Abstracts

Rationale: The Sabril® (vigabatrin, VGB) Registry, which was part of an FDA-mandated Risk Evaluation and Mitigation Strategy (REMS) program, was designed to characterize the prevalence, incidence, time to onset, progression, and severity of vision loss during VGB treatment. The demographics, disposition, and vision loss in patients enrolled in the registry from August 21, 2009 to July 26, 2016 are presented. Methods: Participation in the Sabril Registry was required for all US VGB patients/prescribers. Patient characteristics, including age, epilepsy indication, prior/concomitant epilepsy treatments, and discontinuation rates, were collected as part of required registry documentation. In addition, ophthalmologic assessments, including visual field (VF) and visual acuity (VA) tests, were required within 1 month of taking VGB and every 3 months thereafter (with a 3-month grace period). Initial vision test was considered baseline regardless of whether patient had already taken VGB. Predefined criteria for vision loss were reduction by 3 lines from baseline in VA, or reduction from baseline in VF, in two consecutive assessments using consistent methods. Results: Of 9423 patients enrolled during the 7 years of the Sabril Registry, 2592 (27.5%) were treated for RCPS and 6087 (64.6%) for IS; 7741 (82.2%) were pediatric ( < 17 years). At enrollment, 84.8% of pediatric patients and 87.1% of adults (≥17 years) were VGB-naïve. Most patients were using multiple antiepileptic treatments when VGB was initiated. Prior to enrollment, most pediatric patients had attempted ≤3 antiepileptic treatments (including ACTH/steroids for IS) and most adults had attempted ≥4. Over the 7-year period, 6361 (67.5%) patients discontinued treatment, most of whom (53.1%) reported they had completed treatment. Other commonly reported reasons for discontinuation were prescriber choice (16.6%), patient deceased (7.3%), patient choice (5.6%), and lack of efficacy (3.7%). Less than 1% of patients (n=20, 0.3%) reported discontinuation due to visual field defect. A total of 35.9% of patients could not undergo initial vision testing. The median duration (range) of VGB treatment was 8.9 (0.0-86.7) months. Of 456 patients with a baseline examination and ≥2 follow-up acuity tests, 28 (6.1%) developed VA loss; some of these patients returned to baseline VA after initially meeting acuity-loss criteria. A total of 38 (0.4%) patients were able to undergo consistent VF testing; of these, 7 developed treatment-emergent VF loss. Conclusions: Most registry patients were receiving concomitant AEDs when VGB was initiated. Prior to enrollment, the majority of pediatric registry patients had attempted ≤3 epilepsy treatments, while most adults had attempted ≥4. Most patients who discontinued VGB did so because they had completed treatment; less than 1% reported discontinuing due to vision loss. Reductions in VA and VF were detected in a small subset of patients who underwent testing; however, too few patients were able to undergo VF testing to reliably determine the frequency of VGB-associated VF changes. Funding: Lundbeck LLC