Predictors of Response to Placebo in Phase III Trials of Adjunctive Eslicarbazepine Acetate (ESL) (P1.031)

Objective: To evaluate factors that impact response to placebo in studies involving patients with partial-onset (focal) seizures (POS). Background: Rates of response to placebo in studies of antiepileptic drugs (AEDs) have increased progressively; this may explain, at least in part, the failure of some studies to demonstrate significant differences between the effects of investigational drugs and placebo. Identifying factors that predict response to placebo may facilitate improvements in study design and the ability to distinguish active therapies from placebo. Design/Methods: Data from three randomized, controlled studies of once-daily adjunctive ESL (093-301, -302, and -304) were analyzed. Patients were ≥16 years old, with ≥4 POS per month, and receiving ≥1 AED. Each study comprised an 8-week baseline, a 2-week titration and a 12-week maintenance period. The primary endpoint was standardized seizure frequency (SSF) per 4-weeks during the maintenance period. This analysis used the modified intent-to-treat (mITT) population (randomized patients; ≥1 dose of study drug; ≥1 post-baseline seizure frequency assessment). Analysis of covariance models were used to assess the impact of different factors on log[SSF] during the maintenance period. Results: A total of 418 patients in the pooled mITT population received placebo. Baseline SSF was higher in patients with longer time between the date of enrollment of the first patient and date of randomization (t E-R ). After adjusting for baseline SSF, the effect of t E-R on log[SSF] during maintenance was significant (p=0.0408). There was a negative relationship between t E-R and log[SSF] during maintenance, i.e., patients recruited later had lower log[SSF]. Other factors with a significant effect were: age when diagnosed (p=0.0286), baseline body weight (p=0.0118), and use of valproic acid (p=0.0309) and lamotrigine (p=0.0307). Conclusions: Among placebo patients, those recruited later after study initiation had lower seizure frequency during maintenance. The occurrence of prolonged recruitment periods may contribute to the high placebo response rates in some AED trials. Study Supported by: Sunovion Pharmaceuticals Inc. Disclosure: Dr. Jung has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Employment: Sunovion Pharmaceuticals Inc. Dr. Blum has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Employment: Sunovion Pharmaceuticals Inc. Dr. Grinnell has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Employment: Sunovion Pharmaceuticals Inc. Dr. Cheng has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Employment: Sunovion Pharmaceuticals Inc.