Send to

Choose Destination
Epilepsia. 2018 Aug;59(8):1557-1566. doi: 10.1111/epi.14516. Epub 2018 Jul 16.

Genome-wide association study: Exploring the genetic basis for responsiveness to ketogenic dietary therapies for drug-resistant epilepsy.

Author information

Department of Clinical and Experimental Epilepsy, UCL Institute of Neurology, London, UK.
UCL Great Ormond Street Institute of Child Health, London, UK.
NIHR University College London Hospitals Biomedical Research Centre, UCL Institute of Neurology, London, UK.
Chalfont Centre for Epilepsy, Chalfont St Peter, UK.
European Molecular Biology Laboratory, Wellcome Genome Campus, European Bioinformatics Institute, Cambridge, UK.
Wellcome Genome Campus, Congenica Ltd, Cambridge, UK.
Children's Neurosciences Centre, Guy's and St Thomas' NHS Foundation Trust, London, UK.
Department of Paediatrics, The University of Melbourne, Royal Children's Hospital, Melbourne, Vic., Australia.
Murdoch Children's Research Institute, Melbourne, Vic., Australia.
Epilepsy Research Centre, Department of Medicine, The University of Melbourne, Austin Health, Melbourne, Vic., Australia.
Austin Health, Florey Institute of Neurosciences and Mental Health, Melbourne, Vic., Australia.
Stichting Epilepsie Instellingen Nederland (SEIN), Heemstede, The Netherlands.
Great Ormond Street Hospital for Children, London, UK.
Young Epilepsy, Lingfield, UK.



With the exception of specific metabolic disorders, predictors of response to ketogenic dietary therapies (KDTs) are unknown. We aimed to determine whether common variation across the genome influences the response to KDT for epilepsy.


We genotyped individuals who were negative for glucose transporter type 1 deficiency syndrome or other metabolic disorders, who received KDT for epilepsy. Genotyping was performed with the Infinium HumanOmniExpressExome Beadchip. Hospital records were used to obtain demographic and clinical data. KDT response (≥50% seizure reduction) at 3-month follow-up was used to dissect out nonresponders and responders. We then performed a genome-wide association study (GWAS) in nonresponders vs responders, using a linear mixed model and correcting for population stratification. Variants with minor allele frequency <0.05 and those that did not pass quality control filtering were excluded.


After quality control filtering, the GWAS of 112 nonresponders vs 123 responders revealed an association locus at 6p25.1, 61 kb upstream of CDYL (rs12204701, P = 3.83 × 10-8 , odds ratio [A] = 13.5, 95% confidence interval [CI] 4.07-44.8). Although analysis of regional linkage disequilibrium around rs12204701 did not strengthen the likelihood of CDYL being the candidate gene, additional bioinformatic analyses suggest it is the most likely candidate.


CDYL deficiency has been shown to disrupt neuronal migration and to influence susceptibility to epilepsy in mice. Further exploration with a larger replication cohort is warranted to clarify whether CDYL is the causal gene underlying the association signal.


CDYL ; biomarker; genetics; high-fat; low-carbohydrate

Supplemental Content

Full text links

Icon for Wiley Icon for PubMed Central
Loading ...
Support Center