Protocol Amendment 1, dated 25 Sep 2012, provided the following key changes. The primary purpose of this substantial amendment was to identify significant changes to the study design including inclusion and exclusion criteria, as well as the addition of an independent data monitoring committee (IDMC). In addition, study secondary variables were changed to more accurately represent the study design. • The protocol was updated to include randomization stratified by age at informed consent (≥12 to <18 years of age vs ≥18 years of age). This was in order to maintain balance within each treatment arm and within the existing baseline PGTCS frequency (≤2 per 28 days vs >2 per 28 days for the 16-week Combined Baseline Period prior to randomization) as well as to provide greater control for variability when analyzing the primary efficacy variable (time to second seizure). • An inclusion criterion was added to include study participants of normal intelligence for age in the judgment of the investigator. Exclusion criteria were added or modified to further exclude study participants who had a diagnosis of developmental delay or mental retardation or a history of status epilepticus. The rationale for adding the inclusion/exclusion criteria was that study participants with developmental delay or mental retardation are more likely to have symptomatic rather than idiopathic seizures. • The inclusion criterion regarding the lower limit body weight for male and female study participants ≥12 years of age was changed from ≥ 30kg to ≥ 50kg. This was based on the observations collected to date from pediatric PK/tolerability studies.

- Continuation 1 - • The text regarding the use of benzodiazepines was revised to clarify that up to 3 marketed antiepileptic drugs (AEDs) were allowed for study inclusion provided at least 1 of the 3 AEDs was a benzodiazepine regardless of indication. The intermittent use of a benzodiazepine was permitted for epilepsy indications only and limited to 2 doses per 28 days; intermittent use for non-epilepsy indications was prohibited. The rationale for this change was based on data from SP0961 in which approximately 20% of PGTCS study participants with idiopathic generalized epilepsy (IGE) used benzodiazepines. Study participants were not excluded from the study due to benzodiazepine use unless they were using benzodiazepines intermittently for indications other than epilepsy. This change was consistent with other PGTCS studies in regards to benzodiazepine use. • The secondary efficacy endpoint time to first PGTCS during the 24-week Treatment Period was added as a key secondary endpoint as UCB considers that LCM will improve the rate of seizure freedom in this difficult-to-treat population. A gatekeeping strategy was used to control Type I error. • For the assessment of the effect of LCM on quality of life in pediatric study participants, the SP0982 protocol was updated to include the Pediatric Quality of Life Inventory™ (PedsQL). The PedsQL is widely used in epilepsy and other therapeutic areas, and allows for comparison with other diseases. In addition, the PedsQL also allows for consistency across age groups and development programs, and is available in many languages for global clinical studies. • For the assessment of neurobehavior and cognition in pediatric study participants, the SP0982 protocol was updated to include the Achenbach Child Behavior Checklist (CBCL) and the Behavior Rating Inventory of Executive Function® (BRIEF), respectively. The addition of these assessments also allows for consistency across age groups and development programs.

- Continuation 2 - • To ensure study participant safety, interim assessments for safety and futility were performed using an independent data monitoring committee (IDMC) due to the novel study design and primary endpoint in this patient population. Furthermore, a minority of study participants (~10%) in SP0961 showed an increase in absence seizures that, in this uncontrolled study, could not be distinguished between the drug vs the natural course of the disease and required additional examination in the current study. • SP0982 was an event-driven study where the statistical properties were based upon the number of events, not the number of study participants as is typical in most epilepsy studies (the primary efficacy variable was time to second PGTCS). As a result, the protocol was amended to reflect this focus. Furthermore, a maximum sample size of 250 study participants was introduced if 125 events were not observed on or before the 200th study participant randomized. This is standard for event-driven studies and represents an approximate 25% increase from the projected sample size, in case the event rate was lower than anticipated.

Protocol Amendment 2, dated 11 Jul 2014, provided the following key changes. The primary purpose of this substantial amendment was to identify significant changes to the study design and the inclusion of pediatric study participants (≥4 to <12 years of age). In addition, the key study secondary efficacy variable was changed to more accurately represent the study design. • The protocol was updated to include randomization stratified by age at informed consent for study participants ≥4 to <12 years of age. This was in order to maintain balance within each treatment arm and within the existing baseline PGTCS frequency (≤2 per 28 days vs >2 per 28 days for the 16-week Combined Baseline Period prior to randomization) as well as provide greater control for variability when analyzing the primary efficacy variable (time to second seizure). • An inclusion criterion was modified to include study participants ≥4 years of age. Exclusion criteria were modified in regards to study participants who had a diagnosis of developmental delay or mental retardation or a history of status epilepticus. The rationale for modifying the exclusion criteria was that study participants with developmental delay or mental retardation are more likely to have symptomatic rather than idiopathic seizures. The rationale for modifying the exclusion criteria for study participants with a history of status epilepticus was to align with the withdrawal criteria (ie, a study participant may have been withdrawn from the study due to an episode of status epilepticus, a prolongation of seizure duration, a worsening of seizure frequency, or emergence of a new seizure type considered by the investigator to require intervention).

Protocol Amendment 4, dated 08 Jun 2016, provided the following key changes. The primary purpose of this substantial amendment was to clarify elements of the study design including inclusion and exclusion criteria, exit criteria, and withdrawal criteria, as well as the duration of the Baseline Period, procedure for dividing the daily dose for the tablet formulation without breaking tablets, permitted and prohibited concomitant medication, consent for prescreening electroencephalogram (EEG), seizure count, temperature for storage of plasma samples, and determination of sample size. The protocol was also updated according to the new UCB protocol template, for example, with the addition of text regarding potential drug-induced liver injury (PDILI).

Protocol Amendment 5, dated 07 Nov 2017, provided the following key changes. The primary purpose of this substantial amendment was to stop the study participants’ study participation once 125 events had been observed to avoid exposing study participants to Placebo unnecessarily and allow for flexible dosing of the treatment in the open-label follow-up study EP0012. In addition, the following changes were made: • The initial plan was to enroll 20% pediatric study participants out of a total sample size of approximately 200 study participants, which would have resulted in 40 pediatric study participants enrolled in the study. Due to a fluctuating event rate, the study was changed to enroll study participants until the 125th event occurred (an event was defined as the occurrence of the second PGTCS during the 24-week Treatment Period). Changing the number of pediatric study participants from 20% to 40 absolute ensured that at least 40 pediatric study participants were recruited while also limiting the number of required pediatric study participants in case more than 200 study participants were enrolled. This was considered appropriate in light of the extremely challenging recruitment of pediatric study participants in this study. Additionally, from the biostatistical point of view, increasing the number of pediatric study participants from 40 to 45 or 50 is very unlikely to yield significantly new or different safety information. • To update the introduction with regulatory information on the marketing authorization of Vimpat and to provide an update on the LCM clinical program. • To appropriately align the Inclusion Criterion 7.a with the examples given in the supportive table. • To clarify that Visit 5 was the beginning of the Maintenance Period (18 weeks). • To remove some blood sampling details from the protocol and refer to the laboratory manual.

- Continuation - • To clarify the potential drug-induced liver injury (PDILI) criteria requiring immediate and permanent discontinuation of the investigational medicinal product (IMP). • To clarify that if the monitor had no direct access to the source electronic medical records, certified copies should have been generated by the investigator and verified by the monitor against the original medical record. • To make a minor clarification to the sentence about pooling strategies for age strata for consistency with the Statistical Analysis Plan (SAP).