Complete Thesis

University of Groningen

Genotyping and phenotyping epilepsies of childhood Vlaskamp, Danique

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record

Publication date: 2018

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA): Vlaskamp, D. (2018). Genotyping and phenotyping epilepsies of childhood. Rijksuniversiteit Groningen.

Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license. More information can be found on the University of Groningen website: https://www.rug.nl/library/open-access/self-archiving-pure/taverne- amendment.

Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

Download date: 02-10-2021 Genotyping and phenotyping epilepsies of childhood

Danique Vlaskamp

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 1 Layout Douwe Oppewal Cover design Christien Rondaan Printed by Ipskamp Printing, Enschede, the Netherlands ISBN 978-94-034-1144-6 (printed version) 978-94-034-1143-9 (e-book)

© Copyright 2018 D.R.M. Vlaskamp, Groningen, the Netherlands All rights reserved. No part of this publication may be reproduced or transmitted in any forms or by any means, without prior written permission of the author. This research was funded by the Junior Scientifi c Masterclass. Unrestricted travel grants were received from the Dutch Society of Child Neurology, Jo Kolk Foundation (VVAO), the Research School of Behavioural and Cognitive Neurosciences (BCN). Conference attendances were fi nancially supported by the BCN. Printing of this thesis was fi nancially supported by the the University of Groningen and the BCN.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 2 Genotyping and phenotyping epilepsies of childhood

Proefschrift

ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen op gezag van de rector magniﬁcus prof. dr. E. Sterken en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

woensdag 05 december 2018 om 14.30 uur

door

Danique Rienke Maria Vlaskamp

geboren op 15 mei 1991 te Raalte

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 3 Promotores Prof. dr. C.M.A. van Ravenswaaij-Arts Prof. dr. O.F. Brouwer

Copromotores Dr. P.M.C. Callenbach Dr. P. Rump

Beoordelingscommissie Prof. dr. V.V.A.M. Knoers Prof. dr. H.P.H. Kremer Prof. dr. K.P.J. Braun

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 4 COVER

The tree on the cover of this thesis symbolizes the relationship between the genotype and the phenotype of epilepsies of childhood. The beautiful tree has deeply anchored, widely spreading roots (genotype) that grow a new ﬂowering tree with a unique pattern of branches, leaves and blossoms (phenotype).

Painted by Christien Rondaan

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 5 6

Paranimfen Nienke te Grootenhuis Wieke Eggink

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 6 7

CONTENTS

Chapter 1 Introduction 9

PART I - THE YIELD OF GENETIC TESTING

Chapter 2 Copy number variation in a hospital-based cohort of children with epilepsy 27 Published as: Epilepsia Open. 2017; 2: 244–254. Chapter 3 Empowerment and anxiety of patients and parents 49 during genetic counseling for epilepsy Manuscript ready to submit

PART II - GENOTYPE-PHENOTYPE STUDIES

Chapter 4 SYNGAP1 encephalopathy: a distinctive generalized 71 developmental and epileptic encephalopathy Accepted for publication in Neurology Chapter 5 Genotype-phenotype correlation of 248 individuals with 95 GRIN2A-related disorders identifies two distinct phenotypic subgroups associated with different classes of variants, protein domains and functional consequences Accepted for publication in Brain Chapter 6 Schizophrenia is a later-onset feature of PCDH19 Girls Clustering Epilepsy 123 Revised version under review Chapter 7 Haploinsufficiency of the STX1B gene is associated with 145 myoclonic astatic epilepsy Published as: European Journal of Paediatric Neurology. 2016; 20: 489-492 Chapter 8 PRRT2-related phenotypes in patients with a 16p11.2 deletion 157 Published as: European Journal of Medical Genetics. 2018, in press. Chapter 9 Positive effect of sodium channel blocking anti-epileptic 177 drugs on neonatal infantile epilepsy due to a SCN2A mutation Translated from the Dutch paper published as: Epilepsie. 2017; Periodiek voor Professionals 15, nr.3

Chapter 10 General discussion and a proposal for a diagnostic algorithm 185 for genetic testing for epilepsy

Chapter 11 Nederlandse samenvatting 201

Dankwoord 208 Curriculum vitae 213

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 7 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 8 9

Chapter 1

Introduction

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 9 10 CHAPTER 1

I am about to discuss the disease called “sacred.” It is not, in my opinion, any more divine or more sacred than any other diseases, but has a natural cause…

Its origin, like that of other diseases, lies in heredity... The fact is that the cause of this affection is... the brain... My own view is that those who first attributed a sacred character to this malady were like the magicians, purifiers, charlatans, and quacks of our own day... So that there is no need to put the disease in a special class and to consider it more divine than the others... Each has a nature and a power of its own; none is hopeless or incapable of treatment.1

Hippocrates, 460–370 BC, on epilepsy

Hippocrates was revolutionary for his time in arguing that the origin of epilepsy lies in ‘heredity’. Although hereditary does not always equal genetic, we now have robust evidence that genetic variants can cause or contribute to epilepsy. However, we still have a long way to go to fully understand the intriguing role of genetics in epilepsy.

To comprehend the current challenges in epilepsy genetics addressed in this thesis, it is helpful to learn more about epilepsy phenotyping and genotyping. Whereas epilepsy phenotyping concentrates on describing the clinical presentation, epilepsy genotyping focuses on identifying the underlying genetic cause. Integration of information on epilepsy phenotype and genotype is important in clinical practice and research settings in order to ﬁnd new genes and risk factors for epilepsy, to better understand the disease and its presentation and to improve its management and outcome.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 10 INTRODUCTION 11

EPILEPSY PHENOTYPING 1 An epileptic seizure is deﬁned as a transient occurrence of signs and/or symptoms due to abnormal excessive or synchronous neuronal activity in the brain.2 Seizures can be provoked by acute disturbances in the body, such as hypoglycemia or acute infection, or can occur unprovoked. Patients are diagnosed with epilepsy if they present with any of the following three conditions: 1. at least two unprovoked seizures occurring >24 hours apart, 2. one unprovoked (or reﬂex) seizure and a probability of further seizures in the next 10 years similar to the general recurrence risk after two unprovoked seizures (at least 60%), or 3. a diagnosis of an epilepsy syndrome.3 Epilepsy is diagnosed in 41-87 children per 100,000 children each year.4

Epilepsy is not a single disease entity, but should be considered as a group of disorders. Its presentation, etiology, management and prognosis can vary significantly for the different epilepsies. Epilepsy therefore warrants further classification for clinical and research purposes. The International League Against Epilepsy (ILAE) has made several classification guidelines.5–11 In the most recent guideline from 2017, they recommended classifying epilepsy at three consecutive levels: 1. seizure type, 2. epilepsy type, and 3. epilepsy syndrome.12 The flowchart in Figure 1 demonstrates how to classify epilepsies using this three-step method. Depending on the availability of clinical information, epilepsy can be classified up to level 1, 2 or 3.

1. Seizure type. Epileptic seizures should be classified based on their onset as focal, generalized or unknown.12,13 Epilepsy is a disorder involving neuronal networks and all seizures are hypothesized to occur somewhere within a network.14,15 Focal seizures are limited to networks within one hemisphere, while generalized seizures rapidly engage bilaterally distributed networks.16 Both clinical and electroencephalogram (EEG) data can help to distinguish focal and generalized seizures. If more detailed clinical information on the seizures is available, they can be further classified based on the level of awareness (for focal seizures only) and by describing their first feature.12,13 Focal seizures can evolve to bilateral tonic-clonic seizures (previously called ‘secondary generalized seizures’).12,13,17 2. Epilepsy type. Since individuals can have both focal and generalized seizures, a second level of classification has been introduced: epilepsy type. Epilepsy type can be focal, generalized or combined generalized and focal.12 3. Epilepsy syndrome. To diagnose the epilepsy syndrome, one should also consider the setting in which seizures occur. This setting includes the patient’s gender, age at seizure onset, developmental course, family history, and results from additional investigations such as EEG and imaging. Recognition of a distinctive clinical and EEG pattern allows a syndrome diagnosis.12

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 11 12 CHAPTER 1 z & / z d > ^ ^ z E / W E h > Z ' / t E K K h d / > K / K h d d > 12,13 / W E E D ^ K < d h / z & D E Z ' d / E d h D / ^ , 2017. t d > Z K d ^ / E z K ^ D ^ t D K E W K Ͳ K ƌƌĞƐƚ Z E E E t > ĞŚĂǀŝŽƌĂů / ' K < K W E E E / E z h < ^ E Z hŶŬŶŽǁŶ Žƌ h K d ƵŶĐůĂƐƐŝĨŝĂďůĞ ƐĞŝǌƵƌĞ K ƐƉĂƐŵƐ D ͬĞƉŝůĞƉƚŝĐ dŽŶŝĐͲĐůŽŶŝĐ Z > K d ^ K ' K E D / & Ͳ Z E ĂďƐĞŶĐĞͬ ŵǇŽĐůŽŶŝĂ h E K E / / E dǇƉŝĐĂů ŽƌĂƚǇƉŝĐĂů ŵǇŽĐůŽŶŝĐ ͬĞǇĞůŝĚ & Z ^ , / d > D / K > K Z W Z ZĞƐƵůƚƐĨƌŽŵ ŝŶǀĞƐƚŝŐĂƚŝŽŶƐ W Z z Z E z d ďŽƚŚ ŚĞŵŝƐƉŚĞƌĞƐ ;ŝŵĂŐŝŶŐ͕ĞůĞĐƚƌŽĞŶĐĞƉŚĂůŽŐƌĂŵͿ K E E d z d ' z ^ K ^ ƐƉĂƐŵ Z ' W z D ŽƚŚĨŽĐĂůĂŶĚ ŐĞŶĞƌĂůŝǌĞĚ ƐĞŝǌƵƌĞƐ h ^ > / W / W ^ĞŝǌƵƌĞ ǁŝƚŚŝŶǀŽůǀĞŵĞŶƚŽĨŶĞƚǁŽƌŬƐŝŶ > ͬŵǇŽĐůŽŶŝĐ ͬŵǇŽĐůŽŶŝĐͲ ƚŽŶŝĐͲĐůŽŶŝĐ ͬŵǇŽĐůŽŶŝĐͲ / ĂƚŽŶŝĐ ͬĂƚŽŶŝĐ ͬĞƉŝůĞƉƚŝĐ dŽŶŝĐͬĐůŽŶŝĐ ͬƚŽŶŝĐͲĐůŽŶŝĐ W ͘ ^ ͘ Ϯ ϭ ͘ ϯ Ğ ƌ Z Ƶ K Z ǌ ŝ d d Ğ K Ɛ y t / Đ D ŝ > Ͳ d Ŷ d ƐĞŶƐŽƌǇ E E Ž ů Z Z K ƵƚŽŶŽŵŝĐ ͬ K Đ K Ͳ E h E ďĞŚĂǀŝŽƌ ĂƌƌĞƐƚͬ Đ E ŝ / Ŷ ĐŽŐŶŝƚŝǀĞ ͬĞŵŽƚŝŽŶĂů ͬ Ž ' ƚ ^ ů ĚĞǀĞůŽƉŵĞŶƚ͕ŝŶƚĞůůĞĐƚ͕ > Ă ƌ ĐŽŵŽƌďŝĚŝƚŝĞƐ͕ĨĂŵŝůǇŚŝƐƚŽƌǇ 'ĞŶĚĞƌ͕ĂŐĞ ĂƚƐĞŝǌƵƌĞ ŽŶƐĞƚ͕ Ğ KŶůǇŐĞŶĞƌĂůŝǌĞĚ ƐĞŝǌƵƌĞƐ ƚ K Ă ů Z ŝ & ŝŶŽŶĞ ŚĞŵŝƐƉŚĞƌĞ Z K ď d Ž K ƚ t ů D Ă Đ ŚǇƉĞƌŬŝŶĞƚŝĐ ͬ Ž ĐůŽŶŝĐ ͬƐƉĂƐŵƐ ͬ ^ĞŝǌƵƌĞ ǁŝƚŚŝŶǀŽůǀĞŵĞŶƚŽĨŶĞƚǁŽƌŬƐ ŵǇŽĐůŽŶŝĐ ͬƚŽŶŝĐ & ƵƚŽŵĂƚŝƐŵ ͬĂƚŽŶŝĐ ͬ > z ƚ Ɛ Ğ Ğ ^ Ɛ ƌ Ğ Ɛ ƌ K Ɛ W Ğ Ƶ ƚ W Ƶ & Ŷ Ŷ Ă z Ž > Ğ K / ƌ Ğ d W Ă Ĩ ƚ ǁ Ɛ KŶůǇĨŽĐĂůƐĞŝǌƵƌĞƐ

ƌ ŝ

z d / / Z K D K z & / d E / ^ z t > Flowchart for the classification of epilepsies following a three step method. Modified after Fisher et al. 2017 & Scheffer et al. et al. 2017 & Scheffer et al. Modified after method. Fisher step a three the classification of epilepsies following 1: Flowchart for Figure

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 12 INTRODUCTION 13

Comorbidities In parallel with classifying the epilepsy from seizure type to epilepsy syndrome, it is important 1 to identify comorbidities. Examples of comorbidities include intellectual disability, movement disorders, autism spectrum disorders and other psychiatric and behavioral disorders. Pro-active screening for the presence of comorbidities is vital because it not only helps in epilepsy syndrome and etiology diagnosis, it also enables appropriate management of comorbidities that can improve quality of life of patients and their families.

Etiology During the whole process of phenotyping epilepsies, it is also essential to try to identify the etiology. Understanding etiology can prevent unnecessary further diagnostic investigations and may lead to better counseling, management and, possibly, outcome. Six epilepsy etiology groups have been proposed: genetic, structural, infectious, metabolic, immune and unknown.12 For some epilepsies, etiology can be classiﬁed into more than one etiology category. For example, pyridoxine-dependent epilepsy due to a pathogenic ALDH7A1 variant has a metabolic and genetic etiology while DEPDC5- gene-related epilepsy with a cortical malformation has a genetic and structural etiology. For the genetic etiology, epilepsies can be considered genetic based on the presenting phenotype or family history even when the variant itself has not yet been identiﬁed.12

Developmental and epileptic encephalopathy In some epilepsies, “the epileptic activity itself may contribute to severe cognitive and behavioral impairments that is beyond what might be expected from the underlying pathology alone”11, defined as an epileptic encephalopathy (EE). There are many different EEs that can occur in infancy and childhood, and these are often genetic.18 However, developmental slowing and regression often precede seizure onset and cognitive deficits can persist after seizure remission. Therefore, the term developmental and epileptic encephalopathy (DEE) was introduced to describe the encephalopathies where the underlying genetic condition and the seizure activity together lead to developmental plateauing or regression.12 An example of a DEE is SYNGAP1-encephalopathy, which is studied in this thesis.

EPILEPSY GENOTYPING

Although the role of genetics in epilepsies has been appreciated since antiquity, research on this role only started the last century and the growth in knowledge since then has been revolutionary. What follows is a historical overview of the diﬀerent consecutive study types that have led to our current understanding of the complex role of genetics in epilepsies.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 13 14 CHAPTER 1

Family studies The first evidence for epilepsy as a hereditary disease came from clinical family studies performed in the end of the last century. Family aggregation studies showed a 3-7 times higher incidence of epilepsy in siblings and offspring of patients with epilepsy than expected by chance.19–21 Of patients with epilepsy, 5-21% have at least one first-degree relative sharing a diagnosis of epilepsy.22–25 Furthermore, twin studies revealed a higher concordance for epilepsy among monozygotic versus dizygotic twins.26–28 These studies together indicated that epilepsy has a genetic basis, but its nature and extent are unknown. Nowadays, family studies are mainly performed to evaluate the phenotypic spectrum of familial epilepsy syndromes, e.g. genetic epilepsy with febrile seizures plus29, or to identify new familial syndromes, e.g. familial posterior quadrant epilepsies.30

Searching for epilepsy candidate genes Following these family studies, diﬀerent approaches were used to identify the epilepsy candidate genes and their position within the human genome (called locus). Two important study types, linkage analysis and association studies, have been performed since the eighties and nineties of the last century, respectively.

Linkage studies. Linkage analysis can be used in families to evaluate the segregation of epilepsy with genetic markers with a known position in the genome. This technique is based on the principle that genes that reside closely together are linked during meiosis. By identifying the genetic marker that is carried by affected relatives but not by non-affected relatives, the linked gene or locus for epilepsy can be identified. The first locus for epilepsy (juvenile myoclonus epilepsy on chromosome 6) was identified in 1989, and this was followed by the discovery of many other loci for this and other epilepsy syndromes.31,32 Linkage analysis is still used to identify new loci for epilepsy.32

Association studies. Association studies aim at identifying alleles that increase the risk for developing epilepsy, called ‘susceptibility alleles’. In these studies, the statistical association between a specific allele and an epilepsy syndrome is calculated by comparing the presence of this allele in large numbers of patients versus controls. Alleles found significantly more often in patients are considered susceptibility alleles. The many association studies performed to date have consistently identified a few susceptibility loci for different generalized epilepsies (1q43, 2p16.1, 2q22.3 and 17q21.32), focal and generalized epilepsy and febrile seizures (2q24.3 including SCN1A), febrile seizures (2q24.3, 11p14.2 and 12q21.33) or focal and generalized epilepsy (4p15.1).33 Focusing on epilepsy treatment, an HLA-B*1502 allele has been associated with the occurrence of carbamazepine-induced Stevens-Johnson syndrome.34 Unfortunately, many other findings from association studies could not be replicated, mainly due to low statistical power. Future collaboration between epilepsy research groups is necessary to increase the number of participants and thereby increase the statistical power to identify new loci or genes for epilepsy or drug resistance or efficacy.33,35,36

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 14 INTRODUCTION 15

Molecular genetic studies Advances in molecular genetic techniques that disentangle the human genome in greater detail 1 together with more adequate epilepsy phenotyping have resulted in the identiﬁcation of many new genes for epilepsy. Table 1 gives an overview of the diﬀerent molecular techniques that are currently available to detect genetic variants at all levels of the human genome. Here, we will discuss the three most important techniques in terms of identifying new genes and loci for epilepsy.

Sanger sequencing. In 1995, the first disease-causing gene variant for epilepsy (previously called a mutation) was identified using Sanger sequencing in a locus identified by linkage analysis. A missense mutation in the CHolinergic Receptor Nicotinic Alpha 4 Subunit gene (CHRNA4) was found in a large Australian family with autosomal dominant nocturnal frontal lobe epilepsy.37 Sanger sequencing, invented in 1977, has dominated genetic diagnostics for decades.38 Many disease-causing variants in other genes (SCN1B39, KCNQ240, KCNQ341, SLC2A142, SCN1A43, GABRA144, and GABRG245) for other dominantly inherited epilepsies have been found following identification of their loci using linkage analysis. Currently, Sanger sequencing is mainly used to sequence a single gene selected based on a family history with variants in this gene or on an urgent need for specific treatment (such as SCL2A1 for glut1 deficiency requiring treatment with a ketogenic diet).

Microarray. Microarray analysis has been a widely available technique since the early 2000’s. Microarray analysis detects losses (deletions) or gains (duplications) of relatively small parts of the chromosomes.46 These deletions and duplications are called copy number variants (CNVs). There are two types of CNVs that can underlie epilepsy: recurrent and non-recurrent CNVs.

Recurrent CNVs occur at speciﬁc sites of the genome due to the recombination of DNA between highly homologous chromosomal regions (non-allelic homologous recombination). During meiosis, these highly homologous regions can misalign and an unequal crossover can occur, resulting in deletions or duplications of the DNA. These deletions and duplications can include single or multiple genes.47 Recurrent CNVs are associated with diﬀerent syndromes. For example, a 15q11.2-q13 microdeletion is associated with Angelman and Prader-Willi syndrome (OMIM 105830 and 176270, respectively) and a 22q13.33 microdeletion with Phelan-McDermid Syndrome (OMIM 606232). Epilepsy might occur in the context of these microdeletion or microduplication syndromes.

Non-recurrent CNVs are unique for every individual. These CNVs occur due to unequal crossover of DNA during meiotic (or less frequently mitotic) DNA repair processes with inconsistent breakpoints throughout the genome in regions with limited homology.47 Identifying a critical region of overlap (CRO) between such non-recurrent CNVs of patients with similar phenotypes can elucidate new genes for epilepsy. For example, GRIN2A was identiﬁed as a candidate gene for epilepsy based on the identiﬁcation of a CRO betewen 16p13.2 deletions in three patients with similar phenotypes.48

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 15 16 CHAPTER 1

Subsequently, disease-causing variants in GRIN2A were identiﬁed in patients with epilepsy within the epilepsy-aphasia-spectrum.49–51

Several studies have shown that causal CNVs are present in ~5% of patients from diﬀerent cohorts with generalized epilepsy, focal epilepsy or epileptic encephalopathy.52–55 In addition, a few recurrent CNVs have been found signiﬁcantly more often in patients with epilepsy compared to controls, and these are now referred to as susceptibility CNVs because they can increase the risk of developing epilepsy. Deletions on chromosomes 1q21.1, 15q11.2, 15q13.3, 16p11.2 and 16p13.11 and duplications on chromosome 16p11.2 are well-known examples of susceptibility CNVs.52,55–57

Next generation sequencing. With the introduction of Next Generation Sequencing (NGS) in 2008-2009, we entered a new, very exciting era in epilepsy genetics research and diagnostics. NGS enables sequencing of multiple genes, of the whole exome or of the whole genome in a single test run. This has enabled two major advances. First, variants can be identified in genes that would otherwise not have been selected for genetic testing. NGS may therefore identify many novel candidate genes for epilepsy and, indeed, the list of epilepsy candidate genes has grown tremendously since its introduction.58–60 Second, gene-related phenotypic spectra can be expanded when gene variants are found in patients with phenotypes that were not yet associated with this gene. For example, SYNGAP1 was identified as a gene for an epileptic encephalopathy after it was first identified as a gene for intellectual disability.61,62

GENETIC DATA INTERPRETATION

The genome of each individual, affected or non-affected, differs at 4-5 million sites from the reference human genome and harbors 2,100 to 2,500 structural genetic variants.63 Not surprisingly, the genome-wide genetic techniques discussed above also identify numerous variants that are not associated with disease. Therefore, our challenge is to interpret these variants and pinpoint the disease-causing variant in our patients. International guidelines have been established to classify variants based on their pathogenicity as benign, likely benign, of unknown significance, likely pathogenic or definitely pathogenic.64 Variants are first filtered based on their prevalence in national and international population databases such as the genome of the Netherlands (GoNL)65, genome Aggregation Database (gnomAD)66, 1000 Genomes63, dbSNP67 and the database for genomic variants (DGV)68. A variants is considered likely benign if it is present in more than 1% of the population.64 For the variants that are not considered likely benign, the variant’s inheritance, location, in silico predicted or functionally tested effects, presence in other patients with similar phenotypes are evaluated to determine its pathogenicity.64 With the introduction of whole exome sequencing, and possibly whole genome sequencing in the near future, the interpretation of genetic testing is becoming more difficult and time-consuming as ever more variants are being identified.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 16 INTRODUCTION 17

1 detect intragenic CNVs detect intragenic diseases other than epilepsydiseases rearrangements other than epilepsydiseases necessary is presentation genes individually frequently CNVs detect intragenic variants of intronic relevance other than epilepsydiseases - Low resolution coverage of genome resolution coverage - Low for ndings Possibility fi of incidental - balancedMisses chromosomal - for ndings Possibility fi of incidental - Pre-selection of gene based on clinical - Time- and sequence cost-expensive to - - Not detect possible to intragenic CNVs Gene panels need be to updated - Some laboratories: possible not to - Some laboratories: possible not to - Not the yet possible reliably interpret to - for ndings Possibility fi of incidental - chromosomes and balanced rearrangements of genome coverage ndings fi incidental careful due to selection of genes ndings fi incidental careful due to selection of genes of genes if no clinical clue of genes if no clinical clue detect intronic variants and exonic numbercopy variants incidental fi ndings ndings fi incidental careful due to selection of genes, which can be extended easily - Able to detectAble to ring - High resolution - No possibility of - No need for selection - No need for selection - Data available to - Typical indication indication Typical Advantages of a Suspicion chromosomal disorder Disadvantages Genetic epilepsy possibilityLow of - only applied Currently settings research in Epilepsy “plus” Epilepsy “plus” delay, (developmental problems behavioral or dysmorphisms) syndrome Epilepsy known be to caused gene by a single variant Genetic epilepsy possibilityLow of - Genetic epilepsy without a gene ed identifi variant a gene panelusing Variations detected chromosomal rearrangements Sequence variants of within a subset genes Sequence variants in the and CNVs whole genome (introns + exons) homozygosity, homozygosity, uniparental isodisomy Sequence variants a or up to in single few genes Sequence variants in the exomes, but analyzesonly of ofsubset genes Sequence variants within in the exomes Karyotyping Large Epilepsy gene panel Whole genome sequencing Microarray of regions CNVs, Sanger sequencing Whole exome sequencing with epilepsy lter fi Whole exome sequencing Genetic techniques and their indications, advantages, disadvantages advantages, and their indications, 1: Genetic techniques Table detecting variants for level genome Human Test

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 17 18 CHAPTER 1

CONCEPTUALIZING EPILEPSY GENETICS

Over the last few decades our understanding of the genetic basis of epilepsies has increased tremendously and this knowledge can be summarized in three important concepts.

Epilepsy gene products are part of different pathways When the first ion-channel-coding genes for epilepsies were discovered, the idea was raised that epilepsy could be channelopathy.60 This concept has however been overturned by the discovery in 2002 of LGI1, a non-ion-channel-coding gene for autosomal dominant lateral temporal lobe epilepsy.69 Following the identification of many non-ion-coding-channel genes for epilepsy, we realized that epilepsy gene products can be part of different pathways and thereby affect neuronal function differently. These pathways include DNA repair, transcriptional regulation, axon myelination, metabolite and ion transport, peroxisomal function, cell-cell adhesion, neurite formation, interneuron migration, apoptosis and blood-brain barrier transport.70 Notably, gene products within the same pathway can often result in a similar epilepsy phenotype. The identification of DEPDC5 as the causal gene for focal epilepsy and cortical brain malformations in the mammalian target of rapamycin (mTOR) pathway, known to be involved in tuberous sclerosis, nicely illustrates this.71,72 Even more striking, sequencing of other genes in the mTOR pathway identified variants in NPRL2 and NPRL3, both associated with a similar phenotype.73

Epilepsy encompasses different modes of inheritance We have learned that epilepsy can be inherited following diﬀerent modes of inheritance. Although some epilepsies are caused by a single gene variant (monogenic inheritance), the majority of epilepsies follow a complex inheritance where many genes (polygenic inheritance) or a combination of genes and environmental factors (multifactorial inheritance) increase the risk of developing epilepsy.74 In monogenic epilepsies, the disease-causing gene variant does not necessarily lead to disease in all relatives carrying this variant, a phenomenon called incomplete penetrance. Alternatively, a relative might be aﬀected yet not carry the disease-causing variant, a condition referred to as a phenocopy.

Phenotypic and genetic heterogeneity play a key role in epilepsy Phenotypic heterogeneity (also called pleiotropy) refers to a single genetic disorder leading to different epilepsy syndromes: e.g. mutations in KCNQ2 can cause benign neonatal and infantile epilepsy as well as severe neonatal epileptic encephalopathy.75,76 Genetic heterogeneity means that a single epilepsy syndrome can be caused by pathogenic variants in different genes. For example, benign infantile epilepsy can by caused by variants in KCNQ2, KCNQ3, SCN2A or PRRT2.75 For some genes, such as SCN1A, clear genotype-phenotype correlations have been identified that explain the SCN1A phenotypic heterogeneity by focusing on the mutation type and location of SCN1A variants in relation to the phenotype.77 For other genes, such as SYNGAP1 and GRIN2A, we still lack a clear genotype-phenotype correlation.78,79

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 18 INTRODUCTION 19

ADVANCES AND CHALLENGES IN EPILEPSY GENETICS IN CURRENT CLINICAL PRACTICE 1

The yield of genetic testing Not long ago it was unimaginable that we would now know of 700 genes related to epilepsy.58 Even so, we still cannot identify the genetic cause in a signiﬁcant number of epilepsy cases that are suspected to be genetic. Only one large and two very small studies addressed the yield of microarray in clinical hospital cohorts of patients with all types of epilepsies: the yield varied between 9% and 40%.55,80,81 In one of these studies, new loci of interest for epilepsy were identiﬁed.55 In chapter 2, we describe the diagnostic yield of microarray in a large clinical hospital cohort of children with epilepsy in the Netherlands and the possible novel CNVs for epilepsy.

With our increasing ability to identify the underlying genetic causes of epilepsy, genetic testing is oﬀered more and more in clinical practice. Using these techniques, we hope not only to improve epilepsy care but also to provide answers to longstanding questions from patients and parents about the epilepsy cause and prognosis. However, the impact on patients and their parents of genetic counseling and receiving a genetic diagnosis has not yet been studied.82,83 In chapter 3, we present the results of a study evaluating empowerment and anxiety during a genetic counseling trajectory in patients with epilepsy or their parents in whom genetic causes were and were not identiﬁed.

Genotype-phenotype studies With the introduction of NGS, many new genes for epilepsy have been identiﬁed in diﬀerent cohorts of patients selected based on their phenotype. Due to this ascertainment, the clinical features associated with these genes is often biased in the original papers. To unravel the full gene-related phenotypic spectrum and the genotype-phenotype correlations, patients should be selected based on their genotype and subsequently be phenotyped (a process called reverse phenotyping). In chapter 4 and chapter 5, we report the phenotypic spectra and genotype-phenotype correlations for SYNGAP1 and GRIN2A, respectively. In chapter 6, we show that psychotic disorders are a new, later-onset manifestation of PCDH19 Girls Clustering Epilepsy. Chapter 7 describes a case report of a boy with a deletion of the STX1B gene and a new phenotype: epilepsy with myoclonic- atonic seizures (previously called myoclonic astatic epilepsy).

In Chapter 8, we evaluate the presence of PRRT2-related phenotypes in patients with 16p11.2 microdeletions including PRRT2 and the 16p11.2 microdeletion syndrome.

All efforts to identify genes for epilepsy are driven by the hope of improving epilepsy care. For now, however, the identification of new epilepsy genes has been translated into better management of seizures in only a few cases. Currently, ‘personalized’ medicine for epilepsy has been established in only two scenarios: a ketogenic diet for patients with SLC2A1 variants and treatment with pyridoxine for patients with ALDH7A1 variants.84 In chapter 9, we report the efficacy of carbamazepine as a

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 19 20 CHAPTER 1

sodium-channel blocker in the case of a girl with infantile epilepsy due to a SCN2A sequence variant and. This case report is an example of precision medicine.

In chapter 10 we discuss our results in a general perspective and we propose a diagnostic algorithm for genetic testing for epilepsy in clinical practice.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 20 INTRODUCTION 21

REFERENCES

1. Jones WHS [translator]. Hippocrates, vol II. Cambridge: 17. Blumenfeld H, Meador KJ. Consciousness as a useful 1 Harvard; 1923. concept in epilepsy classification. Epilepsia 2014; 55: 2. Fisher RS, van emde Boas W, Blume W, et al. Epileptic 1145–1150. seizures and epilepsy: Definitions proposed by the 18. McTague A, Howell KB, Cross JH, Kurian MA, Scheffer IE. International League Against Epilepsy (ILAE) and the The genetic landscape of the epileptic encephalopathies International Bureau for Epilepsy (IBE). Epilepsia 2005; 46: of infancy and childhood. Lancet Neurol 2016; 15: 304- 470-472. 316. 3. Fisher RS, Acevedo C, Arzimanoglou A, et al. ILAE 19. Lennox WG. The heredity of epilepsy as told by relatives Official Report: A practical clinical definition of epilepsy. and twins. JAMA 1951; 146: 529–536. Epilepsia. 2014; 55: 475–482. 20. Annegers J, Hauser W, Anderson V, Kurland L. The risks 4. Camfield P, Camfield C. Incidence, prevalence and of seizure disorders among relatives of patients with aetiology of seizures and epilepsy in children. Epileptic childhood onset epilepsy. Neurology 1982; 32: 174–179. Disord. 2015; 17: 117–123. 21. Ottman R, Annegers JF, Hauser WA, Kurland LT. Seizure 5. Classification PI, Seizures E. A Proposed International Risk in Offspring of Parents with Generalized Versus Classification of Epileptic Seizures. Epilepsia. 1964; 5: Partial Epilepsy. Epilepsia 1989; 30: 157–161. 297–306. 22. Aziz H, Ali SM, Frances P, Khan MI, Hasan KZ. Epilepsy 6. Gastaut H. Clinical and Electro-encephalographical in Pakistan: a population-based epidemiologic study. Classification of Epileptic Seizures. Epilepsia 1970; 11: Epilepsia 1994; 35: 950–958. 102–112. 23. Monetti VC, Granieri E, Casetta I, et al. Risk-Factors for 7. Merlis JK. Proposal for an international classification of Idiopathic Generalized Seizures - a Population-Based the epilepsies. Epilepsia 1970; 11: 114–119. Case-Control Study in Copparo, Italy. Epilepsia 1995; 36: 8. Commission on Classification and Terminology of 224–229. the International League Against Epilepsy. Proposal 24. Jain S, Padma MV, Puri A, Jyoti, Maheshwari MC. for Revised Classification of Epilepsies and Epileptic Occurrence of epilepsies in family members of Indian Syndromes: Commission on Classification and probands with different epileptic syndromes. Epilepsia Terminology of the International League Against 1997; 38: 237–244. Epilepsy. Epilepsia 1981; 22: 489–501. 25. Callenbach PM, Geerts AT, Arts WFM, et al. Familial 9. Commission on Classification and Terminology of the occurrence of epilepsy in children with newly diagnosed International League Against Epilepsy. Proposal for the multiple seizures: Dutch Study of Epilepsy in Childhood. classification of epilepsies and epileptic syndromes. Epilepsia 1998; 39: 331–336. Epilepsia 1985; 26: 268–278. 26. Berkovic SF, Howell RA, Hay D, Hopper J. Epilepsies in 10. Commission on Classification and Terminology of twins: genetics of the major epilepsy syndromes. Ann the International League Against Epilepsy. Proposal Neurol 1998; 43: 435–445. for Revised Classification of Epilepsies and Epileptic 27. Corey LA, Berg K, Pellock JM, Solaas MH, Nance WE, Syndromes. Epilepsia 1989; 30: 389–399. DeLorenzo RJ. The occurrence of epilepsy and febrile 11. Berg AT, Berkovic SF, Brodie MJ, et al. Revised terminology seizures in Virginian and Norwegian twins. Neurology and concepts for organization of seizures and epilepsies: 1991; 41: 1433–1436. Report of the ILAE Commission on Classification and 28. Sillanpaa M, Koskenvuo M, Romanov K, Kaprio J. Genetic Terminology, 2005-2009. Epilepsia 2010;51:676–685. factors in epileptic seizures: evidence from a large twin 12. Scheffer IE, Berkovic S, Capovilla G, et al. ILAE classification population. Acta Neurol Scand 1991; 84: 523–526. of the epilepsies: Position paper of the ILAE Commission 29. Zhang YH, Burgess R, Malone JP, et al. Genetic epilepsy for Classification and Terminology. Epilepsia 2017; **: with febrile seizures plus. Neurology 2017; 89: 1210-1219. 1–10. doi: 10.111/epi.13709 30. Abou-Khalil B, Afawi Z, Allen AS, et al. Phenotypic analysis 13. Fisher RS, Cross JH, D’Souza C, et al. Instruction manual of 303 multiplex families with common epilepsies. Brain for the ILAE 2017 operational classification of seizure 2017; 140: 2144–2156. t ypes. Epilepsia 2017; **: 1–9. doi: 10.1111/epi. 13670. 31. Greenberg D, Delgado-Escueta A, Widelitz H, et al. 14. Laufs H. Functional imaging of seizures and epilepsy: Juvenile myoclonic epilepsy (JME) may be linked to the evolution from zones to networks. Curr Opin Neurol BF and HLA loci on human chromosome 6. Am J Med 2012; 25: 194–200. Genet 1988; 31: 185–192. 15. Abela E, Rummel C, Hauf M, Weisstanner C, 32. Chen T, Giri M, Xia Z, Subedi YN, Li Y. Genetic and Schindler K, Wiest R. Neuroimaging of epilepsy: Lesions, epigenetic mechanisms of epilepsy: A review. networks, oscillations. Clin Neuroradiol 2014; 24: 5–15. Neuropsychiatr Dis Treat 2017; 13: 1841–1859. 16. Berg AT, Scheffer IE. New concepts in classification of the 33. Koeleman BPC. What do genetic studies tell us about epilepsies: Entering the 21st century. Epilepsia 2011; 52: the heritable basis of common epilepsy? Polygenic or 1058-1062. complex epilepsy? Neurosci Lett 2017; doi: 10.1016/j. neulet. 2017.03.042.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 21 22 CHAPTER 1

34. Chung W-H, Hung S-I, Hong H-Sh, et al. A marker for 52. Mefford HC, Muhle H, Ostertag P, et al. Genome-wide Stevens-Johnson syndrome. Nature 2004; 428: 486. copy number variation in epilepsy: novel susceptibility 35. Tan NCK, Mulley JC, Berkovic SF. Genetic association loci in idiopathic generalized and focal epilepsies. PLoS studies in epilepsy: “The truth is out there.” Epilepsia Genet 2010; 6: 1–9. 2004; 45: 1429–1442. 53. Heinzen EL, Radtke RA, Urban TJ, et al. Rare Deletions at 36. Risch NJ. Searching for genetic determinants in the new 16p13.11 Predispose to a Diverse Spectrum of Sporadic millennium. Nature. 2000; 405: 847-856. Epilepsy Syndromes. Am J Hum Genet 2010; 86: 707–718. 37. Steinlein OK, Mulley JC, Propping P, et al. A missense 54. Mefford HC, Yendle SC, Hsu C, et al. Rare copy mutation in the neuronal nicotinic acetylcholine number variants are an important cause of epileptic receptor alpha 4 subunit is associated with autosomal encephalopathies. Ann Neurol 2012; 70: 974–985. dominant nocturnal frontal lobe epilepsy. Nat Genet 55. Olson H, Shen Y, Ph D, et al. Copy number variation plays 1995; 11: 201–203. an important role in clinical epilepsy. Ann Neurol 2014; 38. Sanger F, Nicklen S. DNA sequencing with chain- 75: 943–958. terminating. 1977; 74: 5463–5467. 56. De Kovel CGF, Trucks H, Helbig I, et al. Recurrent 39. Wallace RH, Wang DW, Singh R, et al. Febrile seizures and microdeletions at 15q11.2 and 16p13.11 predispose to generalized epilepsy associated with a mutation in the idiopathic generalized epilepsies. Brain 2010; 133: 23–32. Na+-channel beta1 subunit gene SCN1B. Nat Genet 1998; 57. Mullen SA, Carvill GL, Bellows S, et al. Copy number 19: 366–370. variants are frequent in genetic generalized epilepsy 40. Singh NA, Charlier C, Stauffer D, et al. A novel potassium with intellectual disability. Neurology 2013; 81: 1507– channel gene, KCNQ2, is mutated in an inherited epilepsy 1514. of newborns. Nat Genet 1998; 18: 25–29. 58. Wang J, Lin Z-J, Liu L, et al. Epilepsy-associated genes. 41. Charlier C, Singh NA, Ryan SG, et al. A pore mutation in a Seizure 2017; 44: 11–20. novel KQT-like potassium channel gene in an idiopathic 59. Møller RS, Dahl HA, Helbig I. The contribution of next epilepsy family. Nat Genet 1998; 18: 53–55. generation sequencing to epilepsy genetics. Expert Rev 42. Seidner G, Alvarez MG, Yeh JI, et al. GLUT-1 deficiency Mol Diagn 2015; 15: 1531-1538. syndrome caused by haplo-insufficiency of the blood- 60. Helbig I, Abou Tayoun AN. Understanding genotypes brain barrier hexose carrier. Nat Genet 1998; 18: 188–191. and phenotypes in epileptic encephalopathies. Mol. 43. Escayg A, MacDonald BT, Meisler MH, et al. Mutations Syndromol 2016; 7: 172-181. of SCN1A, encoding a neuronal sodium channel, in two 61. Hamdan FF, Gauthier J, Spiegelman D, et al. Mutations families with GEFS+2. Nat Genet 2000; 24: 343–345. in SYNGAP1 in Autosomal Nonsyndromic Mental 44. Cossette P, Liu L, Brisebois K, et al. Mutation of GABRA1 Retardation. N Engl J Med 2009; 360: 599–605. in an autosomal dominant form of juvenile myoclonic 62. Carvill GL, Heavin SB, Yendle SC, et al. Targeted epilepsy. Nat Genet 2002; 31: 184–189. resequencing in epileptic encephalopathies identifies 45. Baulac S, Huberfeld G, Gourfinkel-An I, et al. First genetic de novo mutations in CHD2 and SYNGAP1. Nat Genet evidence of GABA(A) receptor dysfunction in epilepsy: a 2013; 45: 825-830. mutation in the gamma2-subunit gene. Nat Genet 2001; 63. Auton A, Abecasis GR, Altshuler DM, et al. A global 28: 46–48. reference for human genetic variation. Nature 2015; 526: 46. Mefford HC. CNVs in Epilepsy. Curr Genet Med Rep 2014; 68–74. 2: 162–167. 64. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster 47. Hastings PJ, James R Lupski SMR and GI. Mechanisms of J. Standards and Guidelines for the Interpretation of change in gene copy number. Nat Rev Genet 2010; 10: Sequence Variants: A Joint Consensus Recommendation 551–564. of the American College of Medical Genetics and 48. Reutlinger C, Helbig I, Gawelczyk B, et al. Deletions in Genomics and the Association for Molecular Pathology. 16p13 including GRIN2A in patients with intellectual Genet Med 2015; 17: 405–424. disability, various dysmorphic features, and seizure 65. Francioli LC, Menelaou A, Pulit SL, et al. Whole- disorders of the rolandic region. Epilepsia 2010; 51: 1870– genome sequence variation, population structure and 1873. demographic history of the Dutch population. Nat 49. Endele S, Rosenberger G, Geider K, et al. Mutations in Genet 2014; 46: 818–825. GRIN2A and GRIN2B encoding regulatory subunits of 66. Lek M, Karczewski KJ, Minikel E V, et al. Analysis of protein- NMDA receptors cause variable neurodevelopmental coding genetic variation in 60,706 humans. Nature 2016; phenotypes. Nat Genet 2010; 42: 1021–1026. 536: 285–291. 50. Lemke JR, Lal D, Reinthaler EM, et al. Mutations in GRIN2A 67. Sherry S, Ward M, Kholodov M, et al. dbSNP: the NCBI cause idiopathic focal epilepsy with rolandic spikes. Nat database of genetic variation. Nucleic Acids Res 2001; 29: Genet 2013; 45: 1067-1072. 308–311. 51. Lesca G, Rudolf G, Bruneau N, et al. GRIN2A mutations 68. MacDonald JR, Ziman R, Yuen RKC, Feuk L, Scherer SW. in acquired epileptic aphasia and related childhood The Database of Genomic Variants: A curated collection focal epilepsies and encephalopathies with speech and of structural variation in the human genome. Nucleic language dysfunction. Nat Genet 2013; 45: 1061–1066. Acids Res 2014; 42: 986–992.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 22 INTRODUCTION 23

69. Kalachikov S, Evgrafov O, Ross B, et al. Mutations in 77. Zuberi SM, Brunklaus A, Birch R, Reavey E, Duncan J, LGI1 cause autosomal-dominant partial epilepsy with Forbes GH. Genotype-phenotype associations in SCN1A- auditory features. Am J Hum Genet 2008; 30: 335–341. related epilepsies. Neurology 2011; 76: 594–600. 70. McTague A, Howell KB, Helen Cross J, Kurian MA, 78. Mignot C, Stülpnagel C Von, Nava C, et al. Genetic and 1 Scheff er IE. The genetic landscape of the epileptic neurodevelopmental spectrum of SYNGAP1 -associated encephalopathies of infancy and childhood. Lancet intellectual disability and epilepsy. J Med Gen 2016; 53: Neurol 2016; 15:304–316. 511-522. 71. Dibbens LM, De Vries B, Donatello S, et al. Mutations in 79. Myers KA, Scheffer IE. GRIN2A-Related Speech Disorders DEPDC5 cause familial focal epilepsy with variable foci. and Epilepsy. GeneReviews®. Washington: Seattle. 2016. Nat Genet 2013; 45: 546–551. 80. Faheem M, Nasser M, Chaudhary A, et al. Array- 72. Scheffer IE, Heron SE, Regan BM, et al. Mutations in comparative genomic hybridization analysis of a cohort mammalian target of rapamycin regulator DEPDC5 of Saudi patients with epilepsy. CNS Neurol Disord Drug cause focal epilepsy with brain malformations. Ann Targets 2015; 14: 468–475. Neurol 2014; 75: 782–787. 81. Naseer MI, Faheem M, Chaudhary AG, et al. Genome wide 73. Ricos MG, Hodgson BL, Pippucci T, et al. Mutations in analysis of novel copy number variations duplications / the mammalian target of rapamycin pathway regulators deletions of different epileptic patients in Saudi Arabia. NPRL2 and NPRL3 cause focal epilepsy. Ann Neurol 2016; BMC Genomics 2015; 16: S10. 79: 120–131. 82. Shostak S, Ottman R. Ethical, legal, and social dimensions 74. Berkovic SF, Mulley JC, Scheffer IE, Petrou S. Human of epilepsy genetics. Epilepsia 2006; 47: 1595–1602. epilepsies: interaction of genetic and acquired factors. 83. Shostak S, Zarhin D, Ottman R. What’s at stake? Genetic Trends Neurosci 2006; 29:391–397. information from the perspective of people with 75. Zara F, Specchio N, Striano P, et al. Genetic testing in epilepsy and their family members. Soc Sci Med 2011; 73: benign familial epilepsies of the first year of life: Clinical 645–654. and diagnostic significance. Epilepsia 2013; 54: 425–436. 84. Balestrini S, Sisodiya SM. Pharmacogenomics in epilepsy. 76. Weckhuysen S, Mandelstam S, Suls A, et al. KCNQ2 Neurosci Lett 2017. doi: 10.1016/ j.neulet.2017.01.014. encephalopathy: Emerging phenotype of a neonatal epileptic encephalopathy. Ann Neurol 2012; 71: 15–25.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 23 24

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 24 2525

PART I

The yield of genetic testing

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 25 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 26 27

Chapter 2

Copy number variation in a hospital-based cohort of children with epilepsy

Published as: DRM Vlaskamp, PMC Callenbach, P Rump, et al. Copy number variation in a hospital-based cohort of children with epilepsy. Epilepsia Open, 2017; 2: 244–254.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 27 28 CHAPTER 2

Danique RM Vlaskamp1,2, Petra MC Callenbach1*, Patrick Rump2*, Lucia AA Giannini2, Trijnie Dijkhuizen2, Oebele F Brouwer1, Conny MA van Ravenswaaij-Arts2

1 University of Groningen, University Medical Center Groningen, Department of Neurology, Groningen the Netherlands. 2 University of Groningen, University Medical Center Groningen, Department of Genetics, Groningen, the Netherlands. * These authors contributed equally to this work

Acknowledgements. P.M.C. Callenbach received an unrestricted research grant from UCB Pharma BV, the Netherlands. UCB Pharma BV had no role in the study design, data collection, and analysis, decision to publish, or preparation of the manuscript. We thank K. McIntyre for editing the manuscript and J. Anderson for helping identifying CNVs for epilepsy reported in the literature.

Disclosures. None of the authors has any conflicts of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 28 COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 29

ABSTRACT

Objective. To evaluate the diagnostic yield of microarray analysis in a hospital-based cohort of children with seizures and to identify novel candidate genes and susceptibility loci for epilepsy.

Methods. Of all children who presented with their ﬁrst seizure in the University Medical Center 2 Groningen (January 2000 through May 2013) (n = 1,368), we included 226 (17%) children who underwent microarray analysis before June 2014. All 226 children had a deﬁnite diagnosis of epilepsy. All their CNVs on chromosomes 1-22 and X that contain protein-coding genes and have a prevalence of <1% in healthy controls were evaluated for their pathogenicity.

Results. Children selected for microarray analysis more often had developmental problems (82% vs 25%, p < 0.001), facial dysmorphisms (49% vs 8%, p < 0.001) or behavioral problems (41% vs 13%, p < 0.001) than children who were not selected. We found known clinically relevant CNVs for epilepsy in 24 of the 226 children (11%). Seventeen of these 24 children had been diagnosed with symptomatic focal epilepsy not otherwise specified (71%) and five with West syndrome (21%). Of these 24 children, many had developmental problems (100%), behavioral problems (54%) or facial dysmorphisms (46%). We further identified five novel CNVs comprising four potential candidate genes for epilepsy: MYT1L, UNC5D, SCN4B and NRXN3.

Significance. The 11% yield in our hospital-based cohort underscores the importance of microarray analysis in diagnostic evaluation of children with epilepsy.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 29 30 CHAPTER 2

INTRODUCTION

Genetic factors play an important role in the etiology of epilepsy,1 as demonstrated by the large number of genes and regions that cause or predispose to epilepsy newly identified by various genome-wide technologies.2 Chromosomal microarray analysis, in particular, enables the identification of chromosomal deletions (losses) or duplications (gains), called Copy Number Variants (CNVs).3 CNVs may contribute to epilepsy in two ways. First, CNVs that comprise epilepsy-related genes could lead to epilepsy following a Mendelian inheritance. For example, both KCNQ2 sequence variants and whole gene deletions can cause benign familial neonatal seizures.4,5 Second, CNVs that occur more frequently in patients compared to healthy controls may increase an individual’s susceptibility to developing epilepsy, with the responsible haploinsufficient gene(s) often being unknown. Large cohort studies have identified such susceptibility CNVs in several chromosomal regions, including well-known CNVs located at 15q11.2 (BP1-BP2), 15q13.3 and 16p13.11.6-8

Studies using microarray analysis have most often been performed in research cohorts of children who were selected based on their epilepsy diagnosis, e.g. idiopathic generalized epilepsies, focal epilepsies and/or fever-associated epilepsies.6-8 Only a few studies have addressed the yield of microarray analysis in clinical cohorts of all children presenting with any type of seizures in a clinical setting.9-11 We therefore aimed to evaluate the diagnostic yield of microarray analysis in a hospital- based cohort of children with epilepsy for whom detailed phenotypic information was available, with the further goal of identifying novel candidate genes or susceptibility loci for epilepsy.

MATERIAL AND METHODS

Study cohort The study cohort was derived from the childhood seizure database of the University Medical Center Groningen (UMCG), a regional referral center for children with epilepsy. In this database, we retrospectively included all children who presented with their first febrile or afebrile seizure before the age of 18 years between January 2000 and June 2013, and who were seen and/or treated by a child neurologist of the UMCG (n = 1,368). Epilepsy was diagnosed in 91% of these children using the current International League Against Epilepsy (ILAE) practical clinical definition of epilepsy.12 Of the remaining children (9%), 7% had febrile seizures only and 2% had only one afebrile seizure. The UMCG database contains phenotype information and was independently completed by two researchers (DRMV and PMCC). Phenotypic inconsistencies and epilepsy classification were discussed until agreement was reached using the information in the database as well as in the original medical records (PMCC and OFB). Epilepsy syndromes and seizure types were classified according to the 2006 ILAE classification.13 Children were included in this study if they underwent microarray analysis in the context of their diagnostic work-up before June 2014.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 30 COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 31

Formal independent review board evaluation was waived by the Institutional Medical Ethical Committee of the UMCG because of the retrospective and observational character of this study.

Chromosomal microarray analysis and data interpretation Microarray analyses were performed using an oligonucleotide array (Agilent 105K or 180K custom HD-DGH microarray; Agilent Technologies, Santa Clara, USA) or a single nucleotide polymorphism 2 (SNP) array (Illumina Omni Express 12-V1.0; Illumina, San Diego, USA). Cartagenia Bench Lab CNV software was used for storage, analysis and reporting of the structural genomic data (Cartagenia, Leuven, Belgium; part of Agilent Technologies, Santa Clara, USA). The chromosomal coordinates of CNVs were reported relative to the Genome Reference Consortium Human Reference genome version 37 (GRCh37/hg19).

CNVs on chromosome 1-22 or X identified by at least three (SNP microarray) or four (oligonucleotide microarray) consecutive probes were evaluated for their pathogenicity (Figure 1). CNVs were excluded from further analysis when they did not contain (protein-coding) genes or had ≥90% overlap with CNVs seen in ≥1% of healthy controls. The prevalence of CNVs in healthy controls was calculated using the International Database of Genomic Variance (n = 14,316, last updated February 2013),14 the Low Lands Consortium database of oligonucleotides (n = 2,402, last updated December 2012) and SNP microarray results of healthy parents of children who underwent microarray analysis in five Dutch genetic centers (n = 749, last updated October 2014). Remaining CNVs were categorized into two groups: (1) CNVs with <90% overlap with CNVs observed in healthy controls and (2) CNVs with ≥90% overlap with CNVs observed in <1% of the healthy controls (Figure 1). CNVs in both groups were marked as potentially clinically relevant if they had overlap with genetic regions previously associated with epilepsy. These regions were identified by performing a literature search using PubMed, complemented with information from the Decipher database and Cartagenia Bench Lab CNV software. The remaining CNVs in both groups were evaluated for novel candidate genes or susceptibility loci for epilepsy. CNVs with <90% overlap with CNVs of healthy controls were of interest if they contained a gene with an expression or function in the brain or a gene associated with an autosomal dominant or X-linked neuropsychiatric disease, and if they occurred in at least one (for deletions) or two (for duplications) unrelated children in our cohort. In the group of CNVs with ≥90% overlap with CNVs observed in <1% of the healthy controls, overlapping regions between CNVs in at least two unrelated children were of interest if these regions contained protein-coding genes and were 10 times more prevalent in our cohort compared to healthy controls.

Statistical analyses SPSS Statistics Version 22.0 (IBM Corporation, NY, USA) was used to perform descriptive and comparative statistics. Diﬀerences in categorical and ordinal phenotypic data between children were analyzed using Fisher’s exact and Mann-Whitney U tests, respectively.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 31 32 CHAPTER 2

Figure 1: Flowchart for evaluating Copy Number Variants (CNVs) in our hospital-based cohort of children with epilepsy.

ϭ͕ϵϴϮEsƐ ŝŶϮϮϲĐŚŝůĚƌĞŶ

ϭϴϰEsƐǁŝƚŚŶŽŐĞŶĞƐŝŶϭϯϬĐŚŝůĚƌĞŶ ϰϯϳEsƐǁŝƚŚŶŽƉƌŽƚĞŝŶĐŽĚŝŶŐŐĞŶĞƐŝŶϭϵϯĐŚŝůĚƌĞŶ

ϭ͕ϯϲϭŐĞŶĞͲĐŽŵƉƌŝƐŝŶŐ EsƐ ŝŶϮϮϰĐŚŝůĚƌĞŶ

ϵϱϯEsƐǁŝƚŚшϵϬйŽǀĞƌůĂƉŝŶшϭйŽĨŚĞĂůƚŚǇ ĐŽŶƚƌŽůƐ

ϰϬϴEsƐ ŝŶϭϴϭĐŚŝůĚƌĞŶ

EsƐǁŝƚŚфϵϬйŽǀĞƌůĂƉŝŶŚĞĂůƚŚǇĐŽŶƚƌŽůƐ EsƐǁŝƚŚшϵϬйŽǀĞƌůĂƉŝŶфϭйŚĞĂůƚŚǇĐŽŶƚƌŽůƐ ϴϬEsƐŝŶϱϯĐŚŝůĚƌĞŶ ϯϮϴEsƐŝŶϭϳϬĐŚŝůĚƌĞŶ

<ŶŽǁŶĐůŝŶŝĐĂůůǇƌĞůĞǀĂŶƚEsƐ <ŶŽǁŶĐůŝŶŝĐĂůůǇƌĞůĞǀĂŶƚEsƐ ϭϱEsƐŽĨϭϭƌĞŐŝŽŶƐŝŶϭϰĐŚŝůĚƌĞŶ ϭϬEsƐŽĨϲƌĞŐŝŽŶƐŝŶϭϬĐŚŝůĚƌĞŶ

EŽǀĞůEsƐ EŽǀĞůEsƐ;ϭϬǆŵŽƌĞĨƌĞƋƵĞŶƚͿ ϱEsƐŽĨϰƌĞŐŝŽŶƐŝŶϱĐŚŝůĚƌĞŶ ϭϲEsƐŽĨϴƌĞŐŝŽŶƐŝŶϭϲĐŚŝůĚƌĞŶ

ϲϬEsƐŽĨƵŶŬŶŽǁŶ ƐŝŐŶŝĨŝĐĂŶĐĞŝŶϯϵĐŚŝůĚƌĞŶ ϯϬϮEsƐŽĨƵŶŬŶŽǁŶƐŝŐŶŝĨŝĐĂŶĐĞŝŶϭϲϭĐŚŝůĚƌĞŶ

RESULTS

Characteristics of the study cohort In 226 (17%) children, microarray analysis was performed in the context of their diagnostic work-up. Their phenotypic characteristics are summarized in Table 1. All children had a deﬁnite diagnosis of epilepsy, except for one who had a single febrile status epilepticus.

