FormalPara Key Summary Points

Why carry out this study?

It is of clinical relevance to evaluate the efficacy and tolerability of brivaracetam (BRV) when added to specific antiseizure medications (ASMs) because it provides important information on how BRV can potentially improve the care of patients with focal-onset seizures

This was a post hoc analysis of pooled data from three phase 3, double-blind, placebo-controlled trials to investigate the efficacy and tolerability of BRV (50–200 mg/day) in patients on carbamazepine (CBZ), lamotrigine (LTG), valproate (VPA), oxcarbazepine (OXC), topiramate (TPM), phenytoin (PHT), lacosamide (LCM), clobazam (CLB), and phenobarbital (PHB)

What was learned from the study?

Efficacy outcomes were consistently higher in patients randomized to BRV (50–200 mg/day) compared with placebo for each subgroup of patients on specific concomitant ASMs

The overall incidence of treatment-emergent adverse events was generally similar across subgroups by specific concomitant ASMs in patients receiving BRV (50–200 mg/day) or placebo

These data show that BRV provides additional efficacy to patients receiving a broad range of specific ASMs for focal-onset seizures

Introduction

Brivaracetam (BRV) is an antiseizure medication (ASM) with high and selective affinity for synaptic vesicle protein 2A. BRV is approved in multiple regions globally, including the USA [1], the European Union [2], and several countries across North and South America and the Asia Pacific region. In the phase 3 pivotal, placebo-controlled, double-blind trials of adjunctive BRV [3,4,5], adults with uncontrolled focal-onset seizures who were taking one or two concomitant ASMs were randomized to BRV or placebo without titration as adjunctive therapy. Findings from these double-blind trials established the tolerability and efficacy of BRV as adjunctive therapy for focal-onset seizures.

Previous analyses from these three double-blind trials have demonstrated efficacy of BRV (50–200 mg/day) with concomitant lamotrigine (LTG) or topiramate (TPM) independent of the number of concomitant ASMs, in groups that were mutually exclusive [6]. Pooled data from these trials has also previously been analyzed to examine the association between BRV-induced elevation of carbamazepine (CBZ) epoxide levels and tolerability and efficacy in patients who were taking concomitant CBZ at trial entry [7]. However, only a small proportion of patients in the double-blind trials were receiving one concomitant ASM, with most receiving at least two [3,4,5]. A post hoc pooled analysis of data from these trials was conducted in patients randomized to BRV 50–200 mg/day or placebo receiving one specific concomitant ASM, which provided a low sample size and limited the interpretation of the data [8]. Therefore, to better inform clinical practice, the objective of this analysis was to evaluate the efficacy and tolerability of BRV when added to specific ASMs without restriction on the number of concomitant ASMs.

Methods

Trials and Populations

This was a post hoc analysis of pooled data of patients from the placebo-controlled, double-blind trials of adjunctive BRV (N01252/NCT00490035 [5], N01253/NCT00464269 [3], N01358/NCT01261325 [4]). In these trials, patients with focal-onset seizures were randomized to BRV (5, 20, 50, 100, or 200 mg/day) without titration or placebo, and were taking one or two concomitant ASMs. Patients were eligible if they were male or female, and 16–70 years of age.

These trials were conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonisation notes for Guidance on Good Clinical Practice. At all trial sites, institutional review boards approved the trial protocols. Before enrollment, patients provided written informed consent.

This current analysis included patients from the three double-blind trials who were randomized to placebo or BRV at a dose of 50–200 mg/day, who were on the most common concomitant ASMs at trial drug initiation (ASMs taken by at least 50 patients in the combined BRV 50–200 mg/day group). The concomitant ASM subgroups were not mutually exclusive, as the majority of randomized patients were on two concomitant ASMs at trial drug initiation.

