Abstracts

Rationale: TS is a potentially devastating disorder caused by mutations in TSC1 or TSC2 that result in constitutive mTOR activation. Lesions in the brain include subependymal giant-cell astrocytomas (SEGAs) and neuronal tubers, epileptogenic cerebral cortex abnormalities. Recently, an open-label, phase II trial (NCT00411619) of everolimus, an orally bioavailable selective mTOR inhibitor, demonstrated a significant reduction in SEGA and tuber volume and a decreased frequency of seizures. This subgroup analysis investigated potential changes in the normal-appearing white matter in patients from this study. Methods: Patients aged ?3 years with a definitive TS diagnosis and evidence of serial SEGA growth (on MRI) were treated with everolimus 3 mg/m²/d orally (titrated to tolerability to achieve target trough concentrations of 5-15 ng/mL). MRI scans were performed at baseline and at 3, 6, and 12 months. For this subanalysis, we reviewed patients who received DTI as part of their imaging protocol. DTI imaging (15 directions, 1.5 Tesla [GE], 3 3 3 mm, interpolated to 1.5 1.5 3 mm) was used to calculate fractional anisotropy (FA) and axial, radial, and mean diffusivity within regions of interest in the normal-appearing white matter. Treatment effect was measured by comparing these data from patients after 12-18 months of treatment to baseline imaging, using a paired t-test. Results: Twenty-eight patients were enrolled; median duration of treatment was 21.5 months (range, 4.7-34.4). Twenty subjects had sufficient DTI data at multiple time points, including 17 subjects with baseline DTI imaging and 3 subjects with first DTI imaging at 3 months into treatment. Comparing these baseline values to 12-18 months after the treatment, a significant change in the FA was observed in this group in the corpus callosum, internal capsule, and inferior longitudinal fasciculus (all p < 0.05). Median change in FA was 0.05 and was mainly driven by a significant decrease in radial diffusivity in all 3 areas; no significant changes in mean and axial diffusivity were observed. Conclusions: A significant change in FA and radial diffusivity was observed in patients with TS treated with everolimus. These findings support improvement in structural integrity of the measured white matter tracts after treatment with everolimus. As radial diffusivity has been associated with myelin integrity, these changes could represent improvements in myelination. Alternatively, everolimus may decrease abnormal cells in the white matter resulting in a more uniform directional organization of neurons within these tracts. This improvement in white matter integrity may be partly responsible for observed improvements in seizure frequency with everolimus in this population. These findings raise the possibility that the genetic defect of TS in the brain may be modified, even in radiographically normal-appearing white matter.