Children with epilepsy who underwent microarray analysis had significantly more developmental problems (82% vs 25%, p < 0.001), facial dysmorphisms (49% vs 8%, p < 0.001) or behavioral problems (41% vs 13%, p < 0.001) when compared to children in our database who did not undergo microarray analysis. To reduce bias, our comparisons were limited to children with epilepsy onset after December 31, 2005, when microarray analysis was introduced in our center (n = 158 for children with array; n = 271 for children without array or another identified genetic cause). The presence of a positive family history for epilepsy, known in 141/158 children who did and in 206/271 children who did not undergo microarray, did not differ significantly (33% vs 30%, p = 0.56) between the two groups.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 32 COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 33

Table 1: Characteristics of the study cohort (n = 226)

Characteristics Male (%) 132 (58.4) Deceased (%) 17 (7.5) Median age at evaluation (range) 8y 10mo (1y 8mo – 23y 3mo) Median age at epilepsy onset (range) 1y 1mo (0 days – 15y 11mo) Seizure types (%) 22 GTCS 10 (4.4) Absences 5 (2.2) Myoclonic seizures 18 (8.0) Epileptic spasms 36 (15.9) Atonic seizures 12 (5.3) Focal seizures 176 (77.9) Secondarily generalized seizures 123 (54.4) Neonatal seizures 33 (14.6) Unclassified 1 (0.4) Status epilepticus (%) 74 (32.7) Epilepsy syndrome (%) Benign (familial) neonatal seizures 7 (3.1) Neonatal seizures (not benign) 1 (0.4) Ohtahara syndrome 1 (0.4) Benign familial infantile seizures 4 (1.8) West syndrome 36 (15.9) Myoclonic epilepsy in infancy 3 (1.3) Myoclonic encephalopathy in nonprogressive disorders 1 (0.4) Benign epilepsy with centrotemporal spikes 5 (2.2) Childhood absence epilepsy 1 (0.4) Epilepsy with myoclonic absences 1 (0.4) CSWS / Landau-Kleffner syndrome 9 (4.0) Lennox-Gastaut syndrome 5 (2.2) Juvenile absence epilepsy 1 (0.4) Symptomatic focal epilepsies n.o.s. 148 (65.5) Localization-related cryptogenic epilepsy 24 (10.6) Other symptomatic generalized epilepsy 2 (0.9) Epilepsy with both generalized and focal seizures 3 (1.3) Febrile seizures plus 5 (2.2) Febrile infection related epilepsy syndrome 1 (0.4) One febrile status epilepticus 1 (0.4) One seizure likely to re-occur 2 (0.9) Epilepsies of unknown cause 1 (0.4) Seizure free* (%) 89/192 (46.4) Family history of epilepsy** (%) 68/200 (34.0) Developmental problems in speech, language, motor skills and/or cognition (%) 195 (86.3) Behavioral/psychiatric problems (%) 104 (46.0) Microcephaly (≤ -2SD) (%) 40 (17.7) Macrocephaly (≥2 SD) %) 13 (5.8) Short stature (≤ -2SD) (%) 39 (17.3) Tall stature (≥2 SD) %) 13 (5.8) Facial dysmorphisms (%) 109 (48.2) Congenital anomalies (%) 72 (31.9) MRI abnormalities (%) 118/218 (54.1) Epilepsy syndromes and seizure types were classified according to the International League Against Epilepsy (ILAE) classification of 2006.13 * Seizure freedom was defined as present if a patient had no clinical seizures for at least one year at the time of evaluation. ** Family history of epilepsy was positive if a first or second degree relative has epilepsy. Abbreviations: CSWS = continuous spike during slow-wave sleep, GTCS = generalized seizures with tonic and/or clonic manifestations, n.o.s. = not otherwise specified, SD = standard deviation.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 33 34 CHAPTER 2

Diagnostic yield of microarray analysis Microarray analysis revealed 1,982 CNVs in 226 children (Figure 1). After excluding CNVs that contained no (protein-coding) genes and/or were identiﬁed as most likely benign polymorphisms (≥90% overlap in ≥1% of controls), 408 CNVs in 181 children remained to be evaluated for their pathogenicity. These 408 CNVs included 233 (57.1%) duplications with a median size of 198.8 kb (range 17.9 kb - 21.0 Mb) and 175 (42.9%) deletions with a median size of 168.4 kb (range 22.4 kb - 21.0Mb). Inheritance could be analyzed for 102 (25.0%) CNVs, with 23 (22.5%) occurring de novo and 79 (77.5%) being inherited.

Known clinically relevant CNVs for epilepsy were identified in 24 of the 226 (11%) children with epilepsy (Figure 1, Tables 2 and 3). Their epilepsy was most often classified as symptomatic focal epilepsy not otherwise specified (71%) or West syndrome (21%). All children had developmental problems (100%), and many had behavioral problems (54%) and facial dysmorphisms (46%). Overall, no significant differences were found between children with and without clinically relevant CNVs for epilepsy syndrome diagnosis or the presence of other phenotypic characteristics (data not shown).

In 14 (7%) children, 15 known clinically relevant CNVs were found that do not occur in healthy controls (Table 2). Ten of these CNVs occurred de novo. For the remaining CNVs, inheritance was unknown. The phenotypes of these 14 children were compatible with previously reported phenotypes associated with these CNVs and included the well-established diagnoses: chromosome 1p36 deletion syndrome (MIM 607872), chromosome 2q23.1 deletion syndrome (MIM 156200), chromosome 18q deletion syndrome (MIM 601808), Angelman syndrome (MIM 105830), chromosome 15q11q13 duplication syndrome (MIM 608636), chromosome 16p11.2 deletion syndrome (MIM 611913), lissencephaly type 1 (MIM 607432), Phelan-McDermid syndrome (MIM 606232), and Juberg-Hellman syndrome (MIM 300088; epilepsy, female restricted, with mental retardation) (Table 2). One patient had a 2.6 Mb 8q22 deletion, comparable deletions have been published in six other cases.16,17 In 10 (4%) children, known clinically relevant CNVs were found that also occur in healthy controls, albeit in less than 1% (Table 3). Six of these CNVs (60%) were inherited from an affected (n = 2) or non-affected (n = 4) parent while two (20%) CNVs occurred de novo (the deletions including NRXN1 and CHRNA7). In three (30%) children, another cause for epilepsy - one not associated with the CNVs - was identified. These were developmental anomalies of cerebral structure (patient 822), a GLUT1-deficiency (patient 225) and polymicrogyria (patient 761) (Table 3).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 34

COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 35

MRI abnormalities MRI

Congenital anomalies Congenital

Facial dysmorphisms Facial

Tall stature (≥2 SD) (≥2 stature Tall

Short stature (≤ -2SD) (≤ stature Short 2

Macrocephaly (≥2 SD) (≥2 Macrocephaly

Microcephaly (≤ -2SD) (≤ Microcephaly

Behavioral problems Behavioral

problems

Developmental Developmental ++-----+- Age at last seizure last at Age

2y 2y

3mo Age at epilepsy onset epilepsy at Age y.y+-+-+-++- 1y3.5y†+ 1y4.5y++----+++ 2y 2y yy+-+-+-++U 2y4y†+ 1moNA+-----+++ 7moNA+------+U m9o - - -+-+++ 2mo9mo+ 3mo

k k k k k k Epilepsy syndrome Epilepsy Sm2m+ - -+-+- - WS3mo21mo+-

WS, WS,

focal focal Focal Focal Focal Relevant genes Relevant MBD5

UBE3A KLHL17 KLHL17 KLHL17 GABRD GABRD GABRD GABRD GABRB3 GABRB3 KCNAB2

Unknown Undet. Unknown Focal Unknown Unknown

(parental phenotype) (parental Inheritance

De novo novo De De novoDe MBD5 De novo De De novo De De novo De

Unknown Unknown CNV size in kilobases in size CNV 227 649 7,541 Unknown 4,950 6,655 4,849 Unknown 21,089 21,088 c

e b

a b d c Microarray results Microarray

18q21.32q23 (56,921,091-78,010,172)x1 (56,921,091-78,010,172)x1 18q21.32q23 arr 13q31.3q34(94,017,655-115,105,959)x3, arr 13q31.3q34(94,017,655-115,105,959)x3,

Sex, age (y) age Sex, Patient 590 590 10 F, arr 1p36.33p36.23(564,224-8,104,812)x1 183 183 M, 11 arr 2q23.1(148,775,316-149,002,634)x1 575 575 6 F, arr 8q22.3(101,795,020-104,406,406)x1 2,611 1032 M, 8 arr 1p36.33p36.31(746,419-5,696,745)x1 1105 1105 2 F, arr 2q22.3q23.3(146,506,579-151,355,790)x1 1037 M, 4† 1079 M, 4 arr 15q11.2q13.1(22,285,091-28,940,239)x1 1040 1040 3.5† F, Arr 1p34.1p33(46,089,475-46,738,333)x3 Clinically relevant CNVs with <90% overlap with CNVs observed (n = 14) controls with <90% overlap with CNVs in healthy CNVs 2: Clinically relevant Table

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 35 36 CHAPTER 2

b MRI abnormalities MRI 15

Other Other

l Congenital anomalies Congenital Facial dysmorphisms Facial

Phelan-McDermid i Tall stature (≥2 SD) (≥2 stature Tall

Angelman syndrome syndrome Angelman e

Short stature (≤ -2SD) (≤ stature Short

Macrocephaly (≥2 SD) (≥2 Macrocephaly Microcephaly (≤ -2SD) (≤ Microcephaly

Previously published by some of us. of some by published Previously a Behavioral problems Behavioral

hg19). hg19). Developmental problems Developmental

Lissencephaly type 1 (MIM 607432). Lissencephaly type 607432). 1 (MIM

h Age at last seizure last at Age ased. Symptomatic focalSymptomatic epilepsy not otherwise speciﬁed.

he child. Abbreviations: = benign BFIS familial infantile seizures, Age at epilepsy onset epilepsy at Age y4+- - -+- - 7y14y++-- 2yNA+-----++- 9yNA++----++- 3moNA+-+-----+ 8moNA++------

Chromosome 18q deletion syndrome (MIM 601808). 601808). (MIM syndrome deletion 18q Chromosome l d k k k Epilepsy syndrome Epilepsy BFIS4mo4y++------

Focal Focal Focal Undet.

WS, focal Relevant genes Relevant

UBE3A UBE3A GABRB3 GABRB3 GABRB3 GABRB3 SHANK3

(parental phenotype) (parental Inheritance Chromosome 16p11.2 deletion syndrome (MIM 611913). deletion syndrome 611913). (MIM Chromosome 16p11.2 g

De novoDe PRRT2 De novo PCDH19 De novo De De novo PAFAH1B1 De novo De CNV size in kilobases in size CNV (GRCh37/ 37 version genome Reference Human Consortium Reference 88 94 Unknown 579 967 6,392 6,377 f f g i j h

Chromosome 2q23.1 deletion syndrome (MIM 156200). 156200). (MIM syndrome deletion 2q23.1 Chromosome c Microarray results Microarray Juberg-Hellman syndrome (MIM 300088; epilepsy, female restricted with mental retardation (EFMR)). (EFMR)). retardation mental with restricted female epilepsy, 300088; (MIM syndrome Juberg-Hellman j

Chromosome 15q11-15q13 duplication syndrome 608636). (MIM Chromosome 15q11-15q13 Sex, age (y) age Sex, f Patient 319 13 F, arr 16p11.2(29,620,489-30,199,507)x1 831 11 F, arr Xq22.1(99,582,921-99,671,028)x1 201 M, 16 arr 22q13.3(51,125,351-51,219,150)x1 356 M, 16 arr 15q11.2q13.1(22,668,852-29,045,487)x3 1081 M, 4 arr 17p13.3(2,355,353-3,322,779)x1 1062 M, 5 arr 15q11.2q13.1(22,668,852-29,060,634)x3 Continuation table 2 Continuation Chromosome 1p36 deletion syndrome (MIM 607872). 607872). (MIM syndrome deletion 1p36 Chromosome (MIM 105830). The chromosomal coordinates are reported the relative Genome to undetermined epilepsy with both generalized and focal seizures. + Phenotype is present in the child. – Phenotype is absent in t mo M = male, F = female, = months, NA = not applicable seizure-free), (not U = unknown, WS = West syndrome, y = years, † = dece syndrome (MIM 606232). (MIM syndrome

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 36

COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 37

MRI abnormalities MRI

Congenital anomalies Congenital Facial dysmorphisms Facial

Tall stature (≥2 SD) (≥2 stature Tall gene associated with 22

Short stature (≤ -2SD) (≤ stature Short SLC2A1

Macrocephaly (≥2 SD) (≥2 Macrocephaly Microcephaly (≤ -2SD) (≤ Microcephaly

Symptomatic focal Symptomatic epilepsy not otherwise a Behavioral problems Behavioral

hg19). hg19).

Developmental problems Developmental Age at last seizure last at Age epsy, M = male, mat. M = male, = maternal, MAE = epilepsyepsy, with myoclonic

psy with centrotemporal spikes, CAE = childhood absence epilepsy, Age at epilepsy onset epilepsy at Age 4yNA+------6yNA++------11y13y++-----+- 3.5yNA+-+-+-+++ 1mo8y++------+ mN+++--+ 6moNA+++-+-- 11mo6y++------+

a a a a a a Epilepsy syndrome Epilepsy

MAE1.5yNA+-----+--

CSWS 4yNA++------+ CSWS Focal Focal Focal Focal Focal Focal One FS 2.5y 2.5y + + - - - + - - U Relevant genes Relevant This patient carries also a likely pathogenic sequence variant in the c NIPA1, NIPA1, NIPA1, NIPA1, NDE1 NDE1 NIPA2 NIPA2 PRRT2 PRRT2 CHRNA7

gene. gene. CYFIP1, CYFIP1,

(parental phenotype) (parental Inheritance De novo novo De CHRNA7 De novo novo De NRXN1

Reference Consortium Human Reference genome (GRCh37/ version 37 CNV size in kilobases in size CNV 612 519 Pat (none)

c b

Microarray results Microarray Sex, age (y) age Sex,

F, 8 arr 15q13.3(30,921,717-32,515,121)x1 1,593

This patient also carries a chromosome 14q31.1 deletion including the NRXN3 This patient also carries a chromosome 14q31.1 b M, 16M, 15q11.2(20,279,343-23,300,438)x1 arr 3,021 (none) Pat Patient

761 13 F, arr 16p11.2(29,620,488-30,198,752)x3 578 Unknown 337 13 F, arr 2p16.3(50,968,252-51,579,862)x1 270 M, 14 arr 16p13.11(14,944,359-16,525,488)x1 1,581 (FS) Pat 372 M, 9 arr 15q13.3(30,833,546-32,861,767)x3 2,028 (none) Mat 730 M, 12 arr 16p11.2(29,592,582-30,198,752)x3 606 Unknown 822 6 F, arr 1q21.1(145,395,197-146,089,261)x3 694 (none) Pat Unknown WS, focal 225 M, 11 arr 15q11.2(22,698,322-23,217,655)x1 1099 1000 1045 M, 6 arr 16p13.11(14,944,360-16,561,292)x1 1,617 Mat (FS) specified. specified. Clinically relevant CNVs with ≥90% overlap with CNVs observed (n = 10) controls with ≥90% overlap with CNVs in <1% of healthy CNVs 3: Clinically relevant Table The chromosomal coordinates are reported the relative Genome to GLUT-1 deficiency. + Phenotype is presentGLUT-1 in the child. – Phenotype is absent in the child. Abbreviations: BECTS = benign epile CSWS = continuous spike waves during slow-wave sleep syndrome, = febrile FS F = female, seizures, JME = juvenile myoclonic epil absences, mo = months, NA = not applicable seizure-free), (not U = unknown, pat = paternal, WS = West syndrome, y = years.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 37 38 CHAPTER 2

Novel CNVs of interest for epilepsy In five (2%) children, we identified novel CNVs of interest that were not found in healthy controls (Table 4). These CNVs comprised four potential candidate genes for epilepsy: MYT1L, UNC5D, SCN4B and NRXN3 (see Supplemental Table 1 for more information on these genes). In 16 (7%) children, eight different overlapping deletions (n = 4) and duplications (n = 4) occurred at a 10 times higher frequency in our cohort than in healthy controls (Supplemental Table 2). However, another cause for epilepsy was identified in seven (44%) of these children, with CNVs involving six of the eight regions. The two remaining regions did not contain genes that seem of particular interest for epilepsy (Supplemental Table 2).

DISCUSSION

Our study was performed in a university hospital-based cohort of children with epilepsy, who were selected to undergo microarray analysis as part of their diagnostic work-up based on their doctors’ preference. We found that microarray analysis yielded known clinically relevant CNVs for epilepsy in 11% of the children. We further identiﬁed ﬁve novel CNVs of interest for epilepsy in 2% of the children.

Diagnostic yield of microarray analysis The 11% yield of microarray analysis in our cohort is comparable with the 9% yield found in another clinical cohort of American individuals with epilepsy.9 Higher yields of 36-40% have been reported in smaller cohorts of Saudi individuals with epilepsy.10,11 Diﬀerences in yield between studies is probably due to the selection of children for microarray analysis, which is often based on the presence of additional features other than epilepsy. For example, a higher yield of microarray analysis is found in individuals with epilepsy when the epilepsy is accompanied by global developmental delay or cognitive dysfunction.18 In our database, children who underwent microarray analysis more often had developmental problems, facial dysmorphisms and behavioral problems when compared to those who did not, i.e. the presence of such comorbidities prompted the treating physicians to request a microarray analysis. Probably due to this selection, we found no diﬀerences in epilepsy syndrome diagnosis or the presence of other phenotypic characteristics between children with and without clinically relevant CNVs.

We found a 2.6 Mb 8q22 deletion in one child. She (patient 575) is the seventh individual reported with such a deletion so far and shares the combination of absence seizures and focal seizures with two of the previously reported children.16,17 An eighth individual, listed in the DECIPHER database (case 2846), also has a 8q22 deletion and absence seizures. Thus, both focal and generalized (especially absences) seizures may occur in patients with 8q22 deletions. The smallest region of overlap harbors two candidate genes for epilepsy, NCALD (MIM 606722) and RRM2B (MIM 604712), which both have a function in the brain (Supplemental Figure 1).17

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 38

COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 39 MRI abnormalities MRI Diﬀuse Diﬀuse magna Normal Normal possible MTS and dysplasia dysplasia left cortical abnormality white matter white Megacisterna

breviations: bifront. = = bifront. breviations: 2 Additional features features Additional fissures dyskinesia clinodactyly fold, fifth finger Developmental Developmental Developmental problems, autism, autism, problems, pubertas praecox, pubertas praecox, Perinatal asphyxia, asphyxia, Perinatal pectus excavatum pectus abnormal position microcephaly, thick microcephaly, developmental anddevelopmental Developmental and eyes, entropion, thin eyes, entropion, of ears, shortof ears, stature, behavioral problems behavioral upslanting palpebralupslanting eyebrows, deeplyeyebrows, set behavioral problems, problems, behavioral PDD-NOS, epicanthal lips, high nasal bridge, lips, high nasal bridge, problems, hearing loss, hearing problems, This patient also carries a chromosome 2p16.3

b b b b b Epilepsy syndrome Epilepsy

hg19). hg19).

Focal Focal Focal

(effectiveness) (effectiveness) roic acid, y = years. L = left, = lamotrigine, max. LEV M = male, = levetiracetam, LTG Anti-epileptic drugs drugs Anti-epileptic None Focal Focal None ETX (-) LEV (+) CBZ (+) CBZ eﬀects) eﬀects) TPM (+) (+) TPM VPA (side- VPA VPA (side- VPA

LTG (unknown) LTG

on EEG (localization) EEG on Epileptiform activity activity Epileptiform Normal Normal temp., L>R) L>R) temp., Focal sharp waves (occ.) (occ.) waves SWC (fronto-

bifront.), focal focal bifront.), SWC (L temp.) temp.) (L SWC Gen. 3Hz SWC, Gen. SWC, 3Hz

Gen. (max. SWC focal and spikes sharp and waves Age at active epilepsy epilepsy active at Age 3y – 14y – 14y 3.5y – 3.5y

2y - 4y Normal (+) VPA Focal 3y - 12y

ongoing ongoing unknown

seizures)

(estimated number of of number (estimated Seizure types types Seizure sz. (4) sz. (10) sz. (10) Focal sz. Focal sz. gen. sz. (1) gen. sz. (1) (unknown)

sz. (2-4/ month) sz. month) (2-4/ (unknown), sec. (unknown), Focal SE (1), focal Focal (1), SE Focal SE (1), focal Focal (1), SE Frontal absences absences Frontal (3/day), sec. gen. (3/day), Relevant genes Relevant

MYT1L SCN4B NRXN3 NRXN3 UNC5D CNV size in kilobases in size CNV 71 319 376 238 3,207 Symptomatic focal epilepsy not otherwise speciﬁed. + >50% seizure frequency reduction.- <50% seizure frequency reduction. Ab seizure frequency <50% reduction.- seizure >50% speciﬁed. otherwise + not epilepsy focal Symptomatic b

gene. gene.

inheritance inheritance , de novo

de novo de novo a Microarray results, results, Microarray NRXN1 NRXN1 de novo de novo arr 14q24.3q31.1 arr 14q24.3q31.1 arr 8p12 (35,120,621- arr 2p25.3 (1,711,399- arr 2p25.3 2,078,557)x1, 2,078,557)x1, (father has migraine) (father has migraine)

35,358,315)x1, paternal paternal 35,358,315)x1, arr 14q31.1 (79,335,493- arr 14q31.1 79,654,245)x1

arr 11q23.3 (117,951,629- arr 11q23.3 (76,621,116-79,828,269)x1, (76,621,116-79,828,269)x1, 118,022,700)x1, unknown

(sex, age in years) in age (sex, Patient Patient 31 337 969 626 981 (F, 8) (F, (F, 9) (F, (F, 13) (F, (M, 16) (M, (M, 23) (M, bifrontal, = carbamazepine, CBZ ETX = ethosuximide, fronto-temp. F = female, = fronto-temporal, gen. = generalized, Hz = Hertz, = maximum, mo = months, MTS = mesiotemporal sclerosis, occ. = occipital, Pat = paternal, PDD-NOS = pervasive developmental disorder not otherwise speciﬁc, R = right, = status SE epilepticus, sec. gen. = secondarily generalized, SWC = spike-wave complexes, sz. = seizures, = valp = temporal, temp. TPM = topiramate, VPA Novel CNVs in our cohort with epilepsy (n = 5) CNVs 4: Novel of children Table The chromosomal coordinates are reported the relative Genome to Reference Consortium Human Reference genome (GRCh37/ version 37 deletion including the

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 39 40 CHAPTER 2

A large proportion of identiﬁed CNVs are also observed in <1% of the healthy controls. Among these CNVs were chromosome 15q11.2 and 15q13.3 deletions in three children with symptomatic focal epilepsy, while similar deletions are known to predispose to idiopathic generalized epilepsies (Table 3).6-8,19,20 We also found a chromosome 15q13.3 duplication in a child with focal seizures. An association between 15q13.3 duplications and epilepsy has only been reported in a few cases so far.9,21 The observations in our cohort suggest that chromosome 15q11.2 deletions and 15q13.3 deletions and duplications might predispose to both generalized and focal epilepsies. Although these CNVs are regarded as susceptibility CNVs for epilepsy, one should always consider that other causes of epilepsy may also be present, as seen in 30% of the children with susceptibility CNVs in our cohort (Table 3).

Novel CNVs of interest for epilepsy In ﬁve children, we identiﬁed novel CNVs that comprised four candidate genes for epilepsy (MYT1L, UNC5D, SCN4B and NRXN3) with either expression or function in the brain or a previous association with neurodevelopmental disorders (Supplemental Table 1).

We found a 376 kb deletion involving the MYT1L gene (MIM 613084) in a child with focal epilepsy and intellectual disability. MYT1L codes for a transcription factor that has an important role in the diﬀerentiation of cells to functional neurons.22 It has been identiﬁed as a candidate gene for intellectual disability in patients with 2p25.3 deletions,23,24 and seizures have been reported in 8/21 patients with such a deletion.7,23,25 The DECIPHER database includes another individual (case 259324) with absence seizures, intellectual disability, autism and a large chromosome 2p25 deletion including MYT1L. Thus, based on our and previous observations, MYTL1 deletions are not only associated with intellectual disability but also with epilepsy.

We found a deletion of UNC5D in a child with focal epilepsy, mesiotemporal sclerosis and developmental and behavioral problems, as well as in his father who had migraine. UNC5D (MIM 616466) on chromosome 8p12 has been shown to be involved in cortical development and p53- dependent apoptosis in neuroblastoma cells.26,27 UNC5D was considered as a candidate gene for neurodevelopmental phenotypes in a family with a t(6;8) balanced translocation that disrupted this gene in two aﬀected siblings with developmental delay (one with schizencephaly) and their asymptomatic mother.28 Further conﬁrmation that deletion of the UNC5D gene may cause or predispose to epilepsy and developmental problems is, however, needed.

In one child from our cohort with focal epilepsy and developmental problems, we found a deletion of the sodium channel voltage gated type IV beta subunit gene, SCN4B (MIM 608256). SCN4B is expressed in rat brain and spinal cords, and its protein has been shown to influence SCN2A by altering channel properties and shifting the voltage dependence of activation in the hyperpolarizing direction.29 Variants in the SCN2A gene are a well-known cause of benign familial neonatal and infantile seizures5 and early infantile epileptic encephalopathy.30 Haploinsufficiency of SCN4B may cause epilepsy indirectly by influencing SCN2A function.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 40 COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 41

Lastly, the NRXN3 gene (MIM 600567) was deleted in two unrelated children in our cohort. We had described one of them in a previous study: she has a severe developmental delay and a concomitant deletion of NRXN1 with no second NRXN1 sequence variant on the other allele (patient 337).31 NRXN3 encodes a polymorphic cell surface protein, neurexin III, that is expressed in neurons and necessary for neurotransmission. Deletions and variants in the neurexin I gene, NRXN1, have been associated with moderate to severe intellectual disability, language delay, autism spectrum 2 disorder and seizures.31 The severe phenotype of our patient was not in line with the milder phenotypes previously reported in children with heterozygous NRXN1 deletions so far, and we speculated that her severe phenotype might be explained by the additional deletion of NRXN3.31 In the current study, we found a second NRXN3 deletion in another unrelated child with symptomatic focal epilepsy and a pervasive developmental disorder not otherwise specified (Table 2). NRXN3 has been associated with bipolar disorder32 and autism spectrum disorder.33 Recently, NRXN3 deletions were reported in four individuals of three different families with epilepsy (unclassified in three, progressive myoclonic epilepsy in one), behavioral problems and developmental delay, or intellectual disability.10 The observation of NRXN3 deletions in two children in our cohort supports the idea that NRXN3 haploinsufficiency can be associated with epilepsy.

For all four candidate genes, MYT1L, UNC5D, SCN4B and NRXN3, additional patients with compatible genotypes and phenotypes, and/or supporting evidence from functional studies, are needed to conﬁrm their role in the pathophysiology of epilepsy.

In eight diﬀerent regions, we found CNVs that occurred 10 times more often in our study cohort than in healthy controls. The pathogenicity of these CNVs in epilepsy is doubtful because these children either had other identiﬁed epilepsy causes and/or the CNVs lacked genes of interest for epilepsy.

CONCLUSION

Our study demonstrates the importance of microarray analysis in the diagnostic work-up of epilepsy in childhood. We identified known clinically relevant CNVs for epilepsy in 11% of the children investigated. This yield was obviously influenced by the clinical selection of children, which was largely based on the presence of additional developmental or behavioral problems and/or facial dysmorphisms. Furthermore, we identified novel CNVs that include four new candidate genes for epilepsy: MYT1L, UNC5D, SCN4B and NRXN3. Analysis of these genes in larger study cohorts is warranted to further confirm their role in the etiology of epilepsy.

LINKING TO DATABASES

UCSC Genome Bioinformatics DatabaseE of genomiC varIation and Phenotype in Humans using Ensembl Resources (DECIPHER) Online Mendelian Inheritance in Man (OMIM) Database of Genomic Variants (DGV)

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 41 42 CHAPTER 2

REFERENCES

1. Thomas RH, Berkovic SF. The hidden genetics of 18. Mercimek-Mahmutoglu S, Patel J, Cordeiro D, et epilepsy-a clinically important new paradigm. Nat Rev al. Diagnostic yield of genetic testing in epileptic Neurol 2014; 10: 283-292 encephalopathy in childhood. Epilepsia 2015; 56: 707-716. 2. Myers CT, Mefford HC. Advancing epilepsy genetics in the 19. Shinawi M, Schaaf CP, Bhatt SS, et al. A small recurrent genomic era. Genome Med 2015; 7: 91 deletion within 15q13.3 is associated with a range of 3. Schaaf CP, Wiszniewska J, Beaudet AL. Copy number and neurodevelop-mental phenotypes. Nat Genet 2009; 41: SNP arrays in clinical diagnostics. Annu Rev Genomics 1269-1271. Hum Genet 2011; 12: 25-51. 20. Helbig I, Mefford HC, Sharp AJ, et al. 15q13.3 4. Heron SE, Cox K, Grinton BE, et al. Deletions or duplications microdeletions increase risk of idiopathic generalized in KCNQ2 can cause benign familial neonatal seizures. J epilepsy. Nat Genet 2009; 41: 160-162. Med Genet 2007; 44: 791-796. 21. Moreira DP, Griesi-Oliveira K, Bossolani-Martins AL, et al. 5. Zara F, Specchio N, Striano P, et al. Genetic testing in Investigation of 15q11-q13, 16p11.2 and 22q13 CNVs in benign familial epilepsies of the first year of life: Clinical autism spectrum disorder brazilian individuals with and and diagnostic significance. Epilepsia 2013; 54: 425-436. without epilepsy. PLoS One 2014; 9: e107705. 6. Mefford HC, Muhle H, Ostertag P, et al. Genome-wide 22. Pang ZP, Yang N, Vierbuchen T, et al. Induction of human copy number variation in epilepsy: Novel susceptibility neuronal cells by defined transcription factors. Nature loci in idiopathic generalized and focal epilepsies. PLoS 2011; 476: 220-223. Genet 2010; 6: e1000962. 23. Stevens SJ, van Ravenswaaij-Arts CM, Janssen JW, et al. 7. de Kovel CG, Trucks H, Helbig I, et al. Recurrent MYT1L is a candidate gene for intellectual disability in microdeletions at 15q11.2 and 16p13.11 predispose to patients with 2p25.3 (2pter) deletions. Am J Med Genet idiopathic generalized epilepsies. Brain 2010; 133: 23-32. A 2011; 155A: 2739-2745. 8. Jahn JA, von Spiczak S, Muhle H, et al. Iterative 24. De Rocker N, Vergult S, Koolen D, et al. Refinement of phenotyping of 15q11.2, 15q13.3 and 16p13.11 the critical 2p25.3 deletion region: The role of MYT1L in microdeletion carriers in pediatric epilepsies. Epilepsy intellectual disability and obesity. Genet Med 2015; 17: Res 2014; 108: 109-116. 460-466. 9. Olson H, Shen Y, Avallone J, et al. Copy number variation 25. Mayo S, Rosello M, Monfort S, et al. Haploinsufficiency of plays an important role in clinical epilepsy. Ann Neurol the MYT1L gene causes intellectual disability frequently 2014; 75: 943-958. associated with behavioral disorder. Genet Med 2015; 17: 10. Faheem M, Naseer MI, Chaudhary AG, et al. Array- 683-684. comparative genomic hybridization analysis of a cohort 26. Sasaki S, Tabata H, Tachikawa K, et al. The cortical of saudi patients with epilepsy. CNS Neurol Disord Drug subventricular zone-specific molecule Svet1 is part of the Targets 2015; 14: 468-475. nuclear RNA coded by the putative netrin receptor gene 11. Naseer MI, Faheem M, Chaudhary AG, et al. Genome wide Unc5d and is expressed in multipolar migrating cells. Mol analysis of novel copy number variations duplications/ Cell Neurosci 2008; 38: 474-483. deletions of different epileptic patients in saudi arabia. 27. Wang H, Wu Q, Li S, et al. Unc5D regulates p53-dependent BMC Genomics 2015; 16: S10. apoptosis in neuroblastoma cells. Mol Med Rep 2014; 9: 12. Fisher RS, Acevedo C, Arzimanoglou A, et al. ILAE official 2411-2416. report: A practical clinical definition of epilepsy. Epilepsia 28. Utami KH, Hillmer AM, Aksoy I, et al. Detection 2014; 55: 475-482. of chromosomal breakpoints in patients with 13. Engel J,Jr. Report of the ILAE classification core group. developmental delay and speech disorders. PLoS One Epilepsia 2006; 47: 1558-1568. 2014; 9: e90852. 14. MacDonald JR, Ziman R, Yuen RK, et al. The database 29. Yu FH, Westenbroek RE, Silos-Santiago I, et al. Sodium of genomic variants: A curated collection of structural channel beta4, a new disulfide-linked auxiliary subunit variation in the human genome. Nucleic Acids Res 2014; with similarity to beta2. J Neurosci 2003; 23: 7577-7585. 42: D986-992. 30. Ogiwara I, Ito K, Sawaishi Y, et al. De novo mutations of 15. Hanemaaijer N, Dijkhuizen T, Haadsma M, et al. A 649 voltage-gated sodium channel alphaII gene SCN2A in kb microduplication in 1p34.1, including POMGNT1, intractable epilepsies. Neurology 2009; 73: 1046-1053. in a patient with microcephaly, coloboma and 31. Bena F, Bruno DL, Eriksson M, et al. Molecular and clinical laryngomalacia; and a review of the literature. Eur J Med characterization of 25 individuals with exonic deletions Genet 2009; 52: 116-119. of NRXN1 and comprehensive review of the literature. 16. Kuechler A, Buysse K, Clayton-Smith J, et al. Five Am J Med Genet B 2013; 162B: 388-403. patients with novel overlapping interstitial deletions in 32. Noor A, Lionel AC, Cohen-Woods S, et al. Copy number 8q22.2q22.3. Am J Med Genet A 2011; 155A: 1857-1864. variant study of bipolar disorder in canadian and UK 17. Kuroda Y, Ohashi I, Saito T, et al. Refinement of the populations implicates synaptic genes. Am J Med Genet deletion in 8q22.2-q22.3: The minimum deletion size at B 2014; 165B: 303-313. 8q22.3 related to intellectual disability and epilepsy. Am J 33. Vaags AK, Lionel AC, Sato D, et al. Rare deletions at the Med Genet A 2014; 164A: 2104-2108. neurexin 3 locus in autism spectrum disorder. Am J Hum Genet 2012; 90: 133-141.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 42 COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 43

SUPPLEMENTAL DATA

2 and one individual in the DECIPHER 2 to intellectualto disability and epilepsy. Am J Med Genet m J Med Genet A 2011; 155A: 1857-1864. 155A: m J Med Genet A 2011; one individual reported by Kuroda et al et Kuroda by reported individual one 1 2014; 164A: 2014-2108. 2014; Supplemental Figure 1: Overview Figure 8q22 deletions. Supplemental of chromosome 2. et al. Ohashi Reﬁnement I, Saito Kuroda T, of the Y, deletion in 8q22.2-q22.3: The minimum deletion size at 8q22.3 related The 8q22 deletions observed in one patient of our cohort, ﬁve individuals reported by Kuechler et al, References 1. Kuechler A, Buysse K, Clayton-Smith et al. J, Five patients with novel overlapping interstitial deletions in 8q22.2q22.3. A database (https://decipher.sanger.ac.uk/). The chromosomal coordinates are reported the relative Genome to Reference Consortium Human Reference genome (GRCh37/ version 37 The smallest region of overlap is indicated by the highlighted zone.hg19). Only protein-coding genes are shown.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 43 44 CHAPTER 2

Supplemental Table 1: Information on novel candidate genes

Gene (protein) MIM Information on expression, function and/or disease association MYT1L 613084 - Expression in embryonic rat brains from day 15, with maximal (myelin transcription expression at parturition and subsequent decline to low but factor 1-like) detectable levels in adult rat brains 1 - Functions as transcription factor to convert non-neuronal human somatic cells or pluripotent stem cells into functional neurons 2 - Identified as candidate gene for intellectual disability in patients with 2p25.3 deletion syndrome 3,4 UNC5D 616466 - Expression in multipolar cells of the subventricular zone throughout (unc-5 homolog D) cortical development 5. The cortical subventricular zone-specific molecule Svet1 is part of the nuclear RNA coded by the putative netrin receptor gene Unc5d and is expressed in multipolar migrating cells. - Functions as transcriptional target of pro-apoptotic p53 and might induce p53-dependent apoptosis by ser-15 phosphorlation in neuroblastoma cells 6 - Disruption of UNC5D is reported in a patient with schizencephaly and developmental delay due to a t(6;8) balanced translocation 7 SCN4B 608256 - Expression in neurons in rat brain and spinal cord and some sensory (Sodium channel, neurons 8 voltage-gated, type IV, - Whole-cell recording in co-transfected embryonic kidney cells showed beta subunit) that SCN4B alters the channel properties of SCN2A. There is a shift in voltage dependent activation in the hyperpolarizing direction, but no shift in the voltage dependent inactivation 8 NRXN3 600567 - Polymorphic cell surface protein that is expressed in neurons 9 (Neurexin III) - Alpha-neurexin I, II and/or III knockout mice showed impaired function of synaptic calcium channels suggesting that alpha-neurexins organize presynaptic terminals 10 - A clinical study including 4 families with at least one member with autism suggested that deletions comprising NRNX3 may predispose to developing autism spectrum disorder 11 - NRXN3 deletions have been found in four individuals of three different families with unclassified epilepsy (n = 3) or progressive myoclonic epilepsy (n = 1) 12 Abbreviations: MIM = Mendelian Inheritance in Men database number (https://omim.org) References

1. Kim JG, Armstrong RC, Agoston D, et al. Myelin 6. Wang H, Wu Q, Li S, et al. Unc5D regulates p53- transcription factor 1 (Myt1) of the oligodendrocyte dependent apoptosis in neuroblastoma cells. Mol Med lineage, along with a closely related CCHC zinc finger, Rep 2014; 9: 2411-2416. is expressed in developing neurons in the mammalian 7. Utami KH, Hillmer AM, Aksoy I, et al. Detection central nervous system. J Neurosci Res 1997; 50: 272-290. of chromosomal breakpoints in patients with 2. Pang ZP, Yang N, Vierbuchen T, Induction of human developmental delay and speech disorders. PLoS One neuronal cells by defined transcription factors. Nature 2014; 9: e90852. 2011; 476: 220-223. 8. Yu FH, Westenbroek RE, Silos-Santiago I, et al. Sodium 3. Stevens SJ, van Ravenswaaij-Arts CM, Janssen JW, et al. channel beta4, a new disulfide-linked auxiliary subunit MYT1L is a candidate gene for intellectual disability in with similarity to beta2. J Neurosci 2003; 23: 7577-85. patients with 2p25.3 (2pter) deletions. Am J Med Genet 9. Ushkaryov YA, Petrenko AG, Geppert M, et al. Neurexins: A 2011; 155A: 2739-2745 synaptic cell surface proteins related to the alpha- 4. De Rocker N, Vergult S, Koolen D, et al. Refinement of latrotoxin receptor and laminin. Science 1992; 257: 50-56. the critical 2p25.3 deletion region: the role of MYT1L in 10. Hishimoto A, Liu QR, Drgon T, et al. Neurexin 3 intellectual disability and obesity. Genet.Med 2015; 17: polymorphisms are associated with alcohol dependence 460-66 and altered expression of specific isoforms. Hum Mol 5. Sasaki S, Tabata H, Tachikawa K, et al. The cortical Genet 2007; 16: 2880-91. subventricular zone-specific molecule Svet1 is part of 11. Vaags AK, Lionel AC, Sato D, et al. Rare deletions at the the nuclear RNA coded by the putative netrin receptor neurexin 3 locus in autism spectrum disorder. Am J Hum gene Unc5d and is expressed in multipolar migrating Genet 2012; 90: 133-141. cells. Mol Cell Neurosci 2008; 38: 474-483. 12. Faheem M, Naseer MI, Chaudhary AG, et al. Array- comparative genomic hybridization analysis of a cohort of Saudi patients with epilepsy. CNS Neurol Disord Drug Targets 2015; 14: 468-75.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 44 COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 45 MYTL1

pathogenic pathogenic 2 this region less likely Reason for interpreting interpreting for Reason Other epilepsy likely cause:carbohydrate glycoprotein deﬁcient syndrome type IC (patient 719) Other epilepsy likely cause: duplication a 15q11.2-q13 356) (patient No genes of interest Other epilepsy likely cause: duplication a 15q11.2-q13 1062) (patient No genes of interest Another more likely epilepsy cause: 981) deletion (patient No genes of interest No genes of interest patient , developmental , developmental , developmental , developmental , developmental , developmental a a a a Phenotype second second Phenotype Patient 719 (M, 8y): focal 8y): (M, 719 Patient epilepsy and behavioral problems, problems, behavioral and brain hypoplasia on MRI and inverted nipples problems, behavioral and hypotonia and inverted nipples abnormalities problems, on MRI corresponding to perinatal asphyxia and microcephaly Patient 981 (M, 16y): focal 16y): (M, 981 Patient epilepsy problems, autism, pubertas autism, problems, praecox and up slanting palpebral ﬁssures CSWS, 6y): (M, 1070 Patient problems, developmental ataxia, high forehead, tremor, deeply set eyes and thin upper lip Patient 1116 (F, 2y): focal 2y): (F, 1116 Patient epilepsy Patient 1044 (M, 6y): focal 6y): 1044 (M, Patient epilepsy , , and a a a patient patient Phenotype first first Phenotype Patient 356 (M, 16y): 16y): (M, 356 Patient focal epilepsy Patient 99 (M, 99 (M, Patient localization 9): cryptogenic epilepsy, developmental problems, autism macrocephaly, and tall stature developmental developmental epicanthal problems, fold and upturned nasal tip Patient 841 (F, 6y): 6y): (F, 841 Patient West Syndrome and developmental problems developmental anddevelopmental problems. behavioral developmental developmental hypotonia, problems, highly autism, arched eyebrows, hypertelorism, full lips widely spaced and teeth Patient 1062 (M, 5y): 5y): (M, 1062 Patient focal epilepsy Patient 574 (F, 16y): 16y): (F, 574 Patient focal epilepsy b PRSS1 USP41 PDE1C PDE1C Genes Genes RIMBP3 FCGR1A GGTLC3 PPIAL4C FAM72C FAM231D FAM230A LRRC4C HIST2H2BF 2 1 2 2 2 (%) (0.08) (0.04) (0.08) (0.08) (0.08) LLC (n=2402) (n=2402) LLC Frequency in in Frequency 3 3 0 2 0 (-) (-) (%) (0.01) (0.02) (0.02) in DGV (n=14316) Frequency Frequency 2 2 2 2 2 (%) (0.88) (0.88) (0.88) (0.88) (0.88) (n=226) (n=226) in cohort in Frequency Frequency CNVs with a ten times higher frequency in the study cohort compared to controls times higher frequency with a ten in the study cohort controls CNVs to compared (93) (145) (157) (363) (465) 49,763,118 32,383,345 20,709,090 40,484,707 40,339,414 - 40,339,414 32,226,319 - 32,226,319 142,486,373 20,345,668 - 20,345,668 (size in kb) 142,393,823 - 142,393,823 149,298,622 - 149,298,622 Coordinates Coordinates 7q34 q21.2 11p12 7p14.3 1q21.1- Region 22q11.21 Duplications (n = 4) = Duplications (n Deletions = 4) (n Supplemental Table 2: Table Supplemental

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 45 46 CHAPTER 2 sociated with with sociated (MIM 606829) FXN pathogenic pathogenic this region less likely Reason for interpreting interpreting for Reason Other epilepsy likely a mito-chondrial causes: respiratory chain defect and 1080) (patient Gomez-Lopez-Hernandez syndrome (patient 1055) Other epilepsy likely cause: a mitochondrial depletion syndrome 1075) (patient No genes of interest patient , Gomez-Lopez- , developmental a a rder not otherwise speciﬁed. Phenotype second second Phenotype Patient 1075 (M,4y): focal (M,4y): 1075 Patient epilepsy BFIS 4y): (F, 1090 Patient No genes of interest Hernandez syndrome, syndrome, Hernandez problems, developmental abnormalities and on MRI, hydrocephaly hypertonia microcephaly, of the legs and unspeciﬁed facial dysmorphisms hypotonia, problems, hypodense in areas mesencephalon and pons tented on MRI, high palate, mouth, chin, small small upliftedears, lobes, ear transversesingle palmar clinodactyly and crease digiti IV Patient 1055 (F, 5y): focal 5y): (F, 1055 Patient epilepsy e waves during slow-wave sleep syndrome, DGV = database of Compound heterozygous repeat expansion of of expansion repeat heterozygous Compound e (MIM 608817) is a member of the netrin family of axon guidance molecules and is is and molecules guidance axon of family netrin the of member a is 608817) (MIM , developmental , developmental a , developmental , developmental , developmental LRR4C4 a a

b Patient 1048 (M, 5y): focal 5y): (M, 1048 Patient epilepsy problems and PDD-NOS. and problems acrocephaly, problems, agenesis callosum andcorpus asymmetric on MRI, ventricles underdeveloped alae nasi, hypertelorism, extra nipple, stature and tall bronchomalacia Patient 1080 (M, 4y): WS and 4y): (M, 1080 Patient focal epilepsy problems, brain atrophy and and atrophy brain problems, hypoplasia on MRI, hypotonia, ballism, tall stature and foldepicanthal Patient 1072 (M, 17y): focal 17y): (M, 1072 Patient epilepsy d c e IL9R IL9R FXN VAMP7 CCL4L1 CCL3L1 CCL4L2 Genes Phenotype first patient TBC1D3C TBC1D3G TBC1D3G TBC1D3H PIP5K1B PRKACG Reference Consortium Human Reference genome version 37 (GRCh37/hg19). Genes in bold are expressed in the brainReference or Consortium Human Reference genome (GRCh37/hg19). version 37 1 2 1 (%) (0.08) (0.04) (0.04) in LLC (n=2402) Frequency Frequency 0 6 0 (-) (-) (%) (0.04) in DGV (n=14316) Frequency Frequency Symptomatic focalSymptomatic epilepsy not otherwise speciﬁed. a (MIM 602745) encodes a functional phosphatidylinositol 602745) (MIM phosphate kinase and has been isolated from the region of the genome as (MIM 176893) have been associated with short-term episodic memory performance. performance. memory episodic short-term with associated been have 176893) (MIM 2 2 2 PIPI5K1B (%) c (0.88) (0.88) (0.88) (n=226) (n=226) in cohort in Frequency Frequency Variations in PRKACG (49) (114) (278) d 71,727,277 34,817,622 155,229,114 71,613,395 - 71,613,395 34,539,666 - - 34,539,666 (size in kb) 155,180,288 - 155,180,288 Coordinates Coordinates Xq28 17q12 9q21.11 Region have beenhave associated Ataxia. with Friedreich’s Abbreviations: = benign BFIS familial infantile seizures, CSWS = continuous spik genomic variants, = Low Lands Consortium LLC database, MRI = magnetic resonance imaging, PDD-NOS = pervasive developmental diso Continuation of Supplemental Table 2 Table of Supplemental Continuation The chromosomal coordinates are reported the relative Genome to associated with a neuropsychiatric disease. expressed in adult and fetal brain. Friedriech Ataxia. Ataxia. Friedriech

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 46 COPY NUMBER VARIATION IN A HOSPITALBASED COHORT OF CHILDREN WITH EPILEPSY 47

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 47 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 48 4949

Chapter 3

Empowerment and anxiety of patients and parents during genetic counseling for epilepsy

Manuscript ready to submit

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 49 50 CHAPTER 3

Danique RM Vlaskamp1,2, Patrick Rump1, Petra MC Callenbach2, Jan S Voorwinden3, Eva H Brilstra4, Mary E Velthuizen4, Oebele F Brouwer2, Adelita Ranchor3, Conny M.A. van Ravenswaaij-Arts1

1 University of Groningen, University Medical Center Groningen, Department of Genetics, Groningen, the Netherlands. 2 University of Groningen, University Medical Center Groningen, Department of Neurology. Groningen, the Netherlands. 3 University of Groningen, University Medical Center Groningen, Department of Health Psychology, the Netherlands. 4 University Medical Center Utrecht, Department of Genetics, Utrecht, the Netherlands.

Acknowledgements. We thank all patients and their parents for their participation in our study. We are thankful for the help of all research assistants, especially Rianne Lieben, Denise Blom, Elise Boersma and Sumalai Sompitak, with data collection and input. We thank Kate Mc Intyre for editing our manuscript.

Disclosures. None of the authors have any conﬂicts of interest to disclose.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 50 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 51

ABSTRACT

Purpose. To study the patient-reported outcome of genetic counseling before and after genetic testing for epilepsy by evaluating empowermenta combination of cognitive, decisional and behavioral control, emotional regulation and hope—and anxiety.

Methods. Patients or their parents (if <16 years old or intellectually disabled) referred to two university hospitals for genetic testing for epilepsy between June 2014 and 2017 were asked to complete three questionnaires: one before pre-test counseling, one after, and one following 3 post-test counseling. Empowerment was measured with the Genetic Counseling Outcome Scale (GCOS-18), anxiety with the short State Trait Anxiety Inventory (STAI-6).

Results. Of 106 participants who had genetic testing, 63 completed all three questionnaires and were included in our study. Empowerment significantly increased during the genetic counseling trajectory with a medium effect size (p<0.001, d=0.57), and a small but significant increase in empowerment was already seen after pre-test counseling (p=0.038, d=0.29). Anxiety did not change significantly during the counseling trajectory (p=0.223, d=-0.24).

Conclusion. Patients with epilepsy or their parents show a clinically relevant increase in empowerment after genetic counseling, which is a key outcome goal of genetic counseling. Empowerment was already increased after pre-test counseling, indicating the importance of counseling before initiating genetic testing for epilepsy.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 51 52 CHAPTER 3

INTRODUCTION

The increasing use of genetic testing in individuals with epilepsy is transforming epilepsy care. Finding a genetic cause for epilepsy, even though this is still only possible for the minority of patients, precludes unnecessary further diagnostic investigations and leads to a better understanding of epilepsy etiology, comorbidities, prognosis and recurrence risks.1–4 For a very few cases, ﬁnding the genetic variant for epilepsy may even improve treatment and outcome.5 Genetic testing in epilepsy is therefore increasingly part of routine diagnostic care.6,7 However, little is known about the psychological outcomes of genetic services from the patient or parent perspective.8–10

Previous qualitative studies showed that patients with epilepsy or their parents have a strong hypothetical interest in genetic testing, if offered, especially in a scenario where knowing the genetic change would improve medical care.11–13 Study participants mentioned both potential benefits (such as better understanding and care in children at risk and more sense of control and less guilt, blame and anxiety with negative test results) and potential concerns (including increased blame, guilt, stigma, discrimination, self-imposed limitations on life goals, and alterations in fundamental conceptions of ‘what epilepsy is’).13 Individuals with a familial epilepsy for which a genetic cause has been identified also expressed both positive and negative feelings on receiving a genetic diagnosis.14 To date, the psychological outcomes of genetic services for epilepsy have not been studied systematically.

In our current clinical practice, genetic testing for epilepsy is preceded and followed by genetic counseling. During pre-test counseling, counselors ﬁrst obtain a medical and family history to decide which genetic test would be most suitable. Subsequently, they inform the patients and their families about genetic testing and encourage them to make an informed choice about whether this testing should be done. During post-test counseling, the test results are explained to the patients and families. The overall aim of genetic counseling is helping people to understand and adapt to the medical, psychological and familial implications of identifying genetic contributions to disease.15 By this means, genetic counseling can lead to increased knowledge, increased perceived personal control, positive health behavior, improved risk perception accuracy and decreased decisional conﬂict, anxiety and worry.16 Studying the psychological outcome of genetic services for epilepsy may help counselors to improve the counseling trajectory in accordance with patient’s and their families’ needs.

These psychological outcomes can be measured by evaluating the change in ‘empowerment’ and anxiety. Empowerment is an all-encompassing patient-reported outcome of genetic counseling, defined as the set of beliefs that a person can make important life decisions (decisional control), has sufficient information about the condition (cognitive control), can make effective use of health and social care systems (behavioral control), is able to manage feelings about having a

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 52 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 53

genetic condition in the family (emotional regulation) and has hope for a fulﬁlling family life (hope).17–19 We aimed to study the outcome of genetic counseling before and after genetic testing for epilepsy by evaluating empowerment and anxiety of patients or their parents.

METHODS

Study cohort and design Our research was part of a larger study on the Dutch version of the Genetic Counseling Outcome 3 Scale (GCOS) and followed the same study design.20 All patients who were referred to a clinical geneticist in the outpatient clinics of the University Medical Center Groningen (UMCG) or the University Medical Center Utrecht (UMCU) in the Netherlands were eligible for inclusion in the study. For our study, all patients who were referred for genetic counseling and testing for epilepsy between June 2014 and June 2017 were eligible for inclusion.

The research had a pre-post observational study design. All patients were asked to complete three questionnaires during the genetic counseling trajectory: 1. before pre-test counseling (T0), 2. around 1-2 weeks after pre-test counseling (T1), and 3. around 1-2 months after genetic testing and post-test counseling (T2, Figure 1). If patients were under 16 years of age or intellectually disabled, one of their parents or caretakers was asked to complete the questionnaires for their child, but from their own perspective. We will use the term ‘participants’ for those patients or parents who completed the questionnaires. For one patient, a legal representative who was not a parent completed the questionnaires, but was included as a parent. We excluded the participants who declined genetic testing or who did not complete all three questionnaires.

Measurement instruments We used two patient-reported outcome measures that were included in the questionnaires. Empowerment was measured using the validated Dutch version of the Genetic Counseling Outcome Scale (GCOS). The Dutch version includes 18 of the original 24 English questions (GCOS-24), categorized into six subscales: hope and coping, knowledge about the condition, knowledge about genetic services, uncertainty about genetic services, negative emotions, and uncertainty about heredity (Supplemental Table 1 and 2).20 The GCOS-18 shows a satisfactory internal consistency (Cronbach’s α = .77) and excellent test-retest reliability (ICC = .92).20 Anxiety was measured with the short 6-item version of the Spielberger State-Trait Anxiety Inventory (STAI), which is a validated questionnaire that can be used to measure the psychological outcome of genetic counseling.16,21,22 Questions on baseline characteristics were included in the ﬁrst questionnaire.

After obtaining the participants’ consent for extracting information from medical records, we evaluated the pre- and post-test counseling letters from the clinical geneticists for the referral

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 53 54 CHAPTER 3

reasons, the presence of seizures, the results from genetic testing performed before and during our study, and the initiation of further genetic testing after completion of our study. The results from genetic testing were categorized into three groups: a disease-associated pathogenic variant, a variant of unknown signiﬁcance, or normal test results.

Outcome Our primary outcomes were empowerment (GCOS) and anxiety (STAI) scores throughout the genetic counseling trajectory. Secondary outcomes concerned the six subscales of empowerment.

Statistical analysis The total and subscores on the GCOS and STAI were calculated by adding all item scores after reversing item scores for negatively formulated questions. STAI scores were converted to the 20-item STAI questionnaire to allow comparison with reference values, as recommended in the manual.21 Missing items of the GCOS or STAI were imputed using the mean of the other GCOS item scores for that individual if ≤20% of the items were missing. If >20% items of the GCOS or STAI were missing, the participant was excluded from the analyses on this questionnaire.