Outcomes

The evaluation period was the 12-week double-blind evaluation period of each trial. Prior ASMs were those discontinued before study drug initiation. In trials N01252 and N01253, data on ASM use in the 5 years before trial entry were collected, and in trial N01358, data on all historical ASM use before trial entry were collected.

Efficacy outcomes were assessed in the pooled intent-to-treat populations (efficacy population), which excluded patients receiving concomitant levetiracetam (LEV) in trials N01252 and N01253 (N01358 did not enroll patients on concomitant LEV). Efficacy outcomes included percentage reduction over placebo in focal-onset seizure frequency per 28 days; median percentage reduction from baseline in focal-onset seizure frequency per 28 days; 50% responder rate (patients with at least a 50% reduction from baseline in focal-onset seizure frequency); and seizure freedom (all seizure types, including focal-onset, generalized, and unclassified seizures). If patients completed the treatment period, did not report any seizures, and had no missed days in their seizure diary, they were classified as seizure free.

Tolerability outcomes were assessed in patients from the pooled safety set populations, which included patients from N01252 and N01253 receiving concomitant LEV (safety population). Tolerability outcomes included the incidence of treatment-emergent adverse events (TEAEs), TEAEs considered drug-related, and TEAEs leading to discontinuation. TEAEs were coded using Version 15.0 of the Medical Dictionary for Regulatory Activities.

Statistical Analyses

Statistical comparisons were performed for efficacy outcomes for patients randomized to BRV 50–200 mg/day versus placebo for the efficacy population. For the percentage reduction over placebo in focal-onset seizure frequency, treatment group comparisons and parametric effect estimates were based on analysis of covariance (ANCOVA) with the outcome being log-transformed treatment period 28-day adjusted focal-seizure frequency, effects for study and treatment, and a continuous covariate of log-transformed baseline focal-onset seizure frequency. For the median percentage reduction from baseline in focal-onset seizure frequency, the effect differences between placebo and each BRV treatment group were analyzed using the Hodges–Lehmann non-parametric effect estimates and corresponding two-sided 95% confidence intervals, and the Wilcoxon–Mann–Whitney test was used to assess treatment group comparisons. For the 50% responder rates, odds ratios, confidence intervals, and treatment group comparisons were conducted using a logistic regression model with the outcome being the 50% responder rate for 28-day adjusted focal-seizure frequency during the treatment period, effects for trial and treatment, and a continuous covariate of log-transformed baseline focal-onset seizure frequency. For seizure freedom, treatment group comparisons were based on Fisher’s exact test by pooling patients with no seizures but who discontinued, with those considered not seizure free. All p values reported are nominal and can only be interpreted in an exploratory manner. Tolerability outcomes were described using descriptive statistics.

Results

Patient Disposition, Demographics, and Baseline Characteristics

In this analysis, nine concomitant ASMs were taken by at least 50 patients in the combined BRV 50–200 mg/day group (including BRV 50, 100, or 200 mg/day) as part of the patients’ treatment regimen: CBZ, LTG, valproate (VPA), oxcarbazepine (OXC), TPM, phenytoin (PHT), lacosamide (LCM), clobazam (CLB), and phenobarbital (PHB).

Baseline demographics by specific concomitant ASMs at trial drug initiation are shown in Table 1. In the safety population, the mean age across subgroups ranged from 33.4 years in patients on concomitant CLB to 41.9 years in those on concomitant LCM. In the efficacy population, the mean duration of epilepsy ranged from 21.3 (patients on VPA) to 26.2 years (patients on PHB). The mean age at onset of epilepsy ranged from 12.8 (patients on CLB) to 19.2 years of age (patients on LCM). The median baseline focal-onset seizure frequency was generally similar in patients on concomitant CBZ, LTG, VPA, OXC, TPM, PHT, and CLB (8.2–9.5 per 28 days) and was slightly numerically higher in patients on concomitant LCM and PHB (both 11.4 per 28 days). The median number of prior ASMs (discontinued before trial drug initiation) was lowest in patients taking concomitant CBZ or VPA (both 2.0), and highest in patients on concomitant LCM (5.0). LEV was taken as the first or second most common prior ASM in all subgroups (ranging from 33.5% of patients on concomitant VPA to 83.3% of patients on concomitant LCM). Across all subgroups, most patients were on two concomitant ASMs at trial drug initiation (75.6–97.9% in the safety population, 74.5–98.9% in the efficacy population).

