Advanced Gamma Knife Treatment Planning of Epilepsy

Mesial temporal lobe epilepsy (MTLE) specifically consists of atrophy and gliosis within the limbic system and is the most frequent cause of medically intractable epilepsy in adults (2). Currently, the standard treatment for epilepsy is the use of anticonvulsants. Medically intractable cases may be treated with temporal lobectomy, consisting of removal of parts of the superior temporal gyrus, temporal portion of the amygdala, and the hippocampus.


Background
Epilepsy is not a single disease, but a broad group of conditions characterized by recurrent seizures resulting from the abnormal firing of cerebral neurons (1). Both medical and surgical treatments have been available for over a century, but there has been recent interest in radiosurgery as an alternative to open surgery for patients with medically intractable epilepsy.
Mesial temporal lobe epilepsy (MTLE) specifically consists of atrophy and gliosis within the limbic system and is the most frequent cause of medically intractable epilepsy in adults (2). Currently, the standard treatment for epilepsy is the use of anticonvulsants. Medically intractable cases may be treated with temporal lobectomy, consisting of removal of parts of the superior temporal gyrus, temporal portion of the amygdala, and the hippocampus.
Other causes of epilepsy include hypothalamic hamartomas (HH) and vascular malformations. Hypothalamic hamartomas are benign lesions composed of varying amounts of glia, neurons, and myelinated fibers. These tumors are often associated with gelastic seizures, precocious puberty, and behavioral problems (3) and often do not respond well to anticonvulsant therapy. As a result, surgical resection and stereotactic radiofrequency thermocoagulation have been used to treat HH (4). Epilepsy is also a symptom of cerebral vascular malformations. In particular, patients with cavernous malformations (CM) frequently present with drug-resistant epilepsy (5).
Given the toxicity of open surgery and poor quality of life for patients with medically intractable epilepsy, stereotactic radiosurgery (SRS) with the Gamma Knife has been proposed as a viable alternative to open surgery for treating these patients. Preliminary data have shown that Gamma Knife radiosurgery is highly promising in terms of safety and efficacy for the treatment of MTLE (6)(7)(8). Potential advantages of Gamma Knife radiosurgery include lower morbidity and lower cost with equal effectiveness. One potential disadvantage of radiosurgery is the latency of response, which may be up to two years (9). Although much of the data are promising, the results of Vojtĕch et al suggest that more study is needed (10). Based on these data, an international clinical trial is currently being conducted for determining the role of radiosurgery for managing MTLE. www.intechopen.com

Radiosurgery for MTLE
Because radiosurgery for MTLE is still an experimental procedure, a multi-center Phase III clinical trial known as the ROSE trial (Radiosurgery or Open Surgery for Epilepsy) was initiated in 2009 to investigate the use of Gamma Knife radiosurgery as an alternative to open surgery for the treatment of medically intractable mesial temporal lobe epilepsy (11,12). The primary objective of this study is to demonstrate the equivalence of radiosurgery with temporal lobectomy in terms of freedom from seizures. Other endpoints include quality of life, neuropsychological outcomes, and cost-effectiveness.
Due to the location of the target volume and proximity of critical structures, Gamma Knife radiosurgery for mesial temporal lobe epilepsy (MTLE) is technically challenging. In this chapter we will discuss the methods used for planning Gamma Knife radiosurgery treatments for epilepsy, the equipment used for delivering these treatments, and challenges specific to the treatment of MTLE and other causes of epilepsy.
Similar to the traditional surgical approach to treatment of MTLE with temporal lobectomy, the proposed radiosurgical treatment target is comprised of the amygdala, anterior 2 cm of the hippocampus, and the parahippocampal gyrus. Images of a representative target volume are shown in FIG. 1. The current target prescription dose is 24 Gy, which is based on the results of an earlier study in which lower doses were shown to result in reduced efficacy (13). The total irradiated volume is maintained to be less than 7.5 mL in order to minimize late radiation sequelae such as festering radiation necrosis.

Radiosurgery for hypothalamic harmatomas and cavernous malformations
Studies have demonstrated that the use of stereotactic radiosurgery for treatment of epilepsy associated with hypothalamic hamartomas is safe and effective (14)(15)(16)(17)(18)(19)(20)(21)(22). In 2000, Régis et al reported the results of a multi-center study that involved ten patients treated at seven sites with Gamma Knife SRS (15). The median follow-up was 28 months and the results demonstrated a clear relationship between dose and efficacy and the authors recommended a margin dose of 18 Gy or more. Selch et al described the use of linac based SRS for the treatment of HH (18). A good summary of the various treatment options for HH including SRS has been published by Régis et al (20). This report also includes the results of a prospective study of Gamma Knife SRS for hypothalamic hamartomas. At the time of publication, the authors had sufficient (3 years) follow-up to report results for 27 patients and found that a very good result was obtained in 60% of the patients.
Data have also been published regarding the use of radiosurgery for treatment of epilepsy associated with cavernous malformations (23)(24)(25)(26)(27)(28). In 1999, Bartolomei et al reported the results of a retrospective study to evaluate the feasibility of using radiosurgery to treat epilepsy associated with cavernous malformations (23,29). Data from forty nine patients were included in this study, with over 70% either seizure-free or with a significant reduction

