Abstracts

Rationale: Deep Brain Stimulation (DBS) for Epilepsy was approved by the FDA as a treatment for refractory partial onset seizures. Results from the SANTE trial demonstrated that DBS was safe and effective when delivered indirectly to the epileptogenic zone (EZ) via the anterior nucleus of the thalamus (ANT), a 'node' within the Circuit of Papez. We investigated if degree of anatomical connectivity between ATN and the EZ played a role in efficacy of DBS. Using data from a single center, we tested the hypothesis that patients with seizures originating inside the hippocampal formation or the cingulate gyrus, two other nodes in the Papez network (IPN), would have higher efficacy than those outside this network (OPN) (e.g., frontal lobe convexity) on seizure control. Methods: Analyses in this study included patients from the University of Kansas Medical Center (KUMC) since in all these subjects (n=11 SANTE and n=4 pilot), the site of seizure origin (IPN or OPN) was ascertained with intra-cranial electrodes. Statistical analysis was conducted using a generalized estimating equations (GEE) model for repeated measures based on a normal distribution (SAS PROC GENMOD). To test the 'network' hypothesis, comparisons of changes in seizure frequency and quality of life (QOL) in patients with seizures emerging IPN vs. OPN were performed. Wilcoxon rank-sum tests, correlations, and paired t-tests were applied to continuous data and Fisher's Exact tests were conducted on categorical data to test for an effect of IPN vs. OPN. Results: Data from 11 subjects were analyzed. At Year 1, median seizure frequency reduction in this cohort was 71% (-24% to -100%) compared with Baseline (GEE chi-square 9.45, p =0.002), with 1/11 subjects achieving complete control. Median seizure reduction was 80.5% (-100% to -40.3%) in 8/11 subjects with seizures IPN, vs. 52.8% (-61.4% to -23.7%) for 3/11 subjects with seizures OPN (2-sided Wilcoxon p = 0.08). Of the 8/11 with seizures IPN, in 7 they originated in the hippocampal formation and 1 in the cingulate gyrus. At year 7, 3/11 subjects with seizures IPN had been seizure free for several years vs. 0/11 subjects with seizures OPN. Addition of 4 subjects from a pilot trial with a nearly identical protocol to SANTE's, increased to 12/15 the number of subjects with seizures IPN. A 2-sided Fisher's Exact test applied to seizure frequency reduction in the 12/15 cohort compared to the 3/15 with seizures OPN, showed significant (p = 0.04) differences in efficacy at the 70% seizure reduction rate. Median QOL scores for subjects with seizures IPN improved by 81% vs. 53% for subjects with seizures OPN. No other factors (e.g., current intensity) had a statistically significant effect on efficacy. Conclusions: The results of these single-center analyses suggest that those patients whose seizures originated inside the Papez network had greater seizure frequency reduction than those outside this network. Degree of anatomical connectivity between stimulation targets and the epileptogenic network plays an important role in therapeutic efficacy. Future DBS studies for epilepsy may benefit from collecting detailed diagnostic information to precisely localize seizure origin. Funding: No funding