We first studied the change in empowerment and anxiety scores during the genetic counseling trajectory in the total study group using repeated measurements ANOVA tests. Indicators for change were statistical significance and effect sizes. Effect sizes were calculated with the formula with pooled SD 7 An effect size ≥0.2 was considered small, ≥0.5 medium, and ≥0.8 large.23 An effect size of >0.5 was considered as the threshold for a minimal clinically important change for our patient-reported outcome measurements.24,25 To evaluate the outcome of genetic counseling on an individual level, we calculated the changes in GCOS and STAI scores between T2 and T0 and their effect sizes for each individual using the formula . Individual changes with an effect size >0.5 were considered a clinically relevant increase, changes with an effect size between 0.5 and -0.5 as stable, and changes with an effect size <-0.5 as a clinically relevant decrease.24,25

Second, we compared the total GCOS and STAI scores at baseline (T0) between participants with diﬀerent demographic characteristics or genetic testing results using ANOVA tests.

Third, a full-factorial repeated measures ANOVA test was used to evaluate the inﬂuence of demographic characteristics and genetic testing results on the course of the GCOS and STAI scores over time. We also compared the number of participants with clinically relevant increased, stable, or decreased GCOS and STAI scores between subgroups of participants based on these same demographic characteristics and genetic testing results using Fisher’s exact tests.

Lastly, Fisher’s exact tests were also used to compare categorical baseline characteristics and independent T-tests for continuous baseline characteristics between participants who accepted

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 54 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 55

and participants who declined genetic testing and between participants who did and participants who did not complete all follow-up questionnaires.

We used SPSS Statistics Version 23.0 (IBM Corporation, NY, USA). Analyses were two-tailed. A p-value <0.05/2 was considered statistically significant for our two primary outcomes, and a p-value <0.05/6 as statistically significant for our two secondary outcomes. In the post-hoc analyses for comparisons between T0-T1, T1-T2, and T0-T2, Bonferroni corrections were applied and p-values <0.05 were considered statistically significant. 3 Ethical statement The Institutional Medical Ethical Committee of the University Medical Center Groningen gave permission for this study (M13.139274). All participants gave written consent for participation in the study.

RESULTS

Characteristics of the study cohort In total, 116 initial participants who were referred for genetic counseling and testing for epilepsy agreed to participate. Of these, 106 (91%) decided to do genetic testing and 63/106 (59%) completed all questionnaires and were included in our study (Figure 1). Table 1 shows the characteristics of all initial participants (n = 116), participants who decided to do genetic testing (n = 106), and participants included in our study cohort (n = 63).

Participants who declined genetic testing had a higher baseline empowerment compared to those who decided to perform genetic testing (p = 0.003). Furthermore, participants who declined genetic testing were more often from the UMCU (p = 0.014), more often a patient (p = 0.001) and lived more often without children (p = 0.016). Among the participants who had genetic testing, follow-up questionnaires were signiﬁcantly more often completed by participants who had a higher education (p = 0.030). Other demographic characteristics, genetic testing results and baseline empowerment and anxiety scores did not diﬀer between those who did and did not decide to do genetic testing and those who did and did not complete the follow-up questionnaires (data not shown).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 55 56 CHAPTER 3

Figure 1: Inclusion and exclusion of participants in our study cohort.

Empowerment The mean empowerment score signifi cantly increased during the genetic counseling trajectory (p < 0.001; Figure 2A, Table 2). The overall change in empowerment had a medium eff ect size (d = 0.57), indicating a clinically relevant increase. Empowerment was already increased after pre- test counseling compared to baseline (p = 0.038), and a further increase was seen after post-test counseling compared to after pre-test counseling (p = 0.033). Both changes had a small eff ect size (d = 0.28 and d = 0.30, respectively). On an individual level, 32/63 (50.8%) participants showed a clinically relevant increase in empowerment and 6/63 (9.5%) a clinically relevant decrease. In the remaining 25/63 (39.6%), empowerment scores remained stable throughout the counseling trajectory.

Anxiety The mean anxiety score decreased during the genetic counseling trajectory, but the eff ect size was small (d = -0.24) and the diff erences were not statistically signifi cant (p = 0.223; Figure 2B, Table 2). These results were based on 58/63 participants with <20% missing STAI items. On an individual level, we observed a clinically relevant decrease in anxiety in 23/58 (39.7%) participants, a clinically relevant increase in anxiety in 14/58 (24.1%) participants, and a stable score in the remaining 21/58 (36.2%) participants.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 56 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 57 5 (7.9) 6 (9.5) 4 (6.3) 4 (6.3) 2 (3.2) 8 (12.7) 9 (14.3) 63 (100) 10 (15.9) 39 (61.9) 55 (87.3) 45 (71.4) 61 (96.8) 18 (28.6) (n = 63) (n 14 (22.2) 44 (69.8) 44 46 (73.0) 46 4/57 (7.0) 4/57 3/57 (5.3) 2/57 (3.5) 2/57 9/58 (15.5) (15.5) 9/58 8/62 (12.7) 8/62 14/57 (24.6) 21/58 (36.2) 34/57 (59.6) 34/57 28/58 (48.3) 91.3 (11.7, 63) (11.7, 91.3 39.6 (11.2, 62) (11.2, 39.6 Study cohort Study

mptomatic relative. mptomatic 3 Different situation includes living with one 2 2 (1.9) 21 (19.8) 21 (19.8) 17 (16.0) 15 (14.2) 74 (69.8) 38 (35.8) 64 (60.4) 64 89 (84.0) 68 (64.2) 68 3/98 (3.1) 106 (100) 106 6/98 (6.1) 6/98 9/98 (9.2) 104 (98.1) 5/105 (4.8) 5/105 (4.8) 5/105 9/105 (8.6) 9/105 58/98 (59.2) 22/98 (22.4) 12/105 (11.4) 27/100 (27.0) 27/100 21/100 (21.0) 16/105 (15.2) 74/105 (70.5) 52/100 (52.0) 91.3 (12.0, 105) (12.0, 91.3 40.8 (13.0, 103) 40.8 (13.0, testing = 106) (n Participants who had genetic genetic had who Participants 2 (1.7) 16 (13.8) 76 (65.5) 76 92 (79.3) 24 (20.7) 70 (60.3) 46 (39.7) 46 114 (98.3) 106 (91.4) (n = 116) (n 7/115 (6.1) 7/115 5/115 (4.3) 5/115 7/108 (6.5) 7/108 9/108 (8.3) 9/108 3/108 (2.8) 3/108 10/115 (8.7) 77/115 (67.0) 77/115 16/115 (13.9) 30/110 (27.3) 30/110 24/110 (21.8) 21/106 (19.8) 21/106 (19.8) 56/110 (50.9) 56/110 16/109 (14.7) 16/109 25/108 (23.1) 25/108 64/108 (59.3) 64/108 64/106 (60.4) 64/106 40.1 (13.0, 113) (13.0, 40.1 92.3 (12.4, 115) (12.4, 92.3 All initial participants initial All 3 2 Mean total score on GCOS at T0 (SD, n) (SD, T0 at on GCOS score Mean total Mean total score on STAI at T0 (SD, n) n) (SD, T0 at on STAI score total Mean - Living together with children - Living together without children - Living alone with children - Single situation - Different - Disease causing variant (%) variant causing - Disease of unknown (%) - Variant significance - Normal (%) - Working Studying - - Unemployed work (disabled) to - Unable - Retired - UMCG (%) - UMCG (%) - UMCU (%) - Low - Intermediate (%) - High (%) - Patient (%) - Patient - Parent (%) * Mother * Father symptomatic is - Patient pre-symptomatic is - Patient - Genetic testing performed between and T2 (%) T1 - Follow-up genetic testing after T2 (%) - - In these participants, a disease-causing gene variant for epilepsy that was already known in the family was tested in a pre-sy 3 1 1 Education Education level Hospital Hospital Marital status Results from testing genetic Employment Employment status Demographic characteristicsDemographic Participants Seizures Baseline empowerment and anxiety and empowerment Baseline Empowerment - Genetic characteristics testing Genetic testing Anxiety Characteristics differed significantly between participants included and who were not (n=57) were (n=63) in the study cohort ta not shown). (da Abbreviations: GCOS-18 = genetic Anxiety counseling = Spielberger outcome scale, STAI InventoryAbbreviations: State-Trait GCOS-18 parent or assisted living. Table 1: Characteristics of participants Table 1

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 57 58 CHAPTER 3

Figure 2: Empowerment and anxiety during the genetic counseling trajectory in the study cohort (Figure A and B) and in three subgroups based on genetic testing results (Figure C and D). ͿŵƉŽǁĞƌŵĞŶƚ ͿŶǆŝĞƚǇ Ϳ Ϳ ϭϮϲ ϴϬ Ϭ ϲ ϴ Ϯ

^ƚƵĚǇĐŽŚŽƌƚ;ŶсϲϯͿ Ͳ ^ƚƵĚǇĐŽŚŽƌƚ;ŶсϱϴͿ ϭ Ϭ Ͳ ϭϮϬ Ϯ ϴ ; ϭ Ğ

; ϳϬ

ƉфϬ͘ϬϬϭ͕ĚсϬ͘ϱϳ ƌ

Ğ ϭϭϰ Ž ƌ Đ Ž Đ Ɛ /

Ɛ ϭϬϴ ^

d ϲϬ K ^

ϭϬϮ ƉсϬ͘ϮϮϯ͕ĚсͲϬ͘Ϯϰ ' ϵϲ ϵϳ͘ϵ ϱϬ ϵϬ ϵϰ͘ϱ ϵϭ͘ϯ ϯϵ͘ϰ ϰϬ ϴϰ ϯϳ͘ϲ ϯϲ͘ϳ ϳϴ Ɖс Ϭ͘Ϭϯϯ͕ĚсϬ͘ϯϬ ϯϬ ϳϮ Ɖс Ϭ͘Ϭϯϴ͕ĚсϬ͘Ϯϴ ϲϲ ϮϬ ĞĨŽƌĞƉƌĞͲƚĞƐƚ ĨƚĞƌƉƌĞͲƚĞƐƚ ĨƚĞƌƉŽƐƚͲƚĞƐƚ ĞĨŽƌĞƉƌĞͲƚĞƐƚ ĨƚĞƌƉƌĞͲƚĞƐƚ ĨƚĞƌƉŽƐƚͲƚĞƐƚ ĐŽƵŶƐĞůŝŶŐ;dϬͿ ĐŽƵŶƐĞůŝŶŐ;dϭͿ ĐŽƵŶƐĞůŝŶŐ;dϮͿ ĐŽƵŶƐĞůŝŶŐ;dϬͿ ĐŽƵŶƐĞůŝŶŐ;dϭͿ ĐŽƵŶƐĞůŝŶŐ;dϮͿ

ͿŵƉŽǁĞƌŵĞŶƚďĂƐĞĚŽŶŐĞŶĞƚŝĐƚĞƐƚŝŶŐƌĞƐƵůƚƐ ͿŶǆŝĞƚǇďĂƐĞĚŽŶŐĞŶĞƚŝĐƚĞƐƚŝŶŐƌĞƐƵůƚƐ ϭϮϲ ϴϬ sĂƌŝĂŶƚŽĨƵŶŬŶŽǁŶƐŝŐŶŝĨŝĐĂŶĐĞ;ŶсϭϬͿ ŝƐĞĂƐĞͲĂƐƐŽĐŝĂƚĞĚǀĂƌŝĂŶƚ;ŶсϭϰͿ ϭϮϬ ŝƐĞĂƐĞͲĂƐƐŽĐŝĂƚĞĚǀĂƌŝĂŶƚ;ŶсϭϰͿ sĂƌŝĂŶƚŽĨƵŶŬŶŽǁŶƐŝŐŶŝĨŝĐĂŶĐĞ;ŶсϭϬͿ EŽƌŵĂůƌĞƐƵůƚƐ;ŶсϯϵͿ ϳϬ EŽƌŵĂůƌĞƐƵůƚƐ;ŶсϯϰͿ ϭϭϰ ϭϬϴ ϲϬ ^d/ƐĐŽƌĞ;ϮϬͲϴϬͿ

'K^ƐĐŽƌĞ;ϭϴͲϭϮϲͿ ϭϬϮ ϵϱ͘ϰ ϵϴ͘ϵ ϵϲ ϵϳ͘ϰ ϱϬ ϵϮ͘ϯ ϵϰ͘ϳ ϵϰ͘ϵ ϰϯ͘ϰ ϵϬ ϵϯ͘Ϭ ϴϵ͘ϴ ϯϴ͘ϱ ϰϬ ϯϴ͘ϴ ϯϵ͘Ϭ ϴϰ ϯϲ͘ϰ ϯϴ͘ϲ ϯϱ͘Ϯ ϳϴ ϯϲ͘ϯ ϯϲ͘Ϭ ϯϬ ϳϮ ϲϲ ϮϬ ĞĨŽƌĞƉƌĞͲƚĞƐƚ ĨƚĞƌƉƌĞͲƚĞƐƚ ĨƚĞƌƉŽƐƚͲƚĞƐƚ ĞĨŽƌĞƉƌĞͲƚĞƐƚ ĨƚĞƌƉƌĞͲƚĞƐƚ ĨƚĞƌƉŽƐƚͲƚĞƐƚ ĐŽƵŶƐĞůŝŶŐ;dϬͿ ĐŽƵŶƐĞůŝŶŐ;dϭͿ ĐŽƵŶƐĞůŝŶŐ;dϮͿ ĐŽƵŶƐĞůŝŶŐ;dϬͿ ĐŽƵŶƐĞůŝŶŐ;dϭͿ ĐŽƵŶƐĞůŝŶŐ;dϮͿ

The mean ± SD scores for empowerment and anxiety are presented. Empowerment significantly increased in the study cohort between T0, T1 and T2 (A). Anxiety scores did not decrease significantly in the study cohort (B). The results of genetic testing did not seem to significantly influence the course of empowerment and anxiety during the genetic counseling trajectory (C and D).

Of the 23 participants with a clinically relevant decreased anxiety score, 15 (65.2%) also had an increased empowerment score, 6 (26.1%) had a stable empowerment score, and only 2 (8.7%) participants had a decreased empowerment score. Furthermore, of the 14 participants with a clinically relevant increased anxiety score, only 2 (14.3%) also had a decreased empowerment score, while 8 (57.1%) had a stable empowerment score and 4 (28.6%) had an increased empowerment score.

Empowerment subscales During the genetic counseling trajectory, signiﬁcant increases in scores were seen in 3/6 subscales of the GCOS-18: knowledge about genetic services (p = 0.008, d = 0.44), uncertainty about genetic services (p = 0.006, d = 0.38), and uncertainty about heredity (p < 0.001, d = 0.63) (Table 2). Higher

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 58 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 59

scores indicated more knowledge and less uncertainty. In line with this, more participants had relevant increases in these three subscales compared to other subscales on an individual level (Table 2).

Empowerment and anxiety in subgroups of participants At baseline, empowerment and anxiety scores did not differ between subgroups of participants based on demographic and genetic testing characteristics (see Table 1 for tested characteristics, data not shown), except that baseline anxiety scores were significantly higher in participants with a low (48.3, SD 15.0) versus an intermediate (37.2, SD 9.5) or high (37.6, SD 9.1) education level (p = 0.020). 3 The changes in empowerment and anxiety scores during the genetic counseling trajectory were not significantly influenced by the genetic testing results (Figure 2C and 2D) or other demographic or genetic testing characteristics (data not shown). Also, the number of participants with relevant increases, decreases, or stable empowerment and anxiety scores did not differ significantly between participants with different demographic and genetic testing characteristics (data not shown).

DISCUSSION

With the increasing use of genetic testing for epilepsy, there is a need to study the outcome of genetic services from the perspective of patients and their parents. We found that patients and their parents show a clinically relevant increase in empowerment after genetic counseling before and after genetic testing for epilepsy, while their feelings of anxiety did not change signiﬁcantly. Some of the increase in empowerment was already seen after pre-test counseling, suggesting that pre-test counseling is an important part of the genetic counseling trajectory.

Empowerment Empowerment is a validated overarching construct that represents many specific outcomes of genetic counseling.19 Empowerment in our participants significantly increased on three of six subscales: knowledge about the genetic services, uncertainty about the genetic services, and uncertainty about heredity. Higher scores indicate more empowerment and less uncertainty. Translating these subscales into theoretical concepts of empowerment, our results indicate that during the genetic counseling trajectory participants made gains in behavioral control (making effective use of health and social care systems) and emotional regulation (managing feelings about having a genetic condition). Decisional control was increased in those who declined genetic testing after pre-test counseling. Knowledge about the disorder had not increased significantly, possibly because the participants had already received a lot of information about epilepsy from the referring clinicians. Also, feelings of hope about the future did not increase or decrease significantly after counseling.

Since our study was part of a larger study on the Dutch version of the GCOS-18 (n=2.194), and followed the same study design, we were able to compare our results.20 We found similar baseline

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 59 60 CHAPTER 3 6 (9.5) 6 (9.5) 6 (9.5) 9 (14.3) 12 (19.0) 12 (19.0) 16 (25.4) 23 (39.7) decrease (%) decrease (%) Participants with clinically relevant Individual level Individual 17 (27.0) 26 (41.3) 20 (31.7) 33 (52.4) 32 (50.8) 32 25 (39.7) 23 (36.5) increase (%) Participants with clinically relevant Effect sizes between T0 and T2 0.12 0.19 0.21 0.38 0.44 wledge about genetic services, uncertaintywledge about genetic about less level level Group Group 0.63 *** 0.63 0.321 0.092 0.222 <0.001* *** 0.57 0.008 ** 0.006 ** <0.001 ** T1 and T2 P-value for differences differences between T0, After 8.7 (3.6) 8.7 14.2 (4.1) 19.2 (2.0) 14.5 (3.3) 18.0 (2.3) 23.4 (3.6) 23.4 post-test post-test counseling (T2) After 7.4 (3.6) 7.4 17.4 (3.1) 14.2 (3.4) 13.5 (3.7) 18.9 (2.2) pre-test 23.2 (3.2) 23.2 counseling (T1) Mean total scores (SD) (T0) 6.4 (3.7) 6.4 17.7 (2.8) 13.1 (4.0) 13.4 (4.2) 18.2 (2.5) 22.6 (4.0) Baseline Baseline N 63 91.3 (11.7) 94.6 (11.1) 97.9 (11.5) 63 63 63 63 63 63 58 (11.4) 39.4 (9.8) 37.6 (10.8) 36.7 0.180 -0.24 (24.1) 14 1 (18-126) (20-80) A higher score means more empowerment for the six subscales: more hope and coping, more knowledge about the condition, more kno more condition, knowledge about the more coping, hope and more subscales: six the for empowerment more means score higher A Total STAI score STAI Total GCOS score on subscales subscales on score GCOS Total GCOS score score GCOS Total - Hope and coping (4-28) - Hope (4-28) and coping - Knowledge(3-21) condition - Knowledge genetic services (3-21) - Uncertainty genetic services (3-21) - Negative (3-21) emotions - Uncertainty heredity (2-14) Effects of genetic testing and counseling on empowerment anxiety of genetic testing 2: Effects in the study cohort Table scores Questionnaires score) possible maximum - (minimum 1 genetic services, less negative emotions, and less uncertainty about heredity. * P-value ** *** P-value is <0.05/2. Medium is <0.05/6. Abbreviations: effect = genetic GCOS counseling size (d>0.5). Anxietyoutcome scale = Spielberger (scale for empowerment), Inventory STAI State-Trait (scale for anxiety).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 60 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 61

empowerment scores in our subcohort (mean 91.3, SD 11.7, n = 63) to those seen in the total Dutch cohort (91.7, SD 12.1, n = 2.194), as well as similar increases in empowerment in both cohorts (d = 0.57 and d = 0.52, respectively).20,26 To date, three other studies evaluating the outcome of genetic counseling using the GCOS have also shown increased empowerment in people referred for genetic counseling and/or testing for any reason,27 attendees at a psychiatric counseling clinic28 , and in participants with suspected inherited retinal dystrophy.29 Further comparison with their results is not possible, since they used the GCOS-24 and not the GCOS-18.

Anxiety 3 Our participants had normal anxiety levels for their situation, and their anxiety did not significantly change during the genetic counseling trajectory. Although their mean baseline anxiety score (39.4) were higher compared to those in the normal adult population (30-35)30–32 , they were still at the proposed cut-off point for clinically significant anxiety symptoms (39-40).33 Further comparison of our results with those in the literature (available for females only) shows that the mean baseline anxiety score in the females in our cohort (39.6) was slightly higher than in females making non- invasive health care decisions such as whether genetic testing should be performed (36-39) and far below scores for females making invasive health care or difficult treatment decisions (50-62).34

Anxiety was not well captured in the concept of empowerment. A third of the participants with decreased anxiety did not feel more empowerment, and 85% of participants with increased anxiety did not experience less empowerment. Previous studies found contrary correlations between anxiety and empowerment.18,20 We therefore recommend also taking anxiety into account in evaluating the outcome of genetic counseling for epilepsy. Although anxiety levels did not change on a group level in our study, individual changes could occur during the genetic counseling trajectory.

The importance of pre-test counseling The results of our study indicate that genetic counseling before initiating genetic testing for epilepsy is important. First, about half of the increase in empowerment was already seen after pre-test counseling aimed at informed decision making. In line with this, a similar increase in empowerment after the first counseling session was also seen in the total Dutch GCOS study cohort.20 Second, our participants were not becoming more anxious when they approached genetic testing, while clinically significant anxiety scores were observed both before (47) and after (50) genetic testing in a previously published cohort without genetic counseling.35 We therefore emphasize the importance of counseling together with genetic testing. Lastly, 9/115 (8%) of the eligible participants who had pre-test counseling decided not to do genetic testing after pre- test counseling, while they initially agreed with referral to the genetic outpatient clinic. Notably, these participants had higher baseline empowerment scores compared to those who did not decline genetic testing. It is possible that participants with higher baseline empowerment scores feel that their psychological wellbeing would benefit less from genetic testing or they are better equipped to refrain from testing after counseling.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 61 62 CHAPTER 3

,QÁXHQFHRIJHQHWLFWHVWLQJUHVXOWVRQHPSRZHUPHQWDQGDQ[LHW\ Empowerment was not significantly influenced by the genetic testing results. Clinicians might be worried about decreasing participant empowerment by reporting a variant of unknown significance, but all ten participants in our cohort with a variant of unknown significance showed increased empowerment after genetic counseling. However, we did observe a trend towards more anxiety in participants with a variant of unknown significance or disease-associated variant.34 Further studies are warranted to confirm whether there is a difference in anxiety between participants with different genetic testing results, since we had relatively small subgroups and saw a large variation in anxiety scores within these subgroups.

Two previous studies have reported the outcome of genetic services in parents of children with developmental problems of whom only a minority had epilepsy. One study reported a higher quality of life in mothers of children with a diagnostic result from microarray versus those with inconclusive array results.36 Another study identiﬁed that the experiences of parents of children with epilepsy with genetic testing vary and are associated with the genetic testing results and the presence of parental depression and anxiety after receiving these results.37

Limitations We have to address two important limitations of this study. First, this study aimed at reporting the outcome of genetic counseling for epilepsy in a single cohort of participants who all underwent the same genetic counseling trajectory. We had no control group of participants who had genetic testing without counseling. Therefore, we cannot exclude the effect of other individual factors (such as life events) apart from the genetic counseling itself on the empowerment of participants during the genetic counseling trajectory. Still, empowerment was measured with the GCOS-18, which has shown to be very stable over time if no counseling occurs with an excellent test- retest reliability.20 The changes in empowerment over time therefore likely reflect the effect of counseling (possibly together with other factors) and not of time itself. Further randomized controlled trials with different forms of counseling in different groups may help to identify which parts of counseling are most effective in terms of gaining empowerment.

Second, although we had an average responder rate of 58% (n=70/120) in the participants of one hospital (UMCG) and an unknown responder rate in the other (UMCU),38 a significant proportion of the responders did not complete the follow-up questionnaires and were not included for the study. This drop out seemed partially explained by education level, since participants with a higher education were more likely to complete the questionnaires, but not by any other demographic or genetic testing variable or by baseline empowerment and anxiety scores. Participants more often completed the last questionnaires if they had a disease-associated variant (87%) or normal test results (75%) compared to having a variant of unknown significance (56%). Although these differences were not statistically significant, the genetic test results may have influenced the willingness of participants to complete the last questionnaire. In addition, parents mentioned

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 62 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 63

that filling out the questionnaires was time-consuming, and this may also have contributed to a lower response rate. As a result of this drop out, the size of our remaining study cohort may have been too small to detect significant differences in the effect of genetic services on empowerment and anxiety between patient subgroups, based on demographic characteristics and genetic testing results.

CONCLUSION AND FUTURE RESEARCH DIRECTIONS 3 Our study provides insight into the outcomes of genetic counseling before and after genetic testing for epilepsy, which may help genetic counselors of patients with epilepsy. Patients with epilepsy or their parents show increased empowerment after genetic counseling both before and after genetic testing, especially in the domains knowledge about genetic services, uncertainty about genetic services, and uncertainty about heredity. This change is independent of the results of genetic testing. Anxiety remained stable during the genetic counseling trajectory. On an individual level, half of the participants showed a clinically relevant increased empowerment. Further research is warranted to identify the individual diﬀerences in the outcome of genetic services based on demographic characteristics and genetic testing results and to identify which aspects of counseling are most eﬀective in terms of increasing empowerment. Such studies may help to further improve the genetic counseling trajectory personalized to the participants’ needs.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 63 64 CHAPTER 3

REFERENCES

1. Scheffer IE. Epilepsy genetics revolutionizes clinical 18. Mcallister M, Wood A, Dunn G, Shiloh S, Todd C. The practice. Neuropediatrics 2014; 45: 70-74. Genetic Counseling Outcome Scale: A new patient- 2. Berkovic SF. Genetics of epilepsy in clinical practice. reported outcome measure for clinical genetics services. Epilepsy Curr 2015; 15: 192-196. Clin Genet 2011; 79:413-424. 3. Myers CT, Mefford HC. Advancing epilepsy genetics in the 19. McAllister M, Dearing A. Patient reported outcomes and genomic era. Genome Med. 2015; 7: 91. patient empowerment in clinical genetics services. Clin 4. Mei D, Parrini E, Marini C, Guerrini R. The Impact of Genet 2015; 88: 114-121. Next-Generation Sequencing on the Diagnosis and 20. Voorwinden JS, Plantinga M, Krijnen W, et al. A validated Treatment of Epilepsy in Paediatric Patients. Mol Diagn PROM in genetic counseling: the psychometric properties Ther 2017; 21: 357-373. of the Dutch version of the Genetic Counseling Outcome 5. Balestrini S, Sisodiya SM. Pharmacogenomics in epilepsy. Scale. Manuscript under review. Neurosci Lett 2018; 667: 27-39. 21. Marteau TM, Bekker H. The development of a six item 6. Miller DT, Adam MP, Aradhya S, et al. Consensus short form of the state scale of the Spielberger State Statement: Chromosomal Microarray Is a First- Trait Anxiety Inventory (STAI). Br J Clin Psychol 1992; 31: Tier Clinical Diagnostic Test for Individuals with 301-306. Developmental Disabilities or Congenital Anomalies. Am 22. Otten E, Birnie E, Ranchor A V, Tintelen JP Van, Langen J Hum Genet 2010; 86: 749-764. IM Van. A group approach to genetic counselling 7. Mefford HC. Clinical Genetic Testing in Epilepsy. Epilepsy of cardiomyopathy patients : satisfaction and Curr 2015; 15: 197-201. psychological outcomes suf fi cient for further implementation. Eur J Hum Genet 2015; 23: 1462-1467. 8. Haddow J, Palomaki G. ACCE: A model process for evaluating data on emerging genetic tests. In: Khoury 23. Cohen J. editor. Statistical Power Analysis for the M, Little J, Burke W, eds. Human Genome Epidemiology: Behavioral Sciences. 2nd ed. New York, USA: Academic A Scientific Foundation for Using Genetic Information Press,1988. to Improve Health and Prevent Disease. Oxford: Oxford 24. Revicki D, Hays RD, Cella D, Sloan J. Recommended University Press; 2004. methods for determining responsiveness and minimally 9. Shostak S, Ottman R. Ethical, legal, and social important differences for patient-reported outcomes. J dimensions of epilepsy genetics. Epilepsia 2006; 47:1595- Clin Epidemiol 2008; 61: 102-109. 1602. 25. Angst F, Aeschlimann A, Angst J. The minimal clinically 10. Ottman R, Hirose S, Jain S, et al. Genetic testing in the important difference raised the significance of outcome epilepsies—Report of the ILAE Genetics Commission. effects above the statistical level, with methodological Epilepsia 2010; 51: 655-670. implications for future studies. J Clin Epidemiol 2017; 82: 128-136. 11. Caminiti CB, Hesdorffer DC, Shostak S, et al. Parents’ interest in genetic testing of their offspring in multiplex 26. Voorwinden JS, Plantinga M, Ausems M, et al. a large epilepsy families. Epilepsia 2016; 57: 279-287. outcome study on genetic counseling in the Netherlands: empowerment and emotional functioning. Manuscript 12. Okeke JO, Tangel VE, Sorge ST, et al. Genetic testing in Preparation. preferences in families containing multiple individuals with epilepsy. Epilepsia 2014; 55: 1705-1713. 27. Costal Tirado A, McDermott AM, Thomas C, et al. Using Patient-Reported Outcome Measures for Quality 13. Shostak S, Zarhin D, Ottman R. What’s at stake? Genetic Improvement in Clinical Genetics: an Exploratory Study. information from the perspective of people with J Genet Couns 2017; 26: 1017-1028. epilepsy and their family members. Soc Sci Med 2011; 73: 645-654. 28. Inglis A, Koehn D, Mcgillivray B, Stewart SE, Austin J. Evaluating a unique, specialist psychiatric genetic 14. Vears DF, Dunn KL, Wake SA, Scheffer IE. “It’s good to counseling clinic: uptake and impact. Clin Genet. 2016; know”: Experiences of gene identification and result 87: 218-224. disclosure in familial epilepsies. Epilepsy Res 2015; 112: 64-71. 29. Davison N, Payne K, Eden M, et al. Exploring the feasibility of delivering standardized genomic care using 15. Resta R, Biesecker BB, Bennett RL, et al. A new definition ophthalmology as an example. Genet Med 2017; 19: 1032- of genetic counseling: National Society of Genetic 1039. Counselors’ Task Force report. J Genet Couns 2006; 15: 77-83. 30. Spielberger CD, Gorsuch RL, Lushene R, Vagg PR, Jacobs GA, editors. Manual for the State-Trait Anxiety Inventory. 16. Madlensky L, Trepanier AM, Cragun D, Lerner B, Shannon Palo Alto, CA: Consulting Psychologists Press, 1983. KM, Zierhut H. A Rapid Systematic Review of Outcomes Studies in Genetic Counseling. J Genet Counsel 2017: 26; 31. Forsberg C, Bjiirvell H. Swedish population norms for the 361-378. GRHI, HI and STAI-state. Qaulity life Res 1993; 2: 349-356. 17. Mcallister M, Payne K, Macleod R, Nicholls S, Dian 32. Knight RG, Waal-Manning HJ, Spears GF. Some norms Donnai, Davies L. Patient empowerment in clinical and reliability data for the State-Trait Anxiety Inventory genetics services. J Health Psychol 2008; 13: 895-905. and th eZung Self-Rating Depression scale. Br J Clin Psychol 1983; 22: 245-249.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 64 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 65

33. Julian LJ. Measures of Anxiety. Arthritis Care 2011; 63: 1-11. 37. Wynn J, Ottman R, Duong J, et al. Diagnostic 34. Bekker HL, Legare F, Stacey D, O’Connor A, Lemyre L. Is exome sequencing in children: A survey of parental anxiety a suitable measure of decision aid eﬀectiveness: understanding, experience and psychological impact. A systematic review? Patient Educ Couns 2003; 50: 255- Clin Genet 2017; 93: 1039-1048. 262. 38. Baruch Y, Holtom BC. Survey response rate levels and 35. Dinc L, Terzioglu F. The psychological impact of genetic trends in organizational research. Hum Relations 2008; testing on parents. J Clin Nurs 2006; 15: 45-51. 61: 1139-1160. 36. Lingen M, Albers L, Borchers M, et al. Obtaining a genetic diagnosis in a child with disability: Impact on parental quality of life. Clin Genet 2016; 89: 258-266. 3

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 65 66 CHAPTER 3

SUPPLEMENTAL DATA

Supplemental Table 1: The English version of the Genetic Counseling Outcome Scale

The Genetic Counseling Outcome Scale (GCOS-24) Using the scale below, circle a number next to each statement to indicate how much you agree with the statement. Please answer all the questions. For questions that are not applicable to you, please choose option 4 (neither agree or disagree).

1 = Strongly disagree 2 = Disagree 3 = Slightly disagree 4 = Neither disagree nor agree

5 = Slightly agree Agree Disagree 6 = Agree agree nor Slightly agree Slightly Strongly agree Strongly Neither disagree 7 = Strongly agree disagree Slightly Strongly disagree Strongly

1 I am clear in my own mind why I am attending the clinical genetics service. 1 2 3 4 5 6 7 I can explain what the condition means to people in my family who may 2 1 2 3 4 5 6 7 need to know. I understand the impact of the condition on my child(ren) / any child I may 3 1 2 3 4 5 6 7 have. 4 When I think about the condition in my family, I get upset. 1 2 3 4 5 6 7 5 I don’t know where to go to get the medical help I / my family need(s). 1 2 3 4 5 6 7 I can see that good things have come from having this condition in my 6 1 2 3 4 5 6 7 family. 7 I can control how this condition affects my family. 1 2 3 4 5 6 7 8 I feel positive about the future. 1 2 3 4 5 6 7 9 I am able to cope with having this condition in my family. 1 2 3 4 5 6 7 I don’t know what could be gained from each of the options available to 10 1 2 3 4 5 6 7 me. 11 Having this condition in my family makes me feel anxious. 1 2 3 4 5 6 7 I don’t know if this condition could affect my other relatives (brothers, 12 1 2 3 4 5 6 7 sisters, aunts, uncles, cousins). In relation to the condition in my family, nothing I decide will change the 13 1 2 3 4 5 6 7 future for my children / any children I might have. I understand the reason why my doctor referred me to the clinical genetics 14 1 2 3 4 5 6 7 service. I know how to get the non-medical help I / my family needs (e.g. 15 1 2 3 4 5 6 7 educational, financial, social support). I can explain what the condition means to people outside my family who 16 1 2 3 4 5 6 7 may need to know (e.g. teachers, social workers). I don’t know what I can do to change how this condition affects me / my 17 1 2 3 4 5 6 7 family. 18 I don’t know who else in my family might be at risk for this condition. 1 2 3 4 5 6 7 19 I am hopeful that my children can look forward to a rewarding family life. 1 2 3 4 5 6 7 20 I am able to make plans for the future. 1 2 3 4 5 6 7 21 I feel guilty because I (might have) passed this condition on to my children. 1 2 3 4 5 6 7 22 I am powerless to do anything about this condition in my family. 1 2 3 4 5 6 7 23 I understand what concerns brought me to the clinical genetics service. 1 2 3 4 5 6 7 I can make decisions about the condition that may change my child(ren)’s 24 1 2 3 4 5 6 7 future / the future of any child(ren) I may have.

(For the GCOS-18 item 6,7,13,15,22,24 need to be removed; all the items of the negative emotions and the two uncertainty scales need to be reversed to calculate a total score for empowerment)

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 66 EMPOWERMENT AND ANXIETY DURING GENETIC COUNSELING FOR EPILEPSY 67

Supplemental Table 2: The subscales of the Dutch version of the GCOS

Subscale Question Hope and coping (HC) 8. I feel positive about the future. 20. I am able to make plans for the future. 9. I am able to cope with having this condition in my family. 19. I am hopeful that my children can look forward to a rewarding family life. Knowledge about the 2. I can explain what the condition means to people in my family who may need to condition (KC) know. 3. I understand the impact of the condition on my child(ren) / any child I may have. 16. I can explain what the condition means to people in my family who may need to know. Knowledge about 1. I am clear in my own mind why I am attending the clinical genetics service. 3 genetic services (KG) 23. I understand what concerns brought me to the clinical genetics service. 14. I understand the reasons why my doctor referred me to the clinical genetics service. Uncertainty about 17. I don’t know what I can do to change how this condition aﬀects me/my children. genetic services (UG) 5. I don’t know where to go to get the medical help I/my family need(s). 10. I don’t know what could be gained from each of the options available for me. Negative emotions 4. When I think about the condition in my family, I get upset. (NE) 11. Having this condition in my family makes me feel anxious. 21. I feel guilty because I (might have) passed this condition on to my children. Uncertainty about 12. I don’t know if this condition could aﬀect my other relatives (brothers, sisters, heredity (UH) aunts, uncles, cousins). 18. I don’t know who else in my family might be at risk for this condition.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 67 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 68 PART II

Genotype-phenotype studies

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 69 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 70 71

Chapter 4

SYNGAP1 encephalopathy: a distinctive generalized developmental and epileptic encephalopathy

Accepted for publication in Neurology

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 71 72 CHAPTER 4

Danique RM Vlaskamp 1,2, Benjamin J Shaw1, Rosemary Burgess1, Davide Mei3, Martino Montomoli3, Han Xie4, Candace T Myers5, Mark F Bennett1,6, Wenshu Xiang Wei4, Danielle Williams7, Saskia M Maas8, Alice S Brooks9, Grazia MS Mancini9, Ingrid MBH van de Laar9, Johanna M van Hagen10, Tyson L Ware11, Richard I Webster12, Stephen Malone13, Samuel F Berkovic1, Renate M Kalnins14, Federico Sicca15, G Christoph Korenke16, Conny MA van Ravenswaaij-arts2, Michael S Hildebrand1, Heather C Meﬀord5, Yuwu Jiang4,17, Renzo Guerrini3, Ingrid E Scheﬀer1,18

1 University of Melbourne, Melbourne, Australia. 2 University of Groningen, Groningen, the Netherlands. 3 University of Florence, Florence, Italy. 4 Peking University First Hospital, Beijing, China. 5 University of Washington, Seattle, USA. 6 The Walter and Eliza Hall Institute of Medical Research, Melbourne, Victoria. 7 Caulﬁeld, Melbourne, Australia. 8 Academic Medical Center, Amsterdam, the Netherlands. 9 Erasmus University Medical Center, Rotterdam, the Netherlands. 10 VU University Medical Center, Amsterdam, the Netherlands. 11 Launceston General Hospital, Australia. 12 The Children’s Hospital at Westmead, Sydney, Australia. 13 Lady Cilento Children’s Hospital, Brisbane, Australia. 14 Austin Hospital, Melbourne, Australia. 15 IRCCS Stella Maris Foundation, Pisa, Italy. 16 Klinikum Oldenburg, Oldenburg, Germany. 17 Beijing Institute for Brain Disorders, Beijing, China. 18 The Florey Institute of Neurosciences and Mental Health, Melbourne, Australia.

Acknowledgements. We thank all patients and their families for their participation. We are grateful for the support of the SYNGAP1 parent support group. We thank the Victorian Institute of Forensic Medicine, Dr. Angelika Riess, Dr. Marjolaine Willems, Dr. Einar Bryne and Dr. Ferdinant Kok for providing clinical information and Dr. Zhixian Yang and Dr. Xiaoyan Liu for EEG review.

Study sponsorship or funding. Danique R.M. Vlaskamp has received an unrestricted travel grant from the Dutch Society of Child Neurology, Jo Kolk Foundation (VVAO), the Research School of Behavioral and Cognitive Neurosciences (BCN). Han Xie is supported by National Natural Science Foundation of China (Youth Foundation) (Grant No. 81501123), Mark F. Bennett by the Victorian State Government Operational Infrastructure Support and Australian Government NHMRC IRIISS and Yuwu Jiang by Chinese National Key Research Project (2016YFC1306200, 2016– 2020). This work was supported in part by a grant from the EU Seventh Framework Programme FP7 under the project DESIRE (grant agreement n° 602531) to Renzo Guerrini.Ingrid E. Scheﬀer is supported by NHMRC Program Grant (1091593, 2016–2020) and Senior Practitioner Fellowship (1104831, 2016– 2020).

Disclosures. D.R.M. Vlaskamp, B.J. Shaw, R. Burgess and M. Montomoli report no disclosures relevant to the manuscript. D. Mei has received scientiﬁc advisory board honoraria from BioMarin Pharmaceutical. Xie, W. XiangWei, C.T. Myers, M.F. Bennett, D. Williams, S.M. Maas, A.S. Brooks, G.M.S. Mancini, I.M.B.H. van de Laar, J.M. van Hagen, T.L. Ware and R.I. Webster report no disclosures relevant to the manuscript. S. Malone has received scientiﬁc advisory board honoraria from UCB and Biomarin. S.F. Berkovic, R.M. Kalnins, F.Sicca, G.C. Korenke, C.M.A. van Ravenswaaij-

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 72 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 73

Arts and M.S. Hildebrand report no disclosures relevant to the manuscript. H. Mefford is funded by NIH grant NINDS 1R01NS069605. Y. Jiang reports no disclosures relevant to the manuscript. R. Guerrini serves on the editorial boards of Neurology® and Seizure; has received speaker and consultancy fees from Eisai Inc., Novartis, Zogenix, BioMarin, UCB Pharma, and GW Pharma; receives/has received research support from the European Community, the Italian Ministry of Health, The Mariani Foundation, The Italian Epilepsy Federation, Telethon, The Tuscany Region, The Pisa Foundation, The Italian Ministry of University and Research. I. Scheffer serves on the editorial boards of Neurology® and Epileptic Disorders; may accrue future revenue on a pending patent re: Therapeutic compound; has received speaker honoraria from Athena Diagnostics, UCB, GSK, Eisai, and Transgenomics; has received scientific advisory board honoraria from Nutricia, UCB and GSK, has received funding for travel from Athena Diagnostics, UCB, and GSK; and receives/has received research support from the NHMRC, ARC, NIH, Health Research Council of New Zealand, 4 March of Dimes, the Weizmann Institute, CURE, US Department of Defense, and the Perpetual Charitable Trustees.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 73 74 CHAPTER 4

ABSTRACT

Objective. To delineate the epileptology, a key part of the SYNGAP1 phenotypic spectrum, in a large patient cohort.

Methods. Patients were recruited via investigators’ practices or social media. We included patients with (likely) pathogenic SYNGAP1 variants or chromosome 6p21.32 microdeletions incorporating SYNGAP1. We analysed patients’ phenotypes using a standardized epilepsy questionnaire, medical records, EEG, MRI and seizure videos.

Results. We included 57 patients (53% male, median age 8 years) with SYNGAP1 mutations (n=53) or microdeletions (n=4). 56/57 patients had epilepsy: generalized in 55, with focal seizures in seven and infantile spasms in one. Median seizure onset age was 2 years. A novel type of drop attack was identiﬁed comprising eyelid myoclonia evolving to a myoclonic-atonic (n=5) or atonic (n=8) seizure. Seizure types included eyelid myoclonia with absences (65%), myoclonic seizures (34%), atypical (20%) and typical (18%) absences and atonic seizures (14%), triggered by eating in 25%. Developmental delay preceded seizure onset in 54/56 (96%) patients of whom early developmental history was available. Developmental plateauing or regression occurred with seizures in 56 in the context of developmental and epileptic encephalopathies (DEEs). 55/57 patients had intellectual disability, which was moderate to severe in 50. Other common features included behavioral problems (73%), high pain threshold (72%), eating problems including oral aversion (68%), hypotonia (67%), sleeping problems (62%), autism spectrum disorder (54%) and ataxia or gait abnormalities (51%).

Conclusions. SYNGAP1 mutations cause a generalized DEE with a distinctive syndrome combining epilepsy with eyelid myoclonia and myoclonic-atonic seizures, and predilection to seizures triggered by eating.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 74 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 75

INTRODUCTION

Mutations of the SYNGAP1 gene were first identified in 2009 in patients with non-syndromic intellectual disability (ID) and autism spectrum disorder (ASD) followed, in 2013, by recognition of their important role in the developmental and epileptic encephalopathies (DEEs).1–3 Most affected individuals have de novo mutations, with truncating mutations predominating, although missense mutations, chromosomal translocations or microdeletions disrupting SYNGAP1 are also described.4–8

SYNGAP1 (MIM *603384) on chromosome 6p21.32 encodes a SYNaptic Ras-GTPase-Activating Protein, mainly expressed in the synapses of excitatory neurons.9,10 SYNGAP1 is a key mediator in the N-Methyl-D-Aspartate receptor activated RAS-signaling cascade regulating the post-synaptic density and the formation, development and maturation of dendritic spines.11,12 Loss-of-function of 4 SYNGAP1 has major consequences for neuronal homeostasis and development, which are crucial for learning and memory.11 Syngap1 null mice die within a week and Syngap1 heterozygous mutant mice have a lower seizure threshold, learning and memory deﬁcits and behavioral problems.13–16

Since our original description of five patients with a SYNGAP1-DEE,3 there have been two additional studies describing the epilepsy in 24 of 27 patients with SYNGAP1 encephalopathy; the specific epilepsy syndrome was described in only four of these cases.8,17 SYNGAP1 was originally identified in 38 patients with ID or ASD; of whom 15 had seizures and only one had an epilepsy syndrome diagnosis made.1,2,4–7,18–28 We aimed to delineate the epilepsy syndromes within the SYNGAP1 phenotypic spectrum in a large international cohort of patients with SYNGAP1 mutations and microdeletions.

METHODS

Study cohort We recruited 66 patients with SYNGAP1 variants via investigators’ practices in Australia, Italy, the Netherlands and China (n=39) and via the SYNGAP1 Facebook group on which parents posted our invitation to participate (n=27). The pathogenicity of all SYNGAP1 variants was evaluated using standard American College of Medical Genetics and Genomics guidelines (Supplemental Table 1). 29 We included 57 (86%) patients with (likely) pathogenic SYNGAP1 variants (n=53) or chromosome 6p21.32 microdeletions including SYNGAP1 and other genes (n=4). Five (8%) patients with a SYNGAP1 variant of unknown signiﬁcance were studied separately. Four (6%) patients with likely benign SYNGAP1 variants were excluded.

Phenotyping Parents or caregivers of all patients were interviewed using a standardized epilepsy questionnaire.30 We analysed medical records, electroencephalogram (EEG), neuroimaging including magnetic resonance imaging (MRI) results and, where available, seizure videos and video-EEG data. Seizure

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 75 76 CHAPTER 4

types and syndromes were classiﬁed using the 2017 International League Against Epilepsy (ILAE) classiﬁcation.31,32 The severity of ID was established using IQ scores (where available) or information regarding the level of functioning in accordance with the Diagnostic and Statistical Manual of Mental Disorders, edition 5 (DSM-5).33

Genotyping SYNGAP1 mutations were described based on the longest isoform 1 of SYNGAP1 (NM_0067772.2)34 and chromosomal microdeletions based on the Genome Reference Consortium Human Reference sequence version 37 (GRCh37/hg19).

Genotype-phenotype correlation We examined phenotype-genotype correlation in four patient groups: 1) truncating mutations which included nonsense and frameshift mutations, 2) splice-site mutations, 3) missense and inframe insertion/deletion mutations, and 4) chromosome 6p21.32 microdeletions including SYNGAP1 and other genes. Splice-site mutations were considered separately since their eﬀects on the protein are variable.35

Ethical statement All parents or legal representatives of the patients gave written informed consent for inclusion and use of photos and videos. This study was approved by the local institutional Ethics Committee (Austin Health reference number H2007/02961).

Data availability Anonymized data will be shared by request from any qualiﬁed investigator.

RESULTS

Cohort Fifty-seven patients (53% male, median age at study 8 years) with (likely) pathogenic SYNGAP1 variants were included: 34 truncating, 8 splice-site and 11 missense/inframe mutations and 4 microdeletions. Forty-six (81%) patients have not been previously reported. Thirty-nine of these patients had novel mutations, that included a total of 35 unique mutations as four were recurrent. The remaining seven individuals had previously reported mutations. Figure 1A depicts the 57 SYNGAP1 mutations and microdeletions found in our patient cohort and Figure 1B the 62 SYNGAP1 variants of all previously reported patients who were not included in our cohort.1,2,4–8,17–28 Inheritance was tested in 53/57 patients and the mutation had arisen de novo in all. Table 1 describes the clinical features for our cohort according to the four genotypic groups. More detailed individual genetic and phenotypic data are summarized in Supplementary Table 1 and 2.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 76 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 77 ) and ) and

Ɖ͘WƌŽϭϯϮϭWƌŽ хх хх хх хх хх ϯϯ͕ϰϮϭ͕ϰϲϲ ϯϯ͕ϰϮϭ͕ϰϲϲ

ϭϴ ϭϵ ϭϴ ϭϵ

Ɖ͘ƌŐϭϮϰϬΎ

Ɖ͘dǇƌϭϮϭϵΎ

Đ͘ϯϱϴϯͲϲ'х Ɖ͘'ůƵϭϭϲϵΎ

ϭϲ ϭϳ ϭϲ ϭϳ

Đ͘ϯϰϬϴнϭ'х͖Ɖ͍͘

ϳ ϭ Ύ Ɛ Ĩ Ž ƌ W ϲ ϯ ϭ ϭ Ŷ ů ' ͘ Ɖ

ϳ ϭ Ύ Ɛ Ĩ Ž ƌ W ϲ ϯ ϭ ϭ Ŷ ů ' ͘ Ɖ Ɖ͘'ůŶϭϬϵϯΎ

lue), mild developmentallue), and epileptic encephalopathy

Ɖ͘ƌŐϭϬϴϱ'ůǇĨƐΎϰϱ

Ɖ͘sĂůϭϬϳϴůĂĨƐΎϱϭ;ϮͿ and omal coordinates were based [hg19] on NCBI Build 37

Ɖ͘WŚĞϵϳϵ^ĞƌĨƐΎϵϴ 4

were outside the region presented ( here. Truncating Ɖ͘WƌŽϵϳϴ,ŝƐĨƐΎϵϵ Ɖ͘ƌŐϵϲϳΎ;ϮͿ

ϭϱ ϭϱ

Ɖ͘'ůŶϵϮϴΎ evelopmental and epileptic moderate encephalopathy to (red),

Ɖ͘>ĞƵϵϮϱWƌŽĨƐΎϭϱϮ

Ɖ͘ƌŐϵϮϮΎ

ϭϭϵϱϭϮϲϱϭϮϵϲ Ɖ͘'ůŶϴϵϯƌŐĨƐΎϭϴϰ ϭϭϵϱϭϮϲϱϭϮϵϲ ^,ϯ

^,ϯ

Ɖ͘dŚƌϴϳϴƐƉĨƐΎϲϬ

Ɖ͘>ĞƵϴϭϯƌŐĨƐΎϮϯϰ

Đ͘ϮϮϵϰнϭ'х͖Ɖ͍͘

Ɖ͘'ůƵϳϯϵĨƐΎϮϬ

Ɖ͘^ĞƌϳϯϴΎ

Ɖ͘ƐŶϳϮϵdŚƌĨƐΎϯϭ;ϮͿ Ɖ͘ƌŐϳϮϲdŚƌĨƐΎϯϯ

Ɖ Ύ Ϯ Ϭ ϳ Ŷ ů ' ͘ Ύ Ϯ Ϭ ϳ Ŷ ů ' ͘ Ɖ

Ɖ͘ƌŐϲϴϳΎ;ϮͿ

Ɖ͘dǇƌϲϲϱΎ

Ɖ͘dƌƉϲϱϳΎ

Ɖ͘/ůĞϲϯϬƐŶ Ɖ͘WŚĞϲϬϴdƌƉĨƐΎϵ

ϳϲϱϳϳϵϭϭϯϳ Ɖ͘>ĞƵϱϵϱǇƐĨƐΎϱϱ;ϮͿ ϳϲϱϳϳϵϭϭϯϳ Ɖ Ύ ϵ ϳ ϱ Ő ƌ ͘

Ύ ϵ ϳ ϱ Ő ƌ ͘ Ɖ Ɖ͘'ůƵϱϳϴůĂĨƐΎϳϰ Ɖ͘ǇƐϱϵϵdƌƉ

Đ͘ϭϲϳϲнϱ'х͖Ɖ͍͘Ϭ

Đ͘ϭϲϳϳͲϮͺϭϲϴϱĚĞů͖Ɖ͍͘ Ɖ͘WƌŽϱϲϮ>ĞƵ Ɖ͘WƌŽϱϲϮ>ĞƵ Ɖ͘WƌŽϱϰϰΎ

ϭϬ ϭϭ ϭϮ ϭϯ ϭϰ ϭϬ ϭϭ ϭϮ ϭϯ ϭϰ Ɖ͘dǇƌϱϭϴƐŶĨƐΎϴ Ɖ͘'ůǇϱϭϭƌŐ

ZĂƐͲ'W ZĂƐͲ'W Ɖ͘dǇƌϱϬϱΎ Ɖ͘dŚƌϰϴϴ^ĞƌĨƐΎϳ ϵ ϵ

Ɖ͘>ĞƵϰϲϱĚƵƉ ƚ;ϲ͖ϮϮͿ;ƉϮϭ͘ϯϮ͖Ƌϭϭ͘ϮϭͿ

Ɖ͘>ǇƐϰϲϱWŚĞĨƐΎϵ Ɖ͘ƐƉϰϲϯͺ>ĞƵϰϲϱĚĞůŝŶƐsĂů

Ɖ͘'ůŶϰϱϲΎ Ɖ͘ůĂϰϱϬƐƉĨƐΎϰ

Ɖ͘dǇƌϰϭϳǇƐ Ɖ͘>ǇƐϰϭϴƌŐĨƐΎϱϰ;ϮͿ

Ɖ͘'ůǇϯϵϭ'ůŶĨƐΎϮϳ Ɖ͘ůĂϰϬϰWƌŽ

Ɖ͘>ĞƵϯϱϯdƌƉĨƐΎϭϯ Ɖ͘dƌƉϯϲϮƌŐ ϴ Ɖ͘sĂůϯϰϴůĂĨƐΎϳϬ ϴ ϲϯϱϭϭϱϱϵϲϯϴϳϬϲ Ɖ͘'ůǇϯϰϰ^Ğƌ ϲϯϱϭϭϱϱϵϲϯϴϳϬϲ Ɖ͘>ĞƵϯϮϯWƌŽ Ɖ͘>ĞƵϯϮϳWƌŽ;ϮͿ

te to severe developmental encephalopathy with no epilepsy developmental severe encephalopathy (b to te

ϳ Ύ Ɛ Ĩ Ŷ ů ' ϳ ϳ Ϯ Ɛ Ǉ > ͘ Ɖ ϳ Ύ Ɛ Ĩ Ŷ ů ' ϳ ϳ Ϯ Ɛ Ǉ > ͘ Ϯ Ɖ

Ɖ͘ǇƐϮϴϮƌŐ Ϯ

Ɖ͘dƌƉϮϲϳΎ

Ɖ͘'ůŶϮϰϳΎ

Ɖ͘WŚĞϮϯϭ>ĞƵĨƐΎϭϰ Ɖ͘ǇƐϮϯϯdǇƌ

Đ͘ϱϭϬͲϭ'х͖Ɖ͍͘ Ɖ͘/ůĞϮϭϰdƌƉĨƐΎϵ Đ͘ϱϬϵнϭ'хd͖Ɖ͍͘

W, W,

Ŷ ů ' Ϭ ϳ ϭ Ő ƌ ͘

Ɖ͘ůĂϭϵϱWƌŽ Ɖ

Ŷ ů ' Ϭ ϳ ϭ Ő ƌ ͘ Ɖ

Ύ ϰ ϲ ϭ Ő ƌ ͘ Ɖ

Ɖ ϰ ϲ ϭ Ő ƌ ͘ Ύ

Ɖ͘ƌŐϭϱϮ'ůŶĨƐΎϭϰ

Ɖ͘>ĞƵϭϱϬsĂůĨƐΎϲ;ϮͿ

Ɖ͘dŚƌϭϰϰ^ĞƌĨƐΎϮϵ

Ϳ Ϯ ; Ύ ϯ ϰ ϭ Ő ƌ ͘ Ɖ

Ύ ϯ ϰ ϭ Ő ƌ ͘ Ɖ

Ɖ͘>ǇƐϭϰϮ'ůƵĨƐΎϯϭ

Ɖ͘>ǇƐϭϯϴΎ;ϮͿ

Đ͘ϯϴϴͲϮхd͖Ɖ͍͘;ϮͿ

ϭϳϬϮϮϮϮϱϱ ϰ Ɖ͘ƌŐϭϯϱΎ ϭϳϬϮϮϮϮϱϱ ϰ

Ɖ͘^ĞƌϭϮϵ^Ğƌ Ɖ͘dǇƌϭϭϲΎ;ϮͿ

Ϳ Ϯ ; Ϭ Ϯ Ύ Ɛ Ĩ ƌ Ğ ^ ϰ ϭ ϭ Ɛ Ǉ > ͘ Ϭ Ϯ Ύ Ɛ Ĩ ƌ Ğ ^ ϰ ϭ ϭ Ɛ Ǉ > ͘ Ɖ mutations and microdeletions in patients of our cohort (A) or previously published in the literature (B). (B). in patients of our cohortSYNGAP1 published in the literature mutations and microdeletions or previously (A) Ɖ

