The study oversight committees unanimously decided to stop the study early on 26/07/2017 due to poor recruitment.

The investigators are undertaking the first European Randomised Controlled Trial (RCT) for epilepsy surgery in children with FCD type II, to prospectively evaluate the role of the KD prior to surgery in improving seizure outcome. The investigators will evaluate the role of KD as a disease-modifying treatment to achieve seizure control and improve neurodevelopment and quality of life. Children age 3 - 15 years with pharmacoresistant epilepsy believed to be the result of focal cortical dysplasia type II, considered to be surgically treatable, will be randomised to either receive 6m treatment presurgery with a ketogenic diet, or to proceed direct to surgery (no pretreatment). Primary outcome will be the time to achieve a period of 6 months of seizure freedom from the date of randomisation. Tissue resected at surgery will also be evaluated with regard to the degree of any methylation of DNA.

Epilepsy surgery is now an accepted and effective management for individuals with drug resistant focal onset epilepsy in carefully selected candidates. This aside, only two RCTs have been performed in adults in temporal lobe epilepsy [1,2], and none in children . Malformations of cortical development are the most common pathology responsible for drug resistant focal epilepsy in childhood [3}. In retrospective studies, successful surgical resection resulting in seizure freedom in malformations of cortical development has been reported in 42-87% of cases [4], dependent on the series reviewed and the completeness of resection. Outcome is related to extent and completeness of resection, but the extent of tissue to be removed remains a challenge, especially in very young patients, where the limits of a lesion may be unclear and eloquent cortex may be involved.

FCD is one specific type of malformation, of localised areas of abnormal cortex that may be subdivided into I, II and III dependent on the pattern determined and cell types involved. This is by definition a histological diagnosis as was recently classified by the International League Against Epilepsy [5]. Specifically FCD type II can usually be suggested by characteristic changes on an MRI scan. These changes include increased cortical thickness, blurring of the cortical-white matter junction, increased signal on T2-weighted images, a radially oriented linear or conical transmantle strip of T2 hyperintensity, cortical thinning and localized brain atrophy. FCD type II may be subdivided histopathologically into IIa and IIb but it can be difficult to distinguish between the two radiologically.

The KD is a high fat diet with a proven efficacy in the treatment of drug resistant epilepsy in children [6]. An RCT performed in children aged 2-16 years diagnosed with drug resistant epilepsy utilising either a classical or medium chain triglyceride KD, demonstrated >50% reduction in seizures in 38% of participants after 3 months compared to 6% where there was no change in standard treatment, with no difference seen between the two diets [7,8]. A further RCT utilising a more relaxed approach of the KD, the modified Atkins diet, demonstrated similar results [9]. One open label study suggested particular efficacy in children with epilepsy due to FCD [10].

Clinical trials of epigenetic pharmacological treatment are promising and already have been approved for cancer [11]. Intriguingly, the most commonly used AED is valproic acid (VPA), in which histone deacetylase (HDAC) inhibitor activity was discovered in 2001 [12]. The HDAC antagonizing effect of VPA is, however, considerably low compared to second generation HDAC inhibitors suberoylanilide hydroxamic acid (SAHA, Zolinza, Vorinostat), LBH598 or Romidepsin [11]. Unfortunately, all HDAC inhibitors bear the risk of severe side effects when systemically administered during pregnancy. Nevertheless, there is also evidence that the epigenetic machinery can be modified by nutrition and dietary concepts [13]; such a dietary concept is the KD. Stimulus-induced DNA methylation changes have been identified in postnatal brain [14], thus it can be anticipated that DNA methylation modifications contribute to the molecular memory of postmitotic neurons also in the epileptogenic network. We propose that therapies such as the KD that are directly or indirectly targeting the epigenetic machinery could be helpful to prevent, delay or retard drug-resistant epilepsy.

Recent experimental data in an animal model has shown that the classical KD attenuates epigenetic chromatin modifications (i.e. DNA methylation), a master regulator for gene expression and functional adaptation of the cell, thereby modifying disease progression [15]. This hypothesis suggests that epigenetic mechanisms play a pivotal role in epileptogenesis and that seizures can by themselves induce epigenetic chromatin modifications, aggravating the epileptogenic condition [15].

We will conduct the first RCT for epilepsy surgery in children with FCD type II, to prospectively evaluate the role of the KD prior to surgery.

120 children aged 3 - 15 years with a diagnosis of Focal Cortical Dysplasia (FCD) type II a or b (with consistent MRI changes), treatment failure of at least two anti-epileptic drugs (AEDs) in controlling continuing seizures, with seizure semiology consistent with focal onset agreed to be surgically treatable through FCD resection and continuing seizures for less than 5 years, will be included in the trial after parental/ legal representative consent. Children will be excluded if they have a history of less than two seizures in 6 months prior to randomisation or they have previously used the Ketogenic diet (KD) or administration of the KD is medically contraindicated. Patients will be recruited from 19 participating European sites in 8 countries undertaking ketogenic diet and epilepsy surgery in children.

Participants will be followed-up for minimum 24 months/maximum 48 months (depending on the timing of randomisation) after randomisation. Primary outcome will be time to 6 month remission (i.e. the time to achieve a period of 6 months of seizure freedom from the date of randomisation). Tissue removed at surgery will be assessed for DNA methylation.](streamdown:incomplete-link)

Inclusion Criteria:

1. Children aged 3 - 15 years;
2. MRI changes consistent with a diagnosis of FCD type II a or b;
3. History of at least two epileptic seizures in the past 6 months before randomisation;
4. Seizure semiology consistent with focal onset, agreed after pre-surgical discussion to be surgically treatable;
5. Parent/ legal representative willing to give consent.

Exclusion Criteria:

1. Previous use of the KD;
2. Not a surgical candidate for FCD resection;
3. Administration of the KD is medically contraindicated.