treatment. Genetic testing should be implemented early in the presurgical evaluation to better predict outcomes for individuals with refractory epilepsy. Equally as important is that individuals with genetic diagnoses receive a complete surgical workup, because focal resection may be a consideration even in the absence of a structural etiology on imaging.