Ɖ͘>ǇƐϭϬϴsĂůĨƐΎϮϱ

Ɖ͘,ŝƐϵϱWƌŽĨƐΎϱ

Ɖ͘dŚƌϴϱ^ĞƌĨƐΎϭϰ

Đ͘ϭϵϬͲϮх'͖Ɖ͍͘ Ɖ͘ƌŐϯϭΎ ;ϮͿ фф фф фф фф фф фф фф фф ϭ Ϯ ϯ ϰ ϱ ϲ ϳ ϭ Ϯ ϯ ϰ ϱ ϲ ϳ ϯϯ͕ϯϴϳ͕ϴϰϳ ϯϯ͕ϯϴϳ͕ϴϰϳ ϭϮϯϲϰϵϵϭϯϬ ϭϮϯϲϰϵϵϭϯϬ Ϯ Ϯ ϭ ϭ ϯ ϯ ͘ ͘ W W ϭ ϭ Ϯ Ϯ Ɖ Ɖ ' ' ϲ ϲ Ğ Ğ E E ŵŝŶŽĂĐŝĚ ŵŝŶŽĂĐŝĚ ŵ ŵ z z Ž Ž Ɛ Ɛ ^ ^ ǀĂƌŝĂŶƚƐ;ŶсϲͿ Ž Ž ǀĂƌŝĂŶƚƐ;ŶсϭϭͿ ŵ ŵ Ž Ž ƌ ƌ dƌĂŶƐůŽĐĂƚŝŽŶ;ŶсϭͿ DŝĐƌŽĚĞůĞƚŝŽŶƐ ;ŶсϰͿ Ś Ś DŝĐƌŽĚĞůĞƚŝŽŶƐ ;ŶсϲͿ DŝƐƐĞŶƐĞĂŶĚŝŶĨƌĂŵĞ DŝƐƐĞŶƐĞĂŶĚŝŶĨƌĂŵĞ ƐƉůŝĐĞͲƐŝƚĞǀĂƌŝĂŶƚƐ;ŶсϴͿ ^ƉůŝĐĞͲƐŝƚĞǀĂƌŝĂŶƚƐ;ŶсϲͿ dƌƵŶĐĂƚŝŶŐǀĂƌŝĂŶƚƐ;ŶсϰϯͿ dƌƵŶĐĂƚŝŶŐǀĂƌŝĂŶƚƐ;ŶсϯϰͿ (light green), mild developmental(light green), encephalopathy with no epilepsy and unknown/unclassiﬁed green) (dark Chromos epilepsy (grey). mutations,SYNGAP1 domains protein and exons on its longest isoform 1 (NM_006772.2). splice-site mutations (>) are presented abovesplice-site the gene mutations and missense (>) and mutations inframe underneath (•) the gene. Bold variants concern variants that were identiﬁed in our cohort but also previously published in the literature The colours of the lines represent the epilepsy syndrome phenotype: moderate severe to d modera developmental with epilepsysevere encephalopathy (orange), Schematic presentation of 1: Schematic presentation Figure microdeletions including are presented SYNGAP1 Chromosome as grey 6p21.32 bars with and indicating << >> that their breakpoints

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 77 78 CHAPTER 4 3 (5.3) 2 (3.5) 2 (3.5) 5 (8.8) 7 (12.3) 8 (14.0) 8 (14.0) 10 (17.5) 10 11 (19.3) 11 35 (61.4) 35 19 (33.3) 19 19 (33.3) 19 14 (24.6) 14 53 (93.0) 53 13 (22.8) 13 37 (64.9) 37 (n = 57) = 57) (n 55 (96.5) 55 30 (52.6) 56 (98.2) 56 17/31 (54.8) 17/31 39/52 (75.0) 39/52 28/52 (53.8) 28/52 26/52 (50.0) 26/52 1 (1.8), 1 (1.8) 1 (1.8), Total cohortTotal

SYNGAP1 ------, - -, 4 (100) 4 (100) 4 (100) (n = 4) = 4) (n 3 (75.0) 3 (75.0) 2 (50.0) 2 1 (25.0) 2 (50.0) 2 (50.0) 1 (25.0) 1 (25.0) 1 (25.0) 3 (75.0) 1/2 (50.0) 1/2 Microdeletions Microdeletions including including 3 y 8 m (2 y 4 m – 7 y 11 m) y 4 m – 7 11 3 y 8 m (2 y) m – 33 8 y 3 m (18 - - - - 1 (9.1) 1 1 (9.1) 10 m) 3 (27.3) 3 (27.3) 2 (18.2) 2 (18.2) 2 (18.2) 5 (45.5) 5 (45.5) 5 (45.5) 6 (54.5) 8 (72.7) 8 1 (9.1), - 1 (9.1), 6 (54.5) 11 (100) 11 (n = 11) (n 10 (90.9) 10 5/9 (55.6) 5/9 3/6 (50.0) 3/6 4/9 (44.4) 4/9 6/9 (66.7) 6/9 mutations mutations Missense/inframe Missense/inframe 8 y 3 m (3 y 2 m – 18 y y 2 m – 18 8 y 3 m (3 ------, 8 (100) 8 (100) 8 (100) 3 (37.5) 3 (37.5) 3 (37.5) (n = 8) = 8) (n 1 (12.5) 1 (12.5) 1 (12.5) 1 (12.5) 6 (75.0) 4 (50.0) 4 (50.0) 4 (50.0) 4 (50.0) 4 (50.0) 2 (25.0) 2 Splice-site mutations mutations and microdeletions microdeletions and mutations SYNGAP1 SYNGAP1 2 (5.9) 2 2 (5.9) 2 (5.9) 3 (8.8) 3 (8.8) 3 6 (17.6) 6 5 (14.7) 5 5 (14.7) 7 (20.5) -, 1 (2.9) -, 33 (97.1) 33 33 (97.1) 33 15 (44.1) 15 11 (32.4) 11 33 (97.1) 33 13 (38.2) 13 (n = 34) (n 26 (76.5) 26 25 (73.5) 18 (52.9) 18 17/31 (54.8) 17/31 15/31 (48.4) 15/31 25/31 (80.6) 25/31 2 y (6 m – 7 y 3 m, 32)2 y (6 m – 6 y 2 m, 8) 2 y (6 11) m – 4 y, 2.5 y (4 4) m – 2.5 y, m (12 20 m – 7 y 3 m, 55) 2 y (4 Truncating mutations mutations Truncating Photosensitivity (clinical, %) Photosensitivity (clinical, Other seizure types: spasms tonic(%), (%) Slow background (%) EEG: GeneralizedEEG: discharges (%) - Typical absences (%) absences - Typical (%) - Myoclonic absences - Myoclonic (%) Photosensitivity (electrical, %) (47.4) 9/19 Focal seizures (%) - Atypical (%) absences (%) - Atonic attacks- Unclassiﬁed drop (%) - EM-atonic (%) (%) - Myoclonic-atonic seizures (%) Tonic-clonic * Reﬂex seizures (%) EEG: (Multi)Focal discharges (%) discharges (Multi)Focal EEG: Seizures (%) - EM (%) (%) - EM-myoclonic-atonic 4 (11.8) Median study age at (range) y) m – 33 y (18 9.5 y 9 m) y 9 m – 16 y (4 9.5 Median age seizure onset at known in n) (range, Generalized seizures (%) Multiple seizure types (%) - Any absences (%) absences - Any Males (%) Table 1: Phenotypes 1: Table in patients with

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 78 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 79 5) 16 (28.1) 16 29 (50.9) 29 (n = 57) = 57) (n 39 (68.4) 39 30 (52.6) 30 38 (66.7) 38 34/55 (61.8) 34/55 41/56 (73.2) 41/56 39/54 (72.2) 39/54 54/56 (96.4) 54/56 Total cohortTotal 6 m – 18 y, 19) y, 6 m – 18 32/56 (57.1, 2 y, 2 y, (57.1, 32/56 5 (8.8), 1/54 (1.9) 1/54 5 (8.8), 10 (17.8, 7.5 y, 3 – 13 y, 7) y, 3 – 13 y, 7.5 (17.8, 10

SYNGAP1 - - - - 4 (100) 4 (100) 4 (100) 4 (100) (n = 4) = 4) (n 3 (75.0) 1 (25.0), - 1 (25.0), 2/3 (66.7) 4 y (-, 1, 1) 1, 4 y (-, 23) 23, y, m – 13 4 y (16 4 (100): 4 (100)4 (100): (87.7) 50 (96.5): 55 Microdeletions Microdeletions

including including 4 is variable. If no denominator is given, there was information on or without absences, EEG = electroencephalogram, m = months, n -, - -, 3 (27.3) 5 (45.5) 7 (63.6) 4 (36.4) 8 (72.7) 6 (54.5) 6 (54.5) 11 (100) 11 (n = 11) (n 4/10 (40.0) 4/10 mutations mutations 11 (100): 9 (81.8) (100): 11 Missense/inframe Missense/inframe 7 (63.6, 13 m, 9 m – 18 m, 9) m, 9 m – 18 13 7 (63.6, 4) m – 7 y, m, 12 18 4 (100, 8 (100) 3 (37.5) (n = 8) = 8) (n 5 (62.5) 6 (62.5) 5 (62.5) 5 (62.5) 4 (50.0) 5/7 (71.4) 5/7 5/7 (71.4) 5/7 8 (100): 7 (87.5) 7 (87.5) 8 (100): 6 (75.0, 5 y 4 m, 6 (75.0, 4 y 9 m – 6 y, 2) 4 y 9 m – 6 y, 1 (12.5), 1/7 (14.3) 1/7 1 (12.5), Splice-site mutations represents the number of patients with known information on th 23 (67.6) 10 (29.4) 10 (n = 34) (n 20 (58.8) 20 22 (64.7) 22 27/33 (81.8) 27/33 31/33 (93.9) 31/33 20/33 (60.6) 20/33 20/34 (58.8) 20/34 3 (8.8), 0/32 (-) 0/32 3 (8.8), 32 (94.1): 30 (88.2) 30 (94.1): 32 2 y (12 m – 6 y, 32, 33) 32, m – 6 y, 2 y (12 m – 4 y m, 8, 8) m (12 21 10) 10, m – 2.5 y, m (12 19 1) 1, 2 y 9 m (-, 52) 51, m – 6 y, 2 y (12 4 y (16 m – 13 y, 14, 14) 14, y, m – 13 4 y (16 4) 3, m – 4.5 y, 4 y (23 4) 5, y – 5 y, 4 y (2 9 m (4 m – 2.5 y, 32, 33) 32, m – 2.5 y, 9 m (4 m – 2 y 1 m, 8, 8) m (5 10 11) m, 11, m – 20 m (6 12 4) m – 2 y 1 m, 4, m (9 19 56) 55, m – 2.5 y, 9 m (4 Truncating mutations mutations Truncating 2 y 3 m (8 m – 11 y, 24, 27) 27) 24, y, m – 11 2 y 3 m (8 6, 7) m – 6 y, m (12 21 9) 9, y, – 11 2 y (18m 2) m – 3 y 1 m, 2, 2 y 5 m (20 4 41, m – 9 y, 2 y (8 4 (11.8, 8 y 4 m, 3 y – 13 y, 4) y, 8 y 4 m, 3 – 13 4 (11.8, -) 1 (12.5, 5 y – 7 9 m, 3) 7 y, 6 (54.5, 15/33 (45.5, 2 y, 6 m – 18 y, 11) y, 6 m – 18 2 y, (45.5, 15/33 Seizure free median (%, age, knownrange, in) If patients had missing information, a denominator is given that all patients. * Reﬂex seizures include seizures triggered by eating, sound or touch. Abbreviations: EM = eyelid myoclonia with MRI = magnetic= number, resonance imaging, y = years. Orthopedic (%) issues - First word (range, age known- First (range, in word n, possible in n) - Sitting (range, age- Sitting known (range, in n, possible in n) age known- Walking (range, in n, possible in n) Developmental (%) delay Other features: tremor (%), (%), tremor Other features: microcephaly (%) Ataxia and/or gait abnormalities Ataxia and/or (%) - Two words together (range, age together (range, words - Two known in n, possible in n) Intellectual %): disability (ID, moderate-severe ID (%) Hypotonia (%) Regression (%, medianRegression (%, age, knownrange, in n) Sleep (%) problems Autism spectrumAutism disorder (%) Eating problems (%) ** (%) problems Eating Severe behavioralSevere (%) problems (84.8) 28/33 High pain thresholds (%)

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 79 80 CHAPTER 4

Seizure types 56/57 patients had epilepsy with a median age at seizure onset of 2 years (range 4 months – 7 years, known in 55). At onset, 50 patients had generalized seizures, one had infantile spasms in the context of West syndrome, three had possible focal seizures and two had unknown seizure types.

Absence seizures occurred in 53/57 (93%) patients: eyelid myoclonia with absences (EM) in 37, atypical absences in 11, typical absences in 10, and myoclonic absences in two. Other seizure types included myoclonic (19/57, 33%), atonic (8/57, 12%), myoclonic-atonic (3/57, 5%) and unclassiﬁed drop attacks (2/57, 4%). Fourteen (25%) patients had tonic-clonic seizures (TCS), of whom two also had focal seizures. Five patients had focal seizures without TCS. 4/57 (7%) patients had febrile seizures prior to epilepsy onset.

Reﬂex seizures were seen, often occurring in clusters, and triggered by eating (25%), sound (7%) or touch (4%). Hunger was also associated with seizures in 7%. Clinical photosensitivity occurred in 13 (23%) patients and electroencephalographic photosensitivity in 17/31 (55%) tested. Three (5%) patients had self-induced seizures with hyperventilation.

Novel seizure type: eyelid myoclonia-myoclonic-atonic seizures Of the 36 patients with EM, we observed a novel seizure type in 13/36 (36%) in which the EM evolved to a drop attack. The nature of the drop attack was an EM-myoclonic-atonic seizure in ﬁve and an EM-atonic seizure in eight patients (Supplementary Video 1). Ictal EEG showed generalized spike-wave (GSW) discharges coinciding with the eyelid myoclonia, followed by a spike-wave complex correlating with the myoclonic (spike) and atonic (slow wave) component (Figure 2). We also observed that EM seizures were accompanied by myoclonic jerks of the limbs in 4/36 (11%) patients. These novel seizure types, EM-myoclonic-atonic and EM-atonic, mainly occurred when EM clustered and at times of poor seizure control.

Neurodevelopment and behavioral issues Developmental delay was identiﬁed soon after birth in 55/56 (96%) patients and preceded seizure onset in all. Development regressed or plateaued with seizure onset in 56 patients. Language was severely impaired with 12 patients being non-verbal, at age 2 to 33 years. ID was present in 55/57 patients, being moderate to severe in 50, and mild in ﬁve.

Behavioral problems were seen in 41/56 (73%) patients and were often severe with oppositional and deﬁant behavior with aggression, self-injury and tantrums. ASD was diagnosed in 30 (53%) patients.

Other features An interesting observation was that the families of 39/54 (72%) patients noted they had a high pain threshold. Patients did not respond to painful events, such as a fractured bone, surgical incision of a lesion, or a foreign object lodged in their foot.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 80 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 81

Difficulties with eating were a major problem in 39 (68%) patients. Twenty-three patients had a poor intake because they were fussy about food type, colour, texture and temperature (16), had oral aversion (5), oral hypersensitivity (3), and poor appetite (3). Four patients needed a nasogastric tube because of the severity of their eating problems. In contrast, seven patients had uncontrolled eating with gorging and three would eat inedible objects. 11/55 patients had difficulties with transition from fluids to solid food in early childhood. 17/56 patients had chewing and swallowing difficulties.

Sleep was disturbed in 34/55 (62%) of patients with difficulties initiating (n=19) and maintaining (n=29) sleep. Improved seizure control led to better sleep in 12 patients. There was no clear correlation with EEG features as there was an increase in epileptiform activity in sleep in 9/23 (39%) patients with sleep difficulties, which was also observed in 6/12 (50%) without sleep difficulties. 4 Sleep also improved with melatonin (n=10), clonidine (n=5) and trazodone (n=1).

Figure 2: Ictal electroencephalogram registration during eyelid myoclonia-myoclonic-atonic (EM- myoclonic-atonic) seizure.

ǇĞůŝĚŵǇŽĐůŽŶŝĂ DǇŽĐůŽŶŝĐͲĂƚŽŶŝĐ ǇĞůŝĚŵǇŽĐůŽŶŝĂ DǇŽĐůŽŶŝĐ The red arrows indicate the diﬀerent phases of the seizure associated with a generalized spike wave (eyelid myoclonia) and a further spike wave correlating with the myoclonic (spike) and atonic (wave) components.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 81 82 CHAPTER 4

([DPLQDWLRQÀQGLQJV Photographs of 31/57 patients were available (Figure 3) and revealed subtle dysmorphic features, not present in all patients. These included full, slightly prominent eyebrows with medial ﬂaring in some; hypertelorism; full nasal tip, slightly upturned nasal tip in younger children; short philtrum; cupid bow upper lip; broad mouth with diastemata of the upper teeth; and small pointed chin.

Neurological examination showed hypotonia in 38 (67%), ataxia or an unsteady gait in 29 (51%), tremor in ﬁve (9%) and microcephaly in 1/54 (2%). Orthopaedic abnormalities included foot or spine deformities in 16 (28%), including pes planus (7), pronated feet (4), scoliosis (3), pes caves (1), congenital hip dislocation (1), or congenital hip dysplasia (1).

EEG EEG results were available for 52/57 patients from when their epilepsy was active. Many children did not have further EEGs because of severe behavioral problems. GSW and generalized polyspike wave (GPSW) were reported in 39/52 (75%) patients. Focal or multifocal epileptiform discharges (ED) were seen in 28/52 (54%), often in addition to GSW or GPSW (21/39, 54%). Slow background activity was seen in 26/52 (50%) patients.

Neuroimaging Magnetic Resonance Imaging (MRI) was reported as normal in 37/53 (70%) patients. One patient had a left frontal subcortical nodular heterotopia. The 15 remaining patients had non-speciﬁc ﬁndings: thin corpus callosum (3; 2/3 also had a thin chiasma opticum), enlarged ventricles or subarachnoid spaces (3), hyperintense T2 signals in the white matter (3), posterior centrum semiovale (2) or fasciculus longitudinalis medialis (1), patent cavum vergae (1), atypical white matter abnormalities (1), mega cisterna magna fossa posterior (1), discrete hippocampal tissue loss (1), or a pineal cyst (1).

Brain pathology Brain pathology was available from patient 14, a 33 year old non-verbal woman with a de novo SYNGAP1 nonsense mutation (Supplementary video 2), who died due to an aspiration pneumonia. The pathology showed cerebellar Purkinje cell loss with associated astrocytosis (Figure 4A). Astrocytosis was predominantly found in the superﬁcial cortical layer, but also in the hippocampi and deep grey nuclei (Figure 4B). Macroscopic evaluation showed a low brain weight (1194 gram, -3.2 SD) with widely but moderately distributed leptomeningeal ﬁbrosis, most evident over the frontal lobes (Figure 4C). Cortical neurons did not show evidence of anoxic ischaemic injury (data not shown).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 82 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 83 P X H

4 G W O AE patient 50, AB) patient 51, AC) patient 52, AD) patient 54 (monozygotic AD) patient AC) 52, twin patient 51, AB) patient 50, , M) patient 21, N) patient 23, O) patient 25, P) patient 27, Q) patient 28, R) patient Q) R) patient patient 28, patient 27, P) patient 25, O) N) patient 23, patient 21, , M) V AD F F F N E U M AC T D L AB AA S C dysmorphology. dysmorphology. JK Z B R SYNGAP1 I A Q Y 29, S) patient 31, T) patient 32, U) patient 33, V) patient 34, W) patient 35, X) patient 37,Y) patient 45, Z) patient 47, AA) Z) patient 47, patient 45, V) T) U) patient X) 33, patient 34, W) patient 37,Y) patient 32, patient 35, S) patient 31, 29, (monozygotic AE) patient 55 twinwith with patient 55), patient 54). Figure 3: Figure E) patient 9 of patient 8 (sibling patient Portrait 9), D), photographs patients from with mutations 30/56 patient SYNGAP1 6 , C) or microdeletions B) patient A) patient 7, 4, including SYNGAP1. L) patient 20 K) patient 19, J) patient 18, I) patient 16, H) patient 14, G) patient 13, F) patient 12, of(sibling patient 8),

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 83 84 CHAPTER 4

Figure 4: Brain pathology in a patient with a SYNGAP1 mutation.

ĞƌĞďĞůůĂƌŵŽůĞĐƵůĂƌůĂǇĞƌ

/ŶƚĞƌŶĂůŐƌĂŶƵůĂƌůĂǇĞƌ

ϰŽĨ ŚŝƉƉŽĐĂŵƉƵƐ

ĞŶƚĂƚĞ ŵŽůĞĐƵůĂƌůĂǇĞƌ

(A) Glial fibrillary acidic protein (GFAP) stained section of the cerebellar cortex showing a single surviving Purkinje cell (arrow points towards this cell) in a field of Purkinje cell loss and mild internal granular layer and cerebellar molecular layer astrocytosis. (B) GFAP staining of dentate gyrus showing no neuronal loss but mild reactive astrocytosis in the dentate granular layer and the dentate molecular layer of the hippocampus. (C) Vertex view of the cerebral hemispheres showing leptomeningeal fibrosis.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 84 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 85

Epilepsy syndrome A distinctive generalized epilepsy phenotype emerged combining features of two well-described syndromes, epilepsy with eyelid myoclonia, reported by Jeavons36 , and epilepsy with myoclonic- atonic seizures, described by Doose.37 The overall gestalt in 56/57 patients was a spectrum of severity of generalized DEE with multiple seizure types including EM, myoclonic, atonic (and combinations of all three). TCS were fairly frequent, whereas, focal seizures occurred in a minority. The majority of patients had a severe outcome with moderate to severe ID (49/56, 88%), while 7/56 (12%) had a milder outcome with normal intellect or mild ID. One child with a mild outcome had West syndrome. The remaining patient with no seizures but epileptiform abnormalities on EEG had a (static) developmental encephalopathy (DE) with a severe outcome. Another distinctive feature of SYNGAP1, seen in a quarter of cases, was the presence of reﬂex seizures triggered by eating. 4

Epilepsy treatment and outcome Overall, 20 diﬀerent anti-epileptic drugs (AEDs) were commenced (Supplementary Table 3). Valproate (VPA) and lamotrigine (LTG) were most commonly prescribed (n=45 and n=22) with ongoing prescription of LTG for 77% of patients and VPA for 64% patients. Of the six patients who had received cannabidiol (CBD), 5/6 (83.3%) remained on it.

35/56 (63%) patients had refractory epilepsy with ongoing seizures after two AEDs. Six became seizure-free on the third to sixth AED. In total, 10/56 (18%) patients became seizure-free at 7 years of age (median, range 3-13 yrs). Of these, seizures were controlled on monotherapy with VPA (3), levetiracetam (LEV, 1), or steroids (1, for infantile spasms). Polytherapy combinations resulting in seizure freedom included LEV and topiramate (TPM) (1), VPA and LTG (1), and CBD, VPA and LTG (1). Two were seizure-free on no treatment. Four patients had no seizures for a period of 6 months to 7 years while treated with carbamazepine (1), LTG (1), VPA (1) or VPA, LTG and clobazam, but seizures recurred.

Genotype-phenotype correlation The phenotypes of patients with truncating, splice-site or missense/inframe mutations and microdeletions were similar overall, except that four patients with microdeletions all had an earlier seizure onset (20 months compared with 2 years in the remaining cohort, Table 1). All four patients had microdeletions that included additional genes (4-13 genes). In terms of the outcome, all 4 cases had severe ID, consistent with the rest of the cohort. Given their similar epileptology and other features (high pain threshold in 2, poor oral intake in 2, seizures triggered by hunger and photosensitivity in 1, and dysmorphic features in 2), we included them in the total cohort.

For evaluating genotype-phenotype correlations, we separated SYNGAP1 encephalopathies into two groups based on severity determined by their intellect. Milder phenotypes included normal intellect or mild ID, severe phenotypes referred to individuals with moderate to severe ID. Of the

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 85 86 CHAPTER 4

seven patients with a milder phenotype (green mutations, Figure 1A), five had a truncating or splice-site mutation of which four were located in exon 1-4. Two other patients had a missense variant in exon 6 and 11. Five previously reported patients also had a milder phenotype (green and dark green mutations, Figure 1B) with 3/5 having mutations in exon 1-4. In the combined cohort of our and previously reported patients, milder phenotypes were significantly more often observed in patients with mutation or deletions in exon 1-4 (50%, n=6/12) compared to exons 5-19 (7%, 6/85; p=0.001; Fisher’s exact test). No significant differences were observed in outcome between patients with different mutation types (data not shown).

Patients with SYNGAP1YDULDQWVRIXQNQRZQVLJQLÀFDQFH Q Not included in our cohort of 56 patients were five individuals with a SYNGAP1 variant of unknown significance. Supplementary Tables 1 and 2 show details of their pathogenicity predictions and associated phenotypes. Three patients, who had variants reported in gnomAD or ExAC and for whom inheritance of their SYNGAP1 variant was unavailable, had phenotypes that were consistent with our large cohort.38 Two had generalized DEE and one had mild ID with EM. Their presence in population databases is difficult to explain although EM can be subtle and easily missed. It remains to be determined whether their SYNGAP1 variant has a role in their presentation. The remaining two had variants with lower conservation and pathogenicity prediction scores and had focal epilepsy with or without ID, so their phenotype was quite different.

DISCUSSION

SYNGAP1 is an important gene for the developmental and epileptic encephalopathies (DEEs), present in 1% of our original cohort of patients with unsolved DEEs3,8,17 , having been ﬁrst identiﬁed as a gene for intellectual disability.1 Here, we describe a distinctive generalized DEE phenotype for SYNGAP1 combining syndromic features of epilepsy with eyelid myoclonia, and epilepsy with myoclonic-atonic seizures (Figure 5). SYNGAP1 encephalopathy was associated with a spectrum of mild (8%) to severe (88%) intellectual disability and a range of other comorbidities (Figure 6). Interestingly, seizures were often triggered by eating. The one patient without a DEE had a static developmental encephalopathy without seizures, but did have epileptiform abnormalities on EEG.

SYNGAP1-DEE shares many of the striking features of the syndrome of epilepsy with eyelid myoclonia (EMA).39 The key seizure type, absences with eyelid myoclonia (EM), together with photosensitivity, were found in the majority (64% and 55% respectively) of our cohort. The onset of EM was earlier for SYNGAP1-DEE (<3 yrs) compared with the classical syndrome of EMA (peak onset 6-8 yrs).39 Also, EMA is often associated with better cognitive outcome, with individuals being of normal intellect or having mild ID. An earlier age of onset (<3 yrs) of EM is associated with poorer intellectual outcome and perhaps some of these individuals have SYNGAP1 mutations.40–43

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 86 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 87

However, there is debate as to whether the diagnosis of EMA allows for the presence of additional seizures types.40 Our SYNGAP1-DEE cohort with EM also had myoclonic (33%) and atonic (8%) seizures. We speculate that the more severe cases of EMA may be explained by SYNGAP1 mutations, especially those individuals with earlier onset of EM, myoclonic or atonic seizures.

SYNGAP1-DEE also resembles the syndrome of epilepsy with myoclonic-atonic seizures (MAE, formerly called myoclonic-astatic epilepsy), as reported in 3/11 patients in a previous series.8 Drop attacks are the hallmark of MAE, due to atonic, myoclonic and myoclonic-atonic seizures. 11/57 patients had features consistent with MAE, but with a more severe overall outcome compared with the variable outcome described in MAE.37,44,45 Furthermore, severe developmental delay prior to the onset of seizures in all but two patients, together with focal and multifocal epileptiform activity in 57% of cases, mitigate against the classical features of MAE.37,44,45 4 Figure 5: SYNGAP1 encephalopathy.

This conceptual diagram highlights that the most common epilepsy phenotype in our cohort was an overlapping syndrome combining the features of two well recognized epilepsy syndromes: epilepsy with eyelid myoclonia and epilepsy with myoclonic-atonic seizures. About one third of the cohort did not ﬁt into either of these syndromes and had non syndromic developmental and epileptic encephalopathy. One case had West syndrome. Other individuals with SYNGAP1 encephalopathy may have developmental encephalopathy (intellectual disability) without epilepsy

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 87 88 CHAPTER 4

Figure 6: Comorbidities of SYNGAP1 encephalopathy.

ϭϬϬ ϵϲ ϵϬ ϴ ϴϬ ϴϴ

WĞƌĐĞŶƚĂŐĞ ϳϯ ϳϮ ϲϴ ϳϬ ϲϳ ϲϮ ϲϬ ϱϯ ϱϭ ϱϬ ϰϬ Ϯϴ ϯϬ ϮϬ ϭϬ Ϭ

The ﬁgure shows the percentage of each comorbidity associated with SYNGAP1 encephalopathy in our cohort. Dark red colour denotes moderate to severe intellectual disability and light red denotes mild intellectual disability.

SYNGAP1 encephalopathy often combines the features of these two epilepsy syndromes, epilepsy with eyelid myoclonia and epilepsy with myoclonic-atonic seizures. An overlapping diagnosis of MAE and EMA was present in 35% of our cohort (Figure 5). Bringing the two syndromic concepts together, we also identiﬁed a novel, distinctive seizure type, a drop attack beginning with eyelid myoclonia followed by a myoclonic-atonic component (EM-myoclonic-atonic) or simply an atonic component (EM-atonic). We considered EM-myoclonic-atonic and EM-atonic as overlapping seizure types, as the myoclonic component could be easily missed (Supplementary Video 1). We distinguished EM-(myoclonic)-atonic seizures from atypical absences, which may also include myoclonic and atonic components. First, the myoclonic jerks and atonic seizures were more abrupt and pronounced than in atypical absence seizures, where subtle jerks and loss of tone can occur. Second, the frequency of generalized spike-wave was faster (>3 Hz) compared with atypical absences (<2.5 Hz). One reported patient had seizures with blinking followed by a head drop and fall with irregular GSW27 , suggestive of our SYNGAP1 seizure type.

One patient in our cohort had West syndrome, which has not been associated with SYNGAP1 before (Figure 5). Although atypical absences, present in 20% of our cohort, may suggest a diagnosis of Lennox-Gastaut syndrome, other diagnostic criteria such as tonic seizures and generalized paroxysmal fast activity on EEG were absent. Only one patient had nocturnal tonic seizures (pt. 19).46

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 88 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 89

We observed a DEE in all but one of our 57 patients, while only 55% of the previously published patients had a DEE (Figure 1, 5).6,17,22 The remaining patients in the literature had a mild to severe developmental encephalopathy with (20%) or without (25%) seizures; it was not possible to determine if some of these patients may have had a DEE. Our findings that such a high proportion of patients with SYNGAP1 encephalopathy had a DEE may be due to ascertainment bias, as perhaps only parents from the SYNGAP1 Facebook group participated in the study if they were concerned that their child had seizures. Equally, EM seizures can be very subtle and misinterpreted as a behavioral mannerism. For example, Patient 24 was a 9 year old boy in whom the parents only recognized that his eyelid fluttering was seizure activity after a SYNGAP1 mutation was identified. Patient 16 had frequent eyelid fluttering from the age of 1 year, but her first EEG was not performed until the age of 10 years and showed frequent epileptiform abnormalities. Her seizures, behavioral and sleep problems all improved after commencing anti- 4 epileptic medication. In eight patients, developmental regression occurred two months to >3 years prior to the recognition of seizures. Diagnosis of seizures in SYNGAP1encephalopathy is thus crucial as appropriate management may potentially improve outcome.

We found that a quarter of patients in our cohort had the distinctive feature that their reﬂex seizures were triggered by eating, observed previously in a single patient with seizures triggered by chewing.8 The underlying pathophysiological mechanism for this is unknown. Interestingly, eating problems, with poor intake or gorging, were also found in more than half of our cohort.

From our historical data, we could not reliably evaluate the efficacy of the prescribed AEDs. Further studies are warranted to explore the efficacy of the drugs particularly those that appear of benefit, such as LTG, VPA and CBD, for SYNGAP1-DEE.

SYNGAP1-DEE is associated with four main co-morbidities: ID, behavioral problems, a high pain threshold and ataxia (Figure 6). ID was moderate to severe in 88% of patients, with only two exhibiting borderline intellect. We might have underestimated the prevalence of milder phenotypes due to ascertainment bias, given our recruitment via Facebook. Severe behavioral problems with oppositional behavior, tantrums, aggression and self-injury were present in almost three quarters of patients, as previously observed.17 One of our most fascinating findings was that three quarters of the families reported a high pain threshold in their child, often leading to a delay in diagnosis of injuries and inflammation. Parker et al. also reported a high pain threshold as an anecdotal observation by some families. This finding of a high pain threshold may suggest a role for SYNGAP1 in sensory processing.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 89 90 CHAPTER 4

The neuropathologic study of the brain of a woman with a truncating SYNGAP1 mutation identiﬁed a clear loss of cerebellar Purkinje cells and associated astrocytosis (Figure 4A). An unsteady or ataxic gait was identiﬁed in 51% of our patients, including this patient. Ataxia or gait abnormalities were also observed in 7/10 and 10/17 patients reported by Parker et al. and Mignot et al., respectively, highlighting that cerebellar features are a key part of the SYNGAP1-DEE phenotype.8,17

Truncating and splice-site mutations occur throughout SYNGAP1, while missense mutations mainly cluster in the protein domains.8 A milder outcome of SYNGAP1 DEE was observed significantly more frequently in patients with mutations in exons 1-4, compared with those with mutations in exons 5-19.8 In line with this, Mignot et al. also showed that patients with mutations in exon 4-5 were more pharmacosensitive compared to those with mutations in exon 6-19.8 SYNGAP1 has several isoforms in humans with different promoter sites and alternative splicing; these may explain the variation in phenotypic severity since variants early within the gene transcript often to lead to complete loss of the gene product due to nonsense-mediated decay. Further studies on the presence and function of different human isoforms of SYNGAP1 are warranted to explain the correlation between mutation location and phenotype. Other environmental and genetic factors may also contribute to phenotypic variability. We identified concordant phenotypes in monozygotic twins (patients 53, 54), similar phenotypes in two siblings (patients 8, 9), and yet variable phenotypes in individuals sharing the same mutation (patients 3, 4; patient 6, 7; patient 24, 25; patient 28, 29; patient 37, 38).

We describe SYNGAP1 DEE as a predominantly generalized DEE with ID, behavioral problems, a high pain threshold and ataxia. SYNGAP1-DEE combines features of both epilepsy with eyelid myoclonia and epilepsy with myoclonic-atonic seizures. In SYNGAP1 DEE we observed a novel type of drop attack with eyelid myoclonia evolving into a myoclonic-atonic or an atonic seizure. Furthermore, a quarter of patients had seizures triggered by eating. Recognition of seizures in the context of SYNGAP1-DEE can be challenging, but is crucial to optimising anti-epileptic therapy and improving long term outcome.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 90 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 91

REFERENCES

1. Hamdan FF, Gauthier J, Spiegelman D, et al. Mutations 16. Clement JP, Aceti M, Creson TK, et al. Pathogenic in SYNGAP1 in Autosomal Nonsyndromic Mental SYNGAP1 mutations impair cognitive development by Retardation. N English J Med 2009; 360: 599–605. disrupting the maturation of dendritic spine synapses. 2. Pinto D, Pagnamenta AT, Klei L, et al. Functional Impact of Cell 2013; 151: 709–723. Global Rare Copy Number Variation in Autism Spectrum 17. Parker MJ, Fryer AE, Shears DJ, et al. De novo, Disorder. Nature 2010; 466: 368-372. heterozygous, loss-of-function mutations in SYNGAP1 3. Carvill GL, Heavin SB, Yendle SC, et al. Targeted cause a syndromic form of intellectual disability. Am J resequencing in epileptic encephalopathies identifies Med Genet Part A 2015; 167A: 2231–2237. de novo mutations in CHD2 and SYNGAP1. Nat Genet 18. Vissers LELM, de Ligt J, Gilissen C, et al. A de novo 2013; 45: 1–16. paradigm for mental retardation. Nat Genet 2010; 42: 4. Klitten LL, Møller RS, Nikanorova M, Silahtaroglu A, 1109 -1112. Hjalgrim H, Tommerup N. A balanced translocation 19. Hamdan FF, Gauthier J, Araki Y, et al. Excess of de disrupts SYNGAP1 in a patient with intellectual disability, novo deleterious mutations in genes associated with speech impairment, and epilepsy with myoclonic glutamatergic systems in nonsyndromic intellectual absences (EMA). Epilepsia 2011; 52: 190–193. disability. Am J Hum Genet 2011; 88: 306-316. 5. Krepischi AC V, Rosenberg C, Costa SS, Crolla JA, Huang 20. de Ligt J, Willemsen MH, van Bon BWM, et al. Diagnostic 4 S, Vianna-Morgante AM. A novel de novo microdeletion Exome Sequencing in Persons with Severe Intellectual spanning the SYNGAP1 gene on the short arm of Disability. N Engl J Med; 367: 1921-1929 chromosome 6 associated with mental retardation. Am 21. Rauch A, Wieczorek D, Graf E, et al. Range of genetic J Med Genet Part A 2010; 152A: 2365-2378. mutations associated with severe non-syndromic 6. Zollino M, Gurrieri F, Orteschi D, Marangi G, Leuzzi V, sporadic intellectual disability: an exome sequencing Neri G. Integrated analysis of clinical signs and literature study. Lancet 2012; 380: 1674-1682. data for the diagnosis and therapy of a previously 22. Berryer MH, Hamdan FF, Klitten LL, et al. Mutations in undescribed 6p21.3 deletion syndrome. Eur J Hum SYNGAP1 Cause Intellectual Disability, Autism, and a Genet 2011; 19: 239-242. Specific Form of Epilepsy by Inducing Haploinsufficiency. 7. Writzl K, Knegt AC. 6p21.3 microdeletion involving the Human Mutation 2013; 34: 385-394. SYNGAP1 gene in a patient with intellectual disability, 23. von Stülpnagel, Funke, Haber, et al. SYNGAP1 Mutation seizures, and severe speech impairment. Am J Med in Focal and Generalized Epilepsy: A Literature Overview Genet Part A 2013; 161A: 1682–1685 and A Case Report with Special Aspects of the EEG. 8. Mignot C, Stülpnagel C Von, Nava C, et al. Genetic and Neuropediatrics. 2015; 46: 287–291. neurodevelopmental spectrum of SYNGAP1-associated 24. Redin C, Gérard B, Lauer J, et al. Efficient strategy for intellectual disability and epilepsy. J Med Genet 2016; 53: the molecular diagnosis of intellectual disability using 511-522. targeted high-throughput sequencing. J Med Genet 9. Chen HJ, Rojas-Soto M, Oguni A, Kennedy MB. A synaptic 2014; 51: 724-736. Ras-GTPase activating protein (p135 SynGAP) inhibited 25. Dyment DA, Tétreault M, Beaulieu CL, et al. Whole- by CaM kinase II. Neuron 1998;20:895–904. exome sequencing broadens the phenotypic spectrum 10. Kim JH, Liao D, Lau LF, Huganir RL. SynGAP: A synaptic of rare pediatric epilepsy: A retrospective study. Clin RasGAP that associates with the PSD-95/SAP90 protein Genet 2015; 88: 34–40. family. Neuron 1998; 20: 683–691. 26. O’Roak BJ, Stessman HA, Boyle EA, et al. Recurrent de 11. Jeyabalan N, Clement JP. SYNGAP1: Mind the Gap. Front novo mutations implicate novel genes underlying Cell Neurosci [online serial] 2016;10:1–16. Available simplex autism risk. Nat Commun 2014; 5: 5595-5595. at: http://journal .frontiersin.org/Article/10.3389/ 27. Prchalova D, Havlovicova M, Sterbova K, Stranecky V, fncel.2016.00032/abstract. Accessed February, 2018 Hancarova M, Sedlacek Z. Analysis of 31-year-old patient 12. Tidyman WE, Rauen KA. Expansion of the RASopathies. with SYNGAP1 gene defect points to importance Curr Genet Med Rep 2016; 4: 57-64. of variants in broader splice regions and reveals 13. Kim JH, Lee HK, Takamiya K, Huganir RL. The role developmental trajectory of SYNGAP1-associated of synaptic GTPase-activating protein in neuronal phenotype: case report. BMC Medical Genetics 2017; 18: development and synaptic plasticity. J Neurosci 2003; 62. 23: 1119–1124. 28. McRae JF, Clayton S, Fitzgerald TW, et al. Prevalence and 14. Komiyama NH, Watabe AM, Carlisle HJ, et al. SynGAP architecture of de novo mutations in developmental regulates ERK/MAPK signaling, synaptic plasticity, and disorders. Nature 2017; 542: 433–438. learning in the complex with postsynaptic density 95 29. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster and NMDA receptor. J Neurosci 2002; 22: 9721–9732. J. Standards and Guidelines for the Interpretation of 15. Guo X, Hamilton P, Reish NJ, Sweatt JD, Miller C, Sequence Variants: A Joint Consensus Recommendation Rumbaugh G. Reduced expression of the NMDA of the American College of Medical Genetics and receptor-interacting protein SynGAP causes behavioral Genomics and the Association for Molecular Pathology. abnormalities that model symptoms of schizophrenia. Genet Med 2015; 17: 405–424. Neuropsychopharmacology 2009; 34: 1659–1672.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 91 92 CHAPTER 4

30. Reutens TC, Howell RA, Gebert KE, Berkovic SF. Validation 38. Lek M, Karczewski KJ, Minikel EV, et al. Analysis of protein- of a Questionnaire for Clinical Seizure Diagnosis Accurate coding genetic variation in 60,706 humans. Nature 2016; and uniform classification of epileptic seizures is vital to 536: 285-291. clinical practice and research in. Epilepsia 1992; 33:1065– 39. Striano S, Capovilla G, Sofia V, et al. Eyelid myoclonia 1071. with absences (Jeavons syndrome): A well-defined 31. Fisher RS, Cross JH, French JA, et al. Operational idiopathic generalized epilepsy syndrome or a spectrum classification of seizure types by the International League of photosensitive conditions? Epilepsia 2009; 50: 15-19. Against Epilepsy. Epilepsia [online serial]. 2017;**:1–9. 40. Striano S. Eyelid myoclonia with absences: an overlooked Available at: www.ilae.org. Accessed January 17, 2018. epileptic syndrome? Neurophysiologie clinique 2002; 32: 32. Scheffer IE, Berkovic S, Capovilla G, et al. ILAE classification 287–296. of the epilepsies: Position paper of the ILAE Commission 41. Caraballo RH, Fontana E, Darra F, et al. A study of 63 cases for Classification and Terminology. Epilepsia [online with eyelid myoclonia with or without absences: Type of serial]. 2017;**:1–10. Available at: www.ilae.org. Accessed seizure or an epileptic syndrome? Seizure 2009; 18: 440- January 17, 2018. 445. 33. American Psychiatric Association. Diagnostic and 42. Sadleir LG, Vears D, Regan B, Redshaw N, Bleasel A, statistical manual of mental disorders: The Diagnostic Scheffer IE. Family studies of individuals with eyelid and Statistical Manual of Mental Disorders, fifth edition. myoclonia with absences. Epilepsia 2012; 53: 2141-2148. th American Psychiatric Association, 5 ed. Washington DC; 43. Fournier-Goodnight AS, Gabriel M, Perry MS. Preliminary 2013. neurocognitive outcomes in Jeavons syndrome. Epilepsy 34. Bateman A, Martin MJ, O’Donovan C, et al. UniProt: A Behav 2015; 52: 260–263. hub for protein information. Nucleic Acids Res 2015; 43: 44. Stephani U. The natural history of myoclonic astatic D204–D212. epilepsy (Doose syndrome) and Lennox-Gastaut 35. Krawczak M, Riess J, Cooper DN. The mutational spectrum syndrome. Epilepsia 2006; 47: 53–55. of single base-pair substitutions in mRNA splice junctions 45. Oguni H, Hayashi K, Awaya Y, Fukuyama Y, Osawa M. of human genes: causes and consequences. Hum Genet Severe myoclonic epilepsy in infants - a review based on 1992; 90: 41–54. the Tokyo Women’s Medical University series of 84 cases. 36. Jeavons PM. Nosological Problems of Myoclonic Brain Dev 2001; 23: 736–748. Epilepsies in Childhood and Adolescence. Dev Med 46. Arzimanoglou A, French J, Blume WT, et al. Lennox- Child Neurol 1977; 19: 3-8. Gastaut syndrome: a consensus approach on diagnosis, 37. Doose H, Gerken H, Leonhardt R, Völzke E, Völz C. assessment, management, and trial methodology. Centrencephalic myoclonic-astatic petit mal. Clinical and Lancet Neurol 2009; 8: 82–93. genetic investigation. Neuropadiatrie 1970; 2: 59–78.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 92 SYNGAP1 ENCEPHALOPATHY: A DISTINCTIVE GENERALIZED DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 93

SUPPLEMENTAL DATA (available on request)

Supplemental Table 1: Evaluation of pathogenicity of all SYNGAP1 variants Supplemental Table 2: Phenotypes of individuals with SYNGAP1 variants Supplemental Table 3: Anti-epileptic drug use in patients with SYNGAP1 mutations or deletions Supplemental Video 1: A 4 year old girl with a SYNGAP1 truncating mutation showing the novel seizure type: an eyelid myoclonia-myoclonic-atonic seizure. Supplemental Video 2: A 33 year old female with a SYNGAP1 nonsense mutation showing a focal impaired awareness seizure.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 93 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 94 95

&KDSWHU

Genotype-phenotype correlation of 248 individuals with GRIN2A-related GLVRUGHUVLGHQWLÀHVWZRGLVWLQFW phenotypic subgroups associated with different classes of variants, protein domains and functional consequences

Accepted for publication in Brain

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 95 96 CHAPTER 5

Danique R.M. Vlaskamp1,2*, Vincent Strehlow3,*, Henrike O. Heyne3,4,5*, Katie F.M. Marwick6, Gabrielle Rudolf7,8,9,10, Julitta de Bellescize11, Saskia Biskup12, Eva H. Brilstra13, Oebele F. Brouwer1, Petra M.C. Callenbach1, Julia Hentschel3, Edouard Hirsch8,9,10,14, Peter C. Kind6,15,23, Cyril Mignot16,17,18, Konrad Platzer3, Patrick Rump2, Paul A. Skehel6, David J.A. Wyllie6,15,60, GRIN2A study group***, Giles E. Hardingham6,15,19, Conny M.A. van Ravenswaaij-Arts5, Gaetan Lesca20,21,22,**, Johannes R. Lemke3,**

1 University of Groningen, University Medical Center Groningen, Department of Neurology, Groningen, The Netherlands. 2 University of Groningen, University Medical Center Groningen, Department of Genetics, Groningen, The Netherlands. 3 Institute of Human Genetics, University of Leipzig Hospitals and Clinics, Leipzig, Germany. 4 Analytic and Translational Genetics Unit, Massachusetts General Hospital, MA, USA. 3 Program for Medical and Population Genetics/ Stanley Center for Psychiatric Research, Broad Institute of Harvard and MIT, Cambridge, MA, USA.. 6 Center for Discovery Brain Sciences, University of Edinburgh, Edinburgh, UK. 7 Institut de Génétique et de Biologie Moléculaire et Cellulaire, Illkirch, France, Illirch France. 8 Institut National de la Santé et de la Recherche Médicale, Illkirch, France. 9 Université de Strasbourg, Illkirch, France. 10 Department of Neurology, Strasbourg University Hospital, Strasbourg, France. 11 Department of Pediatric and Clinical Epileptology, Sleep Disorders and Functional Neurology, University Hospitals of Lyon, Lyon, France. 12 CeGaT GmbH and Praxis für Humangenetik, Tübingen, Germany. 13 University Medical Center Utrecht, Department of Genetics, Utrecht, The Netherlands. 14 Medical and Surgical Epilepsy Unit, Hautepierre Hospital, University of Strasbourg, Strasbourg, France. 15 Simons Initiative for the Developing Brain, University of Edinburgh, Edinburgh, UK. 16 Département de Génétique, Groupe Hospitalier Pitié-Salpêtrière, Paris, France. 17 Center de Référence Déﬁciences Intellectuelles de Causes Rares, Paris, France. 18 GRC Sorbonne Université “Déﬁcience Intellectuelle et Autisme”, Paris, France. 19 UK Dementia Research Institute at The University of Edinburgh, Edinburgh Medical School, Edinburgh, UK. 20 Department of Genetics, Lyon University Hospitals, Lyon, France. 21 Lyon Neuroscience Research Center, Lyon, France. 22 Claude Bernard Lyon I University, Lyon, France. 23 Center for Brain Development and Repair, inStem, Bangalore, India. *contributed equally to this work. **jointly supervised.

Acknowledgements. HOH was supported by stipends of the German Federal Ministry of Education and Research (BMBF), (FKZ: 01EO1501) and the German Research Foundation, (DFG HE7987/1-1/1-2). KFMM was supported by a Wellcome Trust Clinical PhD Fellowship (102838). GM was supported by the National Institute of Neurological Disorders and Stroke, (NINDS K08NS092898) and Jordan’s Guardian Angels. IH was supported by intramural funds of the University of Kiel, the German Research Foundation (DFG HE5415/3-1) within the EuroEPINOMICS framework of the European Science Foundation (DFG HE5415/5-1/6-1). Animal studies were supported by the Department of Biotechnology, Bangalore, India. We thank J McQueen for providing the data illustrated in Figure 6B. We thank G. Derksen-Lubsen ,T. Dijkhuizen, N. Doornebal, F.E. Jansen, E.K. Vanhoutte for their support and helpful discussion.

Disclosures. The authors declare no conﬂicts of interest.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 96 GRIN2A GENOTYPEPHENOTYPE CORRELATION 97

ABSTRACT

Background. Alterations of the N-methyl-D-aspartate receptor (NMDAR) subunit GluN2A, encoded by GRIN2A, have been associated with a spectrum of neurodevelopmental disorders with prominent speech-related features and epilepsy.

Methods. We performed a comprehensive assessment of phenotypes with a standardized questionnaire in 92 previously unreported individuals with GRIN2A-related disorders. Applying the criteria of the American College of Medical Genetics and Genomics to all published variants yielded 156 additional cases with pathogenic or likely pathogenic variants in GRIN2A, adding to a total of 248 individuals.

Findings. The phenotypic spectrum ranged from normal or near-normal development with mild epilepsy and speech delay/apraxia to severe developmental and epileptic encephalopathy, often within the epilepsy-aphasia spectrum. We found that pathogenic missense variants in transmembrane and linker domains (misTMD+Linker) were associated with severe developmental phenotypes, whereas missense variants within amino-terminal or ligand-binding domains

(misATD+LBD) and null variants led to less severe developmental phenotypes, which we confirmed in a discovery (p=1x10-6) as well as validation cohort (p=0.0003). Other phenotypes such as MRI abnormalities and epilepsy types were also significantly different between the two groups.

Notably, this was paralleled by electrophysiology data, where misTMD+Linker predominantly led

to NMDAR gain-of-function, while misATD+LBD exclusively caused NMDAR loss-of-function. With respect to null variants, we show that Grin2a+/- cortical rat neurons also had reduced NMDAR function and there was no evidence of previously postulated compensatory overexpression of GluN2B.

Interpretation. We demonstrate that null variants and misATD+LBD of GRIN2A do not only share the same clinical spectrum (i.e. milder phenotypes), but also result in similar electrophysiological

consequences (loss-of-function) opposing those of misTMD+Linker (severe phenotypes; predominantly gain-of-function). This new pathomechanistic model may ultimately help in predicting phenotype severity as well as eligibility for potential precision medicine approaches in GRIN2A-related disorders.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 97 98 CHAPTER 5

INTRODUCTION

N-methyl-D-aspartate receptors (NMDAR) are expressed throughout the brain, mediating excitatory neurotransmission important for development, learning, memory, and other higher cognitive functions. NMDAR are di- or tri-heterotetrameric ligand-gated ion channels composed of two glycine-binding GluN1 (encoded by GRIN1) and two glutamate-binding GluN2 subunits (GRIN2A-D).1 All GluN subunits are composed of an extracellular, a transmembrane and an intracellular component. The extracellular component consists of the amino-terminal domain (ATD) with binding sites for antagonists such as Zn2+ and the ligand-binding domains (LBD) S1 and S2 specific for agonist binding with glycine and glutamate. The channel pore is formed by three transmembrane domains (TMD) M1, M3, M4, and a re-entrant pore-loop M2. The C-terminal domain (CTD) is involved in mediating signals within the intracellular compartment. Compared with the ubiquitously expressed GluN1 subunit, the GluN2 subunits show specific spatiotemporal expression profiles throughout the central nervous system.2 Whereas GluN2B and GluN2D subunits are predominantly expressed prenatally, expression of GluN2A and GluN2C is low prenatally but significantly increases shortly after birth.3

Four genes encoding NMDAR subunits (GRIN1, GRIN2A, GRIN2B, and GRIN2D) have so far been linked to human disease; GRIN2A appears to be associated with the broadest and best characterized phenotypic spectrum, including a variety of disorders of the epilepsy aphasia spectrum, such as Landau-Kleﬀner syndrome and epileptic encephalopathy with continuous spike-and-wave during slow-wave sleep (CSWS), and developmental and epileptic encephalopathies (DEE).4-6

GRIN2A is a gene with a signiﬁcantly reduced number of missense variants in controls compared to the expected number of variants in a similarly sized gene (missense z-score 3.8).7 The ratio of 31.2 expected versus 3 observed null variants in ExAC and the probability of loss-of-function intolerance of 1.00 (pLI score).7 This suggests that GRIN2A null and missense variants strongly reduce evolutionary ﬁtness.

Investigation of functional consequences of disease-associated GRIN2A missense variants revealed various gain- and loss-of-function effects.5,8-13 Identification of null variants, likely leading to GluN2A haploinsufficiency, further complicated understanding the underlying pathomechanisms. GluN2A haploinsufficiency is expected to cause reduced expression of GluN2A, which was thought to be potentially compensated for by consecutive up-regulation of expression of other GluN subunits, especially GluN2B, leading to an altered NMDAR assembly.14 NMDAR containing GluN2B have slower deactivation times than those containing GluN2A.2,15 Replacement of GluN2A by GluN2B has thus been hypothesized to increase the duration of activation of the NMDAR suggesting a net gain-of-function effect mediated by GRIN2A null variants.

To delineate the phenotypic spectrum of GRIN2A-related disorders, we reviewed previously reported and newly identiﬁed individuals with pathogenic or likely pathogenic variants in GRIN2A.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 98 GRIN2A GENOTYPEPHENOTYPE CORRELATION 99

We provide a comprehensive phenotypic dataset of 248 individuals with variants in GRIN2A and integrate these data with protein domain and electrophysiological data. Speciﬁcally, we aimed at elucidating genetic and functional correlates to the wide phenotypic range of GRIN2A- related disorders, for which we reviewed published electrophysiological data and investigated consequences of Grin2a knock-out in cortical rat neurons.

METHODS

Cohort recruitment Data on 92 previously unreported individuals with (likely) pathogenic GRIN2A variants (ENST00000396573) were collected from several diagnostic and research cohorts. Clinical and genetic information were obtained with a speciﬁc questionnaire tailored to phenotypes previously reported in individuals with GRIN2A variants (Supplemental Table 1). We also ascertained additional more detailed phenotypic information on individuals previously been published (Supplemental table 2). All information about the listed variants has been added to an open-access online database (www.grin-database.de) (see also Supplemental Table 3). This study has been approved by the ethics committee of the University of Leipzig (224/16-ek, 402/16-ek).

5HYLHZRIWKHOLWHUDWXUHDQGYDULDQWFODVVLÀFDWLRQ We searched the literature (www.ncbi.nlm.nih.gov/pubmed) (until March 23rd, 2018) for reports of cases with GRIN2A variants and reviewed the associated clinical and genetic information. Based on the recommendations of the American College of Medical Genetics and Genomics (ACMG),16,17 we classified missense variants fulfilling at least one of the following conditions (in addition to constraint and prediction scores) as likely pathogenic: de novo + absent from controls* OR confirmative functional studies + absent from controls* OR de novo + confirmative functional studies OR present in three or more affected and no healthy individuals of one family + absent from controls* OR novel missense variant at a location that had been classified as pathogenic according to the above conditions + absent from controls*. Furthermore, null variants located in exon 3-14 (until amino acid position 838) were classified as likely pathogenic. *Controls were over 120.000 people without severe pediatric disease compiled in the gnomAD browser | genome Aggregation Database (http://gnomad.broadinstitute.org/). Only variants classified as pathogenic or likely pathogenic were considered for further genotype-phenotype correlations in this study, regardless of the associated phenotype.

