Tumor-related epilepsy and epilepsy surgery

Authors

  • Samden D. Lhatoo,

    Corresponding author
    1. Epilepsy Center, Neurological Institute, University Hospitals Case Medical Center, Cleveland, Ohio, U.S.A
    • Address correspondence to Samden D. Lhatoo, Epilepsy Center, Neurological Institute, University Hospitals Case Medical Center, 11100 Euclid Avenue, Cleveland, OH 44106, U.S.A. E-mail: samden.lhatoo@uhhospitals.org

    Search for more papers by this author
  • Narges Moghimi,

    1. Epilepsy Center, Neurological Institute, University Hospitals Case Medical Center, Cleveland, Ohio, U.S.A
    Search for more papers by this author
  • Stephan Schuele

    1. Comprehensive Epilepsy Center, Department of Neurology, Northwestern University Feinberg School of Medicine, Northwestern University, Chicago, Illinois, U.S.A
    Search for more papers by this author

Summary

Tumor-related epilepsy (TRE) is a major etiologic category of epilepsy. TRE is heterogeneous, and the epidemiology, pathology, pathophysiology, clinical features, treatment, and outcomes vary accordingly. In addition, treatment imperatives vary between almost purely epilepsy considerations and those that are primarily oncologic. Often, there is no clear separation of imperatives, and there is a relatively scant evidence base that underpins management decisions in such cases. Given a diverse molecular as well as clinical landscape and the rapid pace with which new knowledge accrues, there are relatively few recent literature resources on TRE that provide neurologists, neurosurgeons, epileptologists, and oncologists with an up-to-date, state-of-the-art review of the field in all of its important aspects. The proceedings of the Sixth International Epilepsy Colloquium in Cleveland in Ohio, U.S.A., in May 2013 on Tumoral Epilepsy and Epilepsy Surgery address, at least in part, several TRE aspects crucial to modern epilepsy and oncology practice.

Tumor-related epilepsy (TRE) is encountered in a variety of settings in epilepsy practice and comprises a significant proportion of patients with epilepsy; this ranges from 6% of newly diagnosed epilepsy in community-based cohorts (Hart et al., 1990) to 17% of patients who undergo epilepsy surgery where seizure control is the main imperative (de Tisi et al., 2011). When the seizures produced by higher grade tumors, where the treatment imperative is primarily oncologic, are taken into consideration, TRE is more common. Given this prominence as a major etiologic category of epilepsy, and given the pace at which new knowledge accrues on an almost daily basis, there are relatively few recent literature resources on TRE that provide neurologists and epileptologists with an up-to-date, state-of-the-art review of the field in all of its important aspects. We set out to address this by choosing Tumoral Epilepsy and Epilepsy Surgery as the topic of the Sixth International Epilepsy Colloquium in Cleveland in Ohio, U.S.A., in May 2013. This meeting brought together some well-known basic and clinician scientists from around the world, and their colloquium sessions form the basis of this supplement, where both authoritative reviews and original data are presented.

Epidemiology and Pathology

This section deals with epidemiology, classification, and natural history aspects of TRE, topics that should be viewed from a perspective that prominently considers the pathology of brain tumors. The terms “epileptomas” and “long-term epilepsy-associated tumors (LEATs)” (Luyken et al., 2003) are used to describe chronic epilepsy etiologies, where frequently seizures have occurred for >2 years and the oncologic burden is usually indolent, as opposed to malignant tumors where the oncologic burden is not usually indolent. The natural histories, presentations, and course are likely to differ according to histologic subtype (Kerkhof & Vecht, 2013), and the recent focus on associated pathologic findings of focal cortical dysplasia (FCD; Blümcke et al., 2011) has raised questions of their influence on TRE (Palmini et al., 2013): in particular, the questions of whether focal cortical dysplastic changes lie within the neuropathologic “continuum” of developmental tumors, whether they are in fact “dual pathology,” and whether they have any relevance as far as treatment outcomes are concerned (Cossu et al., 2013; Palmini et al., 2013).

Pathophysiology

Tissue analysis of gliomas not only confirms histologic diagnosis but can also determine genetic alterations associated with improved response to chemotherapy or radiotherapy such as O(6)-methylguanine methyltransferase (MGMT) or mutations in isocitrate dehydrogenase-1 (IDH-1; Christman et al., 2011; Ostrom et al., 2013; Rogers, 2013). Gene markers are thus being used increasingly for diagnosis and prognostication in patients with brain tumors; epileptologists should possess practical knowledge of their current implications. The multiple investigational and treatment aspects of TRE demand a systems biology approach; a strong argument for this is made (Mittal et al., 2013) in order to improve the understanding of TRE and brain tumors and to improve diagnosis and outcomes of both. The actual pathophysiologic mechanisms of TRE are still relatively poorly understood. In the case of gliomas in particular, where both medical and surgical therapies may have limited effect, Pallud et al. (2013) throw some light on “tumorocentric” and “epileptocentric” mechanisms of epileptogenesis.