Table 1 Baseline demographics (safety population) and epilepsy characteristics (efficacy population) by specific concomitant antiseizure medication (ASM) at trial drug initiation
Full size table

Overall, most patients completed the trials across subgroups by specific concomitant ASM. In the safety population, in patients randomized to BRV 50–200 mg/day, the proportion of patients who completed the trials was generally similar to the corresponding proportion of patients in the placebo group on concomitant CBZ (BRV: 92.7% [n = 315], placebo: 93.4% [n = 183]), LTG (90.0% [n = 201], 93.3% [n = 120]), VPA (90.7% [n = 182], 90.3% [n = 93]), OXC (94.2% [n = 137], 96.9% [n = 64]), and LCM (94.4% [n = 72], 94.4% [n = 36]). The completion rate was numerically lower in patients randomized to BRV 50–200 mg/day compared with placebo for patients on concomitant TPM (BRV: 87.6% [n = 89], placebo: 95.2% [n = 84]), PHT (81.1% [n = 74], 93.5% [n = 46]), CLB (86.7% [n = 60], 94.3% [n = 35]), and PHB (85.0% [n = 60], 100% [n = 22]). Generally, in patients on BRV 50–200 mg/day, the main reason for discontinuation was adverse events across all ASM subgroups (Supplementary Table 1).

Efficacy

Efficacy outcomes were consistently higher in patients randomized to BRV (50–200 mg/day) compared with placebo for each subgroup of patients by specific concomitant ASM (Fig. 1). Reduction over placebo in focal-onset seizure frequency was apparent in all subgroups by specific concomitant ASM in patients randomized to BRV, ranging from 11.7% on concomitant OXC to 33.5% on concomitant PHB. These reductions were statistically significant (p < 0.05) for patients taking BRV and concomitant CBZ, LTG, VPA, PHT, CLB, and PHB as part of their treatment regimen (Fig. S1).

Fig. 1
figure 1

Efficacy by specific concomitant antiseizure medications at trial drug initiation: a median percentage reduction from baseline in focal-onset seizures per 28 days, b 50% responder ratea, and c seizure freedom (all seizure types)b (efficacy population). aFor the 50% responder rates (patients with a ≥ 50% reduction from baseline in focal-onset seizure frequency), p values are for odds ratios of brivaracetam versus placebo. bPatients were classified as seizure free if they completed the treatment period, did not report seizures of any type (including focal-onset seizures, generalized seizures, and unclassified seizures), and there were no missed days in their seizure diary. BRV brivaracetam, CBZ carbamazepine, CI confidence interval, CLB clobazam, LCM lacosamide, LTG lamotrigine, OXC oxcarbazepine, PBO placebo, PHB phenobarbital, PHT phenytoin, TPM topiramate, VPA valproate

Full size image

In all subgroups by specific concomitant ASM, the median percentage reduction from baseline in focal-onset seizure frequency per 28 days was consistently numerically greater in patients randomized to BRV (range 25.5–42.8%) compared with placebo (range 4.4–21.2%). These differences were significantly greater (p < 0.05) in patients randomized to BRV compared with placebo in those taking concomitant CBZ, LTG, VPA, OXC, TPM, CLB, and PHB (Fig. 1a).

In all subgroups by concomitant ASM, the 50% responder rates were consistently numerically higher in patients randomized to BRV (range 31.9–44.9%) compared with placebo (range 11.4–25.2%). The 50% responder rates (p values for odds ratios of BRV versus placebo) were significantly higher for BRV versus placebo (p < 0.05) in patients taking concomitant CBZ, VPA, OXC, TPM, and CLB (Fig. 1b).