Statistical analysis All statistical analyses were done with the R programming language (www.r-project.org). Fisher’s exact tests for Count Data, Wilcoxon rank-sum tests and Cochran Armitage tests were performed as referenced in the results. P-values were corrected for multiple testing with the Bonferroni method. For Fisher’s exact tests, we reported odds ratios (OR) and 95% conﬁdence intervals (95%-

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 99 100 CHAPTER 5

CI). To investigate variant clustering in different phenotypes, we calculated the distance (linear amino acid sequence) of all possible variant pairs of individuals with the same ID/DD phenotypes to all combinations of different ID/DD phenotypes (mild versus severe). We compared the variant distances of same versus different phenotypes with Wilcoxon rank-sum tests. The R code used to perform the statistical analyses and figures is available upon request.

5DQNLQJVHYHULW\RILQWHOOHFWXDOGLVDELOLW\GHYHORSPHQWDOGHOD\ ,''' We identiﬁed 178 individuals with detailed information about the presence or absence of ID/ DD and the apportioned categories reﬂecting the severity of the phenotype according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5): no ID/DD (0 points), mild ID/DD (1 point), moderate ID/DD (2 points), severe ID/DD (3 points), and profound ID/DD (4 points) (Supplemental table 4). The terms ID and DD are used interchangeably here.

Neuronal culture, generation of the Grin2a–/–UDW51$TXDQWLÀFDWLRQ Cortical rat neurons were cultured as described18 at a density of between 9-13 x 104 neurons per cm2 from E20.5 rats with Neurobasal growth medium supplemented with B27 (Invitrogen, Paisley, UK). Experiments were performed at days in vitro (DIV) 7-16 as indicated. To generate the Grin2a–/– rat, single cell Long Evans Hooded rat embryos underwent pronuclear microinjection of mRNA encoding the enzyme Cas9 and small guide RNAs (sgRNA) binding to the 5’ and 3’ end of exon 8 of Grin2a, before being implanted into pseudopregnant mothers. The resulting live births were screened by PCR for genomic deletions due to repair by non-homologous end joining of double stranded breaks targeted to either side of exon 8. A 1065bp deletion spanning exon 8 (which encodes key pore forming domains of GluN2A) was identified, and confirmed by sequencing (data not shown). Genotyping was performed using primer pairs P1 (AGGGAAGAAGGGAACAGGAG) with P2 (TCTCTGGGATTCAGTGCAGA) and P3 (AAGGCAGAGAGAGAGACAAAG) with P4 (ATGGCAGTTCCCAGTAGCAT). P1 and P3 bind to the 5’ end of the deletion, P2 binds to the 3’ end of the deletion, and P4 binds within the deletion. The sgRNA design and generation of the founder animals was performed by Horizon Discovery Group plc (St Louis, MO, USA). All the experiments were performed using wild-type, heterozygous, and homozygous littermate 6 matched animals. Animals were treated and all experiments performed in accordance with UK Animal Scientific Procedures Act (1986) following local ethical review.

RNA was isolated from cultured neurons using the Roche High Pure RNA Isolation Kit (including DNase treatment), according to manufacturer’s instructions (Roche, Hertfordshire, UK). 3 wells from a 24-well plate were pooled for each animal. cDNA was synthesized from 13 mg RNA using a Transcriptor First Strand cDNA Synthesis Kit (Roche), according to manufacturer’s instructions, then stored at -20 oC. For real time PCR (RT-PCR), cDNA was diluted to the equivalent of 6 ng of initial RNA per 15 ml qPCR reaction, per gene of interest. Real-time PCR was performed in a Stratagene Mx3000P QPCR System (Agilent Technologies, Waldbronn, Germany), using the FS universal SYBR Green MasterRox mix (Roche), according to manufacturer’s instructions. The required

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 100 GRIN2A GENOTYPEPHENOTYPE CORRELATION 101

amount of template was mixed with water, SYBR Green MasterRox mix and forward and reverse primers (200 nM each final concentration) to the required reaction volume. Primers used were: Grin2a AGCCAGAGACCCCGCTAC & TGGGGTGCACCTGGTAAC; Gadph: AGAAGGCTGGGGCTCACC & AGTTGGTGGTGCAGGATGC. Technical replicates as well as no template and no RT negative controls were included. The qRT-PCR cycling programme was 10 min at 95 oC, then 40 cycles of 30 sec at 95 oC, 40 sec at 60 oC, with detection of fluorescence and 1 min at 72 oC, followed by 1 cycle (for dissociation curve) of 1 min at 95 oC, and 30 sec at 55 oC, with a ramp up to 30 sec at 95 oC, (ramp rate: 0.2 oC /sec) with continuous detection of fluorescence on the 55-95 oC ramp. Data were normalized to Gadph expression.

Cell culture electrophysiological recording and analysis Coverslips containing cortical neurons were transferred to a recording chamber perfused (at a ﬂow rate of 3-5 ml/min) with an external recording solution composed of (in mM): 150 NaCl, 2.8

KCl, 10 HEPES, 2 CaCl2, 10 glucose and 0.1 glycine, pH 7.3 (320-330 mOsm). Tetrodotoxin (300 nM) was included to block action-potential driven excitatory events. Patch-pipettes were made from thick-walled borosilicate glass (Harvard Apparatus, Kent, UK) and filled with a K-gluconate-based internal solution containing (in mM): potassium gluconate 141, NaCl 2.5, HEPES 10, EGTA 11; pH 7.3 with KOH). Electrode tips were fire polished for a final resistance ranging between 3-5 MΩ. All NMDAR currents were evoked by 150 μM NMDA and 100 μM glycine, both applied using a perfusion system. Currents were recorded at room temperature (21 ± 2°C) using an Axopatch 200B amplifier (Molecular Devices, Union City, CA). Neurons were voltage-clamped at –65 mV and recordings were rejected if the holding current was greater than –100 pA or if the series resistance drifted by more than 20% of its initial value (<20 MΩ). Whole-cell currents were analyzed using WinEDR v3.2 software (John Dempster, University of Strathclyde, UK). To determine the ifenprodil- sensitivity of neurons, whole-cell NMDAR currents were recorded followed by the inclusion of 3 μM ifenprodil in the recording solution for a blocking period of 90 seconds. The whole-cell NMDAR current was re-assessed with 3 μM ifenprodil included, and the % block was calculated. To determine spermine potentiation neurons were voltage-clamped at –30 mV and switched to a low sodium recording solution composed of (in mM): 70 NaCl, 60 choline chloride, 2.8 KCl, 20 HEPES, 10 glucose, 0.1 glycine, 0.1 diethylenetriaminepentaacetic acid, pH 6.5 with NaOH. NMDA currents were evoked by 150 mM NMDA then re-assessed in the presence of 100 mM spermine. Only cells with NMDA-evoked currents greater than 40 pA were included.

Western blotting Neurons were lysed in 1.5x Lithium Dodecyl Sulfate sample buffer (NuPage, Life Technologies) and boiled at 100°C for 10 min. Approximately 10 μg of protein was loaded onto a precast gradient gel (4-12%) and subjected to electrophoresis. Briefly, western blotting onto a Polyvinylidene fluoride (PVDF) membrane was then performed using the Xcell Surelock system (Invitrogen) according to the manufacturer’s instructions. Following the protein transfer, the PVDF membranes were blocked for 1 hour at room temperature with 5% (w/v) non-fat dried milk in Tris buffered

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 101 102 CHAPTER 5

saline with 0.1% Tween 20. The membranes were incubated at 4°C overnight with the primary antibodies diluted in blocking solution: anti-GluN2A (N-terminus, 1:1000, Invitrogen), and anti- beta actin (1:200000, Abcam) or anti-GluN2B C-terminus (1:8000, BD Biosciences) and anti-beta actin. For visualisation of Western blots, Horse radish peroxidase-based secondary antibodies were used followed by chemiluminescent detection on Kodak X-Omat ﬁlm.

RESULTS

We reviewed data on 92 unpublished individuals with (likely) pathogenic GRIN2A variants with systemically assessed phenotypes. After re-evaluation of all published GRIN2A variants based on ACMG recommendations,16,17 we additionally included 156 previously reported individuals with (likely) pathogenic variants. Thus, we were able to collectively review genotypes and phenotypes of 248 individuals with GRIN2A-related disorders.

The cohort In our cohort of individuals whose gender was known, 45.1% were female (n = 87) and 54.9% were male (n = 106). Gender was unknown in 55 cases. The youngest individual was 11 months old at evaluation, the oldest was 71 years (median age was 8 years). Of the 248 individuals, 121 (48.8 %) were single cases, including 65 individuals (65/121; 53.7%) where a de novo conﬁrmation of the variant was performed. The remaining 127 individuals (127/248; 51.2%) derived from 36 diﬀerent families. Among 3038 individuals with a neurodevelopmental disorder with epilepsy screened by epilepsy panel sequencing (covering GRIN2A) in the same diagnostic lab, seven a displayed a (likely) pathogenic variant in GRIN2A revealing a prevalence of 0.23% in this disease spectrum.

Variant type and distribution 145 individuals (145/248; 58.5%) had likely protein-truncating variants referred to as null variants. Null variants included 37 (37/145; 25.5%) gross deletions or duplications spanning up to the whole gene but not aﬀecting adjacent genes, 35 (35/145; 24.1%) nonsense variants, 23 (23/145; 15.9%) small frameshift deletions or duplications, 42 (42/145; 29.0%) canonical splice-site variants, 5 (5/145; 3.4%) complex chromosomal rearrangements disrupting GRIN2A, and 3 (3/145; 32.1%) loss-of-start codon variants (Figure 1A). A total of 53 diﬀerent null variants were considered (likely) pathogenic, of which 22 were recurrent.

The remaining 103 individuals (103/248; 41.5%) had missense variants. Of them, 13 individuals (13/103; 12.6%) had variants in the extracellular ATD and 56 individuals (56/103; 54.4%) in the

extracellular LBD S1 or S2, all of which are referred to as misATD+LBD throughout the manuscript (for protein domains see Supplemental table 5). In addition, 6 individuals (6/103; 5.8%) had variants

in linker regions and 28 (28/103; 27.2%) in the three TMD M2-M4, referred to as misTMD+Linker. No variants aﬀecting the C-terminus met the ACMG recommendations for being (likely) pathogenic

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 102 GRIN2A GENOTYPEPHENOTYPE CORRELATION 103 Pathogenic or likely (B) (B) Ex. 14 CTD teria the last exon 14 is spared,teria the last which exon encodes 14 nearly the r C-terminal domain. Appropriately, the number of rare (MAF<0.1%) domain. the Appropriately, number of rare (MAF<0.1%) r C-terminal of loss-of function intolerance 1.00 in ExAC). ExAC). in 1.00 function loss-of of intolerance Ex. 13 S2 LBD M4 S2 LBD luN2A RIN2A G G Ex. 11 Ex. 12 M3 encoding functionally transmembrane ligand important binding and S2 domains domains as well M1-M4 and domains (S1 Ex. 8 Ex. 9 Ex. 10 GRIN2A Ex. 7 S1 LBD S1 LBD M1 M2 Ex. 5 Ex. 6

Ex. 4 ATD Ex. 3 SP Pathogenic or likely pathogenic null variants cri are spread bars) ACMG over nearly according to (red the gene. entire However, 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1469 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600 3800 4000 4200 Figure 1: Distribution of variants. Figure A B linker regions). The density of missense variants in healthy gnomAD controls (MAC=2, black The is highest bars) linker regions). density in the intracellula of missense variants in healthy gnomAD (MAC=2, controls missense variants in ExAC in this region is not lower than expected by chance. pathogenic missense variants cluster bars) in regions (red of (A) complete C-terminal domain. Null variants (black bars) occur primarily in last exon 14 in healthy gnomAD controls (probability

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 103 104 CHAPTER 5

(Figure 1B). A total of 44 diﬀerent missense variants were considered (likely) pathogenic, of which 23 were recurrent.

Intellectual disability/developmental delay In our GRIN2A cohort, cognitive assessment was available for 177 individuals, of which 111 individuals (111/177; 62.7%) had ID/DD. Among the 177 individuals, the level of ID/DD was mild in 35 cases (35/177; 19.8%), moderate in 17 (17/177; 9.6%), severe in 8 (8/177; 4.5%) and profound in 16 (16/177; 9.0%). In 35 cases (35/117; 19.8%), the severity of ID/DD could not be speciﬁed in more detail. Sixty-six individuals (66/177; 37.3%) had normal intelligence (Figure 2A).

6HL]XUHVDQGHOHFWURHQFHSKDORJUDSK\ ((* Information on the epilepsy phenotype was available for 219 cases. The majority of patients (192/219; 87.7%) had seizures, including 121 individuals (121/219; 55.3%) with focal seizures (with or without evolution to bilateral tonic-clonic seizures). Twenty-one individuals (21/219; 9.6%) had tonic clonic seizures of unknown onset, four (4/219; 1.8%) had epileptic spasms and 46 (46/219; 21.0%) had unspeciﬁed seizures. Several individuals displayed a spectrum of diﬀerent seizure types. Twenty-seven individuals (27/219; 12.3%) did not have seizures (Figure 2B).

EEG information was available in 152 individuals and displayed epileptiform discharges in 143 individuals (143/152; 94.1%). In 86 cases (86/152; 56.6%), focal discharges were recorded; 34 cases (34/152; 22.4%) had centrotemporal spikes (CTS) and 28 cases (28/152; 18.4%) had multifocal discharges. Fifty-one individuals (51/152; 33.6%) had CSWS and six individuals (6/143; 4.2%) had generalized discharges. Only nine individuals (9/152; 5.9%) had a normal EEG (Figure 2C).

Recognizable epilepsy syndromes comprised the known spectrum of GRIN2A-associated epilepsy syndromes, such as benign epilepsy with centro-temporal spikes, atypical childhood epilepsy with centrotemporal spikes and Landau-Kleﬀner-syndrome.

Language and speech disorders Information about speech phenotypes was available for 140 cases. The vast majority of individuals presented with speech disorders (129/140; 92.1%). In 120 patients where the type of speech disorder was deﬁned, 55 individuals (57/140; 39.3%) had moderate speech/language impairment, including dysarthria, speech dyspraxia, dysphasia, speech regression with residual impairments, sometimes supplemented by minor impairments such as impaired pitch, hypernasality or imprecise articulation. Twenty-six individuals (26/140; 18.6%) had aphasia (including speech regression with loss of speech) and another 26 (26/140;’18.6%) had isolated delay of speech development. Eight individuals (8.140; 5.7%) presented with temporary speech regression. The type of speech disorder was not further speciﬁed in 14 individuals (14/140; 10.0%). Only 11 individuals (11/140; 7.9%) had normal speech development. Speech disorders were not necessarily linked to EEG abnormalities as 10 out of 11 individuals with normal speech development had

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 104 GRIN2A GENOTYPEPHENOTYPE CORRELATION 105 EEG EEG (C) epilepsy, epilepsy, normal speech delay speech developmental speech disorder moderate speech regression aphasia unspecified speech disorde no seizures seizures focal seizures generalized epileptic spasms unspecified seizure (B) (B) outcome, intellectual (A) 18.6% 55.3% 12.3% 7.9% 39.3%

10.0% 21.0% 5.7% 18.6% 9.6% 1.8% Speech language disorder language Speech n = 140 Epilepsy n = 219 D B normal mild moderate severe profound unspecifie normal focal (CTS) focal multifocal generalized CSWS 22.4% 37.3% 15.7% 5.9% 19.8% 18.4% -related disorders display a broad range of phenotype severity and expressivity with respect to 9.6% 3.9% 19.8% 33.6% 4.5% 9.0% GRIN2A speech or language impairments. language or speech ID/DD n = 177 EEG n = 152 (D) C A Distribution of phenotypes. 2: Distribution of phenotypes. Figure Individuals with patterns, patterns,

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 105 106 CHAPTER 5

an abnormal EEG and eight out of nine individuals with normal EEG still had abnormal speech development (only one individual had normal speech, normal EEG, no epilepsy, no ID/DD) (Figure 2D).

Other neurologic and psychiatric phenotypes Information about tone was available for 139 cases. Forty individuals (40/139; 28.8%) had hypotonia, including 18 individuals (18/139; 11.5%) with mild, 2 (2/139; 1.4%) with moderate and 4 (/139; 2.9%, including one individual with arthrogryposis) with severe hypotonia (Supplemental Figure 2A). Hypotonia was not further speciﬁed in 18 individuals (18/139; 13.0%).

We identified 19 individuals (19/72; 26.4%) with movement disorders, including ataxia (n=10), dystonic/spastic and/or choreatic movement disorders (n=8, including two individuals with complex movement disorders: individual #039 with no ambulation, spasticity, sometimes dystonic, choreatic, athethotic movements and individual #058 with involuntary movements, paroxysmal dyskinesia, movement abnormality of the tongue, abnormality of eye movement, impaired smooth pursuit), and an unspecified movement disorder (n=1) (Supplemental figure 2B).

Information about neuropsychiatric comorbidities was available in 70 cases. Seventeen individuals (17/70; 24.3%) displayed behavioral or psychiatric disorders, such as attention deﬁcit hyperactivity disorder (n=6), autism spectrum disorder (n=6), schizophrenia (n=2) and anxiety disorder (n=1). Two individuals had unspeciﬁed behavioral abnormalities.

0DJQHWLFUHVRQDQFHLPDJLQJ 05, Brain MRI data were available for 85 individuals. Ten individuals (10/85, 11.7%) had a brain abnormality, including focal cortical dysplasia, hypoplasia of corpus callosum with midline lipoma, hippocampal hyperintensity, hippocampal sclerosis, heterotopia, subcortical lesion, hypoplastic olfactory bulb, cerebellar glioma, enlarged Virchow-Robin spaces, delayed myelinisation (n=1 for each). An additional 10 patients (10/5; 11.7%) had generalized volume loss compatible with brain atrophy. Sixty-ﬁve individuals (65/85; 76.5%) had no MRI abnormalities (Supplemental ﬁgure 2C). Abnormal gyral patterns similar to some cases with GRIN1- and GRIN2B-related disorders were not observed19, 20 and were also not expected as knockdown of only GluN1 and GluN2B (and not GluN2A) have been shown to slow down neuronal migration.21

Genotype-phenotype correlations reveal two distinct phenotype groups For 178 out of all 248 individuals with (likely) pathogenic variants in GRIN2A, we obtained detailed information about the presence or absence of ID/DD and ranked severity of intellectual disability into ﬁve categories (see methods). Comparing 59 individuals with missense and 108 with null variants, we found more severe ID/DD in carriers of missense variants (Cochran Armitage Test, p-value 0.00011). However, individuals with missense variants displayed a bimodal distribution of ID/DD severity (Figure 3A). We compared spatial variant clustering between individuals with the same severity of ID/DD (severe or mild) and those with mixed severity of ID/DD (Wilcoxon Rank

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 106 GRIN2A GENOTYPEPHENOTYPE CORRELATION 107

test; p-value 2x10-6 (severe versus mixed ID/DD cases) and 0.5 (comparing mild versus mixed ID/ DD cases). This suggests that missense variants in diﬀerent parts of the protein lead to distinct ID/

DD phenotypes. We observed that 19 individuals with misTMD+Linker had more severe phenotypes

than 40 individuals with misATD+LBD (Figure 3B). To test this observation statistically, we randomly separated missense carriers into a discovery cohort (n=39) and a validation cohort (n=20). Carriers

of misTMD+Linker had signiﬁcantly more severe ranked ID/DD compared to carriers of misATD+LBD, both in the discovery cohort (median score 4 and 0, respectively; p=5x10-8, Cochran Armitage Test) as well as the validation cohort (median score 4 and 0, respectively; p = 0.0003; Cochran Armitage Test).

Figure 3: Severity of ID/DD. A

4 3

2 Severity of ID of Severity

missense Null n = 59 n = 108 B Variant type

2 Severity of ID of Severity 1

ATD S1 M2 Linker M3 S2 M4 T Protein domain Comparison of severity of ID/DD in carriers of variants in diﬀerent protein domains (method: Fisher’s exact test). (A) missense (blue) and null (=truncating) variants (yellow). (B) missense variants in diﬀerent protein domains in the order of the linear amino acid sequence and truncating variants (far right). Here, variants that were inherited are colored black, de novo red and unknown grey. Violins are plotted to have the same maximum width. Bottom, middle and top of boxplots within violins show the 1st, 2nd and 3rd quartile of the data; whiskers maximally extend to 1.5 x interquartile range.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 107 108 CHAPTER 5

Accordingly, all 32 misTMD+Linker were de novo, while only 18 of 47 misATD+LBD were de novo (Fisher’s exact test, OR Inf, 95%-CI 11 to Inf, p-value 2x10 -9). Other variants were inherited; unknown variants were excluded from the test. Notably, carriers of the 107 null variants had a similar degree of ID

(median 1, mild ID) compared to carriers of misATD+LBD (median 0, no ID, Cochran Armitage Test,

p-value 0.3). Furthermore, all 66 individuals with normal intellect were carriers of misATD+LBD or null variants.

We found signiﬁcant diﬀerences for other phenotypes only between individuals with misTMD+Linker

and those with misATD+LBD or null variants, but not between those with misATD+LBD and null variants (Figure 4, all phenotype comparisons in Figure 4 were done with Fisher’s exact tests). Although we

observed no diﬀerence in the presence of epilepsy in individuals with misTMD+Linker and individuals

with misATD+LBD or null variants (Fisher’s exact test, p=0.54), we found signiﬁcant diﬀerences with respect to seizure type as epileptic spasms were only observed in individuals with misTMD+Linker, -6 but not in individuals with misATD+LBD or null variants (Fisher’s exact test, p=2.6x10 ). There were

also significantly more cases with focal seizures in the cohort with misATD+LBD/null variants than -4 in the misTMD+Linker cohort (Fisher’s exact test, p=4.1x10 , OR 5.0, 95%-CI 1.9 to 15.7). There were no significant differences for generalized seizures (Fisher’s exact test, p = 1.0) or for particular EEG

patterns. All individuals with generalized volume loss on MRI were carriers of misTMD+Linker (Fisher’s exact test, p=0.002, OR 5.8, 95%-CI 1.7 to 21.3), while this feature was not observed in any carrier

of misATD+LBD/null variants.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 108 GRIN2A GENOTYPEPHENOTYPE CORRELATION 109

Figure 4: Phenotypes correlated with protein domains.

epilepsy epileptic spasms 0.00022 n= 4, n (epilepsy)=219 epilepsy focal seizures 0.013 n= 121, n (epilepsy)=219 epilepsy generalized seizures 0.74 n= 21, n (epilepsy)=219 epilepsy no seizures 0.54 n= 27, n (epilepsy)=219 epilepsy unspecifiedseizures 0.049 n= 46, n (epilepsy)=219 EEGCSWS 0.021 n= 51, n (EEG)=152 EEGfocal 0.71 n= 24, n (EEG)=152 EEGfocal (CTS) 0.024 n= 34, n (EEG)=152 EEGgeneralized 0.11 n= 6, n (EEG)=152 EEGmultifocal 0.15 n= 28, n (EEG)=152 EEGnormal 0.0059 n= 9, n (EEG)=152 −11 Phenotype × speechlanguage disorder aphasia 4.4 10 n= 26, n (speechlanguage disorder)= 140 speechlanguage disorder moderate speechdisorder 0.00038 n= 55, n (speechlanguage disorder)= 140 speechlanguage disorder speech developmental delay 0.37 n= 26, n (speechlanguage disorder)= 140 speechlanguage disorder speech regression 1 n= 8, n (speechlanguage disorder)= 140 speechlanguagedisorder unspecifiedspeechdisorder 0.22 n= 14, n (speechlanguage disorder)= 140 × −6 hypotonia unspecifie 1.2 10 n= 40,n (hypotonia)= 139 − brain MRIatrophy 2.2×10 7 n= 9, n (brain MRI)=85 movement disorder ataxia 1 n= 10, n (movement disorder)= 72 − movement disorder dystonic/choreatic 1.1×10 7 n= 8, n (movement disorder)= 72 −1 0 1 misTMD+Linker vs null+mis ATD+LBD (log10 Odds Ratio)

Comparison of phenotypes associated with variants in diff erent protein domains (method: Fisher’s exact test). Phenotype diff erences being signifi cant after Bonferroni multiple testing correction for 17*2 test are labelled red. For each gene, Odd’s ratios (OR) with 95%-CI (grey/red bars) and number of patients with the phenotype and number of patients in the phenotype category are shown. For clarity, OR and CIs are cut at ±1.7.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 109 110 CHAPTER 5

Variance of ID/DD phenotype in individuals with the same genetic variant We investigated whether individuals with the same genetic variant had similar ID/DD phenotypes (classiﬁ cation see methods). We investigated 98 individuals carrying 24 unique variants where ID/ DD phenotypes were available in at least two individuals per variant (Figure 5). The mean variance of ID/DD phenotypes per variant was 0.65 (±0.64 SD). Permuting family labels 10,000 times, we found that the real value was lower than the mean variance in 15 of 10,000 permutations (Supplemental Figure 2, empirical p-value 0.0016). This suggests that while considerable phenotype expressivity exists, the same variant leads to similar ID/DD phenotypes. However, more and better ID/DD data (e.g. measured as IQ) is needed to optimally study phenotype expressivity.

Figure 5: Variance of ID/DD phenotype in individuals with the same genetic variant.

F&!* F*!$ F&!7 F&!* F$!* S 3UR$UJ S $UJ+LV S 7KU0HW S 3UR$UJ S $VQ6HU SURIRXQG Vevere PRGHrate PLOG no

F&!$ F&!* F$!* F*!7 F7!& S $VQ/\V S /HXVDO S 0HWVDO S $VST\U S 0HW" SURIRXQG Vevere PRGHrate PLOG no

F*!$ F&!7 F7!$ F*!$ Fí$!* S Tr S S *OQ S /HX SURIRXQG Vevere PRGHrate PLOG ID/DD no

FGHO* F&!7 'HO([RQí Del Exon 4 'HO([RQí S VDOTr SIV S $UJ SURIRXQG Vevere PRGHrate PLOG no 6 'HO([RQí 'HO([RQí 'HO([RQ W ST

SURIRXQG Vevere PRGHrate PLOG no 6 6 6 6 LQGLYLGXDO

ID/DD phenotypes (y-axis) of all recurrent GRIN2A genetic variants are of similar degree (empirical p-value 0.0016) suggesting that the same variant leads to similar ID/DD phenotypes despite considerable phenotype expressivity.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 110 GRIN2A GENOTYPEPHENOTYPE CORRELATION 111

Both phenotype groups correspond to opposing electrophysiological consequences Of the 44 (likely) pathogenic missense variants included in this study (ATD: 4, LBD: 20, TMD: 16, Linker: 4), 23 (52%) had been functionally investigated so far (ATD: 3, LBD: 14, TMD2-4: one each, Linker: 3). We compared the published functional data of all 23 missense variants in the diﬀerent 5,8-13,22-24 domains (Supplemental table 3). Six misTMD+Linker variants display predominantly gain-of-

function effects (5x gain-of-function and 1x loss-of-function) while 17 extracellular misATD+LBD variants show exclusively loss-of-function activity (Fisher’s exact test, p-value = 2x10-4, OR Inf, 95%-CI 5.3 to Inf). We conclude that these opposing electrophysiological consequences are the most likely explanation for the significantly different degree of severity of ID/DD as well as other

phenotypic diﬀerences associated with misATD+LBD and misTMD+Linker. The current single exception

to this pattern is the loss-of-function misTMD+Linker variant c.1642G>A, p.(Ala548Thr), found in an individual with moderate ID.

Rat model and electrophysiological analysis Our phenotypic data suggest that the clinical consequences of GRIN2A null variants are similar to

the clinical consequences of misATD+LBD loss-of-function variants. However, it has previously been hypothesized that GRIN2A null variants could ultimately result in NMDAR with gain-of-function (or altered function) through compensatory increased expression of other NMDAR subunits, particularly GRIN2B. We therefore sought to determine whether homozygous or heterozygous loss of Grin2a results in compensatory up-regulation of Grin2b expression. We employed a newly created Grin2a knockout rat, and compared NMDAR currents in cortical neurons cultured from these rats with those from their wild-type and heterozygous littermates (litters generated by Het-Het crosses (Figure 6A). In this model, there is no detectable compensatory increase in the expression level of GluN2B protein (Figure 6B). We analyzed NMDAR currents at two developmental stages: after 7-8 DIV when currents are almost exclusively GluN2B dominated, and at 15-16 DIV when there is a significant proportion of GluN2A (Grin2a)-containing NMDAR (Supplemental Figure 3).21 Analysis of NMDAR current density at DIV 7-8 revealed no genotype dependent difference, consistent with the near-exclusive presence of GluN2B-containing diheteromeric NMDARs (Figure 6C). Analysis of currents at DIV 15-16 showed an age dependent increase of currents, as expected, but also a deficit in currents in Grin2a+/– and Grin2a–/– neurons, relative to wild-type. This suggests that any compensation of GluN2A deficiency through an increase in other NMDAR subunits is insufficient to rescue currents to wild-type levels.

We next investigated whether there was any evidence of compensation through GluN2B up regulation that could be detected via electrophysiological assessment. If there was compensation, then the proportion of whole cell currents dependent on GluN2B would be expected to be higher in Grin2a+/– and Grin2a–/– neurons, relative to wild-type, at DIV 15-16. We measured the portion of the whole cell currents sensitive to the GluN2B-selective antagonist ifenprodil (Figure 6D), but found no diﬀerence in the magnitude of ifenprodil-sensitive current at DIV 15-16 (or

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 111 112 CHAPTER 5

Figure 6: No compensation by GluN2B. GluN2B-mediated currents do not increase to compensate for GluN2A deﬁciency in cortical eurons from a Grin2a knock-out rat.

(A) Western blot and quantification confirming the absence of GluN2A expression in Grin2a–/– neurons, and an intermediate expression level in Grin2a+/– neurons at DIV15 (DIV=days in vitro). Tukey’s test reveals a significant difference between Grin2a+/+ vs. Grin2a+/– (P=0.0196) and vs. Grin2a–/– (P=0.0014). Grin2a+/+: n=6; Grin2a+/–: n=5; Grin2a–/–: n=5. (B) Western blot and quantification confirming no changes in GluN2B expression in either Grin2a+/– or Grin2–/– neurons compared to Grin2a+/+ neurons at DIV15. Grin2a+/+: n=6; Grin2a+/–: n=5; Grin2a–/–: n=5. (C) NMDA (150 μM) evoked currents were measured in cortical neurons of the indicated genotypes and periods of culture. Currents were calculated and normalized to cell capacitance to give a value for the current density within the neuron. 2-way ANOVA reports a significant developmental stage effect (P<0.0001) and a significant genotype effect (P=0.013) as well as a significant interaction between the two (P=0.0059). Sidak’s post-hoc test reveals a significant difference between Grin2a+/+ vs. Grin2a+/– (P=0.0007) and vs. Grin2a–/– (P=0.0006). Grin2a+/+: n=38 (DIV7-8), 38 (DIV15-16) cells, 8 animals; Grin2a+/–: n=40 (DIV7-8), 35 (DIV15-16) cells, 9 animals; Grin2a–/–: n=48 (DIV7-8), 31 (DIV15-16) cells, 10 animals. (D) NMDA (150 μM) evoked currents were measured in cortical neurons of the indicated genotypes and periods of culture before and after the application of the GluN2B-selective antagonist ifenprodil (3 μM). The ifenprodil-sensitive current was calculated and normalized to cell capacitance. 2-way ANOVA reports a significant developmental stage effect (P<0.0001) but no significant genotype effect (P=0.880) nor a significant interaction between the two (P=0.154). Grin2a+/+: n=13 (DIV7-8), 13 (DIV15-16) cells, 4 animals; Grin2a+/–: n=13 (DIV7-8), 11 (DIV15-16) cells, 5 animals; Grin2a–/–: n=17 (DIV7-8), 16 (DIV15-16) cells, 5 animals. (E) At DIV15-16 the percentage inhibition of NMDA (150 μM) evoked currents by ifenprodil (3 μM) was significantly greater (Tukey’s test) in Grin2a–/– neurons compared to Grin2a+/+ neurons (P<0.0001) and Grin2a+/– neurons (P=0.0038).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 112 GRIN2A GENOTYPEPHENOTYPE CORRELATION 113

DIV 7-8). Nevertheless, the percentage of total NMDAR currents sensitive to ifenprodil block was increased in Grin2a–/– neurons as would be expected for neurons where GluN2A expression is absent (Figure 6E). Thus, within this experimental system there appears to be no evidence for increases in GluN2B expression to compensate for loss of GluN2A expression due to Grin2a allelic deletion.

DISCUSSION

We present a comprehensive investigation of GRIN2A-related phenotypes, comprising 248 aﬀected individuals with pathogenic or likely pathogenic variants in GRIN2A.

Variant distribution We observed a clustering of disease-causing missense variants in the highly conserved LBD S1 and S2 as well as TMD and linker domains, which is similar to our previous observations in GRIN1 and GRIN2B19,26 and may assist in predicting pathogenicity of variants of uncertain signiﬁcance by its location (Figure 1). No missense variants in the intracellular C-terminal domain of GluN2A (beyond amino acid position 838) have been found to fulﬁl ACMG criteria for being pathogenic or likely pathogenic.

Previous reports of alleged disease-associated C-terminal variants may therefore be revised, as this region is also the only region in GRIN2A that shows no evidence for regional depletion as the number of observed variants in ExAC was not higher than expected by a mutational model7,27, similar to GluN1 and GluN2B19,26. As the C-terminus of GluN2A is tolerant to genetic variation in the general population we conclude that most missense variants in the C-terminus likely have no eﬀects.

Phenotypic range Our comprehensive analysis shows that the GRIN2A-related phenotypic spectrum does not only comprise well-established epilepsy-aphasia disorders, but is much broader and ranges from normal or near-normal development to non-speciﬁc DEE. Notably, only three individuals had an apparently normal phenotype with no ID, no epilepsy and no speech disorder (two of them also had EEG investigation, both with normal result), all were relatives from more severely aﬀected individuals. Moreover, 5 individuals with (likely) pathogenic variants are listed in gnomAD and can therefore be considered normal as well, even though very minor phenotypic abnormalities cannot be excluded. Thus, reduced penetrance appears to be possible but not a very common phenomenon among carriers of pathogenic or likely pathogenic GRIN2A variants. Epilepsy and speech disorders, each seen in >80 % of individuals, occur independent of intellectual disability, which is present in 62.4% of individuals and was mild in nearly half of those cases. This is in stark contrast to phenotypes related to GRIN1, GRIN2B and GRIN2D that are associated with marked ID in nearly 100 % of cases.19,26,28 Among all currently known GRIN-associated phenotypes, GRIN2A-

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 113 114 CHAPTER 5

related disorders display the most recognizable epilepsy spectrum, comprising focal or multifocal epilepsy with or without CTS as well as CSWS (Figure 2). As normal and near-normal development are part of the phenotypic range, it can be assumed that individuals with milder phenotypes are more likely to pass on their pathogenic variants, which may explain why 60% of variants of known origin are inherited and do not exclusively occur de novo as is the rule for disorders related to GRIN1, GRIN2B and GRIN2D. 19,26,28

Genotype-phenotype correlation In contrast to previous studies29, our systematic analyses of phenotype and molecular data of a large cohort of individuals with GRIN2A variants identiﬁed two distinct phenotype groups

corresponding to the location of variants in diﬀerent protein domains (Figure 4). MisTMD+Linker

are associated with severe DEE phenotypes, whereas misATD+LBD are associated with speech abnormalities and/or seizures with mild to no ID only. Strikingly, both phenotypic groups are signiﬁcantly correlated with opposing electrophysiological consequences of the NMDAR, even though the complex functional alterations caused by a GRIN2A variant cannot always easily be

reduced to a binary description such as loss- or gain-of-function. It appears plausible that misLBD

may impede agonist binding and thus reduce channel activity, whereas a misTMD+Linker may affect formation of the ion channel pore mediating a gain-of-function effect by e.g. disrupted channel inhibition by Mg2+.8,11-13 However, more electrophysiology data of mutated NMDAR is needed to clarify the exact pathomechanisms of variants in the different protein domains.

Pathomechanistic model

We observed that individuals with extracellular misATD+LBD (displaying exclusively loss-of-function eﬀects) have a comparable phenotypic range to individuals with null variants that is substantially

less severe than the phenotypes of individuals with membrane-associated misTMD+Linker (displaying

predominantly gain-of-function effects). We therefore hypothesize that loss-of-function misATD+LBD and null variants mediate similar pathomechanistic effects. In agreement with our phenotype data but in contrast to previous hypotheses, we observed that Grin2a-/+ and Grin2a-/- cultured rat neurons show lower current density indicating that any compensatory increase in expression of other GluN subunits is not sufficient to match the current normally mediated by GluN2A- containing NMDAR in rats. Furthermore, application of the GluN2B-specific blocker ifenprodil to these neurons did not give any evidence for compensatory increase of GluN2B in NMDAR assembly in GluN2A-deficient cells. Our data thus contradict the hypothesis of a compensatory gain-of-function effect due to GluN2A haploinsufficiency and in fact suggest loss-of-function, in agreement with our phenotype-based observations. Namely, GRIN2A null variants are associated with comparable clinical consequences as misATD+LBD (resulting in loss-of-function) and with markedly less severe clinical consequences than misTMD+Linker (resulting in predominantly gain-of-function). With our pathomechanistic model, we predict that individuals with DEE due to misTMD+Linker are prone to having an underlying gain of NMDAR function and represent promising candidates for treatment with NMDAR blockers, such as memantine.12 However,

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 114 GRIN2A GENOTYPEPHENOTYPE CORRELATION 115

there is currently still little data available on clinical treatment of GRIN2A-related disorders with memantine.12 Conversely, individuals with variants leading to complete or partial loss of channel function (misATD+LBD or null variants) may potentially respond to positive allosteric modulators of the NMDAR.11,30

Our study illustrates how systematically investigating clinical phenotypes in a large cohort of individuals with a monogenic disease cannot only reveal novel genotype-phenotype correlations, but also contribute to a better understanding of the underlying functional mechanisms being a prerequisite for the development of precision medicine approaches.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 115 116 CHAPTER 5

REFERENCES

1. Traynelis SF, Wollmuth LP, McBain CJ, et al. Glutamate 16. Richards S, Aziz N, Bale S, et al. Standards and guidelines receptor ion channels: structure, regulation, and for the interpretation of sequence variants: a joint function. Pharmacol Rev 2010; 62: 405-496. consensus recommendation of the American College of 2. Paoletti P, Bellone C, Zhou Q. NMDA receptor subunit Medical Genetics and Genomics and the Association for diversity: impact on receptor properties, synaptic Molecular Pathology. Genet Med 2015; 17: 405-424. plasticity and disease. Nat Rev Neurosci 2013; 14: 383- 17. Nykamp K, Anderson M, Powers M, et al. Sherloc: a 400. comprehensive refinement of the ACMG-AMP variant 3. Bar-Shira O, Maor R, Chechik G. Gene Expression classification criteria. Genet Med 2017; 19:1105-1117. Switching of Receptor Subunits in Human Brain 18. Baxter PS, Martel MA, McMahon A, Kind PC, Hardingham Development. PLoS Comput Biol. 2015; 11: e1004559. GE. Pituitary adenylatecyclase-activating peptide 4. Lemke JR, Lal D, Reinthaler EM, et al. Mutations in GRIN2A induces long-lasting neuroprotection through the cause idiopathic focal epilepsy with rolandic spikes. Nat induction of activity-dependent signaling via the cyclic Genet 2013; 45: 1067-1072. AMP response element-binding protein-regulated 5. Lesca G, Rudolf G, Bruneau N, et al. GRIN2A mutations transcription co-activator 1. J Neurochem 2011; 118: 365- in acquired epileptic aphasia and related childhood 378. focal epilepsies and encephalopathies with speech and 19. Platzer K, Yuan H, Schutz H, et al. GRIN2B encephalopathy: language dysfunction. Nat Genet 2013; 45: 1061-1066. 686 novel findings on phenotype, variant clustering, 6. Carvill GL, Regan BM, Yendle SC, et al. GRIN2A mutations functional consequences and treatment aspects. J Med cause epilepsy-aphasia spectrum disorders. Nat Genet Genet 2017; 54: 460-470. 2013; 45: 1073-1076. 20. Fry AE, Fawcett KA, Zelnik N, et al. De novo mutations 7. Lek M, Karczewski KJ, Minikel EV, et al. Analysis of protein- in GRIN1 cause extensive bilateral polymicrogyria. Brain. coding genetic variation in 60,706 humans. Nature 2018 2016;536: 285-291. 21. Jiang H, Jiang W, Zou J, et al. The GluN2B subunit of 8. Swanger SA, Chen W, Wells G, et al. Mechanistic Insight N-methy-D-asparate receptor regulates the radial into NMDA Receptor Dysregulation by Rare Variants in migration of cortical neurons in vivo. Brain Res 2015; the GluN2A and GluN2B Agonist Binding Domains. Am J 1610: 20-32. Hum Genet 2016; 99: 1261-1280. 22. Serraz B, Grand T, Paoletti P. Altered zinc sensitivity of 9. Endele S, Rosenberger G, Geider K, et al. Mutations in NMDA receptors harboring clinically-relevant mutations. GRIN2A and GRIN2B encoding regulatory subunits of Neuropharmacology 2016;109: 196-204 NMDA receptors cause variable neurodevelopmental 23. Ogden KK, Chen W, Swanger SA, et al. Molecular phenotypes. Nat Genet 2010; 42: 1021-1026. Mechanism of Disease-Associated Mutations in the 10. Sibarov DA, Bruneau N, Antonov SM, Szepetowski P, Pre-M1 Helix of NMDA Receptors and Potential Rescue Burnashev N, Giniatullin R. Functional Properties of Pharmacology. PLoS Genet 2017; 13: e1006536. Human NMDA Receptors Associated with Epilepsy- 24. XiangWei W, Jiang Y, Yuan H. De Novo Mutations and Related Mutations of GluN2A Subunit. Front Cell Rare Variants Occurring in NMDA Receptors. Curr Opin Neurosci 2017; 11: 155. Physiol 2018; 2: 27-35. 11. Addis L, Virdee JK, Vidler LR, Collier DA, Pal DK, Ursu D. 25. Edman S, McKay S, Macdonald LJ, et al. TCN 201 Epilepsy-associated GRIN2A mutations reduce NMDA selectively blocks GluN2A-containing NMDARs in a receptor trafficking and agonist potency - molecular GluN1 co-agonist dependent but non-competitive profiling and functional rescue. Sci Rep 2017; 7: 66. manner. Neuropharmacology 2012; 63: 441-449. 12. Pierson TM, Yuan H, Marsh ED, et al. GRIN2A mutation 26. Lemke JR, Geider K, Helbig KL, et al. Delineating the and early-onset epileptic encephalopathy: personalized GRIN1 phenotypic spectrum: A distinct genetic NMDA therapy with memantine. Ann Clin Transl Neurol 2014; 1: receptor encephalopathy. Neurology 2016; 86: 2171- 190-198. 2178. 13. Chen W, Tankovic A, Burger PB, Kusumoto H, Traynelis 27. Samocha KE. Regional missense constraint improves SF, Yuan H. Functional Evaluation of a De Novo GRIN2A variant deleteriousness prediction. bioRxiv. 2017. doi: Mutation Identified in a Patient with Profound Global https://doi.org/ 10.1101/148353 Developmental Delay and Refractory Epilepsy. Mol 28. Li D, Yuan H, Ortiz-Gonzalez XR, et al. GRIN2D Recurrent Pharmacol 2017; 91: 317-330. De Novo Dominant Mutation Causes a Severe Epileptic 14. Balu DT, Coyle JT. Glutamate receptor composition of Encephalopathy Treatable with NMDA Receptor Channel the post-synaptic density is altered in genetic mouse Blockers. Am J Hum Genet 2016; 99: 802-816. models of NMDA receptor hypo- and hyperfunction. 29. Myers KA, Scheffer IE. GRIN2A-Related Speech Disorders Brain Res 2011; 1392: 1-7. and Epilepsy. In: Adam MP, Ardinger HH, Pagon RA, 15. Wyllie DJ, Livesey MR, Hardingham GE. Influence of Wallace SE, Bean LJH, Stephens K, et al., editors. GluN2 subunit identity on NMDA receptor function. GeneReviews (R). Seattle (WA) 1993. Neuropharmacology 2013; 74: 4-17. 30. Zhu S, Paoletti P. Allosteric modulators of NMDA receptors: multiple sites and mechanisms. Curr Opin Pharmacol 2015; 20: 14-23.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 116 GRIN2A GENOTYPEPHENOTYPE CORRELATION 117

SUPPLEMENTAL DATA

Supplemental Table 1: Phenotype questionnaire

Please use the drop down menu

Gene c. Mutation p. Mutation (protein) Origin Genetic testing done Year of Birth Sex Year of Inclusion Birth measures (OFC!) Current body measures (OFC!) Family history Seizures Seizure onset (year) Seizure type at onset Seizure type at onset (please elaborate) Further seizure types Further seizure frequency AED used current AED AED response Seizure free? Seizure outcome (please elaborate) EEG at onset EEG at follow up Last EEG Photosensitivity Developmental delay Intellectual disability Hypotonia Movement disorders (e.g. Dystonia, chorea) Cerebral visual impairment Other neurological features Main speech phenotype Additional speech phenotype Psychiatric phenotype (schizophrenia?) Brain MRI Additional features (e.g. congenital malformations? dysmorphic?)

Supplemental Table 2: Patient table (available on request)

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 117 118 CHAPTER 5

Supplemental Table 3: Functional characterization

Membrane DNA Protein Domain Glutamate Glycine expression Magnesium Proton Zinc c.236C>G p.( Pro79Arg) ATD decreased potency decreased potency reduced NA NA NA c.551T>G p.(Ile184Ser) ATD NA NA reduced NA NA NA c.551T>G p.(Ile184Ser) ATD no effect no effect no effect NA no effect no effect c.692G>A p.(Cys231Tyr) ATD decreased potency decreased potency reduced NA NA NA c.1306T>C p.(Cys436Arg) S1 no response no response no expression NA NA NA c.1306T>C p.(Cys436Arg) S1 increased potency decreased potency reduced NA NA NA c.1447G>A p.(Gly483Arg) S1 decreased potency decreased potency reduced NA NA NA c.1447G>A p.(Gly483Arg) S1 decreased potency no effect reduced NA NA NA c.1510C>T p.(Arg504Trp) S1 no effect no effect NA NA NA NA c.1553G>A p.(Arg518His) S1 no response no response reduced NA NA NA c.1592C>T p.(Thr531Met) S1 no response no response reduced NA NA NA c.1642G>A p.(Ala548Thr) Linker decreased potency decreased potency no effect NA NA NA c.1655C>G p.(Pro552Arg) Linker increased potency increased potency NA NA NA NA c.1845C>A p.(Asn615Lys) M2 no effect no effect NA eliminate block NA NA c.1954T>G p.(Phe652Val) M3 NA NA NA NA NA NA c.2054T>C p.(Val685Gly) S2 decreased potency no effect reduced NA NA NA c.2081T>C p.(Ile694Thr) S2 decreased potency no effect reduced NA NA NA c.2095C>T p.(Pro699Ser) S2 decreased potency no effect reduced NA NA NA reduced (not c.2113A>G p.(Met705Val) S2 decreased potency decreased potency significant) NA NA NA c.2113A>G p.(Met705Val) S2 decreased potency no effect reduced NA NA NA c.2146G>A p.(Ala716Thr) S2 decreased potency no effect reduced NA NA NA c.2179G>A p.(p.Ala727Thr) S2 decreased potency no effect reduced NA NA NA c.2191G>A p.(Asp731Asn) S2 no response no response reduced NA NA NA c.2191G>A p.(Asp731Asn) S2 decreased potency no effect reduced NA NA NA c.2200G>C p.(Val734Leu) S2 decreased potency no effect no effect NA NA NA c.2314A>G p.(Lys772Glu) S2 decreased potency no effect reduced NA NA NA reduced reduced c.2434C>A p.(Leu812Met) Linker increased potency increased potency NA increased block sensitivity sensitivity reduced reduced reduced c.2450T>C p.(Met817Thr) M4 increased potency increased potency NA sensitivity sensitivity sensitivity

Abbreviations: ATD = amino-terminal domain, GoF = Gain of Function, LoF = Loss of Function, NA = not available.

Supplemental Table 4: ID/DD score (available on request)

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 118 GRIN2A GENOTYPEPHENOTYPE CORRELATION 119

Deactivation Rate Amplitude, Peak Calcium Result Reference NA NA NA LoF Addis L, et al. Sci Rep 2017 increased decreased NA LoF Sibarov DA, et al. Front Cell Neurosci 2017 no effect NA NA NO Serraz B, et al. Neuropharmacology 2016 NA decreased NA LoF Addis L, et al. Sci Rep 2017 NA NA NA LoF Addis L, et al. Sci Rep 2017 NA decreased NA LoF Swanger SA, et al. Am J Hum Genet 2016. NA NA NA LoF Addis L, et al. Sci Rep 2017 increased decreased NA LoF Swanger SA, et al. Am J Hum Genet 2016. increased decreased (not significant) NA LoF Swanger SA, et al. Am J Hum Genet 2016. NA decreased NA LoF Swanger SA, et al. Am J Hum Genet 2016. increased no current NA LoF Swanger SA, et al. Am J Hum Genet 2016. NA reduced NA LoF Ogden KK, et al. PLoS Genet 2017 increased decreased NA GoF Ogden KK, et al. PLoS Genet 2017 NA NA decreased permeability GoF Endele S, et al. Nat Genet 2010 decreased NA NA GoF Lesca G, et al. Nat Genet 2013 decreased (not significant) decreased NA LoF Swanger SA, et al. Am J Hum Genet 2016. no effect decreased NA LoF Swanger SA, et al. Am J Hum Genet 2016. NA decreased (not significant) NA LoF Swanger SA, et al. Am J Hum Genet 2016. NA NA NA LoF Addis L, et al. Sci Rep 2017 increased (not significant) decreased (not significant) NA LoF Swanger SA, et al. Am J Hum Genet 2016. decreased (not significant) decreased (not significant) NA LoF Swanger SA, et al. Am J Hum Genet 2016. no effect decreased(not significant) NA LoF Swanger SA, et al. Am J Hum Genet 2016. NA NA NA LoF Addis L, et al. Sci Rep 2017 NA decreased decreased permeability LoF Swanger SA, et al. Am J Hum Genet 2016. decreased decreased (not significant) NA LoF Swanger SA, et al. Am J Hum Genet 2016. no effect decreased NA LoF Swanger SA, et al. Am J Hum Genet 2016. NA NA no effect GoF Yuan H, et al. Ann Clin Transl Neurol 2014

NA NA NA GoF Chen W, et al. Mol Pharmacol 2017

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 119 120 CHAPTER 5

Supplemental Table 5: Protein domains

Start End Domain 1 22 Signal peptide 23 404 ATD 405 539 S1 ABD 540 555 Linker 556 576 M1 (TMD) 598 622 M2 (TMD) 634 654 M3 (TMD) 655 660 Linker 661 801 S2 ABD 802 816 Linker 817 837 M4 (TMD) 838 1464 CTD

Abbreviations: ABD = agonist-binding domain, ATD = amino-terminal domain, TMD = transmembrane domain, CTD = C-terminal domain

A Hypotonia B Brain MRI n = 139 n = 85

13.0% 14.1% 2.9% none 1.4% mild 10.6% normal 11.5% moderate atrophy

71.2% severe other unspecifie 75.3%

C Movement disorder n = 72

1.4% 11.1% none 13.9% ataxia dystonic/choreatic unspecifie 73.6%

Supplemental Figure 1: Distribution of phenotypes. Individuals with GRIN2A-related disorders display a broad range of phenotype severity and expressivity with respect to (A) hypotonia, (B) movement disorders and (C) MRI ﬁndings.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 120 GRIN2A GENOTYPEPHENOTYPE CORRELATION 121

Supplemental Figure 2: Variance of ID/DD phenotype in individuals with the same genetic variant.

The mean variance of ID/DD phenotypes per variant was 0.65 (±0.64 SD). Permuting family labels 10,000 times revealed a lower value than the mean variance in 15 of 10,000 permutations (empirical p-value 0.0016).

Supplemental Figure 3: Conﬁrmation of GluN2A expression at DIV15-16.

Evidence for the developmental increase in the expression of GluN2A by DIV15 in Grin2a+/+ neurons. (A) Western blot indicating the increase in GluN2A protein at DIV15 compared to DIV7. (B) Grin2a mRNA levels are significantly increased (P<0.0001, t-test) at DIV15 (n=8) compared to DIV7 (n=8). Similarly GluN2A protein is significantly increased (P<0.0118, t-test) at DIV15 (n=6) compared to DIV7 (n=6). (C) At DIV15-16 there is significantly less potentiation by spermine (100 μM) of NMDA (150 μM) evoked currents as would be expected for neurons where GluN2A expression was present (P=0.011, t-test). DIV7-8: n=25 cells; DIV15-16: n=25 cells, 8 animals.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 121 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 122 123123

Chapter 6

Schizophrenia is a later-onset feature of PCDH19 Girls Clustering Epilepsy

Revised version submitted

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 123 124 CHAPTER 6

Danique R.M. Vlaskamp1,2, Anne S. Bassett3,4,5, Joseph E. Sullivan6, Jennifer Robblee7, Lynette G. Sadleir8, Ingrid E. Scheﬀer1,9,10*, Danielle M. Andrade7,11*

1 Epilepsy Research Center, Department of Medicine, the University of Melbourne, Austin Health, Melbourne, Australia. 2 University of Groningen, University Medical Center Groningen, Department of Neurology and Department of Genetics, Groningen, the Netherlands. 3 Clinical Genetics Research Program, Center for Addiction and Mental Health, and Campbell Family Mental Health Research Institute, Toronto, ON, Canada. 4 Department of Psychiatry, University of Toronto, Toronto, ON, Canada. 5 Dalglish Family 22q Clinic for Adults with 22q11.2 Deletion Syndrome and Toronto General Research Institute, University Health Network, Toronto, ON, Canada. 6 University of California San Francisco, Pediatric Epilepsy Center, Benioﬀ Children’s Hospital, San Francisco, CA, Unites States. 7 Division of Neurology, Toronto Western Hospital, University of Toronto, Toronto, ON, Canada. 8 Department of Paediatrics and Child Health, University of Otago, Wellington, New Zealand. 9 Department of Paediatrics, The University of Melbourne, Royal Children’s Hospital, Victoria, Australia. 10 The Florey Institute of Neurosciences and Mental Health, Melbourne, Victoria, Australia. 11 Epilepsy Genetics Research Program, Krembil Neuroscience Center, University of Toronto, Toronto, ON, Canada. *These authors contributed equally

Acknowledgements. We thank the patients and their families for participating in our research. We thank research assistants at the Epilepsy Research Center for their assistance.

Disclosures. A.S. Bassett receives/has received research support from the Canadian Institutes of Health Research, National Institute of Mental Health, McLaughlin Foundation, and Canada Research Chairs, and reports no conflicts of interest. J.E. Sullivan is a consultant for Epygenix and has received speaking honorarium from Invitae. L.G. Sadleir has received funding for travel from Seqirus and Nutricia. I.E. Scheffer serves on the editorial boards of Neurology® and Epileptic Disorders; may accrue future revenue on a pending patent re: Therapeutic compound; has received speaker honoraria from Athena Diagnostics, UCB, GSK, Eisai, and Transgenomics; has received scientific advisory board honoraria from Nutricia, UCB and GSK, has received funding for travel from Athena Diagnostics, UCB, and GSK; and receives/has received research support from the NHMRC, ARC, NIH, Health Research Council of New Zealand, March of Dimes, the Weizmann Institute, CURE, US Department of Defense, and the Perpetual Charitable Trustees. D.M. Andrade received research support from the Ontario Brain Institute, Brain and Behavior Foundation, McLaughlin Foundation, Dravet.ca, Genome Canada, Toronto Western and General Foundation. D.M. Andrade also serves on the editorial board of Epilepsia and is part of the ILAE Task Force on Intellectual Disability. D.M. Andrade received speaker honoraria from UCB. The remaining authors have no conflicts of interest to disclose.