Investigations

High-resolution magnetic resonance imaging (MRI) scanning is invaluable in the identification and follow-up of epilepsy-associated brain tumors (Duncan & de Tisi, 2013). Despite advances in structural imaging, accurate radiologic tumor classification is not always possible (Rheims et al., 2013), even though therapeutic approaches can depend on this. Functional imaging studies with various radioactive tracers have been of limited usefulness in determining underling pathology (Wehner, 2013). Surgery for seizures is a treatment option in many patients, and this requires appropriate presurgical investigations. The principles of epilepsy surgery in TRE are largely similar to those in non-TRE patients. Seizure semiology and interictal electroencephalography (EEG) findings are able to confirm and further localize the epileptogenic zone in ~50% of extratemporal and ~75% of temporal TRE cases (Kennedy & Schuele, 2013). Invasive video-EEG evaluations using either grid or stereotactic depth electrodes may be best used in patients who fail to respond to initial resection and for highly selected patients with a challenging overlap of lesion, eloquent cortex, and epileptogenic zone (Guerrini et al., 2013; Rosenow & Menzler, 2013; Sweet et al., 2013).

Treatment

Management strategies can vary according to the specialist first encountered (tumor or epilepsy specialist), even though the underlying pathology and seizure presentation may have been similar (Tandon & Esquenazi, 2013). Tumor classification based on MRI can also influence antiepileptic drug (AED) choice (Perucca, 2013). The possibility of chemotherapy should preclude enzyme-inducing AED use where possible. Patients with glioblastoma TRE who are treated with valproic acid may gain modest survival benefit. Patients with TRE related to glioneuronal tumors and low-grade gliomas have excellent seizure outcome postsurgery (Englot et al., 2011, 2012). For different reasons, in glioneuronal tumor and low-grade glioma TRE, early surgery may be advisable. In the former, intractable epilepsy is often the rule; excellent surgical prognosis (Rheims et al., 2013) and an absent seizure burden, particularly in young brains, strengthens the case for this. In the latter (low-grade glioma TRE), an emerging evidence base advocates an aggressive and early surgical approach. Given the high prevalence of seizures at tumor presentation, the opportune time to perform epilepsy surgery may be when the patient is already undergoing oncosurgery in the same area.

In the case of temporal lobe TRE, the inclusion (or not) of mesial temporal structures is a frequent clinical conundrum. A systematic, algorithmic approach has been proposed depending on how involved these structures are in the TRE (Rosenow & Menzler, 2013), although what constitutes functional and structural integrity can be argued. Intraoperative electrocorticography, with spatially accurate sampling through subdural grids and stereotactic depth electrodes covering mesial and lateral temporal structures, might make a difference in intraoperative surgical decision-making (Luther et al., 2011). For some tumors such as dysembryoplastic neuroepithelial tumors (DNTs), an MRI-based scheme might be helpful in determining resection extent (Chassoux et al., 2012; Chassoux & Daumas-Duport, 2013).

In extratemporal lobe TRE, the situation is different. Extended lesionectomies for glioneuronal tumors may not be necessary from either an oncologic or an epilepsy perspective. In low-grade gliomas, however, the recent literature supports a more aggressive approach (Jakola et al., 2012). The idea of supratotal resection of these tumors based on functional boundaries is intriguing and may affect long-term survival (Duffau, 2013).

The relationship between the lesion zone and epileptogenic zone is fascinating. Long-standing tumoral epilepsy may lead to dependent secondary epileptogenicity with mirror foci and contralateral seizures resolving after ipsilateral tumor resection (Rosenow & Menzler, 2013). However, in selected patients, the argument for independent secondary epileptogenicity can also be made (Scholly et al., 2013).

Tumor-associated FCD is a well-known entity and is often posited as the explanation for surgical failure. TRE FCD (FCD type IIIb and TRE with FCD type II) appears to differ significantly from isolated, non-TRE-FCD but not from non-FCD TRE. Cossu et al. (2013) present data that emphasizes equally good seizure outcomes in FCD and non-FCD TRE. They question the clinical value of an “FCD-centric” view, although why the presence of a second, usually epileptogenic, abnormality did not influence outcome requires further study.

The fields of brain tumor chemotherapy and immunotherapy are evolving. An understanding of the field, underpinned where possible by the molecular contexts, is important for epileptologists (Rogers, 2013).

Special Situations

The heterogeneity of TRE renders an all-encompassing text far beyond the scope of this supplement. However, a few clinical entities require special consideration. These include the unique clinical questions posed in the management of hypothalamic hamartoma. Early surgery in such cases may be facilitated by novel MRI-guided laser ablation methods that seem to offer excellent results while avoiding the significant morbidity that conventional hypothalamic surgery can produce (Wilfong & Curry, 2013). Chassoux and Daumas-Duport bring their considerable experience in the diagnosis and management of DNTs to their review, including the MRI and neuropathologic classifications of DNT subtypes and their influence on surgical resection (Chassoux & Daumas-Duport, 2013). Although the tubers of tuberous sclerosis are somewhat outside the remit of a TRE text, we have also included a brief review of tubers long-term outlook (Shahid, 2013).

Acknowledgments

The authors would like to thank the contributors for their efforts in the publishing of an important contribution to contemporary epilepsy literature. The authors would also like to thank the Scientific Panel of the annual International Epilepsy Colloquium series, including, in alphabetical order, Alexis Arzimanoglou, Hajo Hamer, Edouard Hirsch, Hans Lüders, Philippe Kahane, Susanne Knake, Felix Rosenow, and Philippe Ryvlin.

Disclosure

The authors do not have any conflicts to disclose. The authors confirm that they have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Ancillary