CHAPTER 7 Seeking New Paradigms in Epilepsy: Stereotactic Radiosurgery Jean Régis, MD, Romain Carron, MD, Fabrice Bartolomei, PhD, and Patrick Chauvel, PhD R adiosurgery is a neurosurgical approach consisting of the delivery of a high energy confined to a small, sharply lim- ited target with stereotactic accuracy in a single session either to create a lesion or to induce a desired biological effect. 1,2 Since the first attempt by the pioneers in Stockholm, the practice of radiosurgery has changed dramatically. Nowadays, in the vast majority of the indications, Gamma Knife (GK) radiosurgery entails the use of nondestructive low dose to induce subtle biological effects like apoptosis in tumors or endothelial pro- liferation in arteriovenous malformations (AVMs). THE DIFFERENTIAL EFFECT CONCEPT This is the classic clinical observation that in 85% of the AVMs associated with a drug-resistant or medically intractable epilepsy, radiosurgery is followed by seizure cessation or a dramatic improvement of epilepsy occurs long before the occlusion of the AVM itself (Figure 1). Moreover, when the AVM is located in a highly functional area, the seizure cessation is obtained without clinical def- icit. This observation led us to put forth the concept of clinical differential effect in 1992: Radiosurgery can induce a functional effect such as rendering the cortex surrounding the AVM no longer epileptic without destroying the under- lying function, thanks to its capacity to alter some systems specifically while sparing others. The first proof of concept came from the demonstration of the existence of such an effect at the biochemical level in the striatum of rats. 3 With the help of an original rat frame, 4 a group of rats received a single isocenter of 4 mm in the left striatum with a max- imum dose of 50 Gy using the GK (Figure 2). Biochemical analyses demonstrated no change in the level of the gluta- mate decarboxylase but a significant decrease in the choline acetyltransferase level, indicating an injury of the cate- cholaminergic system with concomitant sparing of the GABAergic system. Similarly, GABA was unchanged despite a major decrease in the amino excitatory acids (glu- tamate and aspartate). This experimental demonstration led us to consider radiosurgery as a neuromodulation therapy 1-3 and encouraged us to organize several prospective clinical trials. 5-9 More recently, the demonstration of the existence of a differential effect at the cellular level came from the Charlottesville group. 10 In epileptic rats irradiated with 40 Gy in the temporal lobe with the GK, immunohisto- chemical study suggested that at least 1 subtype of hippo- campal interneurons was selectively vulnerable to GK radiosurgery. Neuronal cells appear to have undergone a phenotypic shift with respect to calbindin and glutamate decarboxylase-67 expression. There is now a growing body of evidence in favor of a neuromodulatory effect of radiosurgery. 1,2,11 A series of successive clinical trials has been carried out in Marseille to evaluate GK surgery in epilepsy. In 1993, we organized a phase II prospective trial in 4 mesial temporal lobe epilepsy (MTLE) patients with a goal of dose ranging and toxicity evaluation. 7,9 In 1995, the good safety and impressive efficacy in the patient receiving the 24-Gy dosage led us to organize a phase III prospective single- center study in 4 MTLE patients (24 Gy, 7-8 cm 3 ) to eval- uate the reproducibility of the efficacy. 5 In 1996, we orga- nized a prospective multicentric European study (21 MTLE patients), confirming the reproducibility of the safety effi- cacy. 8 In 1998, a dose de-escalation study (24, 20, 18 Gy) showed that the efficacy decreased dramatically when the marginal doses were , 24 Gy. 6 Finally, the neurologists from our team performed a long-term evaluation (. 5-year follow-up) of the first 15 consecutive patients treated ac- cording our standard protocol. 12 This study confirmed the good safety and efficacy of GK surgery in this group of patients over the long term with a rate of 60% of Engel I at a mean follow-up of 8 years, 12-16 comparing well with the safety and efficacy of open surgery over the long term. More recently, a multicenter prospective trial in the United States confirmed all our findings. 17 Radiosurgery has been the current practice for selected pure MTLE in our group ever since. 18 There are convincing arguments for such an investiga- tion of the potential role of radiosurgery in epilepsy surgery. We know the following: 1. The safety and efficacy of radiosurgery for the treatment of numerous small, deep-seated intracerebral tumors or mal- formations have been well documented since the 1950s. 2. Radiosurgical treatment of small corticosubcortical lesions associated with epilepsy is known to be associated with seizure cessation in a high percentage of cases (58%-80% in AVMs) long before the expected occlusion of the lesion Copyright © 2012 by The Congress of Neurological Surgeons 0148-396X Clinical Neurosurgery  Volume 59, 2012 59