Ethical statement. We conﬁrm that we have read the Journal’s position on issues in ethical publication and aﬃrm that this report is consistent with those guidelines.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 124 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 125

ABSTRACT

Objective. To investigate the occurrence of psychosis and serious behavioral problems in females with PCDH19 mutations.

Methods. We evaluated whether psychosis and serious behavioral problems had occurred in 60 females (aged 2-75 years) with PCDH19 mutations, belonging to 35 families. Patients were identiﬁed from epilepsy genetics databases in Australia, New Zealand, US and Canada. Neurological and psychiatric disorders were diagnosed using standard methods.

Results. 8/60 (13%) females from 7 families developed a psychotic disorder: schizophrenia (6), schizoaffective disorder (1) or an unspecified psychotic disorder (1). Median age at onset of psychotic symptoms was 21 years (range 11-28 years). In our cohort of 39 females aged 11 years or above, 8 (21%) developed a psychotic disorder. Seven had ongoing seizures at onset of psychosis, with two continuing to have seizures when psychosis recurred. Psychotic disorders occurred in the setting of mild (4), moderate (2) or severe (1) intellectual disability or normal intellect (1). Pre- existing behavioral problems occurred in four females, and autism spectrum disorder in three. Two (3%) additional females had psychotic features with other conditions: an adolescent had recurrent episodes of post-ictal psychosis and a 75 year old woman had major depression with 6 psychotic features. A further three (5%) adolescents with moderate to severe intellectual disability had onset of severe behavioral disturbance, or significant worsening, in adolescence.

Significance. We identify that psychotic disorders, including schizophrenia, are a later onset manifestation of PCDH19 Girls Clustering Epilepsy. Aﬀected girls and women should be carefully monitored for later-onset psychiatric disorders.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 125 126 CHAPTER 6

INTRODUCTION

PCDH19 mutations (OMIM 300460) were initially identified in a family with Epilepsy and Mental Retardation limited to Females (EFMR, OMIM 300088).1,2 However, as about one-quarter of women do not have mental retardation,1,3 the new name of Girls Clustering Epilepsy (GCE) was suggested to aid recognition of this distinctive disorder with an unusual inheritance pattern.4 PCDH19-GCE occurs in heterozygous females, while males are usually unaffected transmitting carriers. Rare males with mosaic mutations are affected, reflecting the female pattern of two X chromosomes and the consequent co-existence of cells with and without mutant PCDH19.2,5,6 Girls with PCDH19 mutations present with recurrent clusters of seizures triggered by fever in infancy or early childhood. Intellect can range from normal to severe intellectual disability (ID).1,2,7

While psychiatric features were noted in a few females reported in the original studies of PCDH19- GCE (called EFMR in these studies),1,8,9 behavioral and psychiatric comorbidities are common, and may pose the most disabling problem in adulthood.3 An array of psychiatric co-morbidities has been described including severe behavioral problems, obsessive features and autism spectrum disorder. In adult life, we previously reported two women with a schizophreniform psychosis1 and an additional woman with psychosis has been subsequently reported.10

We hypothesized that psychosis was a later-onset feature of PCDH19-GCE. We therefore analysed the presence of psychosis and serious behavioral problems in a cohort of females with PCDH19 mutations.

METHODS

Study cohort Patients were ascertained from the epilepsy genetics databases of the authors located in Australia and New Zealand, the United States, and Canada. These databases include all patients presenting with genetic epilepsy and their relatives who consent to research participation. The database includes information on (likely) pathogenic variants. Families A, B, E and G were previously published1,2,11,12, however, their psychiatric phenotype has been evaluated in detail for this study. We analyzed the frequency of psychosis in adolescent and adult women with a PCDH19 mutation, using the youngest age of onset of psychosis observed in our cohort to delineate the group of interest.

Phenotyping It has been suggested that PCDH19-related epilepsy should be called PCDH19-Girls Clustering Epilepsy (PCDH19-GCE), to assist clinicians in considering this diagnosis early in infant of young girl’s presentation.4,13,14 We examined whether each female had experienced psychosis and

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 126 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 127

serious behavioral problems and the evolution of these symptoms over time. We also evaluated the epilepsy phenotype, developmental course and intellect, magnetic resonance imaging (MRI), and genetic results, and the family history. Information was obtained using a validated seizure questionnaire,15 clinical interviews and review of medical records. Psychotic disorders were diagnosed using the fifth edition of the Diagnostic and Statistical Manual of Mental disorders (DSM-V) by an academic psychiatrist (ASB), after local assessment by a psychiatrist at time of the patient’s presentation.16 The level of intellectual disability was based on IQ scores obtained through formal psychometric assessment. Seizure types and epilepsy syndromes were classified according to the 2017 International League Against Epilepsy Classification of the Epilepsies.17,18

Genotyping In the research genetic epilepsy database, only (likely) pathogenic variants (mutations) were included, based on the criteria of the American College of Medical Genetics.19 We present the data on (likely) pathogenic varians identiﬁed in females with a psychotic disorders in Supplemental Table 1. PCDH19 mutations or deletions were reported in accordance with its longest isoform 1 (NM_001184880.1).

Ethical statement Patients, or their parents or legal guardians in the setting of minors or those with intellectual 6 disability, gave written informed consent for inclusion in our study. This study was approved by the local institutional Human Research Ethics Committee (Austin Health reference H2007/02961), the Health and Disability Ethics Committee (New Zealand) and the Research Ethics Board (Toronto Western Hospital 15-9512).

RESULTS

We studied 60 females with PCDH19 mutations, from 35 families. Females were aged between 2 and 75 years (median age 18 years).

PCDH19 Girls Clustering Epilepsy with FKURQLFSV\FKRWLFGLVRUGHUV Q Eight of our cohort of 60 females developed a psychotic disorder. Median age at study of these eight females was 31 years (range 21-64 years). Psychosis began in adolescence and adult life at a median age at onset of 21 years (range 11-28 years). When we analysed our cohort aged 11 years (youngest age at onset of psychosis in our cohort) or older, we found that 8/39 (21%) females were aﬀected with a psychotic illness.

The eight females came from seven families (Figure 1). Six (15%) had schizophrenia, of whom two were sisters (family D), one (3%) had a schizoaﬀective disorder and one (3%) an ‘unspeciﬁed schizophrenia spectrum and other psychotic disorder’ (Table 1). One female had childhood onset

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 127 128 CHAPTER 6

schizophrenia from age 11 years. Two had a co-existing psychiatric diagnosis; one had post- traumatic stress disorder (after a sexual assault) and another had a major depressive disorder.

In 7 women, the initial episode of psychosis markedly varied in duration between 1-2 weeks and 5 years. The remaining woman (E:II:4) had frequent ongoing hallucinations from 11 to 27 years. Six women had recurrent psychotic episodes. Psychotic symptoms included disorganised motor behavior (8) including catatonia in 1, hallucinations (7), delusions (6), negative symptoms (5) and disorganised thinking (3). Antipsychotic medication was commenced in all eight females and was eﬀective in all but one (E:II:4). Two females also had a positive response to a selective serotonin- reuptake inhibitor (SSRI, sertraline) or to a selective serotonin-norepinephrine reuptake inhibitor (SNRI, duloxetine).

Three women had autism spectrum disorder (ASD) diagnosed at age 3.5 - 12 years, 7 to 12 years prior to their presentation with psychosis; ﬁve women did not have a diagnosis of ASD. Behavioral problems in childhood or adolescence were already present before the onset of psychotic symptoms in four females, and included aggression and anxiety (3), and behavioral problems with suicidal gestures (1). Only one female (C:II:1) was on psychotropic medication, an anti-depressant, and this was discontinued when she discovered she was pregnant, prior to onset of psychosis later in pregnancy.

The psychotic disorders occurred with co-morbid ID in 7/8 females which was mild in four, moderate in two and severe in one. The remaining patient had borderline intellect. A decline in scholastic abilities preceded psychotic symptoms in one adolescent (D:IV:2) aged 18 years with mild ID. A 19 year old adolescent (G:IV:4) with moderate ID had cognitive regression associated with romantic obsessions; following an episode of status epilepticus, she developed erotomanic delusions. In four patients, social withdrawal was observed with the initial or recurrent episode of psychosis.

All eight females had a history of seizures. In one woman, seizures had resolved 15 years prior to psychosis onset. The remaining seven had ongoing seizures and were on anti-epileptic medication at the onset of psychosis. Seizures had resolved in three women one to four years before their psychosis recurred. For the 27 year old woman (E:II:4) with childhood onset schizophrenia beginning at 11 years, seizures settled at age 15 years, although hallucinations persisted throughout adult life. The remaining three females had ongoing seizures including bilateral tonic-clonic seizures (3), focal impaired awareness seizures (2), tonic seizures (2), absences (1) and febrile seizures (1) with seizure frequencies varying from every few weeks to breakthrough seizures only associated with medication change. Neuroimaging performed in 7/8 patients (CT–2, MRI–5) was normal.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 128 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 129 ŵƵƚĂƚŝŽŶ 'ŝƌůƐ'ůƵƐƚĞƌŝŶŐ ƉŝůĞƉƐǇ W,ϭϵ 'ŝƌůƐ'ůƵƐƚĞƌŝŶŐ ƉŝůĞƉƐǇ W,ϭϵ W,ϭϵ W,ϭϵ GCE had epilepsyGCE severe W,ϭϵ͚DŝůĚ͛ ͚^ĞǀĞƌĞ͛ &ĞďƌŝůĞ ƐĞŝǌƵƌĞƐ hŶĐůĂƐƐŝĨŝĞĚ ƐĞŝǌƵƌĞƐ WƐǇĐŚŽƚŝĐ ĚŝƐŽƌĚĞƌ ƵƚŝƐƚŝĐ ĨĞĂƚƵƌĞƐ ^ĞǀĞƌĞďĞŚĂǀŝŽƵƌĂů ƉƌŽďůĞŵƐ ĞƉƌĞƐƐŝŽŶ ĞƉƌĞƐƐŝŽŶ ǁŝƚŚ ƉƐǇĐŚŽƚŝĐ ĨĞĂƚƵƌĞƐ DƵƚĂƚŝŽŶ ŝŶ EŽŵƵƚĂƚŝŽŶ ŝŶ DŽƐĂŝĐ ĐĂƌƌŝĞƌ

6 GCE hadGCE mild epilepsy and intellectual disability and those with a severe PCDH19- PCDH19- and moderate severe to intellectual disability. Pedigrees of 8 females with PCDH19 Girls Clustering epilepsy and psychotic disorders in 7 families. in 7 families. with PCDH19 Girls disorders Clustering epilepsy and psychotic of 8 females 1: Pedigrees Figure Females with a psychotic disorder are marked by a red box. Girls with a mild

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 129 130 CHAPTER 6

Table 1: Schizophrenia and other psychotic disorders in 8 females with PCDH19 mutations

A:III:3 1 B:III:4 1 C:II:1 Age at study (y) 25 64 35 Mutation, inheritance p.Val441Glu, pat p.Gln85*, UK p.Lys708Argfs*9, UK Age at seizure onset: 12 m: GTCS 12 m: FS, GTCS 9 m: FMS, FIAS, CSE seizure types Development Always delayed - Delay since 9 m Regression (age) - - Yes (9 m)

Intellectual disability Mild No, but borderline intellect Mild (IQ) (68) (77; VIQ 85, PIQ 72) Psychiatric and behavioral - UK Depression, anxiety, aggression, self-injury, problems before psychosis running away (21y) onset (age at onset)

Age at psychosis onset 23 y 21 y 28 y 10 m Ongoing seizures at psychosis None for 15 y: Yes, 2 per year: Yes: onset: seizure treatment no AED AED CBM, LEV

Duration of initial episode Few months 12 months admitted UK - Delusions Persecutory Religious and grandiose UK - Hallucinations Auditory Auditory UK - Disorganized thinking Poor cognitive insight Irritational, poor cognitive Threatened to cut out fetus insight

- Grossly disorganized or motor behavior Sentenced for aggression UK abnormal motor behavior to her child

- Negative symptoms Deterioration self-care, -Dysthymia social withdrawal, alogia, flat affect - Other symptoms related to Suicidal ideation, anxiety, Grossly disorganized affect UK psychotic symptoms agoraphobia, panic

Treatment commenced after TFP, VLF, CLP, CBM (non- TDZ (+) RSP (non-compliant), RSP consta (+), OLZ psychosis onset compliant), admitted, TFP (+, but brieﬂy) (eﬀect) 3 depot (+/-), SRT (+)

Recurrent psychotic episodes 24 y 7 m: Auditory 25 y: Religious delusions, - 30 y 4 m: Auditory hallucinations, hallucinations with suicidal delusions of reference, aggression, social withdrawal, anhedonia ideation hypomania and with suicidal attempts, less sleep, anxiety, aggression. feelings of guilt - 33 y 10 m: delusions (bizarre, persecutory, somatic, of reference) poor cognitive insight, incoherent speech

Seizures at last recurrence None for 16 y None for >1 year None for >3 years

Treatment commenced after RSP (galactorrea), CLP (+) RSP (+), DLX (+), QTP (-), PLP (UK), HLP psychosis recurrences (eﬀect) 3 admitted, OLZ (+) with benztropine (+)

Psychotic diagnosis Schizophrenia Schizoaffective disorder Schizophrenia

Psychiatric outcome Multiple episodes, Multiple episodes Multiple episodes, currently in remission currently in remission with with progressive with treatment. treatment. Ongoing PTSD worsening, currently in with panic attacks after remission on treatment sexual assault (occasional outbursts of aggressiveness) 1 Patient has previously been published by Scheffer et al. 2008. 2 Mother is mosaic for PCDH19 mutation. 3 Treatment effect: - ineffective, + effective. Underlined treatment is currently used. Bold and italic treatment concerns an anti-psychotic drug. Mutations were reported according to transcript NM_001184880.1. Abbreviations: Abs = absences, AED = anti-epileptic drugs, AH = auditory hallucinations, APM = alprazolam, APZ = aripriprazole, ASD = autism spectrum disorder, At = atonic seizures, AtAbs = atypical absences, CBM = carbamazepine, CLP = chlorpromazine, CLZ = clonazepam, CSE = convulsive status epilepticus, DLX = duloxetine, FBTS = focal evolving to bilateral tonic seizures, FMS = focal motor seizures, FIAS = focal impaired awareness seizures, FS = febrile seizures, FSIQ = full-scale intellectual

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 130 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 131

D:IV:1 D:IV:2 E:II:4 F:II:1 G:IV:4 29 21 27 23 27 Exon 6 deletion, mat Exon 6 deletion, mat p.Leu25Pro, mat 2 p.Ser845Asn, not mat p.Arg958Gln, pat 14 m: FBTCS 19 m: T, FIAS, FBTCS 7 m: FS, GTCS, At, T, 8 m: TCS 7 y: Abs, T, GTCS, SE, FIAS AtAbs, FIAS Delay since 14 m Delay since 19 m Yes - Delay noted at 3 y Language (14 m), delay since Yes (19 m ) Yes (7m and language - Language (7 y) at 2 y) Mild Mild Severe Moderate Moderate (62; VIQ 61, PIQ 72) (20 single words) (48-49) ASD (12y) Aggression and Aggression, self-injury, Challenging behaviors - anxiety (<6y), ASD anxiety (early age), and suicidal gestures (6y) ASD (3.5 y), running (15y) away (5 y) 20 y 4 m 18 y 2 m 11 y 19 y 19 - 20 y Yes, 2-4 clusters yearly: CLB, TPM, PHE Yes, single GTC last Yes, few yearly until 15 Yes, breakthrough Yes, every few weeks: year: AED y: AED seizures with AED TPM, CLB, LMT, VPA change: AED 1-2 weeks >1.5 years Continuous UK 5 years, following a SE - Making up stories - - Erotomanic Auditory - Visual Auditory Auditory - - 14 y: clear thought -Yes disorder for 2 weeks while on RSP - Catatonia (stands still Behavioral -Aggression for hours, excitable) deterioration with severe aggression - Avolition, social -- - withdrawal, alogia - Decline in - - Psychotic event 6 functioning and preceded by romantic scholastic abilities obsession with cognitive regression - RSP (+), OLZ (-), QTP (-), RSP (-). UK (+) HLP ( sz.), PHE (-), LEV (+) Currently on: OLZ, QTP (+), PRZ (UK) CLZ, FVX (ongoing visual hallucinations) - 21.5 y: Persecutory and religious - NA - 23 y: Two relapses Frequent relapses delusions, auditory and visual with hallucinations when trying to hallucinations, ﬁre-setting behavior, and behavioral reduce QTP social withdrawal, deterioration deterioration self-care and alogia (mimicking a - 23 y: Disorientation, depression) for 1.5 years following a for which she was cluster of seizures admitted a month - 26-28 y: twice psychosis deterioration when trying to switch OLZ into APZ None for >4 years NA NA Yes, breakthrough Yes, every few weeks seizures CLZ (+), RSP (+/-), VPA (-), OLZ (+, NA NA LMP (-), ZPM (-). - weight), LEV (UK) APZ (-), CLP (+), Currently stable on HLP (+) VPA, LEV, HLP with procyclidine, APM Schizophrenia Schizophrenia Schizophrenia Unspecified Schizophrenia with catatonia Multiple episodes, currently in Single long episode, Continuous Multiple episodes, Multiple episodes, remission with treatment. Ongoing currently in remission hallucinations currently in partial currently in remission major depressive disorder. with treatment. remission; persistent with antipsychotic auditory hallucinations; treatment sedated

quotient, FPT = flupenthixol, FVX = fluvoxamine, GTCS = generalized tonic clonic seizures, HLP = haloperidol, ID = intellectual disability, IQ = intellectual quotient, LEV = levetiracetam, m = month, mat = maternally inherited, LMP = levomepromazine, LMT = lamotrigine, NA = not applicable, NVLD = non-verbal learning disability, OLZ = olanzapine, pat = paternally inherited, PHE = phenytoin, PIQ = performal intellectual quotient, PLP = paliperidone, PRZ = prazosin, PTSD = post-traumatic stress disorder, QTP = quetiapine, SE = status epilepticus, SRT = sertraline, sz. = seizure, T = tonic seizures, TDZ = thioridazine, TFP = trifluoperazine, TPM = topiramate, UK = unknown, VLF = venlafaxine, VPA= valproate, VH = visual hallucinations, VIQ = verbal intellectual quotient, y = year, ZPM = zolpidem.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 131 132 CHAPTER 6

)HPDOHVZLWKVHFRQGDU\SV\FKRWLFV\PSWRPV Q Two women in our cohort had psychotic symptoms that did not fulﬁl a diagnosis of a ‘schizophrenia spectrum or other psychotic disorder’. A 17 year old adolescent with a de novo p.Glu201Pro PCDH19 mutation (patient H) and moderate ID developed recurrent post-ictal psychosis, comprising auditory and visual hallucinations, following clusters of seizures from the age of 14 years. She has not yet received anti-psychotic medication, because her psychotic episodes have been post-ictal and self-limiting.

The matriarch of family A (Figure), patient A:I:3, underwent abdominal surgery for a bowel obstruction at 75 years and developed a major depression. Associated with this depression, she had ongoing psychotic features including visual hallucinations of people from her past hitting her. Treatment with citalopram, topiramate and levetiracetam (the latter two for seizures) did not improve her depression or psychotic symptoms.

$GROHVFHQWVZLWKVHULRXVEHKDYLRUDOSUREOHPV Q In the remaining 50/60 females without a psychotic disorder or psychotic symptoms, severe behavioral problems were present in 15/50 (30%). For three girls with moderate to severe ID, behavioral problems were extremely severe and disruptive to the families’ life in adolescence and adult life, rendering them housebound; and resulting in police visits to ensure safety due to the severity of the patient’s aggression (Table 2). Behavioral problems began in childhood, but became much more severe at age 10-12 years. Anti-psychotic treatment was eﬀective in one, ineﬀective in another, and whether anti-psychotic treatment was prescribed for the third was not known. It was not possible to classify these behavioral problems as a schizophrenia spectrum or other psychotic disorder due to the lack of obvious signs of hallucinations or delusions in these adolescents with severe intellectual disability. Another girl (A:III:13) with severe ID developed extremely severe violent behavior twice following a midazolam intravenous treatment for anaesthesia (<14 years) and for seizures (14 years).

Genotype-phenotype correlation No genotype-phenotype correlation for the occurrence of PCDH19-related psychotic and serious behavioral problems could be identiﬁed. The mutations of the eight females with the psychotic disorders occurred throughout the gene, and did not diﬀer in location from those without psychotic disorders. The pedigrees show striking inter- and intrafamilial phenotypic heterogeneity of features including a spectrum of severity for epilepsy, intellectual disability, behavioral and psychiatric problems including psychotic disorders (Figure 1).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 132 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 133

Table 2: Grossly disorganised behavior in 3 females with PCDH19 mutations

Patient B:V:20 Patient E:II:5 Patient I Age at study (y) 17 15 13 Mutation, inheritance p.Gln85*, paternal p.Leu25Pro, mother p.Asp412Gly, maternal 1 mosaic Development Delay since 11 m Delay since 8 m Always delayed Regression (age) Yes (11 m) Yes (8 m) Frequent regression with SE Intellectual disability; speech Severe: 20-30 single Moderate: limited Severe: single words words speech Behavioral problems before Tantrums, autistic ADHD, OCD, ASD, Aggression and anxiety deterioration features (since early age) aggression (<13 y) (since early age), ASD (age at onset) (12 y ) Age at seizure onset: seizure types 11 m: GTCS, FIAS, Abs, FS 8 m : T, FBTCS, aura, Abs, 2 w: Abs, M, FIAS, UTCS, GTCS, CSE, FS DA, spasms, SE Age at onset behavioral 10 y 11-12 y 12 y deterioration Seizures at behavioral Yes: CBM, CLB Yes, clusters every 6-8 Yes: VPA, ZNS, CLB deterioration onset: seizure weeks till 13 y: UK treatment Initial presenting symptoms Intermittent episodes Cognitive and Cognitive and of anger and difficult functioning functioning behavior characterized deterioration with deterioration from by tantrums, aggression, aggression to the point age 12 y with severe restlessness and self- of requiring police difficulties with 6 injury intervention at home on transition, violence and multiple occasions and aggression to the point suspension from school that the family cannot leave the house. Evolution of symptoms 16 y: easily frustrated, Continuous intermittent Continuous labile, inappropriate aggression and running intermittend aggression sexual touching of away herself and others Treatment commenced for HLP (+), CLN (-), SRT (-), RSP (UK). Currently on: RSP (high prolactin), behavioral deterioration (effect) 2 QTP (-) QTP (-), FVX (-),CLZ (+), APZ (-) TPM (+)

1 This girl also has a p.Ser218Leu mutation in CACNA1A that was not maternally inherited. 2 Treatment effect: - ineffective, + effective. Underlined treatment is currently used. Bold and italic treatment concerns an anti-psychotic drug. Mutations were reported according to transcript NM_001184880.1. Abbreviations: Abs = absences, ADHD = attention deficit hyperactivity syndrome, APZ = aripriprazole, ASD = autism specrum disorder, CBM = carbamazepine, CLB = clobazam, CLN = clonidine, CLZ = clonazepam, CSE = convulsive status epilepticus, CT = computotomotography, DA = drop attacks, ED = epileptic discharges, FBTCS = focal evolving to bilateral tonic-clonic seizures, FIAS = focal impaired awareness seizures, FS = febrile seizures, FSIQ = full-scale intellectual quotient, FVX = fluvoxamine, GTCS = generalized tonic clonic seizures, HLP = haloperidol, ID = intellectual disability, L = left, M = myoclonic seizures, m = month, ODD = obsessive compulsive disorder, QTP = quetiapine, RSP = risperidone, SE = status epilepticus, SRT = sertraline, sz. = seizure, T = tonic seizures, TPM = topiramate, UTCS = unknown onset tonic-clonic seizures, VPA = valproate, w = weeks, y = year, ZNS = zonisamide.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 133 134 CHAPTER 6

DISCUSSION

PCDH19-GCE has a distinctive presentation in infant girls with clusters of seizures, often triggered by fever. However, the psychiatric features in childhood and adolescence of severe behavioral problems, obsessional features and ASD often prove more disabling for patients and families, even though seizures may have abated.1 Here we describe eight females with PCDH19-GCE who present with adolescent or adult onset psychotic disorder with prominent decline in functioning, of whom six satisﬁed a diagnosis of schizophrenia.

Psychotic disorders in PCDH19-GCE We studied the presence of psychotic disorders in 60 females with PCDH19-GCE. As psychosis presented from the age of 11 years, we calculated the frequency in our 39 older females aged 11 years or older. We found that 21% of our older cohort developed psychotic disorders, most often classiﬁed as schizophrenia (15% of total). This 21% ﬁgure is a minimum estimate of the frequency of psychotic disorders. Fourteen (36%) of the 39 females are still younger than 29 years of age, which was the oldest age of onset of psychosis in our cohort.

Diagnosis of psychotic disorders in patients with comorbidities can be challenging. First, psychotic episodes in patients with ongoing seizures are often considered post-ictal phenomena. However, 4/8 females developed their initial or recurrent psychotic symptoms when they had been seizure- free for at least a year, such that their psychosis could not be considered post-ictal. Guidelines for the diagnosis of post-ictal psychosis indicate that the duration of psychosis should be less than two months yet, for three of our remaining four females with psychotic illness, the psychosis lasted from 18 months to 16 years.20,21 Second, three other women had concurrent behavioral or social deterioration with psychosis, supporting a diagnosis of a ‘schizophrenia spectrum or other psychotic disorder’.

Third, psychotic symptoms such as changes in thinking and speech are more diﬃcult to recognize in patients with ID. While the women with mild ID or borderline intellect presented with a typical thought disorder, the woman with severe ID (E:II:4) had more subtle observable changes such as looking and smiling at the corners of a room. This was coupled with a major change in behavior at age 11 years, presenting with severe aggression and running away from school. Not included in the eight patients with psychotic disorders was a 19 year old non-verbal adolescent with severe ID who developed odd post-ictal behavioral changes such as laughing at the toilet bowl. This behavior was only observed on two occasions and was insuﬃcient to meet diagnostic criteria for a psychotic disorder.

Only one other female with PCDH19-GCE and psychosis has been reported10 , in addition to our original report of two cases; the latter two are included in the eight patients reported here (Table 1)1,2 The lack of additional published cases with psychotic disorders may be explained by

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 134 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 135

the young age of the majority of girls with PCDH19-GCE. Recognition of this disease is rapidly increasing with greater access to genetic testing for children with infantile-onset seizures. In addition, many adolescent and adult women with ID and seizures are not oﬀered genetic testing, despite recommendations for clinical testing.22,23 Moreover, the association of psychosis in a woman with a past history of seizures in childhood may not be considered as related to an underlying PCDH19 mutation, even where the mutation is known. Currently, there are nine PCDH19 mosaic males reported in the literature, of whom seven had behavioral problems, but they are too young (only three were aged 13 and 14 years) to determine whether they are at risk of psychosis.5,6,24,25 Longer term follow-up of women and mosaic men with PCDH19-GCE will clarify the risk of later-onset psychotic disorders.

It is possible that the severe behavioral problems seen in PCDH19-GCE could be a precursor to later-onset psychotic disorders in some patients (Table 2). In three additional adolescents with moderate to severe ID, we observed horrific and grossly disorganized behavioral problems. It is uncertain whether their behavioral regression in puberty could reflect thought disorder that is difficult to decipher given the severity of their cognitive impairment.

Frequency of psychotic disorders in PCDH19-GCE The frequency of psychotic disorders (21%), including schizophrenia (15%), in our cohort of females 6 with PCDH19 mutations was far higher than general population estimates of psychotic disorders (3.5%) and schizophrenia (1.4%) in the Finnish National Population Register.26 The frequency is also higher than that found in cohorts with ID, ASD or epilepsy, where risk is elevated. In individuals with ID, the prevalence of psychotic disorders was 4.4% in a cohort of 1023 patients27 and the prevalence of schizophrenia was 3.6% of 13,295 patients.28 For patients with ASD and normal intellect, schizophrenia spectrum disorders occurred in 6% of 713 patients.29 Turning to epilepsy, psychosis (not further classiﬁed) was diagnosed in 5.6% of patients in a meta-analysis of 56 studies including more than 40,000 participants, and was most frequent in those with temporal lobe epilepsy (7%).30

Risk for psychotic disorders in females with PCDH19-GCE appears to be more comparable to that of individuals with rare recurrent microdeletions and microduplications that are associated with schizophrenia.31 These include chromosome 22q11.2 deletions where schizophrenia and schizophrenia spectrum disorders develop in up to 25% of patients,32 and 15q13.3 deletions where 10.2% are reported to have schizophrenia.33 Other lower risk copy number variants have been reported in large scale population studies.34,35

Why only a ﬁfth of women with PCDH19-GCE develop psychotic disorders remains to be elucidated. Our patients had mutations throughout PCDH19: in exon 1 (4), 2 (1) and 5 (1) and deletions in exon 6 (2). Larger numbers of aﬀected women may enable genotype-phenotype correlations to emerge.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 135 136 CHAPTER 6

Possible shared mechanisms between psychosis and PCDH19-GCE There may be shared disease mechanisms explaining the increased risk of psychosis in PCDH19- GCE, based on observations relating to hormonal manipulation and brain connectivity. First, psychotic disorders often begin in adolescence, when hormonal levels in the hypothalamic- pituitary-adrenal axis change.36 Lower blood levels of allopregnanolone, an important neurosteroid, are found in patients with psychosis and also in females with PCDH19-GCE.37–39 Furthermore, differences in neurosteroid levels between patients with PCDH19 mutations and controls become more evident in puberty, when psychosis has its onset.39 Pilot clinical trials of allopregnanolone showed a positive effect on schizophrenia-related symptoms and cognition in patients with schizophrenia.40 A single open-label pilot study of ganaxolone, a synthetic neurosteroid, in 11 children with PCDH19-GCE showed improved seizure control, but the effect on behavior and psychiatric symptoms has not been studied.41 Second, reduced connectivity in perceptual and executive networks has been shown in schizophrenia.42 PCDH19 encodes a cell-cell adhesion molecule and mosaic Pcdh19 murine expression results in differential adhesion of neural progenitor cells and abnormal arrangement and sorting of cells during cortical development that likely disturbs connectivity of the brain.4 Brain network connectivity studies in females with PCDH19 mutations may shed light on the human correlate and the networks involved may aid in predicting risk for the development of schizophrenia.

Psychotic disorders as later-onset manifestation of genetic disease The concept of a later-onset phenotype has been recognized in other genetic disorders. For instance, males who carry the Fragile X pre-mutation are unaﬀected until late adult life when they may develop fragile X-associated tremor/ataxia syndrome.43 Adolescents with Dravet syndrome due to a SCN1A mutation often develop a crouch gait, while adults may develop Parkinsonian features in their thirties.44,45 Such later-onset disorders will only become recognized with more access to genetic testing of older individuals, and with longitudinal observation of patients with speciﬁc genetic diseases as they age.

Limitations This study has several limitations. It would be ideal to have the same psychiatrist see all cases, however, this is impossible as the patients are in diﬀerent regions of the world, and the episodes of psychosis have occurred over many years. Also, we do not have scores on neuropsychological subscales (verbal, nonverbal, memory, processing speed and attention) and this would be very challenging in some of the patients given their severe behavioral disturbance and severity of cognitive impairment. Looking at changes in these domains before, during and after psychotic episodes would help to understand the course of these disorders in patients with PCDH19-GCE.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 136 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 137

CONCLUSIONS

We identiﬁed schizophrenia and other psychotic disorders as a later-onset manifestation of PCDH19-GCE, occurring in a ﬁfth of older adolescents and women. Monitoring, recognition and appropriate treatment of later-onset schizophrenia in females with PCDH19-GCE should become part of routine care in this population.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 137 138 CHAPTER 6

REFERENCES

1. Scheffer IE, Turner SJ, Dibbens LM, et al. Epilepsy and 15. Reutens TC, Howell RA, Gebert KE, et al. Validation of a mental retardation limited to females: An under- Questionnaire for Clinical Seizure Diagnosis Accurate recognized disorder. Brain 2008; 131: 918–927. and uniform classification of epileptic seizures is vital 2. Dibbens LM, Tarpey PS, Hynes K, et al. X-linked to clinical practice and research in. Epilepsia 1992; 33: protocadherin 19 mutations cause female-limited 1065–1071. epilepsy and cognitive impairment. Nat Genet 2008; 40: 16. American Psychiatric Association. Diagnostic and 776–781. statistical manual of mental disorders: The Diagnostic 3. Camacho A, Simón R, Sanz R, et al. Cognitive and and Statistical Manual of Mental Disorders, fifth Edition. behavioral profile in females with epilepsy with PDCH19 American Psychiatric Association, 5th ed. Washington DC; mutation: Two novel mutations and review of the 2013. literature. Epilepsy Behav. 2012; 24: 134–137. 17. Fisher RS, Cross JH, French JA, et al. Operational 4. Pederick DT, Richards KL, Piltz SG, et al. Report Abnormal classification of seizure types by the International Cell Sorting Underlies the Unique X-Linked Inheritance League Against Epilepsy. Epilepsia [online serial] 2017; of PCDH19 Epilepsy. Neuron 2017; 97: 1–8. **(*):1–9. Available at www.ilae.org. Accessed March 2018 5. Depienne C, Bouteiller D, Keren B, et al. Sporadic infantile 18. Scheffer IE, Berkovic S, Capovilla G, et al. ILAE epileptic encephalopathy caused by mutations in classification of the epilepsies: Position paper of the ILAE PCDH19 resembles dravet syndrome but mainly affects Commission for Classification and Terminology. Epilepsia females. PLoS Genet 2009; 5: e1000381. [online serial] 2017; **(*):1–10. Available at www.ilae.org. 6. de Lange IM, Rump P, Neuteboom RF, et al. Male patients Accessed March 2018. affected by mosaic PCDH19 mutations: five new cases. 19. Richards S, Aziz N, Bale S, et al. Standards and Guidelines Neurogenetics 2017; 18: 147–153. for the Interpretation of Sequence Variants: A Joint 7. Higurashi N, Nakamura M, Sugai M, et al. PCDH19-related Consensus Recommendation of the American College female-limited epilepsy: Further details regarding early of Medical Genetics and Genomics and the Association clinical features and therapeutic efficacy. Epilepsy Res for Molecular Pathology. Genet Med 2015; 17: 405–424. 2013; 106: 191–199. 20. Logsdail S, Toone B. Post-ictal psychoses. A clinical and 8. Fabisiak K, Erickson RP. A familial form of convulsive phenomenological description. Br J Psychiatry 1988; 152: disorder with or without mental retardation limited to 246–252. females: extension of a pedigree limits possible genetic 21. Devinsky O. Postictal Psychosis: Common, Dangerous, mechanisms. Clin Genet 1990; 38: 353–358. and Treatable. Epilepsy Curr 2008; 8: 31–34. 9. Ryan SG, CHance, Phillip F, Zou C-H, et al. Epilepsy and 22. Miller DT, Adam MP, Aradhya S, et al. Consensus mental retardation limited to females: an X-linked Statement: Chromosomal Microarray Is a First-Tier Clinical dominant disorder with male sparing. Nat Genet 1997; Diagnostic Test for Individuals with Developmental 17: 91–95. Disabilities or Congenital Anomalies. Am J Hum Genet 10. Depienne C, Trouillard O, Bouteiller D, et al. Mutations 2010; 86: 749–64. and deletions in PCDH19 account for various familial or 23. Andrade DM, Bassett AS, Bercovici E, et al. Epilepsy: isolated epilepsies in females. Hum Mutat 2011; 32: 1959– Transition from pediatric to adult care. Recommendations 1975. of the Ontario epilepsy implementation task force. 11. Dibbens LM, Kneen R, Bayly MA, et al. Recurrence risk Epilepsia 2017; 58: 1502–1517. of epilepsy and mental retardation in females due to 24. Thiffault I, Farrow E, Smith L, et al. PCDH19-related parental mosaicism of PCDH19 mutations. Neurology epileptic encephalopathy in a male mosaic for a 2011; 76: 1514-1519. truncating variant. Am J Med Genet Part A 2016; 170: 12. Carvill GL, Heavin SB, Yendle SC, et al. Targeted 1585–1589. resequencing in epileptic encephalopathies identifies 25. Terracciano A, Specchio N, Darra F, et al. Somatic de novo mutations in CHD2 and SYNGAP1. Nat Genet mosaicism of PCDH19 mutation in a family with low- 2013; 45: 1–16. penetrance EFMR. Neurogenetics 2012; 13: 341–345. 13. Homan CC, Pederson S, To TH, et al. PCDH19 regulation 26. Perälä O, Suvisaari A, Saarni SI, et al. Lifetime Prevalence of of neural progenitor cell differentiation suggests Psychotic and Bipolar I Disorders in a General Population. asynchrony of neurogenesis as a mechanism Arch Gen Psychiatry 2007; 64: 19–28. contributing to PCDH19 Girls Clustering Epilepsy. 27. Cooper S-A, Smiley E, Morrison J, et al. Mental ill-health Neurobiol Dis 2018; 116: 106-119. in adults with intellectual disabilities: prevalence and 14. Kolc KL, Sadleir LG, Scheffer IE, et al. A systematic review associated factors. Br J Psychiatry 2007; 190: 27–35. and meta-analysis of 271 PCDH19-variant individuals 28. Morgan VA, Leonard H, Bourke J, et al. Intellectual identifies psychiatric comorbidities, and association of disability co-occurring with schizophrenia and other seizure onset and disease severity. Mol Psychiatry (in psychiatric illness: Population-based study. Br J press 2018). Psychiatry 2008; 193: 364–372.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 138 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 139

29. Marín JL, Rodríguez-Franco MA, Chugani VM, et al. 38. Tan C, Shard C, Ranieri E, et al. Mutations of Prevalence of Schizophrenia Spectrum Disorders in protocadherin 19 in female epilepsy (PCDH19-FE) lead to Average-IQ Adults with Autism Spectrum Disorders: A allopregnanolone deficiency. Hum Mol Genet 2015; 24: Meta-analysis. J Autism Dev Disord 2018; 48: 239–250. 5250–5259. 30. Clancy MJ, Clarke MC, Connor DJ, et al. The prevalence 39. Trivisano M, Lucchi C, Rustichelli C, et al. Reduced of psychosis in epilepsy; a systematic review and meta- steroidogenesis in patients with PCDH19-female limited analysis. BMC Psychiatry 2014; 14: 75. epilepsy. Epilepsia 2017; 58: e91-e95. 31. Lowther C, Costain G, Baribeau DA, et al. Genomic 40. Marx CE, Bradford DW, Hamer RM, et al. Pregnenolone as a Disorders in Psychiatry—What Does the Clinician Need novel therapeutic candidate in schizophrenia: Emerging to Know? Curr Psychiatry Rep 2017; 19: 82. preclinical and clinical evidence. Neuroscience. 2011; 191: 32. Bassett AS, Chow EWC, Husted J, et al. Clinical features 78–90. of 78 adults with 22q11 deletion syndrome. Am J Med 41. Lappalainen J, Chez M, Sullivan J, et al. A multicenter, Genet 2005; 138 A: 307–313. Open-label Trial of Ganaxolone in Children with PCDH19 33. Lowther C, Costain G, Stavropoulos DJ, et al. Delineating Epilepsy. Presented at the 69th Annual AAN Annual the 15q13.3 microdeletion phenotype: a case series and Meeting, April 2017, Boston. comprehensive review of the literature. Genet Med 2015; 42. Li P, Fan TT, Zhao RJ, et al. Altered Brain Network 17: 149–157. Connectivity as a Potential Endophenotype of 34. Stone JL, O’Donovan MC, Gurling H, et al. Rare Schizophrenia. Sci Rep 2017; 7: 5483. chromosomal deletions and duplications increase risk of 43. Hagerman RJ, Leehey M, Heinrichs W, et al. Intention schizophrenia. Nature 2008; 455: 237–241. tremor, parkinsonism, and generalized brain atrophy in 35. Marshall CR, Howrigan DP, Merico D, et al. Contribution of male carriers of fragile X. Neurology 2001; 57: 127–130. copy number variants to schizophrenia from a genome- 44. Rodda JM, Scheffer IE, McMahon JM, et al. Progressive wide study of 41,321 subjects. Nat Genet 2017; 49: 27–35. gait deterioration in adolescents with Dravet syndrome. 36. Holtzman CW, Trotman HD, Goulding SM, et al. Stress Arch Neurol 2012; 69: 873–878. and neurodevelopmental processes in the emergence 45. Fasano A, Borlot F, Lang AE, et al. Antecollis and of psychosis. Neuroscience 2013; 249: 172–191. levodopa-responsive parkinsonism are late features of 37. Cai HL, Zhou X, Dougherty GG, et al. Pregnenolone- Dravet syndrome. Neurology 2014; 82: 2250–2251. progesterone-allopregnanolone pathway as a potential therapeutic target in first-episode antipsychotic-naïve 6 patients with schizophrenia. Psychoneuroendocrinology 2018; 90: 43-51.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 139 140 CHAPTER 6

SUPPLEMENTAL DATA

Case histories of two women with PCDH19 Girls Clustering Epilepsy who presented with a psychotic disorder

1. Woman with mild intellectual disability and schizophrenia with PCDH19 pathogenic variant (p.Lys708Argfs*9) (case C:II:1)

This 35 year old woman had focal seizures beginning at age 9 months. Developmental milestones were delayed. She attended a normal school and had mild intellectual disability. In her early twenties, she started living semi-independently and worked as a cleaner.

At the age of 21 years, she developed depression and anxiety and was treated with an anti- depressant. She was aggressive towards herself and her parents. She received a domestic violence order for assaulting her mother. At 28 years, she ran away to the ‘big city’ and her parents lost contact with her for 14 months. She lived on the streets where she was assaulted and raped. When her parents and the police ﬁnally found her, she was pregnant, and was very aggressive towards her mother and threatened to cut out her fetus with a knife. She was diagnosed with schizophrenia and commenced on risperidone, which was eﬀective. Her anti-depressant medication was ceased in view of her pregnancy. She returned home to live with her parents and gave birth to a girl, who inherited her PCDH19 p.Lys708Argfs*9 mutation.

She had two further episodes of psychosis. At 30 years, she developed auditory hallucinations, aggression, anxiety with sleep disorder, and anhedonia with suicidal ideation. She had been non-compliant with her risperidone. After re-introducing risperidone and commencing an anti- depressant, her psychosis settled. Her last seizure occurred at 30 years.

Three years later at age 33 years, she developed bizarre delusions. She thought she was pregnant with all the relevant investigations showing that she was not and she stated she has several children and a husband, whereas she only had one child and no husband. She also had persecutory delusions, where she felt that her parents were ‘imposters’. She was commenced on anti-psychotic drugs paliperidone and haloperidol; haloperidol was eﬀective. Her schizophrenia went into remission.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 140 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 141

2. Woman with severe intellectual disability and a continuous psychotic disorder with PCDH19 pathogenic variant (p.Leu25Pro) (case E:II:4)

This 27 year old woman had clusters of seizures from age 7 months until 20 years. With the onset of seizures, her development regressed. Her speech further regressed at the age of 2 years when she stopped speaking. She has severe intellectual disability and speaks 20 single words.

Since an early age, she had severe behavioural problems. She was very obsessive and was diagnosed with autism spectrum disorder at 3.5 years. Later, she became extremely aggressive smashing windows and hurting relatives to the extent that the police were called. She was often highly anxious and sometimes tried to escape from her home.

At age 11 years, she developed hallucinations. Her parents observed that she would suddenly look at the corners of the room or at the ceiling, and laugh. Her parents believed that she saw imaginary friends. The hallucinations lasted a few minutes. At 17 years, she had frequent hallucinations with episodes of thought disorder where she escaped from her school on two occasions, and walked along a busy road necessitating police involvement. No other changes in her thinking, sleeping or eating were seen at the time of her hallucinations, however, such changes would be diﬃcult to identify given her severe intellectual disability. She still experiences 6 hallucinations.

She was treated with several antipsychotic drugs including quetiapine, risperidone and lithium, which were all ineﬀective. She is currently on olanzapine, ﬂuvoxamine and clonazepam, but continues to have hallucinations. Her aggression and anxiety are still present, but have improved from age 25 years.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 141 142 CHAPTER 6

Supplemental table 1: Data on pathogenicity of PCDH19 variants

Family A Family B Family C Family D Previously published in Scheffer et al, 2008; Scheffer et al, 2008; -- Dibbens et al, 2009 Dibbens et al, 2009 Mutation nomenclature (GRCh37/ g.99661472_ hg19) g.99662274A>T g.99663343G>A 99661473del NA Base change (NM_001184880.1) c.1322T>A c.253C>T c.2123_2124del NA Amino acid change (NM_001184880.1) p.Val441Glu p.Gln85* p.Lys708Argfs*9 Exon 6 deletion Genetic test Sanger sequencing Sanger sequencing NGS panel MLPA Exon Exon 1 Exon 1 Exon 1 Exon 6 PCDH19 Girls Clustering Epilepsy phenotype? yes yes yes yes maternal Inheritance paternal unknown unknown (unaffected) Allele frequency in other databases Control databases (ExAC) no no no - Control databases (gnomAD) no no no - Patient databases (ClinVar) yes, pathogenic yes, pathogenic no - Predicted effect for missense variants Mutation Taster Disease causing - - - SIFT Deleterious - - - Probably PolyPhen damaging --- Variant Effect Scoring Tool 3.0 (scores >0.5 pathogenic) 0.991 - - - Combined Annotation Dependent Depletion (scores >20 pathogenic) 26.2 - - - Conserving scores for missense variants Nucleotide High - - - Amino acid (organism) High (tetraodon) - - - GVGD Class C65 - - -

References Scheﬀer IE, Turner SJ, Dibbens LM, et al. Epilepsy and mental retardation limited to females: An under-recognized disorder. Brain 2008; 131: 918–927. Dibbens LM, Tarpey PS, Hynes K, et al. X-linked protocadherin 19 mutations cause female-limited epilepsy and cognitive impairment. Nat Genet 2008; 40: 776–781. Dibbens LM, Kneen R, Bayly MA, et al. Recurrence risk of epilepsy and mental retardation in females due to parental mosaicism of PCDH19 mutations. Neurology 2011; 76:1514-1519. Carvill GL, Heavin SB, Yendle SC, et al. Targeted resequencing in epileptic encephalopathies identiﬁes de novo mutations in CHD2 and SYNGAP1. Nat Genet 2013; 45: 1–16.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 142 SCHIZOPHRENIA IS A LATERONSET FEATURE OF PCDH19 GIRLS CLUSTERING EPILEPSY 143

Family E Family F Family G Patient H Patient I Dibbens et al, 2011 - Carvill et al. 2013 --

g.99663522A>G g.99605644C>T g.99551849C>T g.99662994T>G g.99662361T>C c.74T>C c.2675G>A c.2873G>A c.602A>C c.1235A>G p.Leu25Pro p.Ser892Asn p.Arg958Gln p.Gln201Pro p.Asp412Gly Sanger sequencing MIPs Multigene panel MIPs MIPs Exon 1 Exon 3 Exon 6 Exon 1 Exon 1 yes yes yes yes yes maternal (mosaic) not maternal paternal de novo maternal (aﬀected)

no no 21/85587 (0.02%) no no no no 35/177431 (0.01%) no no no no yes, likely benign no no

Disease causing Disease Causing Benign Disease causing Disease causing Deleterious Deleterious Deleterious Deleterous Deleterious 6 Probably Probably Probably Probably damaging damaging Disease Causing damaging damaging 0.979 0.348 0.169 0.789 0.961

27.6 29.3 22.8 24.4 25.4

High High Moderate Moderate High High (tetraodon) High (Chicken) High (Chicken) High (tetraodon) High (tetraodon) Class C0 Class C0 Class C0 Class 65 Class 65

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 143 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 144 145

Chapter 7

+DSORLQVXIÀFLHQF\RIWKHSTX1B gene is associated with myoclonic astatic epilepsy

Published as: DRM Vlaskamp, P Rump, PMC Callenbach, YJ Vos, B Sikkema-Raddatz, CMA van Ravenswaaij-Arts, OF Brouwer. Haploinsuﬃ ciency of the STX1B gene is associated with myoclonic astatic epilepsy. European Journal of Peadiatric Neurology, 2016: 20; 489-492.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 145 146 CHAPTER 7

Danique R M Vlaskamp,1,2 Patrick Rump,2 Petra M C Callenbach,1 Yvonne J Vos,2 Birgit Sikkema- Raddatz,2 Conny M A van Ravenswaaij,2 Oebele F Brouwer1

Acknowledgements. P.M.C. Callenbach received an unrestricted research grant from UCB Pharma BV, the Netherlands. UCB Pharma BV had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. We thank the parents of the patient for giving their informed consent to report his case, and J. Senior and K. Mc Intyre for editing the manuscript.

Disclosures. None of the authors has any conﬂict of interest to disclose.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 146 HAPLOINSUFFICIENCY OF THE STX1B GENE IS ASSOCIATED WITH MYOCLONIC ASTATIC EPILEPSY 147

ABSTRACT

We describe an 18-year-old male patient with myoclonic astatic epilepsy (MAE), moderate to severe intellectual disability, behavioral problems, several dysmorphisms and a 1.2-Mb de novo deletion on chromosome 16p11.2. This deletion results in haploinsufficiency of STX1B and other genes. Recently, variants in the STX1B gene have been associated with a wide spectrum of fever- related epilepsies ranging from single febrile seizures to severe epileptic encephalopathies. Two previously reported patients with a STX1B missense variant or deletion were diagnosed with MAE. Our observation of a STX1B deletion in a third patient with MAE therefore supports that STX1B gene variants or deletions can be involved in the aetiology of MAE. Furthermore, STX1B encodes for syntaxin-1B, of which interaction with the protein encoded by the STXBP1 gene is essential for the regulation of the synaptic transmission of neurotransmitters. STXBP1 gene variants have been identified in patients with many different types of epilepsy, including Dravet syndrome and epileptic encephalopathies, suggesting STX1B plays a similar role. We recommend that analysis of STX1B should be considered in the diagnostic work-up of individuals with MAE.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 147 148 CHAPTER 7

INTRODUCTION

Myoclonic Astatic Epilepsy (MAE), or Doose syndrome, is classiﬁed as an idiopathic generalized epilepsy.1 Although likely pathogenic sequence variants in the SCN1A, SCN1B, SLC2A1 and GABRG2 genes have been identiﬁed in some patients with MAE, the cause of MAE remains unknown in most patients.2

Recently, sequence variants in the STX1B gene (MIM 601485) have been shown to cause a broad spectrum of fever-associated epilepsy syndromes, ranging from simple febrile seizures to severe epileptic encephalopathies.3 Two patients in this study with a de novo missense sequence variant or a de novo 0.8-Mb deletion including STX1B had MAE.3 Here, we present a new patient with MAE who has a 1.2-Mb de novo 16p11.2 deletion that only has a 160-kb overlap with the previously published deletion, yet it also resulted in the loss of the STX1B gene.

CASE STUDY

This 18-year-old male was born after a normal pregnancy and delivery, as the ﬁrst child of healthy Dutch parents. At one year of age, he had two generalized clonic seizures followed, a few days later, by some focal motor seizures. After having been seizure-free on valproic acid until the age of 20 months, his seizures re-occurred and resembled infantile spasms. An EEG showed high- amplitude polyspike waves and he was treated with vigabatrin and prednisone.

At two years of age, he presented with several seizure types including staring and head nodding, myoclonic, myoclonic-astatic and astatic seizures, as well as focal seizures with epigastric uprising and impairment of consciousness. He became seizure-free on a combination of valproic acid, lamotrigine and clobazam, but diurnal atonic seizures re-occurred after six months. An EEG showed slow polyspike-wave paroxysms with normal background activity. Based on these seizure types and the EEG ﬁndings, his epilepsy was classiﬁed as MAE.

From age two and a half to eight years, he only experienced some generalized seizures during fever. Both lamotrigine and clobazam were successfully withdrawn. During tapering oﬀ of the valproic acid at the age of 11 years, tonic-clonic seizures re-occurred. Currently, his epilepsy is well controlled with valproic acid monotherapy.

Since one year old, developmental delay became apparent. Formal neuropsychological testing at age 23 months showed delays in communication, social behavior, adaptation, fine motor skills (developmental age 14 months) and gross motor skills (developmental age 15 months). Later, at five years and three months old, his developmental age was between two and two and a half years. He attended a special school for children with severe learning difficulties. Over this time,

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 148 HAPLOINSUFFICIENCY OF THE STX1B GENE IS ASSOCIATED WITH MYOCLONIC ASTATIC EPILEPSY 149

behavioral problems became manifest with temper tantrums, attention diﬃculties, and sleeping problems. During puberty, he showed compulsive behavior and frequent episodes of aggression with screaming, beating and kicking. Several behavioral therapies were ineﬀective.

At the age of 17 years, he has a normal height, weight and head circumference. His appearance is characterized by dark curly hair, prominent eyebrows with a mild synophrys, mild hypertelorism, short alae nasi, overfolded helices of both ears, retrognathia with a dental overbite, wide-spaced teeth and a primary failure of tooth eruption with only a few elements of his permanent dentition visible. Additionally, he has a pectus carinatum, scoliosis, pes plano valgus and a hyperpigmented macula on his left ankle. He also has a rotational nystagmus, but good visual acuity. In his family, one cousin (the son of a maternal uncle) had epilepsy between 18 months and 16 years of age and an attention deﬁcit hyperactivity disorder, but normal development. Further information about the epilepsy in this cousin was not available. The family history was otherwise unremarkable.

Initial diagnostic investigations, including a brain CT-scan at age one year, a metabolic screen, a muscle biopsy to evaluate respiratory chain defects and mitochondrial DNA sequence variants, standard karyotyping, and DNA analysis for fragile-X syndrome (MIM 300624) did not reveal the underlying cause of his epilepsy, developmental delay and behavioral problems. At the age of 17 years, targeted next-generation sequencing was performed using an epilepsy gene panel (Agilent Sure Select Design 0501411, Agilent Technologies Inc., Santa Clara, CA, USA) including 94 genes associated with early infantile and other epileptic encephalopathies, syndromes with epilepsy and intellectual disability (ID) and fever-associated epilepsies (see Supplemental Table 1 for the genes included in this panel). Sanger sequencing of the PTHR1 gene (MIM 168468, associated 7 with primary failure of tooth eruption) showed no likely pathogenic variants. Microarray analysis (Illumina Omni Express 12-V1.0, Illumina, San Diego, CA, USA) showed a de novo deletion of chromosome 16p11.2 (arr[hg19]16p11.2 (30,943,951-32,151,753)x1) in the patient that included the STX1B gene (Figure 1).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 149 150 CHAPTER 7 the protein-coding genes are shown. << and >> indicate that the the and protein-coding >> genes are shown. << and in the two DECIPHER patients. and in the two DECIPHER patients. 3 breakpoints of the deletions are outside the region presented here. Figure 1. The 16p11.2 deletions observed in our patient, the patients reportedFigure by Schubert et al. and the smallest regionThe of overlap is indicated coordinates are based by the [hg19] highlighted on zone. NCBI All Build 37

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 150 HAPLOINSUFFICIENCY OF THE STX1B GENE IS ASSOCIATED WITH MYOCLONIC ASTATIC EPILEPSY 151

DISCUSSION

We report a 18-year-old patient with MAE, moderate to severe ID, behavioral problems and several dysmorphic characteristics, in whom we found a 1.2-Mb de novo 16p11.2 deletion that includes the STX1B gene. Involvement of the STX1B gene in the aetiology of MAE was previously reported by Schubert et al.3 They described two patients with MAE with a STX1B deletion and a de novo missense variant, respectively. Furthermore, atonic seizures, which are considered to be part of the MAE phenotype, were also reported by the same authors in six relatives in two other families with either a nonsense sequence variant or a complex insertion/deletion sequence variant in the STX1B gene.3 Lastly, seven (24%) of the 29 individuals reported with STX1B sequence variants or deletions had no febrile seizures and three (10%) others experienced afebrile seizures at the onset of their epilepsy and later also developed seizures during fever. Our patient did not have febrile seizures prior to epilepsy onset. Together, these ﬁndings support the hypothesis that STX1B is not only involved in the aetiology of fever-associated epilepsies, but also in the aetiology of non-fever-related epilepsies, including MAE.