Seizure freedom (all seizure types) during the 12-week evaluation period (from the first day of study treatment) was also consistently numerically higher in patients randomized to BRV compared with placebo in all subgroups by concomitant ASM. In patients randomized to BRV, the seizure freedom rates ranged from 1.4% (patients on concomitant PHT) to 12.5% (patients on concomitant LCM; p < 0.05). In patients randomized to placebo, the seizure freedom rate was 0% in patients on concomitant CBZ, LTG, OXC, PHT, LCM, CLB, and PHB, 1.1% on concomitant VPA, and 1.2% on concomitant TPM (Fig. 1c).

Tolerability

The overall incidence of TEAEs was generally similar across subgroups by specific concomitant ASMs in patients randomized to BRV 50–200 mg/day (range 60.8–74.5%) or placebo (range 53.8–66.7%) (Table 2). Drug-related TEAEs were numerically higher across all subgroups by concomitant ASM in patients randomized to BRV 50–200 mg/day (range 35.2–48.3%) compared with placebo (range 23.9–37.1%). Across subgroups by concomitant ASM, discontinuations due to TEAEs ranged from 2.9% to 13.3% in patients randomized to BRV 50–200 mg/day, and from 0% to 5.7% for patients taking placebo; serious TEAEs ranged from 1.6% to 5.0% on BRV and from 0% to 5.7% on placebo; severe TEAEs ranged from 3.2% to 11.7% on BRV and from 2.5% to 9.1% on placebo. No deaths were reported among the patients included in this analysis.

Table 2 Incidence of treatment-emergent adverse events (TEAEs) with onset during the treatment period by specific concomitant antiseizure medication (ASM) at trial drug initiation (safety population)
Full size table

The most frequently reported TEAEs (reported by at least 10% of patients in any subgroup) were somnolence, dizziness, headache, and fatigue with BRV; and somnolence, dizziness, headache, fatigue, and nasopharyngitis with placebo. In patients on BRV, a numerically higher incidence of somnolence in patients on concomitant TPM, headache on concomitant OXC or PHB, and fatigue on concomitant LCM, compared with the other concomitant ASM subgroups was observed (Table 2). The incidence of somnolence in patients randomized to BRV 50–200 mg/day ranged from 9.9% (VPA subgroup) to 22.5% (TPM subgroup), and from 4.5% (PHB subgroup) to 11.1% (LCM subgroup) with placebo. The incidence of dizziness in patients randomized to BRV ranged from 6.7% (CLB subgroup) to 12.5% (LCM subgroup), and from 2.9% (CLB subgroup) to 10.9% (OXC subgroup) in patients randomized to placebo. In patients randomized to BRV, the incidence of headache ranged from 5.6% (TPM subgroup) to 15.3% (OXC subgroup), and from 4.3% (VPA and PHT subgroups) to 22.7% (PHB subgroup) in patients randomized to placebo. The incidence of fatigue in patients randomized to BRV ranged from 3.3% (CLB subgroup) to 18.1% (LCM subgroup), and from 0% (PHB subgroup) to 11.1% (LCM subgroup) in patients randomized to placebo.

Discussion

The results of this post hoc analysis demonstrate that BRV (50–200 mg/day) was efficacious and generally well tolerated regardless of the concomitant ASMs used as part of their treatment regimen. These results are supported by the similarity of effect observed for the efficacy and tolerability outcomes across subgroups by specific concomitant ASM. A reduction in seizures with adjunctive BRV (50–200 mg/day) treatment was observed regardless of the concomitant ASM used, including ASMs of first (CBZ, VPA, PHT, CLB, and PHB), second (LTG, OXC, and TPM), and third generation (LCM). Notably, there were no differences in efficacy or tolerability with BRV in combination with sodium channel blockers (e.g., PHT, CBZ, OXC, LTG, LCM) and non-sodium channel blockers (e.g., VPA, TPM, PHB, CLB). Efficacy outcomes were consistently higher in patients randomized to BRV compared with placebo across all ASM subgroups assessed, and no specific trends were observed across the outcomes measured.

