Abstracts

Rationale: The aim of this study was to investigate the relative safety profile of levetiracetam (LEV) extended release (XR) as compared to LEV immediate release (IR) using data from phase III clinical trials. Methods: Data from patients with partial onset seizures, included in three phase III RCTs of adjunctive LEV XR 1000 mg once daily (N01235-12 weeks) and LEV IR 500 mg bid (N051-16 weeks, N132-18 weeks) were used. Intention-to-treat populations were included in this analysis. Treatment-emergent adverse events reported during the up titration (for LEV IR only) and maintenance periods (LEV IR and LEV XR) were classified using MedDRA 9.0. System Organ Class (SOC) level and, where a significant difference was found, preferred term (PT) side-effects were considered. Indirect comparisons of placebo-adjusted rates (Risk Differences, RD) for adverse events were obtained using intergroup heterogeneity testing. Epilepsy-related adverse effects were not included in the analysis to avoid contamination by efficacy outcomes. Tables and analyses were abstracted and analyzed using SAS® version 9.1.3 (SAS Institute Inc.), alpha risk being set at 10%, the normative value for such meta-comparisons. Results: 555 subjects from trials N051, N132, and N01235 (204 LEV IR, 70 LEV XR, and 281 placebo) aged 16-70 years, with similar characteristics and baseline median seizure rates ranging from 1.77 to 2.82 per week were included in the meta-analysis. Mean follow-up times were 12 weeks for LEV XR and 17 weeks for LEV IR, or corresponding placebo (PBO). The RDs for any adverse events were similar across groups: LEV IR/PBO: RD=0.02 (-0.06; 0.09); LEV XR/PBO: RD=-0.05 (-0.21; 0.10); intergroup difference, RD=7%, p=0.44. LEV XR showed statistically significantly lower relative rates of nervous system (RD=17.93%, difference: p=0.031), psychiatric disorders (RD=10.53%, difference: p=0.08), and metabolic and nutrition (RD=3.75%, difference p=0.08) as compared to LEV IR at the SOC levels. Within the nervous system organ class, the only statistically significant determinant was headache (RD=10.62%, difference: p=0.08). Within psychiatric disorders and metabolism and nutrition disorder SOCs, there was a continuum of trends among preferred terms that drove the SOCs to significance, although no particular PT reached statistical significance. Conclusions: Limitations include the lack of statistical power of randomized controlled trials to document rare occurrences of adverse events and the current lack of understanding of the synergistic or antagonistic effects of adjunctive AED therapies on the safety profile of the initially prescribed drug. Notwithstanding, although some underlying differences might remain, their systematic impact is likely to be trivial in such a comparative setting. Levetiracetam extended release may provide, on the basis of this exploratory analysis, statistically significantly lower rates of nervous system, psychiatric, nutrition and metabolism adverse events as compared to levetiracetam immediate release bid.