When used as adjunctive treatment of refractory partial-onset seizures (POS), eslicarbazepine acetate (ESL) is strongly associated with significant reductions in seizure frequency [Sperling MR et al. Epilepsia. 2015]. T. Christopher Bond, PhD, and investigators from Sunovion Pharmaceuticals Inc, Marlborough, Massachusetts, USA, and Covance Market Access Services, Gaithersburg, Maryland, USA, reported in a poster that the effect of ESL was not significantly modified by baseline depressive symptoms or use of concomitant antiepileptic drug (AED) therapy.

The purpose of this pooled analysis of 3 randomized double-blind phase 3 clinical trials [Sperling MR et al. Epilepsia. 2015; Ben-Menachem E et al. Epilepsy Res. 2010; Elger C et al. Epilepsia. 2009] was to determine whether the impact of adjunctive ESL on seizure frequency reduction (SFR) was influenced by the presence of depressive symptoms at baseline or by use of concomitant AEDs in patients with POS. A secondary assessment addressed the association between measures of depressive symptoms at baseline and SFR.

Patients with uncontrolled POS were randomized to either ESL (800 or 1200 mg QD) or placebo. Two binomial regression models were developed to evaluate the associations between baseline factors and treatment response. One included the 10-item Montgomery-Åsberg Depression Rating Scale (MADRS), in which scores ≥ 20 indicate moderate to severe depressive symptoms. The second was based on the 5-item Emotional Well-Being (EWB) subscale of the Quality of Life in Epilepsy Inventory–31 self-assessment questionnaire, in which a score ≤ 52 indicates the presence of depressive symptoms. A ≥ 50% SFR between baseline and the end of the 12-week maintenance period was defined as an ESL treatment response. Both models were controlled for multiple baseline factors and interactions. Demographic and baseline characteristics of the patients are shown in Table 1.

Mean number of AEDs used at baseline was 1.8, with carbamazepine being the most common (48.1%). With the MADRS model, < 10% of patients were classified as having moderate to severe depressive symptoms, while almost 35% had depressive symptoms based on the EWB model. The presence of baseline depressive symptoms was significantly associated with treatment response in the MADRS model (OR, 1.57; 95% CI, 1.01 to 2.45; P = .044) but not in the EWB model (OR, 1.18; 95% CI, 0.86 to 1.60; P = .311). The effect of ESL was not significantly modified by depressive symptoms or baseline AED use in either model. Furthermore, treatment with ESL (both doses) was significantly associated with greater likelihood of treatment response on the basis of both models compared with placebo (P < .001; Table 2).

Treatment with adjunctive ESL significantly reduced seizure frequency, which was not modified by the presence of depressive symptoms or the use of other AEDs. The association between depressive symptoms and treatment response differed according to the instruments used to assess these symptoms.