Protocol Amendment 1 was finalized on 04 Apr 2008, before any subjects were enrolled, and resulted in the following: • Replacement of the option for enrollment in N01199 and N01125 with the option of enrollment in N01315. This change was based on feedback received from ethics committees and regulatory authorities in some European countries. • Revision of the methods for handling missing data in acknowledgement of the fact that no details were known regarding the way missing data were handled, if at all, in the historical-control studies. Thus, the application of any imputation rule for the assessment of Exit Criterion 2 during the study (and implemented in the study EDC system) leading to subjects being prematurely or incorrectly exited from the study was avoided. Final sensitivity analyses were to include these revised methods. The possibility for the Investigator exiting subjects on the basis of Exit Criterion 4 was considered an adequate safeguard against subjects remaining too long in the study if missing data interfered with the assessment of Exit Criterion 1 or 2.

Protocol Amendment 1 was finalized on 04 Apr 2008, before any subjects were enrolled, and resulted in the following: • Addition of text describing randomization stratification by region to provide clarity with regard to the percentage of the subjects anticipated for enrollment in the US. • Clarification that the total duration of the Baseline Period was 8 weeks +/-1 week. • Replacement of exclusion criteria referring to “clusters” of seizures with “seizure patterns being too frequent or indistinctively separated to reliably be counted” in order to more clearly define the term “clusters.” • Addition of a recommendation for the sites to call subjects weekly to promote good completion of the subject diary. Text referring to this recommendation was added to the Informed Consent. • Replacement of the interactive voice response system (IVRS) with EDC for screening subjects. This change was made in accordance with recommendations of the IVRS and Seizure Frequency EDC system provider. • Replacement of the subject’s daily record card (DRC) with the CRF as the location for recording the “Investigator Seizure Assessment.” • Correction of typographical errors.

Protocol Amendment 2 was finalized on 02 Nov 2009, and resulted in the following: • Addition of a recruitment hold and interim analysis. This change was motivated by ongoing monitoring that suggested a higher than expected number of subjects discontinuing either for predefined exit criteria or other reasons. The interim analysis was to include efficacy information for all subjects who had an opportunity to complete 112 days of treatment after initiation of concomitant AED tapering by the time of the defined clinical cut-off date. • Establishment of an IDMC to review primary efficacy, sensitivity, and safety data for the purpose of making a recommendation to the Sponsor regarding study continuation. • Elimination of the requirement for restricted database access of the study team before approval of the final Statistical Analysis Plan (SAP) in accordance UCB Standard Operating Procedures (SOPs) revised subsequent to study initiation. Due to this change and because the primary efficacy analysis was fully specified in the original and amended protocols, this requirement was eliminated from this protocol. • Updates of UCB study personnel and the List of Abbreviations. Following the recruitment hold, a decision was made to require subjects who were in the Baseline Period, the BRV Add-On Period, and the AED Tapering Phase to terminate the study. Subjects who had already progressed to the Monotherapy Phase were permitted to remain in the study.

Protocol Amendment 3 was finalized on 16 Nov 2009 (after the recruitment hold) and resulted in the following: • Provision for enrollment in N01315 for subjects discontinued from the BRV Add-On Period or the Baseline AED Tapering Phase due to recruitment hold and interim analysis. This enrollment option was made available to provide continued treatment with BRV to these subjects as deemed beneficial by both subjects and Investigators. • Correction of an error in Amendment 2 that placed information pertaining to database access and SAP finalization in an incorrect section of the protocol.