The STX1B gene encodes for the syntaxin-1B protein (STX1B). Syntaxins are cellular receptors for vesicle transport. In rats, the stx1b protein is involved in the fusion of presynaptic vesicle membranes to release their neurotransmitters. STX1B consists of an α-helical domain, which can be unfolded (‘open’ conformation) or folded (‘closed’ conformation), a soluble N-ethylmaleimide- sensitive factor-attachment protein receptor (SNARE) motif and a transmembrane region.4 ‘Open’ stx1b is able to bind other SNARE proteins and syntaxin binding protein-1 (stxbp1) resulting in presynaptic vesicle exocytosis in rats.5, 6 ‘Closed’ stx1b can also bind to stxbp1, which is essential for 7 the regulation of presynaptic vesicle release.4, 7 Stxbp1 thus executes and mediates the synaptic transmission by binding to the ‘open’ and ‘closed’ conformations of stx1b, respectively.7 Stxbp1 is encoded by the STXBP1 gene (MIM 602926). STXBP1 sequence variants have been associated with a wide spectrum of epilepsies, including fever-related Dravet syndrome, focal epilepsies in children with ID, and early onset epileptic encephalopathies like Ohtahara syndrome and infantile spasms.8-12 As the interaction between STX1B and STXBP1 is essential for the synaptic transmission of neurotransmitters, it is understandable that alterations in one of these genes can lead to epilepsy. The diversity of the related epilepsy syndromes, however, has not yet been explained.

Remarkably, STXBP1 sequence variants have been associated with infantile spasms, which were not observed in the previously reported patients with STX1B sequence variants or deletions.12 The patient presented here had seizures resembling infantile spasms at a relatively old age (20 months). Unfortunately, original EEGs were no longer available to conﬁrm the presence of hypsarrhythmia or allow the diagnosis of West syndrome.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 151 152 CHAPTER 7

The DECIPHER database includes two other patients with STX1B deletions but no reported epilepsy. Patient 291666 (chr16:26,746,122-31,394,579x1) had ID and a double outlet of the right ventricle. Patient 291768 (chr16:29,159,416-33,286,258x1) had ID, aplasia or hypoplasia of the corpus callosum, a high-arched palate, hypertelorism, low-set ears and micrognathia. The deletions of these DECIPHER patients are larger – 4.6 and 4.1 Mb, respectively – but include the region of overlap between the deletions in our and Schubert’s patient (see Figure 1).3

Besides epilepsy, a developmental delay, mainly related to speech, was observed in some members of the reported families with STX1B sequence variants.3 A more severe and global developmental delay was observed in our patient and Schubert’s patient with STX1B deletions.3 Furthermore, both patients and one DECIPHER patient had facial dysmorphisms, which were not reported in patients with STX1B sequence variants. Possibly, one or more of the other deleted genes in the region of overlap might be associated with these additional phenotypes (Figure 1). For example, the STX4 gene (MIM 186591) encodes syntaxin-4 (STX4), another protein of the syntaxin protein family. Stx4 has been shown to be expressed in rat brains and to be involved in the recycling of endosomes in dendritic spines.9 No clinical phenotype has been associated with STX4 yet, but we speculate that haploinsuﬃciency of this gene could have contributed to the patients’ developmental problems.

Recently, pathogenic variants in the SLC6A1 gene have also been found in some individuals with MAE.13 As the SLC6A1 gene was not included in our targeted next-generation sequencing epilepsy gene panel, we cannot exclude the possibility of an additional variant in the SLC6A1 gene in our patient.

In conclusion, we report a 18-year-old male patient with MAE and a 16p11.2 deletion that includes the STX1B gene. This gene has recently been associated with fever-related epilepsies and with MAE in two other patients. Our observation of a STX1B deletion in a third patient supports that MAE is indeed part of the STX1B-related epilepsy spectrum. We therefore suggest considering analysis of STX1B in the diagnostic work-up of MAE patients. An association between STX1B and infantile spasms is still doubtful and needs further investigation.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 152 HAPLOINSUFFICIENCY OF THE STX1B GENE IS ASSOCIATED WITH MYOCLONIC ASTATIC EPILEPSY 153

REFERENCES

1. Engel J,Jr, International League Against Epilepsy (ILAE). A 8. Saitsu H, Kato M, Mizuguchi T, et al. De novo mutations proposed diagnostic scheme for people with epileptic in the gene encoding STXBP1 (MUNC18-1) cause early seizures and with epilepsy: report of the ILAE Task Force infantile epileptic encephalopathy. Nat Genet 2008; 40: on Classiﬁcation and Terminology. Epilepsia 2001; 42: 782-788. 796-803. 9. Kennedy MJ, Davison IG, Robinson CG, Ehlers MD. 2. Tang S, Pal DK. Dissecting the genetic basis of myoclonic- Syntaxin-4 deﬁnes a domain for activity-dependent astatic epilepsy. Epilepsia 2012; 53: 1303-1313. exocytosis in dendritic spines. Cell 2010; 141: 524-535. 3. Schubert J, Siekierska A, Langlois M, et al. Mutations 10. Barcia G, Chemaly N, Gobin S, et al. Early epileptic in STX1B, encoding a presynaptic protein, cause fever- encephalopathies associated with STXBP1 mutations: associated epilepsy syndromes. Nat Genet 2014; 46: Could we better delineate the phenotype? Eur J Med 1327-1332. Genet 2014; 57: 15-20. 4. Dulubova I, Sugita S, Hill S, et al. A conformational switch 11. Carvill GL, Weckhuysen S, McMahon JM, et al. GABRA1 in syntaxin during exocytosis: role of munc18. EMBO J and STXBP1: novel genetic causes of Dravet syndrome. 1999; 18: 4372-4382. Neurology 2014; 82: 1245-1253. 5. Jahn R, Scheller RH. SNAREs--engines for membrane 12. Boutry-Kryza N, Labalme A, Ville D, et al. Molecular fusion. Nat Rev Mol Cell Biol 2006; 7: 631-643. characterization of a cohort of 73 patients with infantile 6. Dulubova I, Khvotchev M, Liu S, Huryeva I, Sudhof TC, spasms syndrome. Eur J Med Genet 2015; 58: 51-58. Rizo J. Munc18-1 binds directly to the neuronal SNARE 13. Carvill GL, McMahon JM, Schneider A, et al. Mutations complex. Proc Natl Acad Sci U S A 2007; 104: 2697-2702. in the GABA Transporter SLC6A1 Cause Epilepsy with 7. Gerber SH, Rah JC, Min SW, et al. Conformational switch Myoclonic-Atonic Seizures. Am J Hum Genet 2015; 96: of syntaxin-1 controls synaptic vesicle fusion. Science 808-815. 2008; 321: 1507-1510.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 153 154 CHAPTER 7

SUPPLEMENTAL DATA

Supplemental Table 1: Genes analysed in our patient with MAE using a targeted next generation sequencing gene panel

Genes included in the panel (n = 94) ALDH7A GAMT MED12 RAB39B SLC25A22 ARHGEF9 GATM MEF2C RAI1 SMC1A ARX GPC3 MOCS1 RANBP2 SMS ATP6AP2 GPHN MOCS2 RNASEH2A SPTAN1 ATRX GPR98 NRXN1 RNASEH2B SRGAP2 AUTS2 GRIA3 NSDHL RNASEH2C SRPX2 CASK GRIN2A OFD1 ROGDI ST3GAL3 CDKL5 GRIN2B OPHN1 RPS6KA3 STXBP1 CLCN2 HPRT PAK3 SAMHD1 SYN1 CNKSR2 HSD17B10 PCDH19 SCN1A SYNGAP1 CNTNAP2 IQSEC2 PHF6 SCN1B SYP CUL4B KCNJ10 PHGDH SCN2A TBC1D24 DCX KCNQ2 PLCB1 SCN8A TBCE DYRK1A KCNT1 PLP1 SCN9A TCF4 FGD1 KDM5C PNKP SLC6A8 TREX1 FLNA MAGI2 PQBP1 SLC9A1 UBE2A FOXG1 MAPK10 PRPS1 SLC9A6 UBE3A GABRD MBD5 PSAT1 SLC16A2 ZEB2 GABRG2 MECP2 PSPH SLC19A3

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 154 HAPLOINSUFFICIENCY OF THE STX1B GENE IS ASSOCIATED WITH MYOCLONIC ASTATIC EPILEPSY 155

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 155 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 156 157

&KDSWHU

PRRT2-related phenotypes in patients with a 16p11.2 deletion

Published as: DRM Vlaskamp, PMC Callenbach, P Rump, LAA Giannini, EH Brilstra, T Dijkhuizen, YJ Vos, AMF van der Kevie-Kersemaekers, J Knijnenburg, N de Leeuw, R van Minkelen, CAL Ruivenkamp, APA Stegmann, OF Brouwer, CMA van Ravenswaaij-Arts. European journal of medical genetics, 2018. In press.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 157 158 CHAPTER 8

Danique RM Vlaskamp 1,2, Petra MC Callenbach *1, Patrick Rump *2, Lucia AA Giannini 2, Eva H Brilstra 3, Trijnie Dijkhuizen 2, Yvonne J. Vos 2, Anne-Marie F van der Kevie-Kersemaekers 4, Jeroen Knijnenburg 5, Nicole de Leeuw 6, Rick van Minkelen 5, Claudia AL Ruivenkamp 7, Alexander PA Stegmann 8, Oebele F Brouwer 1, Conny MA van Ravenswaaij-Arts 2

1 University of Groningen, University Medical Center Groningen, Department of Neurology, Groningen, the Netherlands. 2 University of Groningen, University Medical Center Groningen, Department of Genetics, Groningen, the Netherlands. 3 University Medical Center Utrecht, Department of Genetics, Utrecht, the Netherlands. 4 Academic Medical Center, Department of Genetics, Amsterdam, the Netherlands. 5 Erasmus Medical Center, Department of Genetics, Rotterdam, the Netherlands. 6 Radboud University Medical Center, Department of Genetics, Nijmegen, the Netherlands. 7 Leiden University Medical Center, Department of Genetics, Leiden, the Netherlands. 8 Maastricht University Medical Center, Department of Genetics, Maastricht, the Netherlands. * Both authors contributed equally to this work

Acknowledgements. We are grateful to the patients and their parents/caretakers for participating in this study and to the physicians for including their patients. We thank Jackie Senior for editing the manuscript and Rita Dirks for submitting the patients with 16p11.2 deletions to the European Cytogeneticists Association Register of Unbalanced Chromosome Aberrations.

Disclosures. None of the authors have any conﬂicts of interest.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 158 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 159

ABSTRACT

We studied the presence of benign infantile epilepsy (BIE), paroxysmal kinesigenic dyskinesia (PKD), and PKD with infantile convulsions (PKD/IC) in patients with a 16p11.2 deletion including PRRT2 or with a PRRT2 loss-of-function sequence variant. Index patients were recruited from seven Dutch university hospitals. The presence of BIE, PKD and PKD/IC was retrospectively evaluated using questionnaires and medical records. We included 33 patients with a 16p11.2 deletion: three (9%) had BIE, none had PKD or PKD/IC. Twelve patients had a PRRT2 sequence variant: BIE was present in four (p=0.069), PKD in six (p<0.001) and PKD/IC in two (p=0.067). Most patients with a deletion had undergone genetic testing because of developmental problems (87%), whereas all patients with a sequence variant were tested because of a movement disorder (55%) or epilepsy (45%). BIE, PKD and PKD/IC clearly showed incomplete penetrance in patients with 16p11.2 deletions, but were found in all and 95% of patients with a PRRT2 sequence variant in our study and a large literature cohort, respectively. Deletions and sequence variants have the same underlying loss-of-function disease mechanism. Thus, diﬀerences in ascertainment have led to overestimating the frequency of BIE, PKD and PKD/IC in patients with a PRRT2 sequence variant. This has important implications for counseling if genome-wide sequencing shows such variants in patients not presenting the PRRT2-related phenotypes.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 159 160 CHAPTER 8

INTRODUCTION

PRRT2 (MIM 614386) has been identiﬁed as a causal gene for benign infantile epilepsy (BIE), paroxysmal kinesigenic dyskinesia (PKD), and paroxysmal kinesigenic dyskinesia with infantile convulsions (PKD/IC).1–3 These clinical entities reﬂect the core of the PRRT 2-related phenotypic spectrum.4 Other epilepsies, movement disorders and (hemiplegic) migraine have been reported to be possibly related to PRRT2 sequence variants.4

PRRT2 encodes for proline-rich transmembrane protein 2 that interacts with SNAP25 in glutamatergic synapses in the brain to modulate glutamate release.1–3, 5 PRRT2 sequence variants have been shown to result in a loss-of-function of PRRT2, impaired SNAP25 interaction, raised intracellular glutamate levels and increased neuronal hyperexcitability.3, 5

Chromosome 16p11.2 deletions including PRRT2 are associated with the 16p11.2 microdeletion syndrome (MIM 611913). We expected that 16p11.2 deletions are also associated with PRRT2- related phenotypes, because deletions and sequence variants of PRRT2 share an underlying loss-of-function disease mechanism. So far, only six cases with a 16p11.2 deletion and PKD (n=4) or PKD/IC (n=2) have been reported.6–11 Previous large 16p11.2 deletion cohort studies reported seizures in 24-31% of patients, dystonia in 1% and paroxysmal dyskinesia in 5% without classifying these phenotypes as BIE, PKD or PKD/IC.12–14

We systematically evaluated the presence of the PRRT2-related phenotypes BIE, PKD, and PKD/IC in patients with a 16p11.2 deletion including PRRT2, and compared these frequencies with those seen in patients with a PRRT2 sequence variant.

METHODS

We identiﬁed 129 Dutch-speaking patients with a 16p11.2 deletion including PRRT2 in seven Dutch university medical centers (UMCs). Five patients were not approached for participation (two were deceased, two had an additional disease-associated 22q11.2 deletion, and one was lost to follow-up before this study). Forty of 124 (32%) patients agreed to participate, including 33/40 (83%) index patients from 33 families.

Following the same strategy, we identiﬁed 43 patients with a PRRT2 sequence variant in three UMCs. One patient was involved in another study and not contacted. Fifteen (36%) patients agreed to participate, including 12/15 (80%) index patients from 12 families.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 160 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 161

All patients or their parents/caregivers gave written consent for participation and completed a questionnaire containing (1) a PKD Screening questionnaire, (2) a Headache-Attributed Restriction, Disability, Social Handicap and Impaired Participation questionnaire and (3) questions on epilepsy (seizure onset, remission, frequency and semiology), school performance and development, height, weight and family history.15,16 An English version of the questionnaire is available on request.

Phenotypes were compared between index patients with a 16p11.2 deletion (n=33) and a PRRT2 sequence variant (n=12). To increase the validity of any observed diﬀerences, we compared the phenotypes of patients with a 16p11.2 deletion in our cohort (n=33) with those of patients with a heterozygous PRRT2 sequence variant reported in the large review cohort of Ebrahimi-Fakhari (n=1423, excluding patients with bi-allelic PRRT2 sequence variants (n=15) or a 16p11.2 deletion (n=6)).4

Additional information on methods is given in the Supplemental Methods.

RESULTS

We included 33 index patients with the recurrent ~600kb BP4-BP5 16p11.2 deletion (Supplementary Figure 1) and 12 index patients with a disease-associated PRRT2 sequence variant: c.649dupC; p.(Arg217Profs*8) (n=10), c.629dupC; p.(Ala211Serfs*14) (n=1), or c.824C>T; p.(Ser275*) (n=1) (NM_145239.2, Supplementary Table 1). All variants were added to public databases (see Supplementary methods for more information). Patients with a 16p11.2 deletion most often underwent genetic testing because of developmental delay (87%), while those with a PRRT2 sequence variant were tested because of a movement disorder (55%) or epilepsy (45%) (Table 1).

Patients with a 16p11.2 deletion less often had a PRRT2-related phenotype than those with a PRRT2 sequence variant (9% vs. 100%, p<0.001) (Table 1). These phenotypes concerned BIE (9% vs. 33%, p=0.069), PKD (0% vs. 50%, p<0.001) and PKD/IC (0% vs. 17%, p=0.067) (See Tables 2 and 3 for epilepsy and movement disorder phenotypes, respectively). Comparisons between patients with a 16p11.2 deletion in our cohort and those with a PRRT2 sequence variant from the review cohort showed significant differences for all PRRT2-related phenotypes (Table 1). The presence of other epilepsies, movement disorders, hemiplegic migraine and migraine as possible PRRT2-related phenotypes did not significantly differ between the three groups (Tables 1-3, see Supplementary Table 2 for migraine phenotypes).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 161 162 CHAPTER 8 3 NA NA NA NA NA NA NA NA NA NA 0.370 0.534 0.063 0.340 0.010 <0.001 < 0.001 < 0.001 < 0.001 p-value p-value PRRT2 2 5 4 6

NA NA NA NA NA NA NA NA NA NA 34 (2.4) 68 (4.8) 20 (1.4) 19 (1.3) (1.3) 19 51 (3.6) 51 201 (14.1) 201 553 (38.9) 598 (42.0) 598 1352 (95.0) (n=1423) the review cohort the review Patients with a sequence variant from from variant sequence 1

0.152 0.189 0.059 1.000 1.000 1.000 0.067 0.069 0.340 0.001 0.023 0.003 <0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 p-value p-value PRRT2 0 (-) 0 (-) 0 (-) 1 (8.3) 1 (8.3) 1 1 (8.3) 1 (8.3) 4 (33 ) 4 (33 0/11 (-) 0/11 2 (16.7) 4 (33.3) (n=12) 3 (25.0) 6 (50.0) 12 (100) 12 11.5 (9.0 – 12.0, 6) – 12.0, (9.0 11.5 14.5 (11.0 – 22.0, 8) – 22.0, (11.0 14.5 18.0 (16.0 – 30.0, 6) – 30.0, (16.0 18.0 sequence variant variant sequence Patients with a 7 7 7

0 (-) 0 (-) 0 3 (9.1) 3 (9.1) 2 (6.1) 1 (3.0) (n=33) 33 (100) 33 19 (57.6) 19 32 (97.0) 32 28 (84.8) 28 1/31 (3.2) 1/31 4/31 (12.9) 4/31 5/25 (20.0) 5/25 29/32 (90.6) 29/32 12.4 (3.8 – 37.1) (3.8 12.4 – 48.7) (4.5 21.5 Patients with a 10.0 (6.0 – 48.0, 28) (6.0 10.0 7) – 11.0, (7.0 9.0 16p11.2 deletion deletion 16p11.2 17.5 (9.0 – 84.0, 24) – 84.0, 24) (9.0 17.5 19.8 (12.0 – 30.0, 30) 30) – 30.0, (12.0 19.8

-related phenotypes PRRT2 -related phenotypes - walking (range, known- walking in n) (range, Median age at in months known- sitting in n) (range, Migraine or probable migraine (%) 16p11.2 deletion-related phenotypes 16p11.2 (%) Developmental problems - Special education in those >4 (%) years - Problems in motor developmental (%) developmental in motor - Problems - Problems in language (%) development - Problems Male (%) Male (%) Patient characteristics at inclusion in study characteristics in inclusion Patient at Hemiplegic migraine (%) - speaking comprehensively (range, known in n) known in n) (range, - speaking comprehensively 15) – 72.0, 48.0 (18.0 - speaking ﬁrst word (range, known- speaking ﬁrst in) (range, word Other (%) diagnosis disorder movement Obesity (%) Median age (range) in years PRRT2 Diagnosis in BIE – PKD – PKD/IC spectrum in BIE – PKD – PKD/IC Diagnosis (%) Other epilepsy (%) diagnosis - BIE (%) - PKD (%) - PKD/IC (%) - PKD/IC Possible Possible Table 1: Phenotypes of patients with a 16p11.2 deletion or PRRT2 sequence variant in our cohort and the literature Table

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 162 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 163 Three Three 7 3 NA NA NA NA NA NA p-value p-value sequence variant. PRRT2 sequence variant. Fisher’s PRRT2 PRRT2 lity or learning disabilities. lity learning or 2 NA NA NA NA NA NA (n=1423) sequence variant included in the review study the review cohort the review Patients with a sequence variant from from variant sequence Other movement disorder diagnoses included paroxysmal 5 1 . is variable. If no denominator is given, there was information on Patients with intellectual disabi Patients

6 = not available, PKD = paroxysmal kinesigenic dyskinesia, PKD/IC = 0.551 0.085 0.286 0.022 <0.001 <0.001 p-value p-value All patients with a heterozygous PRRT2 PRRT2 2 Other epilepsy diagnoses included epilepsy/seizures not otherwise speciﬁed, febrile seizures 4 0/11 (-) 0/11 0/11 (-) 0/11 (n=12) 1/11 (9.1) 1/11 5/11 (45.4) 5/11 2/11 (18.2) 2/11 6/11 (54.5) 6/11 sequence variant variant sequence Patients with a

p-values of comparisons between patients with a 16p11.2 deletion p-values and patients in our cohort of comparisons between with a patients with a 16p11.2 0/30 (-) 0/30 0/30 (-) 0/30 (n=33) 1 3/30 (10.0) 3/30 3/30 (10.0) 3/30 15/30 (50.0) 15/30 26/30 (86.7) 26/30 represents therepresents number of patients with known information on th walk or talk. Abbreviations: BIE = benign infantile NA epilepsy, Patients with a 16p11.2 deletion deletion 16p11.2 p-values of comparisons between patients with a 16p11.2 deletion and patients p-values from the review cohort of comparisons between with a patients with a 16p11.2 3

- Obesity (%) - Other (dysmorphisms, behavioral- Other problems, (dysmorphisms, family history) (%) Indications for genetic testing genetic for Indications - Epilepsy (%) - Movement disorders (%) - (Hemiplegic) Migraine (%) (%) - Developmental problems non-kinesigenic dyskinesia, paroxysmal exercise-induced dyskinesia, episodic ataxia, writer’s cramp and paroxysmal torticollis. females years† aged years and 29 were not able 6 years, to 21 paroxysmal kinesigenic dyskinesia with infantile convulsions. exact tests were used for categorical data and Mann-Whitney U tests for continuous variables. If patients had missing information, a denominator is given that all patients. P-values in bold were considered signiﬁcant (p < 0.05). by Ebrahimi et al. (see methods also section). plus, absence seizures, Dravet syndrome, generalized epilepsy with febrile seizures, West syndrome and benign Rolandic epilepsy Fisher’s exact tests were used for categorical data and Mann-Whitney U tests for continuous variables.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 163 164 CHAPTER 8 Develop- mental problems DD No MRI normal No Brain imaging Wide peripheral cerebrospinal ﬂuid spaces (L>R) on MRI Ultrasound normal MRI normal No MRI normal DD/ID EEG at onset of epilepsy

Functional disorder Functional disorder left frontotemporal with epileptiform abnormalities? Unknown Unknown Severe DD/ID Epileptiform activity left occ., sometimes right to spreading occ. = benign infantile = benign BFIE epilepsy, familial infantile d awareness seizures, = focal FMS motor seizures, ID = intellectual ic dyskinesia, R = right, UTCS = unknown onset tonic-clonic seizure, Anti-epileptic drugs (effect)

Unknown Unknown Unknown No VPA (+), LTG LTG (+), VPA (unknown) LEV (-), CBZ (-), (+) LTG Seizure typeSeizure

Total of number seizures

Age at remission seizures of

Age at onset of seizures PKD No Unknown 1yNo 28y Unknown 29y Unknown 1-5 Unknown Unknown UTCS Unknown No No 4mNo 4mNo 12m 5mNo 12yNo 4y 10-25 0m 7m No (tonic-clonic) FMS No LEV (+) >25 5mNo † 21y 1-5 UnknownYes 20/day Unknown FBTCS FIAS, 8m 10-50/day 6mYes Unknown FIAS Unknown UTS, FBTCS (atonic), FIAS, FMS (+) VPA 8m Unknown (+) VPA 8mNo 5-10 Unknown 6m Unknown 8m Unknown Normal 5 (tonic) FMS Unknown Unknown 2yDD/ID Unknown 7 LEV (+) Unknown Severe (tonic-clonic) FMS DD/ID (+) PHB Unknown >25 (tonic-clonic) FMS performed Not (+) VPA No Normal FBTCS performed Not No Normal

Epilepsy Epilepsy syndrome diagnosis Unclassified BFIE Focal epilepsy Focal epilepsy Focal BFIE BIE BIE BIE BIE BFIE BFIE BFIE Sex, age in years Pt Pt 16p11.2 deletions (n=5) deletions 16p11.2 7F, 31 7F, 48 48 F, 1M, 13 1M, 33 † 21 F, 10 24 F, 17 F, 13 3F, 15 2 3 F, 3F, 42 29 F, PRRT2 sequence variants (n=7) 6 M, 4 35 22 F, 36 5 F, epilepsy, CBZ = carbamazepine, CBZ epilepsy, DD = developmental = focal FBTCS bilateral to tonic-clonic delay, seizures, = focal FIAS impaire disability, L = left, = lamotrigine, LEV occ. = levetiracetam, = occipital, = phenobarbital, LTG PHB PKD = paroxysmal kinesigen UTS = unknown onset = valproic tonic seizure, acid. VPA Table 2: Epilepsy phenotypes in patients with a 16p11.2 deletion or PRRT2 sequence variant Table eﬀect:Treatment + treatment response, deﬁned seizure frequency as >50% reduction, - no treatment response. Abbreviations: BIE

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 164 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 165

Table 3: Movement disorders in patients with a 16p11.2 deletion or a PRRT2 sequence variant

Sex, Previous Current Pt. age in Movement BIE Age at Age at Motor medication medication 1 years disorder onset remission DD (effect) (effect) 16p11.2 deletions (n=1) Myoclonic dystonia VPA (+, but side- 3F, 15 with cortical Yes Birth No Yes effects) CZP (+) myoclonus PRRT2 sequence variants (n=8 7F, 31 PKD No 9y No No CBZ (+) None 8M, 18PKD No 14y 16y No None CBZ (+) 17 F, 13 PKD Yes 10y 12y No CZP (-) CBZ (+) 20 F, 20 PKD No 8y 19y No None None 35 F, 22 PKD Yes 14y 14y No None OXC (+) L-DOPA (-),CBZ 44 M, 34 PKD No 8y No No (+, but side-effects) None CBZ (+), LEV (-), OXC 49 M, 24 PKD No 10y No No (side-effects), GBP (-) LTG (+) 53 F, 18 PKD No 15y 16y Yes 2 None CBZ (+)

Treatment eﬀect: + treatment response, - no treatment response. 1 At last moment of contact with specialist. 2 Attributed to perinatal asphyxia. Abbreviations: BIE = benign infantile epilepsy, CBZ = carbamazepine, CZP = clonazepam, DD = developmental delay, GBP = gabapentin, L-DOPA = levodopa, LTG = lamotrigine, OXC = oxcarbazepine, PKD = paroxysmal kinesigenic dyskinesia, VPA = valproic acid.

DISCUSSION

We found that only a minority of patients with a 16p11.2 deletion in our cohort suﬀered from BIE (9%) while none had PKD or PKD/IC. In comparison, patients with a PRRT2 sequence variant in our cohort (100%) and a review cohort (95%) had these PRRT2-related phenotypes signiﬁcantly more often.4

It is unlikely that phenotypic differences between the two different genotype cohorts are due to differences in the underlying PRRT2 disease mechanism. First, PRRT2-related phenotypes occurred in both genotype groups in our study and other studies.6–11 Second, both genotypes cause a loss-of-function of PRRT2. The p.(Arg217Profs*8) variant, found in most patients of our (83%) and the review (79%) cohort, results in a PRRT2 loss-of-function without a dominant-negative effect.3–5 Patients with 16p11.2 deletions have a 50% reduced expression of PRRT2 and other genes within the deletion region that probably explains their additional problems.17 In theory, the deletion of these other genes might have had a protective effect on the patients’ phenotypes, but no clear evidence for this hypothesis exists so far.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 165 166 CHAPTER 8

It seems most likely that differences in ascertaining patients underlie the differences in PRRT2- related phenotypes observed in patients with a 16p11.2 deletion versus a PRRT2 sequence variant. The frequency of phenotypes has probably been overestimated in patients with a PRRT2 variant, who most often underwent genetic testing because of a movement disorder or epilepsy. The high frequency of the recurrent c.649dupC; p.(Arg217Profs*8) PRRT2 variant in individuals included in the ExAC Database (1.3%, n=401/32,017; Exac version 0.3.1) seems to support this hypothesis, although the frequency in the gnomAD Database is substantially lower (0.05%, n=8/14,859; gnomAD version r2.0.2).18 This difference might be related to the used data (whole exome versus whole genome data) or whether DNA amplification was performed, as previously suggested by the relatively high frequency of this variant in the Exome Variant Server database that uses amplification.19 A compatible influence of ascertainment might be present in the published review cohort. The inclusion of index cases in calculating penetrance of PRRT2 in other studies has probably resulted in an overestimated disease penetrance for PKD (60%) and BIE (60-90%).20,21 A lower penetrance (48%) has been found in a single family with 23 relatives with PRRT2 sequence variants (Family 1, excluding the index patient).20,22 Reduced penetrance is also known for other clinical features associated with the 16p11.2 BP4-BP5 deletion.12–14 The low frequency of PRRT2-related phenotypes in our patients with 16p11.2 deletions is in line with two large previous studies showing the presence of any seizures in 24% (n=49/195) and 27% (n=22/83), paroxysmal dyskinesia in 5% (n=12/233) and dystonia in 1% (n=1/83), emphasizing the incomplete penetrance of BIE and PKD.12–14 However, in our small cohort, we might have underestimated the presence of PRRT2-related phenotypes. First, BIE occurred long time ago in some patients, leading to recall bias, and may have a low seizure-frequency, leading to missed diagnoses. Second, three patients had unwitnessed incidents, but these occurred too late for a diagnosis of BIE. Last, some patients were too young to fully exclude PKD. It is thus possible that the differences in frequencies between the two patient groups might be partly explained by under-recognition of PRRT2-related phenotypes in the 16p11.2 microdeletion patients.

The observation that the frequency of PRRT2-related phenotypes has thus far been overestimated in patients with a PRRT2 loss-of-function variant is important for counseling, because increasingly used whole exome sequencing (WES) may detect these variants as secondary ﬁndings. Doctors are tempted to use large cohort studies to counsel patients with such unexpected ﬁndings, but should realize that ascertainment bias is always present in these studies.

CONCLUSION

We conclude that 16p11.2 deletions including PRRT2 and PRRT2 sequence variants both lead to the PRRT2-associated phenotypes BIE, PKD or PKD/IC, but with incomplete penetrance. PRRT2- related phenotypes were more commonly found in patients with PRRT2 sequence variants, despite the shared underlying PRRT2 loss-of-function disease mechanism. Ascertainment bias

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 166 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 167

has led to an overestimation of the penetrance of BIE, PKD and PKD/IC in patients with a PRRT2 sequence variant. This study is important for the clinical interpretation of PRRT2 sequence variants found by WES in patients without these speciﬁc phenotypes.

WEB RESOURCES

We used the following URLs for data: Exome Aggregation Consortium (ExAC), http://exac.broadinstitute.org Expression Atlas for gene expression, http://www.ebi.ac.uk Online Mendelian Inheritance in Man (OMIM), https://www.omim.org UCSC Genome Bioinformatics, https://genome.ucsc.edu European Cytogeneticists Association Register of Unbalanced Chromosome Aberrations (ECARUCA), http://ecaruca.radboudumc.nl:8080/ecaruca/ Leiden Open Variation Database (LOVD), http://www.lovd.nl/PRRT2 DatabasE of genomiC varIation and Phenotype in Humans using Ensembl Resources (DECIPHER), https://decipher.sanger.ac.uk/

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 167 168 CHAPTER 8

REFERENCES

1 Heron SE, Grinton BE, Kivity S, et al. PRRT2 mutations 13 Steinman KJ, Spence SJ, Ramocki MB, et al. 16p11.2 cause benign familial infantile epilepsy and infantile deletion and duplication: Characterizing neurologic convulsions with choreoathetosis syndrome. Am J Hum phenotypes in a large clinically ascertained cohort. Am J Genet 2012; 90: 152–60. Med Genet Part A 2016; 170: 2943–2955. 2 Chen W-J, Lin Y, Xiong Z-Q, et al. Exome sequencing 14 Zufferey F, Sherr EH, Beckmann ND, et al. A 600 kb identifies truncating mutations in PRRT2 that cause deletion syndrome at 16p11.2 leads to energy imbalance paroxysmal kinesigenic dyskinesia. Nat Genet 2011; 43: and neuropsychiatric disorders. J Med Genet 2012; 49: 1252–1255. 660–668. 3 Lee HY, Huang Y, Bruneau N, et al. Mutations in the Gene 15 Tan LCS, Methawasin K, Teng EWL, et al. Clinico-genetic PRRT2 Cause Paroxysmal Kinesigenic Dyskinesia with comparisons of paroxysmal kinesigenic dyskinesia Infantile Convulsions. Cell Rep 2012; 1: 2–12. patients with and without PRRT2 mutations. Eur J Neurol 4 Ebrahimi-Fakhari D, Saffari A, Westenberger A, Klein C. 2014; 21: 674–678. The evolving spectrum of PRRT2 -associated paroxysmal 16 Steiner TJ, Gururaj G, Andrée C, et al. Diagnosis, diseases. Brain 2015; 138: 3476–3495. prevalence estimation and burden measurement 5 Li M, Niu F, Zhu X, et al. PRRT2 mutant leads to dysfunction in population surveys of headache: presenting the of glutamate signaling. Int J Mol Sci 2015; 16: 9134–9151. HARDSHIP questionnaire. J Headache Pain 2014; 15: 3. 6 Dale RC, Grattan-Smith P, Fung VSC, Peters GB. Infantile 17 Blumenthal I, Ragavendran A, Erdin S, et al. Transcriptional convulsions and paroxysmal kinesigenic dyskinesia with consequences of 16p11.2 deletion and duplication in 16p11.2 microdeletion. Neurology 2011; 77: 1401–1402. mouse cortex and multiplex autism families. Am J Hum 7 Dale RC, Grattan-Smith P, Nicholson M, Peters GB. Genet 2014; 94: 870-883 Microdeletions detected using chromosome microarray 18 Lek M, Karczewski KJ, Minikel EV, et al. Analysis of protein- in children with suspected genetic movement disorders: coding genetic variation in 60,706 humans. Nature 2016; A single-centre study. Dev Med Child Neurol 2012; 54: 536: 285–291. 618–623. 19 Huguet G, Nava C, Lemière N, et al., Heterogeneous 8 Lipton J, Rivkin MJ. 16p11.2-related paroxysmal Pattern of Selective Pressure for PRRT2 in Human kinesigenic dyskinesia and dopa-responsive Populations, but No Association with Autism Spectrum parkinsonism in a child. Neurology 2009; 73: 479-480. Disorders. PLoS ONE 2014; 9, e88600. doi:10.1371/journal. 9 Silveira-Moriyama L, Gardiner AR, Meyer E, et al. Clinical pone.0088600 features of childhood-onset paroxysmal kinesigenic 20 Callenbach PMC, van den Boogerd EH, de Coo RFM, et dyskinesia with PRRT2 gene mutations. Dev Med Child al. Refinement of the chromosome 16 locus for benign Neurol 2013; 55: 327–334. familial infantile convulsions. Clin Genet 2005; 67: 517–25. 10 Weber A, Köhler A, Hahn A, Neubauer B, Müller U. Benign 21 Van Vliet R, Breedveld G, De Rijk-Van Andel J, et al. PRRT2 infantile convulsions (IC) and subsequent paroxysmal phenotypes and penetrance of paroxysmal kinesigenic kinesigenic dyskinesia (PKD) in a patient with 16p11.2 dyskinesia and infantile convulsions. Neurology 2012; 79: microdeletion syndrome. Neurogenetics 2013; 14: 251– 777–784. 253. 22 De Vries B, Callenbach PMC, Kamphorst JT, et al. PRRT2 11 Termsarasab P, Yang AC, Reiner J, Mei H, Scott SA, Frucht mutation causes benign familial infantile convulsions. SJ. Paroxysmal kinesigenic dyskinesia caused by 16p11.2 Neurology 2012; 79: 2154–2155. microdeletion. Tremor Other Hyperkinet Mov (N Y) 2014; 4: 274. 12 Shinawi M, Liu P, Kang SL, et al. Recurrent reciprocal 16p11.2 rearrangements associated with global developmental delay, behavioural problems, dysmorphism, epilepsy, and abnormal head size. J Med Genet 2010; 47: 332–341.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 168 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 169

SUPPLEMENTAL DATA

SUPPLEMENTAL METHODS

Additional data collection If any problems related to epilepsy, movement disorders or migraine were reported in the questionnaire, we requested further information from their physicians to conﬁrm a diagnosis of epilepsy, movement disorder and/or migraine, If the information was still inconsistent or unclear, we called the patients or their parents/caregivers to further clarify and classify the reported problems. A local coordinator in each center provided information on the patients’ genotypes and indications for genetic testing.

Phenotypes For the purpose of this study, a PRRT2-related phenotype involved a diagnosis of BIE, PKD or PKD/IC. Possible PRRT2-related phenotypes included hemiplegic migraine, migraine, other epilepsies than BIE, and other movement disorders than PKD. The presence of epilepsy was determined using the latest definition of the International League Against Epilepsy (ILAE, 2014).1 We classified seizure types and epilepsy syndromes including BIE in accordance with the recent ILAE classifications for seizure types and epilepsy syndromes.2,3 We defined treatment response to anti-epileptic drugs (AED) as >50% seizure frequency reduction. We confirmed a diagnosis of PKD based on the criteria proposed by Bruno et al. and a diagnosis of migraine based on the guidelines of the International Classification of Headache disorders (third edition).4,5 In accordance with the World Health Organization standards, we defined obesity as a body mass index (BMI) ≥30 (adults), a BMI adjusted for age >2 standard deviations (SD) from the median (for children 5-19 years of age), or a weight-for-height >3 SD above the median (for children 0-5 years) using the Dutch growth curves from the Netherlands Organization for Applied Scientific Research.6,7 A family history was positive for epilepsy, PKD or (hemiplegic) migraine, if this phenotype was present in a first- or second-degree relative.

Genotypes Genes and gene variants were described in accordance with the HUGO Gene Nomenclature Committee (HGNC) and the Human Genome Variation Society (HGVS). The chromosomal coordinates were converted according to the Genome Reference Consortium Human Reference sequence version 37 (GRCh37/hg19). The cDNA positions are presented based on the transcript variant 1 of PRRT2 (NM_145239.2). We added all patients and their variants to public databases. Patients with 16p11.2 deletions were submitted to the European Cytogeneticists Association Register of Unbalanced Chromosome Aberrations (ECARUCA; patient IDs 4545, 4941, 5021, 5034, 5068, 5109, 5197, 5128, 5328-5346; available at http://ecaruca.radboudumc.nl:8080/ecaruca/) or to the DatabasE of genomiC varIation and Phenotype in Humans using Ensembl Resources (DECIPHER; patient IDs 264572, 359643, 359690, 359692, 359693, 360315; available at https://

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 169 170 CHAPTER 8

decipher.sanger.ac.uk/). Patients with PRRT2 sequence variants were submitted to the Leiden Open Variation Database 2.0 (LOVD 2.0; patient IDs 151965, 153034, 153062, 153085, 153086, 153280, 153281, 153283-153287; available at http://www.lovd.nl/2.0).

Analysis and statistics We used SPSS Statistics Version 23.0 (IBM Corporation, NY, USA). Descriptive statistics were applied to our data. To compare phenotypic diﬀerences between two patient cohorts, we used Fisher’s exact tests for categorical data and Mann-Whitney U tests for continuous variables. A p-value < 0.05 was considered signiﬁcant.

Ethics A formal evaluation of the study was waived by the Medical Ethical Committee of UMC Groningen because of its observational character.

References 1 Fisher R. Operational Classification of Seizure Types by the International League Against Epilepsy. J Chem Inf Model 2013; 53: 1689–99. 2 Fisher RS. The New Classification of Seizures by the International League Against Epilepsy 2017. Curr. Neurol. Neurosci. Rep. 2017; 17: 48. 3 Scheffer IE, Berkovic S, Capovilla G, et al. ILAE classification of the epilepsies: Position paper of the ILAE Commission for Classification and Terminology. Epilepsia 2017; 58: 512-521. 4 Bruno MK, Hallett M, Gwinn-Hardy K, et al. Clinical evaluation of idiopathic paroxysmal kinesigenic dyskinesia: new diagnostic criteria. Neurology 2004; 63: 2280–2287. 5 Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013;33:629–808. 6 World Health Organization. Obesity and overweight. 2016. Available at: http://www.who.int/mediacentre/factsheets/ fs311/en/ (accessed 1 Feb 2017). 7 Netherlands Organisation for Applied Scientific Research. Groeidiagrammen in PDF-format. 2010. Available at: https://www.tno.nl/nl/aandachtsgebieden/gezond-leven/prevention-work-health/gezond-en-veilig-opgroeien/ groeidiagrammen-in-pdf-formaat/. (accessed 1 Feb 2017).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 170 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 171

Supplemental Figure 1: 16p11.2 deletions including PRRT2 in our cohort.

The black bars represent the 16p11.2 deletions that include PRRT2 in the patients in our cohort (n=33). 16p11.2 deletion syndrome region is indicated by the blue zone. The chromosomal coordinates of 16p11.2 deletions were based on the Genome Reference Consortium Human Reference sequence version 37 (GRCh37/hg19).

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 171 172 CHAPTER 8 Ref ------) - unlikely does this this does Migraine )- ) ) - ) - unlikely), unlikely), unlikely unlikely ) unlikely unlikely , maternal parents , maternal no maternal parents maternal no, PRRT2 sequence variant Migraine (maternal mother Migraine (maternal Migraine (mother Migraine (grandmother relative share the same genotype same the share relative Family history for epilepsy, movement movement history epilepsy, for Family (relative migraine or disorders (father Migraine (grandmother - Unclassiﬁed epilepsy (sister ------related PRRT2- Focal epilepsy - - Migraine - Possible Possible phenotype - - - - Focal epilepsy - Myoclonic dystonia with cortical myoclonus, hemiplegic migraine - - - - -related -related PRRT2 phenotype De novoDe De novo - novoDe - novoDe - novoDe - De novo - De novo - De novo - De novo novo De - novoDe BIE novoDe BIE novoDe - novoDe - novoDe - De novo - - Inheritance Inheritance Genotype arr[GRCh37] arr[GRCh37] 16p11.2(29656483_30198752)x1 arr[GRCh37] 16p11.2(29567296_30177917)x1 arr[GRCh37] 16p11.2(29624565_30198753)x1 arr[GRCh37] 16p11.2(29652488_30177807)x1 arr[GRCh37] 16p11.2(29503993_30281111)x1 arr[GRCh37] 16p11.2(29567295_30178406)x1 arr[GRCh37] 16p11.2(29564185_30106852)x1 arr[GRCh37] 16p11.2(29432212_30177916)x1 arr[GRCh37] arr[GRCh37] 16p11.2(29656484_30198753)x1) arr[GRCh37] 16p11.2(29503993_30195356)x1 arr[GRCh37] 16p11.2(29620489_30199507)x1 arr[GRCh37] 16p11.2(29477859_30199507)x1 arr[GRCh37] 16p11.2(29621498_30264061)x1 arr[GRCh37] 16p11.2(29656484_30197490)x1 arr[GRCh37] 16p11.2(29656484_30199507)x1 arr[GRCh37] 16p11.2(29567296_30178000)x1 Sex, age years in 21 M, 12 1M, 13 deletions (n=33) 16p11.2 deletions (n=22) De 16p11.2 novo 1M, 25 M, 6 Pt. 33 † 21 F, 29 M, 5 24 M, 9 28 M, 10 30 M, 13 41 M, 8 5M, 14 5M, 3F, 15 3F, 11 9M, 13 F, 7 14 M, 7 16 M, 12 2 3 F, Supplemental Table phenotypes1: Genotypes, and family histories of patients with a 16p11.2 deletion or Table Supplemental

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 172 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 173 ------Ref - )- ? ) - ) unlikely unlikely ) - ) - ? ? does this relative share share relative this does , paternal mother, paternal ) yes no , brother , brother ? , paternal mother, paternal , maternal mother , maternal )- ) - ) - , paternal mother, paternal no no ? no yes unlikely ) ), Migraine (sister Migraine (sister ), yes (father Possibly epilepsy (mother epilepsyPossibly (mother Migraine (mother BIE (sister BIE (sister Migraine (mother Migraine (mother - - Epilepsy but no BIE (mother Family history for epilepsy, movement movement history epilepsy, for Family (relative migraine or disorders ------Migraine (sister Migraine (sister the same genotype same the Migraine (mother Migraine (mother related PRRT2------Migraine - - - Possible Possible phenotype - - - - Probable migraine- - - Migraine Migraine Migraine Migraine -

PRRT2 related phenotype Unknown - UnknownUnknown - Unknown - maternalNot - - De novoDe novoDe BIE De novo - De novo - novoDe - De novo - - InheritedPaternal - Maternal - Paternal - Maternal - Paternal - - Inheritance Inheritance arr[GRCh37] arr[GRCh37] 16p11.2(29503993_30195356)x1 arr[GRCh37] arr[GRCh37] 16p11.2(29567295_30177999)x1 arr[GRCh37] 16p11.2(29567296_30177917)x1 arr[GRCh37] 16p11.2(29656684_30199351)x1 arr[GRCh37] 16p11.2(29656484_30198753)x1 arr[GRCh37] 16p11.2(29432213_30177917)x1 arr[GRCh37] 16p11.2(29567296_30177917)x1 arr[GRCh37] 16p11.2(29345903_30264061)x1 arr[GRCh37] 16p11.2(29503993_30281111)x1 arr[GRCh37] 16p11.2(29673203_30177807)x1 arr[GRCh37] 16p11.2(29478059_30199351)x1 arr[GRCh37] 16p11.2(29621498_30264061)x1 arr[GRCh37] 16p11.2(29564185_30106852)x1 arr[GRCh37] 16p11.2(29503993_30195356)x1 arr[GRCh37] 16p11.2(29656484_30198753)x1 arr[GRCh37] 16p11.2(29656484_30199507)x1 arr[GRCh37] 16p11.2(29620688_30198552)x1 Genotype Sex, age years in 12 M, 20 18 M, 21 43 30 F, 46 10 F, 55 M, 8 32 15 F, 39 M, 12 50 6 F, 11 F, 21 31 M, 37 42 29 F, 54 8 F, deletions (n=6) 16p11.2 Inherited 23 22 F, 4M, 10 (n=5) inheritance unknown with deletions 16p11.2 4M, Pt. 34 15 F, 26 M, 11 47 8 F, Continuation Supplemental Table 1 1 Table Supplemental Continuation

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 173 174 CHAPTER 8 1 2-4 Ref - - - - ) , third , third yes ), yes ), Migraine ? )- ? )- ? ) 1 ), Hemiplegic), migraine )- ), Migraine (father ), ? ? yes yes , paternal parent parent paternal , ? does this relative share the same genotype same the share relative this does , father , maternal mother , maternal ? ) )- ? , father yes (NM_145239.2). Abbreviations: BFIE benign familial familial benign BFIE Abbreviations: (NM_145239.2). ) , paternal parents, paternal ? yes in 1, ? inyes 2 in 1, ), Hemiplegic), migraine (mother , maternal father, maternal ), PKD (two sibs ), , paternal parent parent paternal , paroxysmal kinesigenic dyskinesia, TTH tension-type yes ? ) - , sister , sister ? ? yes , mother yes ? yes in 2, ? in 1 eletions were based on the Genome Reference Consortium yes in 1, ? inyes 1 in 1, (father emiplegic migraine and benign familial infantile convulsions. and infantile 777-784. 79: convulsions. Neurology. 2012; 90. ) no Eur J PaediatrEur Neurol. 2002; 6: 269-283. BFIE (father (father BFIE BFIE (brother (brother BFIE PKD (father (two sibs Migraine (brother (three sibs sib BFIE (brother (brother BFIE (father BFIE sibs (three BFIE - - - Family history for epilepsy, movement disorders disorders movement epilepsy, history for Family (relative migraine or PKD (father (father PKD related PRRT2- - Unclassiﬁed epilepsy PKD (sister - HM - HM - - - Possible Possible phenotype - -related -related PRRT2 phenotype variants were reported in accordance with the transcript variant 1 of PRRT2 of 1 variant transcript the with accordance in reported were variants PRRT2 UnknownUnknown BFIE PKD MaternalPaternal BFIE PaternalMaternal PKD PKD/ICInherited BFIE BFIE HM UnknownUnknown PKD Unknown PKD/ICUnknown PKD Unknown PKD PKD Migraine Migraine Inheritance Inheritance c.649dupC; p.(Arg217Profs*8) c.649dupC; p.(Arg217Profs*8) c.649dupC; p.(Arg217Profs*8) p.(Ser275Phe*) c.824C>T; c.649dupC; p.(Arg217Profs*8) c.629dup; p.(Ala211Serfs*14) c.649dupC; p.(Arg217Profs*8) c.649dupC; p.(Arg217Profs*8) c.649dupC; p.(Arg217Profs*8) c.649dupC; p.(Arg217Profs*8) c.649dupC; p.(Arg217Profs*8) c.649dupC; p.(Arg217Profs*8) Genotype Sex, age years in Ann Neurol. 54: 2003; 360-366 Pt. 35 22 F, 36 5 F, PRRT2 sequence variants with unknown inheritance inheritance unknown with variants sequence PRRT2 (n=7) 6 M, 4 44 34 M, PRRT2 sequence variants (n=12) variants sequence PRRT2 Inherited PRRT2 sequence variants (n=5) 10 24 F, 7F, 31 7F, 8M, 18 8M, 48 48 F, 20 20 F, 17 F, 13 49 M, 24 53 18 F, 2. De AA, Coo RFM, Vein Callenbach et al. Benign familial P, infantile convulsions: A clinical study of seven Dutch families. 3. Vanmolkot KRJ, Kors associated EE, Hottenga with familial pump et al. gene J, Novel h ATP1A2 mutations K+-ATPase in the Na+, 4. Pelzer N, de Vries B, Kamphorst et al. PRRT2 JT, and hemiplegic migraine: A complex 288-2 association. 83: Neurology. 2014; References 1. van Vliet R, Breedveld G, de Rijk-van Andel et al. J, PRRT2 phenotypes and penetrance of paroxysmal kinesigenic dyskinesia Human Reference sequence version 37 (GRCh37/hg19). Human Reference sequence (GRCh37/hg19). version 37 Continuation Supplemental Table 1 Table Supplemental Continuation ? This genotype relative’s is not known because no genetic testing has been d performed. The chromosomal coordinates of 16p11.2 infantile BIE benign epilepsy, infantile HM hemiplegic epilepsy, migraine, NMO-headache not-medication-overuse headache, PKD headache.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 174 PRRT2RELATED PHENOTYPES IN PATIENTS WITH A 16P11.2 DELETION 175

Supplemental Table 2: Migraine in patients with a 16p11.2 deletion and a PRRT2 sequence variant

Number of days missing Sex, Motor school or work / social Pt Age in Migraine diagnosis Aura weakness activities (in last 3 years months) 16p11.2 deletions (n=5) Visual, 3F, 15 Hemiplegic migraine sensitive, Yes 0 / 0 speech 24 M, 9 Migraine without aura No No 0 / 0 26 M, 11 Migraine without aura No No 0 / 0 32 F, 15 Probable migraine without aura No No 0 / 0 46 F, 10 Migraine without aura Unknown No 2 - 4 PRRT2 sequence variants (n=4) Visual, 10 F, 24 Hemiplegic migraine sensitive Yes 1 / unknown Visual, 17 F, 13 Migraine with aura sensitive No 0 / 0 35 F, 22 Hemiplegic migraine1,2 Speech ? 0 / 0

1,3 Visual, 48 F, 48 Hemiplegic migraine speech Yes 0 / 0

1 Hemiplegic migraine did not cosegregate with PRRT2 sequence variants in these patients’ families (1-4). 2 A paternal sibling had hemiplegic migraine but no PRRT2 variant (1). 3 A causal ATP1A12 sequence variant for hemiplegic migraine was identiﬁed in this patient and her relatives (3-4).

References 1. van Vliet R, Breedveld G, de Rijk-van Andel J, et al. PRRT2 phenotypes and penetrance of paroxysmal kinesigenic dyskinesia and infantile convulsions. Neurology. 2012; 79: 777-784. 2. Callenbach P, De Coo RFM, Vein AA, et al. Benign familial infantile convulsions: A clinical study of seven Dutch families. Eur J Paediatr Neurol. 2002; 6: 269-283. 3. Vanmolkot KRJ, Kors EE, Hottenga J, et al. Novel mutations in the Na+, K+-ATPase pump gene ATP1A2 associated with familial hemiplegic migraine and benign familial infantile convulsions. Ann Neurol. 2003; 54: 360-366. 4. Pelzer N, de Vries B, Kamphorst JT, et al. PRRT2 and hemiplegic migraine: A complex association. Neurology. 2014; 83: 288- 290.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 175 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 176 177

Chapter 9

Positive effect of sodium channel blocking anti-epileptic drugs on neonatal-infantile epilepsy due to a SCN2A mutation

Translated from the Dutch paper published as:

Chania de Cordt, Danique Vlaskamp, Marrit Hitzert, Roelineke Lunsing, Erica Gerkes, Oebo Brouwer. Gunstig eﬀ ect van natriumkanaal blokkerende anti-epileptica bij neonataal-infantiele epilepsie. Epilepsie, 2017; Periodiek voor Professionals 15, nr. 3

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 177 178 CHAPTER 9

Chania De Cordt1, Danique R.M. Vlaskamp2,3, Marrit Hitzert3, Roelineke J. Lunsing2, Erica Gerkes3, Oebele F. Brouwer2

1 Universiteit Antwerpen, Antwerpen. 2 University of Groningen, University Medical Center Groningen, Department of Neurology, Groningen, the Netherlands. 3 University of Groningen, University Medical Center Groningen, Department of Genetics, Groningen, the Netherlands.

Acknowledgements. We thank Katy McIntyre for her help in translating the manuscript to English.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 178 SODIUM CHANNEL BLOCKING DRUGS IN EPILEPSY DUE TO A SCN2A MUTATION 179

Recently, the journal Brain published an article about patients with epilepsy due to a SCN2A mutation.1 In this study, Wolﬀ and his colleagues showed a relation between the type of SCN2A mutation, the age at onset of seizures, and the eﬀect of sodium channel blocking anti-epileptic drugs. Here we present a case that illustrates this correlation.

CASE

A nine-week-old girl was transferred to our paediatric intensive care because of refractory seizures. She was born at term after an uncomplicated pregnancy and delivery. The parents are not consanguine and the mother has a healthy son and daughter from a previous relationship. At the age of five days, an incident occurred that was interpreted as related to aspiration due to reflux. From the age of six weeks, the girl had seizures with smacking, staring and saturation drops that increased in frequency, as well as generalized tonic-clonic seizures. The administering of levetiracetam and pyridoxine did not have the desired effect on the seizures, and the girl was transferred to the University Medical Center Groningen (UMCG).

On admission, the girl was not alert but responded adequately to pain stimuli. She had normal body body lengths and no evident dysmorphisms. On neurological examination, her fontanel was normal and both her pupils were isocoric and responded to light. She showed little spontaneous motor activity but did show symmetrical motor activity. Reﬂexes were symmetrically present. The results of further physical examination of her heart, lungs and abdomen were normal.

Because of persistent seizures, she was subsequently treated with higher doses of levetiracetam (up to 50/mg/kg/day) and pyridoxine (up to 30mg/kg/day), followed by pyridoxalphosphate (50mg/kg/day) and clobazam (up to 1 mg/kg/day), unfortunately without effect. Two loading doses of phenytoin (up to 30mg/kg/day) were temporarily effective, but no maintenance dose was started. Because of on-going seizures, intravenous midazolam was started (up to 0.3mg/kg/ hour), after which she developed respiratory insufficiency and needed intubation and ventilation. 9

An electroencephalogram (EEG) on the fourth day after admission (Figure 1A) showed a diﬀuse slow background pattern with epileptiform function disorders, predominantly left temporal and central. A second EEG one week later (Figure 1B) still showed a similar pattern with multifocal epileptiform abnormalities. During ophthalmologic investigation, a small opacity was noticed in both lenses, and in the list of diﬀerential diagnoses was a subcapsular cataract or Mittendorf dot. Magnetic resonance imaging (MRI) of the brain showed no abnormalities. Metabolic screening of her blood, liquor and urine, including -aminoadipine semialdehyde (-AASA) showed no abnormalities.

The family history revealed that her father also had epilepsy, with an onset at the age of three months. He had focal seizures, sometimes evolving into bilateral tonic-clonic seizures. Phenytoin

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 179 180 CHAPTER 9

and phenobarbital did not have any eﬀect in his ﬁrst year of life. He became seizure-free at the age of two years after introduction of carbamazepine. He tried to stop carbamazepine twice, once at eight years and once at eleven years. Both times he developed focal clonic seizures, once evolving to a bilateral tonic-clonic seizure. At the age of eleven years, an EEG showed focal epileptiform function disorders, predominantly right frontal.

