Abstracts

Rationale: Patients with refractory epilepsy often require antiepileptic drug (AED) combination therapy. It has been hypothesized that AED combination therapy with different mechanisms of action (MOA) could produce synergistic effects and therefore, lead to potentially better outcomes than AED combination therapy with similar MOAs. However, real-world evidence thus far has been limited. The objective of this study was to analyze treatment patterns, compare treatment duration and other health outcomes associated with combination therapies categorized by MOA in patients diagnosed with partial onset seizures (POS). Methods: Using administrative claims from the MarketScan Commercial Database with approximately 96 million covered lives, POS patients (≥ one diagnosis of ICD-9-CM code 345.4x or 345.5x) who initiated AED combination therapy between 1/1/2005 and 3/31/2010 were identified. Patients whose combination therapy treatment lasted ≥90 days with at least 6 months pre-index and 12 months post-index were included in the study. Index date was defined as the first date of AED combination therapy use. Combination therapies were categorized based on MOA of each drug, including sodium channel blockers (SC), GABA analogs (G), SV2A binding (SV2), and broad (multiple) mechanism (B). Baseline demographics and clinical characteristics, including the Deyo-Charlson Comorbidity Index (DCI) and unique 3-digit ICD-9-CM diagnoses (3DDx), were measured in the 6-month pre-index period. Treatment duration was measured in the post-index period from index date to the time of treatment regimen change (switching, addition or discontinuation) or end of patient follow up. Results: 8,615 patients were included in the study. Mean ages (Table 1) ranged from 41.7 years (SC+B) to 48.4 years (G+G) (p<0.001). There were also significant gender variation across combination therapy categories (p<0.001) from 48.8% male (SC+SV2) to 32.1% male (G+B). Pre-index comorbidities varied significantly across combination therapies, evidenced by DCI and 3DDx (p<0.001). Mean duration of therapy was highest for SC+SV2 combinations at 526.7 days and lowest for G+G combinations at 344.3 days (p<0.001) (Table 2). Conclusions: In this study, patients diagnosed with POS receiving AED combination therapy exhibited different treatment durations according to the MOA of the combination, with SC+SV2 combinations the longest and G+G combinations the shortest. However, there were also substantial differences in baseline demographics and clinical characteristics across combination categories, which may have contributed to the observed difference in duration on therapy. It also implies that there may be differential practice patterns for the selection of AED combinations based on patients' comorbidities and characteristics. Multivariate analyses on duration of therapy are being performed to account for the differences in baseline characteristics. Further analyses will focus on the comparison of healthcare utilization and comorbidities across different MOA-based combination therapies.