Some differences in the level of response to BRV and placebo were noted in the subgroups by specific concomitant ASMs. Previous studies have suggested that the number of lifetime ASMs, baseline seizure frequency, and age could affect the response to BRV [9,10,11]. Older age, lower baseline seizure frequency, and fewer lifetime ASMs have been reported as independent predictors of 12-month seizure freedom [11]. In this analysis, although the patient demographics of each subgroup differed (in terms of age, number of previous ASMs, and baseline seizure frequency), the efficacy and tolerability of BRV was generally similar across the groups. Across all ASM subgroups, the median number of previous ASMs was at least two (representing a drug-resistant population). Despite most patients being on two concomitant ASMs at the time of trial enrollment, these patients still experienced greater seizure reductions on BRV compared with placebo. This highlights the unique mechanism of action of BRV which can provide seizure reduction in patients despite various levels of intractability and current use of ASMs with various mechanisms of action.

The efficacy response during adjunctive BRV treatment in patients on concomitant CBZ, LTG, OXC, or VPA (independent of the total number of concomitant ASMs) was consistent with the efficacy response observed in a previous post hoc analysis (data pooled from the same trials) in patients on a single concomitant ASM [8]. This further supports previous findings that demonstrated BRV efficacy regardless of the number of concomitant ASMs to which BRV was added (one or two) [8].

Adjunctive BRV (50–200 mg/day) was generally well tolerated in subgroups of patients by specific concomitant ASM at trial initiation. The overall incidences of TEAEs, drug-related TEAEs, and discontinuation due to TEAEs in the subgroups by specific concomitant ASM in this analysis were generally similar to that observed in the overall pooled safety population [12], and were also similar in patients receiving CBZ, LTG, VPA, and OXC in this analysis to the same tolerability outcomes in the pooled analysis of patients receiving only one specific concomitant ASM and BRV [8]. This confirms that regardless of the type and number of concomitant ASMs, BRV is well tolerated.

Overall, the incidences of the most common TEAEs reported during adjunctive BRV (50–200 mg/day) treatment in the subgroups by specific concomitant ASMs (somnolence [9.9–22.5%], dizziness [6.7–12.5%], headache [5.6–15.3%], and fatigue [3.3–18.1%]) were consistent with the incidences reported for the overall safety population of these studies (somnolence 15.2%, dizziness 11.2%, headache 9.6%, and fatigue 8.7%) [12].

Since most of the patients in this analysis were on two concomitant ASMs at trial initiation, the observed numerically higher incidences of somnolence with BRV and concomitant TPM, headache with concomitant OXC or PHB, and fatigue with concomitant LCM, compared with the other concomitant ASM subgroups, are difficult to attribute to a specific ASM combination. No additional safety concerns were identified in this pooled analysis. The lack of additional safety concerns may in part relate to the unique mechanism of action and lack of significant drug–drug interactions of BRV with other ASMs with differing mechanisms of action.

Limitations

This was a post hoc exploratory analysis and, as such, the analyses were not powered to detect statistically significant differences in the efficacy or tolerability of BRV by specific concomitant ASMs. Although the most frequent concomitant ASMs were selected, the sample size was relatively small in some subgroups.

Conclusion

BRV (50–200 mg/day) provided additional efficacy and was well tolerated in patients with focal-onset seizures on concomitant CBZ, LTG, VPA, OXC, TPM, PHT, CLB, LCM, or PHB (as part of their treatment regimen) supporting the use of BRV as adjunctive therapy across a broad range of ASMs.