Methods and techniques 2.1 Mesial temporal lobe epilepsy
The dosimetric criteria for the ROSE trial included: 1) prescription dose of 24 Gy to the 50% isodose line (i.e., maximum dose of 48 Gy), 2) total volume receiving 24 Gy (V-24 Gy) within the range of 5.5-7.5 cm 3 , 3) maximum brainstem dose of 10 Gy, and 4) maximum dose of 8 Gy to the optical apparatus.

Hypothalamic hamartomas
Planning for hypothalamic hamartomas is challenging due to the proximity of the brainstem and the optical apparatus (FIG. 2). However, compared to MTLE, planning for HH is less challenging due to lower doses and smaller target volumes for hypothalamic hamartomas. In our institution, target doses range from 15-18 Gy. Target volumes have ranged from 0.3 to 2.1 cm 3 , with a mean volume of 0.81 cm 3 (N=4). www.intechopen.com

Benchmarking treatment planning practices
Due to the critical location of these lesions, special radiosurgical treatment planning techniques are applied to minimize the dose to the surrounding normal brain and functional structures. In order to achieve optimal dose fall-off outside of the target, it is preferable to use small collimators (i.e., 4-mm diameter) for treating such lesions. Selective blocking of individual beamlets is a must for enhancing the dose gradient toward a nearby critical structure. However, selective blocking and planning techniques are known to vary significantly among different Gamma Knife models and individual users. To benchmark such differences, a pre-clinical trial quality assurance procedure was developed. The results of such a study are summarized in the following section. The collimator size is selected by switching helmets. All beams have the same diameter, but individual beams can be blocked by "plugging" the helmets. However, switching helmets and changing plugging patterns can be time consuming, so in reality only a limited number of plugging patterns are used. Despite these differences, the mechanical accuracy of all Gamma Knife models has been consistently maintained to be better than 0.4 mm. As a result, Gamma Knife has so far been the only radiosurgical modality reported for managing intractable mesial temporal lobe epilepsy.

Comparison among present gamma knife models for treating MTLE
As part of the physics review process for the ROSE trial, each center was required to submit a treatment plan based on a sample data set for the purpose of quality assurance. Image data for this sample patient were transferred to participating institutions. The target volume was then delineated and a treatment plan was created to satisfy the previously described dosimetric constraints. The plans were then transferred to the review center for a centralized review by the trial director. Plans that did not satisfy the dosimetric constraints were revised and resubmitted. Plans were submitted for the Gamma Knife Perfexion and C/4C models, depending on the device used at the individual centers. These data were analyzed to look for potential differences between the Perfexion and C/4C models.
A total of 13 plans from 8 institutions satisfied the dosimetric constraints and were included in the data analysis. This included seven plans for the Model 4/4C and six plans for the Perfexion. Details of the individual plans are shown in The average 24-Gy volume was larger for Perfexion plans as compared with Model 4/4C plans, suggesting more aggressive targeting of the treatment area while satisfying normal tissue constraints due to the added flexibility of the Perfexion. Plans created for the Perfexion tended to have a higher minimum dose to the target volume (FIG. 3), which is consistent with the above result. In addition, the results showed that the plans for the Perfexion tended to have shorter beam-on times (155±58 vs. 163±62) and used fewer shots (19±5 vs. 20±10). However the differences were not statistically significant due to the small sample size (FIG. 4). In particular, Plan 7 (for the 4C) used only four shots and had an extremely short beam-on time. After excluding Plan 7, the beam-on time and number of shots for the Model 4C increase to 182±41 minutes and 23±8 shots, respectively. The mean gradient index was 3.0 for both the Perfexion and the Model 4C. Fig. 3. Boxplots showing the total volume receiving the prescription dose and the minimum dose received by the target. Differences were not statistically significant, although the results suggest that plans generated for the Perfexion provide better target coverage. www.intechopen.com

Summary
Gamma Knife radiosurgery is being actively investigated as an alternative to open surgery for the treatment of MTLE and other forms of medically intractable epilepsy. Despite notable efficiency and practice differences between the various Gamma Knife models, consistent treatment planning practices and dosimetric parameters are demonstrated for a multi-institutional international trial setting. The final clinical results of such a trial for treating MTLE are forthcoming.