Figure 1: Electroencephalogram (EEG) the fourth day after admission (A) and on the 11th day after admission (B).

Figure (A) shows a diﬀuse slow background pattern with epileptiform abnormalities, predominantly left temporal and fronto- central. Figure (B) shows a similar the background pattern with still multifocal epileptiform abnormalities, maximum over the left hemisphere.

After restarting carbamazepine, he became seizure-free again. The father is still on carbamazepine and does not want to stop this treatment. Only with stress or sleep deprivation does he still note subtle phenomena, including uncertainty with walking and diﬃculties with speaking, but without losing consciousness. On admission of his daughter to the UMCG, the father suggested initiating carbamazepine in her as well.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 180 SODIUM CHANNEL BLOCKING DRUGS IN EPILEPSY DUE TO A SCN2A MUTATION 181

Because of the positive family history and the lack of any other obvious cause for her epilepsy, genetic testing was performed. A Single Nucleotide Polymorphism microarray (SNP-array) was normal. Whole Exome Sequencing (WES) with trio-analysis (father-mother-child) showed a few heterozygous missense variants of unknown signiﬁcance (VUS). The variant found in SCN2A (c.3973G>A, p.Val1325Ile) was inherited from her father. After investigations in the fathers’ parents, who both had no epilepsy based on history, it turned out that the variant had risen de novo in the father. The variant has not been reported before and is likely pathogenic given its segregation with epilepsy in the family and the compatible phenotypes reported for other pathogenic variants in SCN2A. Furthermore, the amino acid change is in a conserved region and prediction programs predict it to have a likely pathogenic eﬀect. The missense variant has not been described in databases of healthy controls.

The girl commenced carbamazepine treatment (15mg/kg/day), and became seizure-free within ﬁve days and was more alert. She was soon transferred to a general hospital. At follow-up at the age of six months, she smiled and babbled interactively, but only followed visual stimuli brieﬂy and did not yet grab. She thus had some developmental delay. In the meantime, she had had a few seizures, but these disappeared after increasing the carbamazepine dose to 18.8mg/kg/ day. On ophthalmological examination, no abnormalities were seen anymore. Given her young age, no conclusions can be made regarding her developmental course and movement disorders, although there was a mild developmental delay at six months.

DISCUSSION

The clinical picture of this case ﬁts within the known SCN2A phenotypic spectrum. Howell and her colleagues divided SCN2A phenotypes into four groups: three neonatal-infantile phenotypes with subcategories based on the severity of the epilepsy, possibly associated developmental problems and movement disorders, and one later onset phenotype.2 The phenotype of our case can be classiﬁed as one of the neonatal-infantile variants. SCN2A is the abbreviation for Sodium 9 Voltage-Gated Channel Alpha subunit 2. The SCN2A gene codes for a voltage sensitive sodium

channel, Nav1.2, that plays an important role in initiating action potentials. On the functional level a distinction is made between gain-of-function and loss-of-function mutations in SCN2A. Gain- of-function mutations cause increased activity of the sodium channel, leading to an increased duration or height of sodium inﬂux. Loss-of-function mutations result in reduced activation of the sodium channel. Both mutation types can cause epilepsy.

Functional studies on missense mutations in a cohort of children with a neonatal-infantile onset of epilepsy (younger than three months) performed by Wolﬀ and colleagues suggest that this cohort has gain-of-function mutations. In the patient cohort with later-onset epilepsy (older than three months), mutations were shown to result in a loss of function, presumably because these mutations

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 181 182 CHAPTER 9

need a more mature brain to disturb the channels. Even though the missense variant in our case has not yet been studied and its functional eﬀect is unknown, we can hypothesize that it probably concerns a gain-of-function mutation given the above-mentioned genotype-phenotype correlation.

Wolff and colleagues also showed a correlation between the type of mutation and the effect of sodium-channel-blocking anti-epileptic drugs. Sodium channel blockers result in a reduction of seizures or even seizure freedom in neonatal-infantile epilepsy with gain-of-function mutations, while other anti-epileptic drugs are often less effective. In children with seizures after the age of three months, the opposite is true and sodium channel blockers can even lead to a higher seizure frequency. This is a nice example of personalized medicine, where treatment with drugs is optimized based on the specific underlying aetiology.

Our case supports the conclusion in the recent article of Wolff and his colleagues. In our patient, there was also a positive effect of the sodium-channel-blocker carbamazepine with complete seizure freedom in a neonatal-infantile epilepsy that was difficult to treat. Carbamazepine was commenced because of the therapy resistance and the positive effect of this drug in her father. Later, the positive effect was explained by the observed SCN2A mutation. An adequate family history for epilepsy, including the effect of treatment, and genetic testing can be very important in the choice of treatment.

Currently, there are no standardized guidelines in the Netherlands for the treatment of neonatal- infantile seizures. In a previous study by Sands and colleagues, they concluded that carbamazepine was a safe and effective drug in benign familial neonatal epilepsy (BFNE).3 This study included a few patients with mutations in genes other than SCN2A, namely KCNQ2 and KCNQ3. Mutations in these potassium-channel-coding genes are also associated with benign familial neonatal seizures. From a neurophysiological point of view, it is understandable that sodium channel blockers can have a similar positive effect in patients with a loss-of-function mutation in KCNQ2 and KCNQ3. Our case confirms that carbamazepine can also be recommended in epilepsy with neonatal-infantile onset. In the study of Wolff and colleagues, phenytoin was the most effective sodium channel blocker. In our patient, a loading dose of phenytoin had effect. In her father, this effect was retrospectively not present, but it was not known if there was a temporary effect or if therapeutic levels of phenytoin were ever reached.

CONCLUSION

Our case illustrates the possible positive eﬀect of sodium channel blockers in neonatal-infantile seizures associated with a SCN2A mutation. A precise family history and careful interpretation of genetic testing results can be an important guideline for making the right diagnosis and initiating eﬀective treatment.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 182 SODIUM CHANNEL BLOCKING DRUGS IN EPILEPSY DUE TO A SCN2A MUTATION 183

REFERENCES

1. Wolﬀ M, Johannesen KM, Hedrich UBS et al. Genetic 3. Sands TT, Balestri M, Bellini G, et al. Rapid and safe and phenotypic heterogeneity suggest therapeutic response to low-dose carbamazepine in neonatal implications in SCN2A-related disorders. Brain 2017. epilepsy. Epilepsia 2016; 57: 2019–2030. doi:10.1093/brain/awx054 2. Howell KB, McMahon JM, Carvill GL, et al. SCN2A encephalopathy: A major cause of epilepsy of infancy with migrating focal seizures. Neurology 2015; 85: 958– 66.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 183 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 184 185185

Chapter 10

General discussion and a proposal for a diagnostic algorithm for genetic testing for epilepsy

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 185 186 CHAPTER 10

Epilepsy is a disorder with a wide range of presentations due to a large variation of underlying etiologies in which genetic variants play a fundamental role. The aims of this thesis on the genetics of epilepsies of childhood were threefold. Our first aim was to elucidate the diagnostic yield and psychological impact of genetic testing in clinical practice. Our second was to evaluate the phenotypic spectra of some specific genetic epilepsies and examine genotype-phenotype correlations. Our third was to illustrate the benefit of precision medicine, an upcoming and promising therapy for genetic epilepsies based on the patient’s genotype.

The yield of genetic testing We learned that microarray analysis is an important diagnostic tool for epilepsies in childhood as we found clinically relevant copy number variants (CNVs) in 11% of the 226 children in our university hospital cohort (chapter 1). Of course, this yield was inﬂuenced by how doctors clinically selected their patients for a microarray. In our cohort, the children who had a microarray more often had epilepsy ‘plus’ developmental problems, facial dysmorphisms or behavioral problems compared to those who did not have a microarray. We concluded that a microarray is an important diagnostic technique in these children with epilepsy ‘plus’.

Equally important, and systematically studied in chapter 2, is the psychological outcome of genetic testing and counseling from the perspective of patients or their parents. In our study, patients with epilepsy and their parents showed a clinically relevant increase in empowerment after the genetic counseling trajectory. Empowerment is a key outcome goal of genetic counseling and is defined as the set of beliefs that a person has decisional, cognitive and behavioral control; can regulate emotions; and has hope.1 Feelings of anxiety did not decrease significantly during the counseling trajectory. The change in empowerment was not influenced by the type of genetic testing result (disease-associated variant, variant of unknown significance, or normal testing result). We also learned to appreciate the importance of pre-test counseling as part of the counseling trajectory, since we observed empowerment already starting to increase after this first consultation.

Genotype-phenotype studies Most genes for epilepsy have been discovered in patients who were selected based on the presence of a certain phenotype. Therefore, our knowledge about gene-related phenotypic spectra is often biased by this ascertainment of patients. To limit this phenotypic ascertainment bias, we performed three study types.

First, we performed reversed phenotyping studies in patients who were selected based on their genotype with mutations in SYNGAP1 (chapter 4), GRIN2A (chapter 5), or PCDH19 (chapter 6). We were able to identify broader gene-related phenotypic spectra, including novel and key phenotypic features. In chapter 4, we described the SYNGAP1 disorder with a distinctive developmental and epileptic encephalopathy with seizures triggered by eating, intellectual

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 186 GENERAL DISCUSSION AND A PROPOSAL FOR A DIAGNOSTIC ALGORITHM FOR GENETIC TESTING FOR EPILEPSY 187

disability, a high pain threshold, and ataxia. In chapter 5, we evaluated the phenotypic spectrum associated with GRIN2A mutations and found that this spectrum ranged from normal or near normal development to severe developmental and epileptic encephalopathy with seizures and disorders of speech development as most prevalent features. We also identiﬁed a clear genotype-phenotype correlation for GRIN2A, which encodes part of the N-methyl-D-aspartate receptor. Clinical and electrophysiological data together showed that patients with missense mutations in the transmembrane or linker domain, associated with a receptor gain-of-function, have a more severe phenotype than patients with truncating mutations or missense mutations in the amino-terminal or ligand-binding domain, associated with a loss-of-function of the receptor. This pathomechanistic model might help in predicting the severity of GRIN2A disorders and facilitating potential precision medicine approaches. In chapter 6, we screened for the presence of psychotic disorders in females with PCDH19 mutations. This research was prompted by our clinical observation of schizophrenia in two females with PCDH19 Girls Clustering Epilepsy (GCE). We found a high frequency of psychotic disorders in 21% of females aged above 10 years, the earliest age at psychosis onset. This indicates that psychotic disorders are a novel, later-onset manifestation of PCDH19-GCE. Our reverse phenotype studies in chapter 4-6 are fundamental to better recognizing, monitoring and treating these genetic epilepsies and for revealing genotype- phenotype correlations, with the eventual goal of improving outcome.

Second, novel phenotypes can also be found unexpectedly in the context of now widely available genome-wide genetic testing. In chapter 7, we presented the case of a boy with epilepsy with myoclonic-atonic seizures and intellectual disability in whom microarray revealed a microdeletion including STX1B. This gene has previously been associated with fever-related epilepsy syndromes with epilepsy with myoclonic-atonic seizures in two other patients. This third published case supports that epilepsy with myoclonic-atonic seizures is part of the STX1B-related epilepsy spectrum.

Third, and last, combining data from microarray and sequencing studies may provide better insight into the phenotypic spectrum of a specific genetic disorder. In chapter 8, we have compared the presence of PRRT2-related phenotypes in patients with disease-associated PRRT2 variants and patients with microdeletions including PRRT2. Although both genotypes result in a loss-of-function of PRRT2, patients with a microdeletion had PRRT2-related phenotypes 10 significantly less often. This led us to realize that the penetrance of PRRT2 mutations seems largely overestimated, probably due to ascertainment of patients based on known PRRT2-related phenotypes. Chromosomal array analysis revealing microdeletions is applied in a wide range of clinical presentations, while sequencing of epilepsy genes or gene panels is mostly restricted to a more selected patient group with epilepsy. Our finding is important, as the same ascertainment bias effect might hold true for other genes associated with familial epilepsies. A clue for such an overestimation of disease penetrance in the literature could be the presence of the gene variant in control databases, as is the case for PRRT2.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 187 188 CHAPTER 10

Precision medicine Ideally, epilepsy genotype and phenotype studies will result in finding a precise treatment that will improve outcome based on the underlying genotype. In chapter 9, we illustrated the benefit of precision medicine in a girl with refractory epilepsy due to a SCN2A variant, probably resulting in a gain-of-function of the channel, who became seizure free on a sodium channel blocker. We learned that obtaining a detailed family history with respect to epilepsy and of the treatment responses in affected relatives may be helpful for identifying the epilepsy etiology and the best treatment option.

GUIDELINES FOR GENETIC TESTING IN CLINICAL PRACTICE

Early diagnosis prevents unnecessary further diagnostic investigations, leads to a better understanding of the epilepsy etiology and prognosis, may help recognition and monitoring of co-morbidities, and–although more rare–could improve treatment and prognosis. The International League Against Epilepsy (ILAE) also recommends that clinicians should aim to determine the underlying epilepsy etiology from the ﬁrst presentation with seizures onwards (see also Figure 1 in chapter 1).2

But when should we think of a genetic etiology? This is a diﬃcult question since the presentation of genetic epilepsies can vary tremendously. Phenotypes as mild as a few seizures in the neonatal period in an otherwise normal child or as severe as a developmental and epileptic encephalopathy in a child with profound intellectual disability, hypotonia, and a movement disorder can both have a genetic cause. The ﬁrst child has a history of self-limiting benign neonatal seizures due to a KCNQ2 mutation, while the second has DNM1 developmental and epileptic encephalopathy.3–6

Although many experts in the ﬁeld of genetic epilepsies have given their recommendations for genetic testing in current clinical practice,7–12 no clinical guidelines for genetic testing for epilepsies exist to date. The most recent European and American guidelines for diagnosing epilepsy from 2016 and 2006, respectively, indicate that genetic testing is not yet part of routine diagnostic care (Table 1).13–17 Our national Dutch guideline suggests when we should think of genetic testing, but it is unclear how this should be performed. With the wide range of genetic tests available (see Table 1 in the introduction of this thesis), it may be challenging to select the most appropriate test and to communicate the results to patients or their parents.

Therefore, we propose a new algorithm that may guide clinicians in when and how to oﬀer genetic testing for their patients with epilepsy. This algorithm is based on the results of the research described in this thesis and on our own clinical experience. The algorithm may help answering three important questions: Which patients should be selected for genetic testing? Which genetic test(s) should be performed? And how should the information about genetic testing and their results be communicated to patients or parents?

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 188 GENERAL DISCUSSION AND A PROPOSAL FOR A DIAGNOSTIC ALGORITHM FOR GENETIC TESTING FOR EPILEPSY 189

Table 1: Recommendations for genetic testing in epilepsy in national and international guidelines

Organization Guideline (year) Recommendations for genetic testing and counseling National guideline Dutch Society of Epilepsy (2016) Consider genetic testing or referral to a clinical geneticist in Neurology (NVN) 13 patients with epilepsy in the following situations: - Therapy-resistant epilepsy - Epilepsy with atypical response to specific medication - Epilepsy plus dysmorphisms, congenital anomalies, mental retardation, regression, psychosis/psychiatric disorders or other neurological problems in the patients or the family (migraine, episodic ataxia, paroxysmal dyskinesia) - Consanguinity of the parents - Suspicion of special epilepsy syndrome - Progressive disease course - Epilepsy in the family - Patient from a genetically isolated population - Suspicion of phacomatosis or neuronal migration disorder - Patient with a wish to have children - Patient has questions about heredity Necessity of genetic testing is not proven for: - Frequent epilepsies without one of the abovementioned criteria - Determining the recurrence risks for idiopathic generalized epilepsy and focal epilepsy Consider POLG-investigation before treating with valproate in patients with clinical suspicion for inborn error of metabolism. Request HLA-typing in the context of pharmacogenetic investigations before starting with carbamazepine in patients from South-East Asia descent. Always think of advantages and disadvantages of genetic testing while making a choice for genetic testing. European guidelines European EFNS guidelines on the “Molecular investigations are possible and may help in some federation of molecular diagnosis of cases to diagnose the condition but cannot be considered as a neurological channelopathies, epilepsies, routine procedure with regard to the large number of different societies migraine, stroke, and mutations in different genes. Furthermore, diagnosis can be (EFNS) 14 dementias (2010) made more easily by clinical and physiological investigations (good practice point). One exception of note is the diagnosis of severe myoclonic epilepsy of infancy, in which mutations are found in SCN1A in 80% of the patients.” National Institute Epilepsies: diagnosis and - In case of pregnancy: “Genetic counseling should be for Health and Care management (2012; updated considered if one partner has epilepsy, particularly if the Excellence (NICE) 15 2016) partner has idiopathic epilepsy and a positive family history of epilepsy” - No recommendations about genetic testing as part of the diagnostic work-up 10 American guidelines American Academy Practice parameter: Future research recommendations: of Neurology Treatment of “Identifying genetic, immune, or imaging markers may (AAN) 16 the child with a first improve prediction of prognosis.” unprovoked seizure (2003) “A goal of pharmacogenetics will be to minimize the likelihood of adverse events from medication.” American Academy Practice Parameter: “There are insufficient data to support or refute whether of Neurology Diagnostic assessment of the genetic testing (chromosomal or molecular studies) should be (AAN) 17 child with status epilepticus done routinely in children with status epilepticus.” (2006)

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 189 190 CHAPTER 10

A PROPOSED DIAGNOSTIC ALGORITHM FOR GENETIC TESTING FOR EPILEPSY

Step 1: Selecting patients We recommend that the search for a genetic cause for epilepsy should always start with deep and careful phenotyping for two reasons. First, for some epilepsies, recognition of a set of features will point towards a specific genetic diagnosis. Second, the likelihood of finding a genetic cause for epilepsy with genetic testing is largely influenced by the selection of patients based on their phenotype. For example, a disease-associated genetic variant is more likely to be found in a child with refractory seizures with developmental regression, autism, and dysmorphic features than in an adolescent with normal development without co-morbidities who had a few tonic-clonic seizures. In contrast, incidental findings (variants in genes for other diseases than epilepsy) and variants of unknown significance can be found in both patients. If patients are not carefully pre-selected for genetic testing, the chance of such incidental findings or findings of unknown significance may overshadow the chance of finding a disease-associated variant. It is a challenge to wisely select our patients for genetic testing. We propose seven questions that might be helpful in this selection process.

Is there a clear non-genetic cause for the epilepsy? If the epilepsy is clearly a consequence of an acquired cause (e.g. perinatal asphyxia, trauma, brain infarction, or meningitis), the epilepsy is not likely to be genetic and further genetic testing is thus not indicated.

Is there developmental regression or plateauing with seizure onset? The epileptic encephalopathies (EE), deﬁned as epilepsies with developmental plateauing or regression associated with epileptiform activity, are most likely to have a genetic origin. High yields up to 60% have been reported in EE-cohorts, but most often it was between 30- 40%.5,18–29 This variable yield is most likely due to diﬀerences in cohort size, selection of patients based on additional features and previous genetic testing results, the type of genetic testing performed in the study, and the number of genes sequenced. In three cohorts of patients with (drug-resistant) epilepsy, the yield was highest in the subgroup of participants with EE.30–32 We therefore recommend to prioritize genetic testing in the etiological workup of patents with a (developmental and) epileptic encephalopathy.

Is there epilepsy ‘plus’ developmental delay, intellectual disability, behavioral or psychiatric problems, congenital anomalies, or dysmorphisms? There are several other features that may increase the likelihood of ﬁnding a genetic cause for epilepsy, which we call ‘plus’ features. In this thesis, we showed that 11% of patients with epilepsy ‘plus’ had a disease-associated CNV identiﬁed with microarray analysis. Other studies on next generation sequencing techniques also reported high yields (11-47%) in their clinically selected

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 190 GENERAL DISCUSSION AND A PROPOSAL FOR A DIAGNOSTIC ALGORITHM FOR GENETIC TESTING FOR EPILEPSY 191

group of patients, in whom such ‘plus’ features were highly prevalent.31–37 Furthermore, two studies showed higher diagnostic yields of any genetic testing in patients with epilepsy ‘plus’ developmental delay versus epilepsy only38 and high yields of microarray in patients with epilepsy ‘plus’ musculoskeletal or cardiac malformations versus epilepsy only.39 We therefore recommend looking for the presence of ‘plus’ features, and oﬀering genetic testing (both microarray and next generation sequencing) if these are present. Some of these features only become evident as time evolves. Developmental delay and behavioral problems might only be noticed when children grow older, movement disorders may not manifest until childhood or adolescence, and psychiatric symptoms might even arise in adolescence and early adulthood, as is shown for PCDH19. Although early genetic testing is recommended, some epilepsies need time before their genetic origin becomes recognizable. The epilepsy implementation task force in Canada also highlights that the transition period between childhood and adulthood is an excellent time to re- evaluate the diagnosis and etiology, when the phenotype is more crystallized and novel genetic techniques might be available.40

Is there a family history for epilepsy or other paroxysmal disorders? A positive family history for epilepsy in close relatives can be an important clue for identifying the epilepsy etiology, as it may point towards a familial genetic epilepsy syndrome, such as a genetic epilepsy with febrile seizures plus.41 In addition, the pedigree could indicate the mode of inheritance (x-linked, autosomal dominant, autosomal recessive, etc.), information that is helpful in the search for the underlying gene. Thomas and Berkovic have published their tips for obtaining a full epilepsy family history by identifying all provoked and unprovoked seizures, by unraveling who is aﬀected in what part of the family, and by understanding the family background (consanguinity, isolated population, etc.).42 We like to emphasize here that it is also important to ask about other paroxysmal disorders in the family, as such disorders can be a presentation of the same underlying disorder. For example, a diagnosis of paroxysmal kinesigenic disorder (PKD) in the father of a patient with benign familial infantile epilepsy may suggest a familial PRRT2 mutation and PKD in an older sibling of a child with early childhood absence epilepsy may be suggestive of a glucose transporter 1 deﬁciency due to a mutation in SLC2A1.43,44 Thus, if the family history is suggestive of a genetic underlying disease, genetic testing is recommended.

Is there an epilepsy syndrome that is likely to be genetic? There are some epilepsy syndromes that are likely to be genetic without fulﬁlling any of the 10 abovementioned criteria. Examples of these are benign neonatal or infantile seizures (due to KCNQ2, KCNQ3, SCN2A, or PRRT2 mutations)4 , focal seizures with auditory features (due to LGI1 mutations)45,46 , nocturnal frontal lobe seizures (due to CHRNA4 or CHRNB2 mutations)47,48 or familial focal epilepsy with variable foci (due to DEPDC5 mutations).49 Although these epilepsy syndromes usually occur in families with an autosomal dominant pattern of inheritance, a family history might be lacking in de novo cases and genetic testing could then still be indicated.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 191 192 CHAPTER 10

So far, our questions have been aimed at identifying patients with epilepsies caused by a single gene mutation that are likely to be recognized with genetic testing. On the other end of the genetic epilepsy etiology spectrum are the epilepsies with a complex or multifactorial inheritance.42 Examples of such epilepsies are childhood absence epilepsy, juvenile myoclonus epilepsy, and epilepsy with generalized tonic-clonic seizures only. These epilepsies are also considered to be genetic. However, we have not yet been able to fully unravel the genotypes that are associated with them. As explained in the introduction, a few susceptibility CNVs have been identiﬁed that increase the risk for epilepsy, but other genetic risk factors for these epilepsies are unknown. Therefore we believe that further genetic testing for these complex inherited epilepsies is not yet indicated.

Do you plan treatment that is known to be influenced by genetics? Another yet largely unknown field in epilepsy genetics is pharmacogenomics. Pharmacogenomics focuses on how individual variation in anti-epileptic drug (AED) response—in terms of effects and side-effects—and dose can be explained by genetic factors. To date, the only robust evidence of these effects in epilepsy pharmacogenomics comes from variants in three genes.50 First, in patients of Han Chinese origin, we recommend screening for the presence of an HLA-B*1502 allele associated with carbamazepine-induced Stevens-Johnson syndrome.51 Second, in individuals of European or Japanese ancestry, the HLA-A*31:01 allele has been linked to an increased risk of carbamazepine-induced hypersensitivity reactions,52,53 and screening for this in patients who are being prescribed carbamazepine for epilepsy has been shown to be cost-effective.54 Third, patients with certain polymorphisms in CYP2C9 (e.g. CYP2C9*2 (rs1799853) and CYP2C9*3 (rs1057910(C)) have a reduced phenytoin metabolism and are at greater risk for developing phenytoin-concentration-dependent neurotoxicity.55–57

Many studies have tried to identify gene variants associated with epilepsy pharmacogenomics, but their ﬁndings need to be replicated in larger cohorts to make them more robust. This future research on pharmacogenomics in epilepsy is important to further personalize epilepsy treatment.

Does the patient have questions about epilepsy recurrence risks or pregnancy risks? As for most genetic disorders, patients and parents might have several important questions about the risks for epilepsy in their relatives and new generations. Furthermore, females who are pregnant or would like to become pregnant may want to learn more about the risks of their epilepsy and epilepsy treatment for their yet unborn child. Genetic counseling, either by the specialist or clinical geneticists, can then be helpful.

Step 2: Pre-test counseling We recommend always providing pre-test counseling before initiating genetic testing. This counseling can either be done by a clinical geneticist or a (pediatric) neurologist with experience in genetics. Clinicians can use this counseling setting to evaluate the need for genetic testing, to

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 192

GENERAL DISCUSSION AND A PROPOSAL FOR A DIAGNOSTIC ALGORITHM FOR GENETIC TESTING FOR EPILEPSY 193

ĞŽĞĂǀĂŝůĂďůĞ ďĞĐŽŵĞ

ĞĞĂƵƚ ŚŶŶǁƋĞƚŽƐŽǇƉŽƐĂƌŝƐĞ ŽƌƐǇŵƉƚŽŵƐ ŶĞǁƋƵĞƐƚŝŽŶƐ ǁŚĞŶ ZĞͲĞǀĂůƵĂƚĞ

ĞĞĂƵƚ ŚŶŶĞǁƚĞĐŚŶŝƋƵĞƐ ǁŚĞŶ ZĞͲĞǀĂůƵĂƚĞ

Ő Ŷ ŝ ƚ Ɛ Ğ ƚ Đ ŝ ƚ Ğ Ŷ Ğ Ő Ž E

Ě Ğ Ĩ ŝ ŝ ƚ Ŷ Ğ Ě ŝ ƚ Ŷ Ă ŝ ƌ Ă ǀ Ő Ŷ ŝ Ɛ Ƶ Ă Đ Ͳ Ğ Ɛ Ă Ğ Ɛ ŝ Ě Ž E EŽ Ő Ŷ ŝ zĞƐ ů Ğ Ɛ EŽ EŽ Ŷ ƌŝƐŬƐ͍ Ƶ ƚĞƐƚŝŶŐ ĞƉŝůĞƉƐǇ ŽĞƐƚŚĞ Ž ƉĂƚŝĞŶƚŚĂǀĞ ŽƌƉƌĞŐŶĂŶĐǇ /Ĩ ŶŽĚŝƐĞĂƐĞͲ ƚŽ ďĞ ŐĞŶĞƚŝĐ ƌĞĐƵƌƌĞŶĐĞƌŝƐŬƐ ƋƵĞƐƚŝŽŶƐĂďŽƵƚ ĐŽŶƚŝŶƵĞŐĞŶĞƚŝĐ ĂƐŚĂƌĞĚĚĞĐŝƐŝŽŶ ĂďŽƵƚ ǁŚĞƚŚĞƌ ƚŽ ƉŝůĞƉƐǇŝƐŶŽƚ ůŝŬĞůǇ ĂƐƐŽĐŝĂƚĞĚ ǀĂƌŝĂŶƚŚĂƐ ďĞĞŶŝĚĞŶƚŝĨŝĞĚ͕ŵĂŬĞ EŽ Ɛ Ɛ Đ ƚ ŝ Đ ƚ ŝ Ŷ Ğ Ğ ŵ Ŷ Ž ŵ Ğ ͍ ƚ zĞƐ Ŷ zĞƐ Ő Ă z Ğ Ǉ Ğ ^ Ő ƌ ď Ž ƚ W Ě Đ Ǉ Ğ Ɛ Ă Đ > Ɖ ŐĞŶĞƚŝĐƐ͍ / Ŷ ŵ Ğ ŽǇŽƵ ƉůĂŶ ů ƌ Ğ ŝ W ŝŶĨůƵĞŶĐĞĚ ďǇ Ă ƚƌĞĂƚŵĞŶƚƚŚĂƚ ŝƐŬŶŽǁŶ ƚŽ ďĞ Ƶ Ɖ ů Ś Ĩ W Ŷ ŝ Z K & EŽ ' E / ŶĚŝƚŝŽŶ d ^ zĞƐ d Ő ŵŽůĞĐƵůĂƌ ĂŶĚ ƉŚĞŶŽƚǇƉŝĐ ĚĂƚĂͿ / Ϳ d Ϳ Ő Ě Ŷ ĞƉŝůĞƉƐǇ ŐĞŶĞƚŝĐ͍ ŝ Ğ /ƐƚŚĞƌĞ ĂŶ ůŝŬĞůǇ ƚŽ ďĞ E Đ Ě Ŷ Ƶ Ǉ ƐǇŶĚƌŽŵĞƚŚĂƚ ŝƐ ů Ğ Ɛ Đ ' Ƶ Ɖ ǆ Ƌ Ğ Z ů Ğ ŝ ͍ Ğ zĞƐ Ɖ Ğ Ɛ Ě & Ğ Ğ ď EŽ Ğ ŝ & Đ Ĩ ŝ ŝ Ž ŶĞĐĞƐƐĂƌǇ ƚ ŵ ƚ ƚ Žƌ ĐůŝŶŝĐĂů ƉƌĂĐƚŝĐĞ Ğ K Ž Ŷ Ɛ Ŷ ǆ ͍ Ğ ƚĂƉƉƌŽƉƌŝĂƚĞ ƚĞƐƚ;ŚŝŐŚĂŶĂůǇƚŝĐĂů ĂŶĚ ĐůŝŶŝĐĂů ǀĂůŝĚŝƚǇͿ Ě Ğ Ő Ě Ğ Ğ ' ŝ ƌ Ŷ ƚ Ğ ŝ t Ž ƚ Ŷ ͬ Ŷ Ɛ Ă ƌ ŝ zĞƐ Ě Ğ > ƌ Ž Ŷ zĞƐ Ă ƚ h Ě Đ Ă ǀ ŝ Ɛ Ğ K Ő ƚ ů ƚ Ŷ Ğ Ğ /ƐƚŚĞƌĞ Ă ƌĞůĂƚŝǀĞƐ͍ Đ ŝ , Ɛ ĨŽƌ ĞƉŝůĞƉƐǇ͕ Ŷ Ŷ Ğ ĨĂŵŝůǇŚŝƐƚŽƌǇ ^ Ƶ Ă Ğ Ɖ ĚŝƐŽƌĚĞƌƐŝŶĨŝƌƐƚ ŽƌƐĞĐŽŶĚĚĞŐƌĞĞ ŽƚŚĞƌ ƉĂƌŽǆǇƐŵĂů Ă Ɛ Ɖ ' Đ t Ƶ Ğ Ğ Ɛ Ŷ Ɛ K Ɛ Ă ŝ Ğ Ğ Ő , Ŷ Ɛ EŽ ŝ Ǉ Ž ŝ Ɛ ƚ Ɖ E Ă Ğ ƚ Ğ ů ŝ Ƶ ŵ Ɖ Ž E Ğ Ğ ƌ ŵ Ś Ě Ğ Ě ƚ Ŷ ŐĞŶĞƚŝĐ ƚĞƐƚŝŶŐ ĂŶĚ ƚŚĞŝƌ ŝŵƉĂĐƚĨ Ŷ Ğ zĞƐ zĞƐ , Ǉ Ŷ ƚ Ğ ŝ Ɛ Ğ Ǉ Ő t Ő Ă Ɛ Ĩ Ğ ƌ Ɖ ů /ƐƚŚĞƌĞ Ă Ğ Ž Ő ů ƚ ƚ ŝ ; Ŷ ǆ ĞƉŝůĞƉƐǇ ͚ƉůƵƐ͛ ďĞŚĂǀŝŽƵƌĂů Žƌ Ɖ ŝ Ğ Ő Ɛ ŽƌĚǇƐŵŽƌƉŚŝƐŵƐ͍ ƚ ;ŝŶƚĞƌƉƌĞƚĂƚŝŽŶ ŽĨƉĂƚŚŽŐĞŶŝĐŝƚǇ ŽĨŝĚĞŶƚŝĨŝĞĚǀĂƌŝĂŶƚƐ ďǇ ĐŽŵďŝŶŝŶ Ŷ ŝŶƚĞůůĞĐƚƵĂů ĚŝƐĂďŝůŝƚǇ͕ ƉƐǇĐŚŝĂƚƌŝĐ ƉƌŽďůĞŵƐ͕ Ĩ ĚĞǀĞůŽƉŵĞŶƚĂů ĚĞůĂǇ͕ Ŷ ĐŽŶŐĞŶŝƚĂů ĂŶŽŵĂůŝĞƐ ŝ ŝ Ŷ Ž Đ Ǉ Ž Đ ů ŝ Ŷ Ɛ Ŷ Ğ Ɛ zĞƐ Ž Ƶ Ƶ ; EŽ Đ Ƌ ƚŽ ĞǆƉůĂŝŶ ƚŚĞƌĞƐƵůƚƐ ĨƌŽŵ ŐĞŶĞƚŝĐ ƚĞƐƚŝŶŐ ĂŶĚ ƉƌŽŵŽƚĞ ĂĚĂƉƚŝŽŶ ƚŽ ƚŚĞ ĐŽ Ő Ɛ Ő Ğ ŝ Ŷ Ő Ɛ Ǉ Ŷ ŝ Ě ŝ Ő Ŷ Ϳ Ś Ɖ ŝ Đ Ŷ Ǉ ƚ Ŷ ů Ě Ǉ ŝ ƌ Ŷ Ɖ Ž ů ƚ Ğ Ŷ ŝ Ă Ž Ğ Ɛ Ğ ƚ Ž ů Ă Ŷ Ɛ ů ŝ Ƶ Ă Ŷ Ă Ŷ ů Ă ƌ Ǉ Ŷ Ś Ƌ Ƶ ƚ Ɖ Ğ zĞƐ Ğ ŝ zĞƐ Ă Ɖ Ƶ Ğ EŽ Ŷ Ž Ś ƌ Đ Ŷ Ğ Ğ Ž ƌ Ɛ ^ Đ Ɖ Đ ŝ Ğ Ă ƌ ƚ Đ ŵ Ŷ Ě Ɛ Ě ƚ ŝ Ğ Ž Ő Ɖ Ğ ƚ Ɛ ƚ ƌ Ğ Ğ Ő ů /ƐƚŚĞƌĞ Ž Đ ƚ ů Ɛ ǆ Ğ Đ ŝ Ƶ Ŷ Ͳ ŝ ƌ ƚ Ğ ƚ Ğ ƚ Ͳ ƌĞŐƌĞƐƐŝŽŶ Žƌ Ă ǀ Ğ Ɖ ƐĞŝǌƵƌĞ ŽŶƐĞƚ͍ Ɛ /ƐƚŚĞƌĞĂĐůĞĂƌ D ĚĞǀĞůŽƉŵĞŶƚĂů ϭ͘ǀĂůƵĂƚĞ ƚŚĞ ĐůŝŶŝĐĂů ƵƚŝůŝƚǇ ŽĨŐĞŶĞƚŝĐ ƚĞƐƚŝŶŐĂŶĚϮ͘ǆƉůĂŝŶ ĐŚŽƐĞƚŚĞŵŽƐ ƚŚĞƉŽƐƐŝďůĞ ŽƵƚĐŽŵĞƐ ŽĨ ϯ͘DĂŬĞĂƐŚĂƌĞĚĂŶĚ ŝŶĨŽƌŵĞĚ ĚĞĐŝƐŝŽŶ ĂďŽƵƚ ŐĞŶĞƚŝĐ ƚĞƐƚŝŶŐ ƉůĂƚĞĂƵŝŶŐ ǁŝƚŚ Ğ Ğ Ğ ǀ D E ĨŽƌƚŚĞĞƉŝůĞƉƐǇ͍ ^ ϭ͘^ĐƌĞĞŶĨŽƌ ŽƚŚĞƌ ŐĞŶĞͲƌĞůĂƚĞĚ ƉŚĞŶŽƚǇƉĞƐ ĂŶĚ ŵŽŶŝƚŽƌĂŶĚ ƚƌĞĂƚ͕ŝĨ Ϯ͘ŝŵ ĨŽƌ ƉĞƌƐŽŶĂůŝƐĞĚƚƌĞĂƚŵĞŶƚďĂƐĞĚ ŽŶƚŚĞĞƉŝůĞƉƐǇ ŐĞŶŽƚǇƉĞ ů ƌ Ž ͘ ͘ ŶŽŶͲŐĞŶĞƚŝĐĐĂƵƐĞ ͘ ŝ

Ğ ; W W Ϯ

Ɖ ϯ ϭ Z Ğ

z ^ W > / W ' E / W z d K E , W z ^ W > / W ' E / W z d K E

' 10 Ϯ ϯ ϰ ϱ ϲ ϭ Ɖ Ɖ Ɖ Ɖ Ɖ Ɖ Ğ Ğ Ğ Ğ Ğ Ğ ƚ ƚ ƚ ƚ ƚ ƚ ^ ^ ^ ^ ^ ^ 'ĞŶĞƚŝĐ ƚĞƐƚŝŶŐ ^ĞůĞĐƚŝŶŐ ƉĂƚŝĞŶƚƐ WƌĞͲƚĞƐƚĐŽƵŶƐĞůŝŶŐ WŽƐƚͲƚĞƐƚĐŽƵŶƐĞůŝŶŐ /ŶƚĞƌƉƌĞƚŝŶŐǀĂƌŝĂŶƚƐ ZĞǀĞƌƐĞĚ ƉŚĞŶŽƚǇƉŝŶŐ A proposed diagnostic algorithm for genetic testing for epilepsy. epilepsy. for genetic testing algorithm for diagnostic 1: A proposed Figure

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 193 194 CHAPTER 10

inform patients or parents about their options, and to promote informed decision making about genetic testing. The results from this thesis have demonstrated that patients and parents show an increase in empowerment after pre-test counseling. Furthermore, a small number of patients declined genetic testing after being informed in more detail during pre-test counseling.

The ILAE has also suggested evaluating four points before deciding to do genetic testing.58 First, analytical validity should be considered to determine whether genetic testing can accurately identify the genotype of interest. Table 1 in the introduction of this thesis may be helpful in determining analytical validity. Second, we should evaluate clinical validity, a term for how accurate a test is for determining whether or not a person is affected with the disorder. Clinical validity is especially important to consider when the disease has an incomplete penetrance and finding a mutation does not necessarily mean that a person is affected. Third, and we think most important to assess, is clinical utility. Do the patient and parents really want to know the etiology? How will a positive test affect treatment or the reproductive choices of the patient and parents? Does a positive test impact the patient’s or parents’ life in terms of decision making? These questions might be easy to answer if a patient is severely affected and parents are craving answers. However, if a patient has a history of self-limiting neonatal seizures, one might question what the additional value is of knowing whether this was due to a particular gene mutation, other than confirmation of the clinically suspected diagnosis? Fourth, and last, the ILAE recommends evaluating the ethical, legal and social implications of genetic testing even though little is known about this in the population of patients with epilepsy. The ILAE also emphasizes the importance of counseling for discussing these points with the families.

Step 3: Genetic testing The next step is the genetic testing itself, with a range of tests being available. We propose a three-step method.

1. Sanger Sequencing is the ﬁrst choice for identifying variants in a single gene, selected based on a strong clinical suspicion for a certain genetic disorder or an aﬀected relative with a known disease-associated variant in this gene.

2. Microarray has shown to be an important diagnostic tool in children with developmental and epileptic encephalopathies or epilepsy in the context of a syndrome. There is (inter)national consensus that microarray is recommended as a ﬁrst-tier diagnostic test for developmental delay, which is often present in these two patient groups.59,60 In some centers, CNV analysis can be done using data from exome sequencing, but in most hospitals where this is not yet possible or where this takes a long time, microarray is still recommended to detect CNVs.

3. Next generation sequencing is used to sequence multiple genes in a single test run. In current clinical practice, whole exome sequencing is often performed. However, to reduce the

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 194 GENERAL DISCUSSION AND A PROPOSAL FOR A DIAGNOSTIC ALGORITHM FOR GENETIC TESTING FOR EPILEPSY 195

chance of incidental findings and variants of unknown significance, we recommend analyzing an epilepsy gene subpanel if the clinician is familiar with these panels and beliefs that the patient’s epilepsy phenotype only fits with genes in this subpanel. If this is not the case, we advise analyzing the full epilepsy gene panel. If this full panel is also negative, the exome can be further analyzed if this is clinically warranted. Whole genome sequencing is not yet a widely available clinical diagnostic test, mainly because it is expensive and interpreting intronic variants is still difficult.

Step 4: Multidisciplinary discussion The interpretation of genetic results can be very challenging and often needs discussion between clinicians and molecular geneticists. As mentioned in the introduction, the American College of Medical Genetics and Genomics and the Association for Molecular Pathology have written their consensus on how to classify variants. The pathogenicity of variants should be determined based on molecular characteristics (type of mutation, predicted effect, conservation, etc.) and its presence in control databases, on one hand, and whether the patient’s phenotype fits with the known gene- related phenotypic spectrum, on the other. It is important to realize that this known phenotypic spectrum could reflect a more severe end of the phenotypic spectrum with an overestimated disease penetrance due to ascertainment bias. On the other hand, current research already results in expansion of these phenotypic spectra, making it more difficult to determine whether the patient’s phenotype fits. Identifying the presence of key phenotypic features can then be helpful.

6WHS3RVWWHVWFRXQVHOLQJ During post-test counseling the results from genetic testing are explained to patients and/or their parents. If no disease-associated variant has been identified, epilepsy can still be genetic and it should be discussed with the families whether further genetic testing is warranted or not. If a disease-associated variant has been identified, the counselor can inform the families about the disorder; the associated phenotypic spectrum; its treatment options, prognosis, mode of inheritance, and recurrence risks; and reproductive options and can refer patients and families to relevant support groups. Using social media, the patients or parents of children with the same disorder can connect, listen, feel heard, help each other, and even initiate and contribute to research. The importance and power of such groups for families should never be underestimated. Overall, post-test counseling aims at helping patients and families to understand and adapt to the medical, psychological, and familial implications of the identified genetic variants that cause or contribute to their epilepsy.61 We have shown that patients or parents experience more 10 empowerment after post-counseling, regardless of the results of genetic testing.

Step 6: Reversed phenotyping Last, but definitely not least important, is the final step in patients in whom a disease-associated variant has been identified: going back to the epilepsy phenotype where we started the genetic journey. This is called reversed phenotyping. Ours and other reversed phenotyping studies in patients with mutations in several epilepsy genes will make it easier to recognize seizures and

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 195 196 CHAPTER 10

co-morbidities associated with these genetic epilepsies. This may lead to better monitoring and treatment of these patients, which is crucial for improving outcome.

RESEARCH TOWARDS THE FOLLOWING STEP: PRECISION MEDICINE

If a genetic epilepsy etiology can be identified, this may offer the opportunity to optimize epilepsy treatment based on the underlying genotype and thereby improve the outcome. This is called precision medicine. In chapter 9, we showed an example of the benefit of a precision medicine application that is available for a few genetic epilepsies to date.

To develop precision medicine for other genetic epilepsies, we first need to disentangle the underlying pathophysiology of these epilepsies. This can be done by studying the function of wild type and mutant epilepsy genes and proteins using in vitro or animal models. We recommend initiating such functional studies based on observations from clinical studies. In chapter 5, a clinically observed GRIN2A genotype-phenotype correlation prompted functional studies on specific missense variants that were associated with different presentations of the disease. These functional studies indeed showed contrasting gain-of-function and loss-of-function effects of the missense mutations located in different domains. Hypothetically, patients with gain-of- function mutations might benefit from receptor blockers while patients with loss-of-function mutations might not. Clinically initiated functional studies may therefore represent an important starting point for precision medicine aimed at these and other genetic epilepsies.

The use of precision medicine can be expanded once we have identified new genetic variants for epilepsy. Although many genes have been discovered for monogenic epilepsies, the underlying genetic etiology for most polygenic epilepsies is still unknown. Furthermore, for many monogenic epilepsies, we cannot fully explain their phenotypic variability based on the single gene variant. Other genetic factors are therefore likely to contribute to the phenotype. The field of epilepsy research may benefit from further world-wide collaborations that will include very large numbers of patients in order to have adequate power to compare the prevalence of genetic variants between patients and controls and to find genetic modifiers that contribute to the clinical penetrance of epilepsy-related variants. The discovery of such new genetic variants might then give rise to new therapeutic insights in the era of precision medicine.

As already discussed in our proposed diagnostic algorithm, another largely unexplored ﬁeld is pharmacogenomics: the role of genetic variants in drug responses. Similar large-scale genetic studies comparing genetic variants between drug responders and non-responders might help us to identify new pharmacogenetic variants and consequently to select the most suitable drug for an individual patient in the context of precision medicine.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 196 GENERAL DISCUSSION AND A PROPOSAL FOR A DIAGNOSTIC ALGORITHM FOR GENETIC TESTING FOR EPILEPSY 197

Finally, we believe that precision medicine should not only focus on providing personalized therapeutic treatment based on an individual’s genotype, but also on providing personalized care in terms of helping them understand and adapt to their disease based on the patient’s lifestyle. We found an increased empowerment in a small cohort of patients and parents who had genetic testing and counseling for epilepsy. Follow-up studies are however warranted to evaluate the individual diﬀerences in the outcome of genetic counseling to personalize the genetic counseling trajectory and improve the psychological outcome of genetic services for patients and their families.

CONCLUSIONS

Epilepsy is a highly heterogeneous disorder, also with respect to its etiology, in which genetic variants play an important role. Our research contributes to this field of genetic epilepsy in three ways. First, we showed that genetic services for epilepsies in childhood not only have important diagnostic value, but that patients and parents also show an increased empowerment after genetic counseling. Second, our reversed phenotyping studies aid in the recognition of some important genetic epilepsies because we identified their expanded phenotypic spectrum including key and novel phenotypic features. Furthermore, the results of these studies made us realize that the phenotypic spectra and the disease penetrance of known genetic epilepsies may be biased due to the selection of patients. For GRIN2A, our clinical and functional data showed a clear genotype-phenotype correlation that may direct research into precision medicine. An example of the benefit of such precision medicine for another genetic epilepsy was given in a case report. Finally, the results from this thesis, together with our clinical experiences, enabled us to propose a new diagnostic algorithm for genetic testing in epilepsy. We emphasize the importance of a careful pre-selection of patients for genetic testing followed by pre-test counseling, genetic testing, post-test counseling and, if a disease-associated variant has been identified, reverse phenotyping.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 197 198 CHAPTER 10

REFERENCES

1. Mcallister M, Payne K, Macleod R, Nicholls S, Dian Donnai, 18. Lemke JR, Riesch E, Scheurenbrand T, et al. Targeted next Davies L. Patient empowerment in clinical genetics generation sequencing as a diagnostic tool in epileptic services. J Health Psychol 2008; 13 :895–905. disorders. Epilepsia 2012; 53: 1387–1398. 2. Scheffer IE, Berkovic S, Capovilla G, et al. ILAE 19. Kodera H, Kato M, Nord AS, et al. Targeted capture and classification of the epilepsies: Position paper of the ILAE sequencing for detection of mutations causing early Commission for Classification and Terminology. Epilepsia onset epileptic encephalopathy. Epilepsia 2013; 54: 1262– [online serial]. 2017;**: 1–10. Accessed at: http://doi.wiley. 1269. com/10.1111/epi.13709. 20. Carvill GL, Heavin SB, Yendle SC, et al. Targeted 3. Singh NA, Charlier C, Stauffer D, et al. A novel potassium resequencing in epileptic encephalopathies identifies channel gene, KCNQ2, is mutated in an inherited epilepsy de novo mutations in CHD2 and SYNGAP1. Nat Genet of newborns. Nat Genet 1998; 18: 25–29. 2013; 45: 1–16. 4. Zara F, Specchio N, Striano P, et al. Genetic testing in 21. Epilepsy Phenome/Genome Project & Epi4K Consortium. benign familial epilepsies of the first year of life: Clinical Copy number variant analysis from exome data in 349 and diagnostic significance. Epilepsia 2013; 54: 425–436. patients with epileptic encephalopathy. Ann Neurol 5. EuroEPINOMICS-RES Consortium, Epilepsy Phenome/ 2015; 78: 323–328. Genome Project, Epi4K Consortium. De Novo Mutations 22. Wirrell EC, Shellhaas RA, Joshi C, Keator C, Kumar S, in Synaptic Transmission Genes Including DNM1 Cause Mitchell WG. How should children with West syndrome Epileptic Encephalopathies. Am J Hum Genet 2014; 95: be efficiently and accurately investigated? Results from 360–370. the National Infantile Spasms Consortium. Epilepsia 6. Von Spiczak S, Helbig KL, Shinde DN, et al. DNM1 2015; 56: 617–625. encephalopathy. Neurology 2017; 89: 385–394. 23. Mercimek-Mahmutoglu S, Patel J, Cordeiro D, et 7. Scheffer IE. Genetic testing in epilepsy: What should you al. Diagnostic yield of genetic testing in epileptic be doing? Epilepsy Curr 2011; 11: 107–111. encephalopathy in childhood. Epilepsia 2015; 56: 707– 8. Poduri A, Sheidley BR, Shostak S, Ottman R. Genetic 716. testing in the epilepsies-developments and dilemmas. 24. Kwong AKY, Ho ACC, Fung CW, Wong VCN. Analysis Nat Rev Neurol 2014; 10: 293–299. of mutations in 7 genes associated with neuronal 9. Berkovic SF. Genetics of epilepsy in clinical practice. excitability and synaptic transmission in a cohort Epilepsy Curr 2015; 15: 192–196. of children with non-syndromic infantile epileptic encephalopathy. PLoS One 2015; 10: 1–14. 10. Mefford HC. Clinical Genetic Testing in Epilepsy. Epilepsy Curr 2015; 15: 197–201. 25. Zhang Q, Li J, Zhao Y, Bao X, Wei L, Wang J. Gene mutation analysis of 175 Chinese patients with early- 11. Myers CT, Mefford HC. Advancing epilepsy genetics in onset epileptic encephalopathy. Clin Genet 2017; 91: the genomic era. Genome Med 2015; 7: 91. 717–724. 12. Poduri A. When should genetic testing be performed in 26. Gokben S, Onay H, Yilmaz S, et al. Targeted next epilepsy patients? Epilepsy Curr 2017; 17: 16–22. generation sequencing: the diagnostic value in early- 13. Nederlandse Vereniging voor Neurology. Richtlijnen voor onset epileptic encephalopathy. Acta Neurol Belg 2017; diagnostiek en behandeling. 2016. Available at: http:// 117: 131–138. epilepsie. neurologie.nl/cmssite/index.php?pageid=40. 27. Olson HE, Kelly M, LaCoursiere CM, et al. Genetics and Accessed on 01-05-2018. genotype–phenotype correlations in early onset 14. Burgunder JM, Finsterer J, Szolnoki Z, et al. EFNS guidelines epileptic encephalopathy with burst suppression. Ann on the molecular diagnosis of channelopathies, Neurol 2017; 81: 419–429. epilepsies, migraine, stroke, and dementias. Eur J Neurol 28. Rim JH, Kim SH, Hwang IS, et al. Efficient strategy for the 2010; 17: 641–648. molecular diagnosis of intractable early-onset epilepsy 15. NICE. Epilepsies: diagnosis and management. 2012. using targeted gene sequencing. BMC Med Genomics Available at: nice.org.uk/ guidance/cg137. Accessed on 2018; 11: 6. 01-05-2018. 29. Ko A, Youn SE, Kim SH, et al. Targeted gene panel and 16. Hirtz D, Berg A, Bettis D. Practice parameter: Treatment genotype-phenotype correlation in children with of the child with a first unprovoked seizure: Report of developmental and epileptic encephalopathy. Epilepsy the Quality Standards Subcommittee of the American Res 2018; 141: 48–55. Academy of Neurology and the Practice Committee of 30. Ream MA, Mikati MA. Clinical utility of genetic testing in the Child Neurology Society. Neurology 2003; 60: 166– pediatric drug-resistant epilepsy: A pilot study. Epilepsy 175. Behav 2014; 37: 241–248. 17. Riviello JJ, Ashwal S, Hirtz D, et al. Practice Parameter: 31. Moller RS, Larsen LHG, Johannesen KM, et al. Gene Diagnostic assessment of the child with Report of the panel testing in epileptic encephalopathies and familial Quality Standards Subcommittee of the American epilepsies. Mol Syndromol 2016; 7: 210–219. Academy of Neurology and the Practice Committee of the Child Neurology Society. Neurology 2006; 67: 1542– 1550.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 198 GENERAL DISCUSSION AND A PROPOSAL FOR A DIAGNOSTIC ALGORITHM FOR GENETIC TESTING FOR EPILEPSY 199

32. Ortega-Moreno L, Giráldez BG, Soto-Insuga V, et al. 48. Liu XR, Huang D, Wang J, et al. Paroxysmal hypnogenic Molecular diagnosis of patients with epilepsy and dyskinesia is associated with mutations in the PRRT2 developmental delay using a customized panel of gene. Neurol Genet 2016; 2: e66. epilepsy genes. PLoS One 2017; 12: 1–13. 49. Dibbens LM, De Vries B, Donatello S, et al. Mutations in 33. Mina E Della, Ciccone R, Brustia F, et al. Improving DEPDC5 cause familial focal epilepsy with variable foci. molecular diagnosis in epilepsy by a dedicated high- Nat Genet 2013; 45: 546–551. throughput sequencing platform. Eur J Hum Genet 50. Balestrini S, Sisodiya SM. Pharmacogenomics in epilepsy. 2015; 23: 354–362. Neurosci Lett 2018; 667: 27–39. 34. Helbig KL, Farwell Hagman KD, Shinde DN, et al. 51. Chung W-H, Hung S-I, Hong H-Sh, et al. A marker for Diagnostic exome sequencing provides a molecular STevens-Johnson syndrome. Nature 2004; 428: 2501. diagnosis for a significant proportion of patients with 52. Ozeki T, Mushiroda T, Yowang A, et al. Genome- epilepsy. Genet Med 2016; 18: 898–905. wide association study identifies HLA-A • 3101 allele 35. Trump N, McTague A, Brittain H, et al. Improving as a genetic risk factor for carbamazepine-induced diagnosis and broadening the phenotypes in early- cutaneous adverse drug reactions in Japanese onset seizure and severe developmental delay disorders population. Hum Mol Genet 2018; 20: 4–6. through gene panel analysis. J Med Genet 2016; 53: 310– 53. McCormack M, Alfirevic A, Bourgeois S et al. HLA-A*3101 317. and Carbamazepine-Induced Hypersensitivity Reactions 36. Segal E, Pedro H, Valdez-Gonzalez K, et al. Diagnostic in Europeans. N Engl J Med 2011; 364: 1134–1143. Yield of Epilepsy Panels in Children With Medication- 54. Plumpton CO, Yip VL, Alfirevic A, et al. Cost-effectiveness Refractory Epilepsy. Pediatr Neurol 2016; 64: 66–71. of screening for HLA-A*31:01 prior to initiation of 37. Butler KM, da Silva C, Alexander JJ, Hegde M, Escayg A. carbamazepine in epilepsy. Epilepsia 2015; 56: 556–563. Diagnostic Yield From 339 Epilepsy Patients Screened 55. Depondt C, Godard P, Espel RS, Cruz AL Da, Lienard P, on a Clinical Gene Panel. Pediatr Neurol 2017; 77: 61–66. Pandolfo M. A candidate gene study of antiepileptic 38. Berg AT, Coryell J, Saneto RP, et al. Early-life epilepsies drug tolerability and efficacy identifies an association and the emerging role of genetic testing. JAMA Pediatr of CYP2C9 variants with phenytoin toxicity. Eur J Neurol 2017; 171: 863–871. 2011; 18: 1159–1164. 39. Hrabik SA, Standridge SM, Greiner HM, et al. The Clinical 56. Lee CR, Goldstein JA, Pieper JA. Cytochrome P450 2C9 Utility of a Single-Nucleotide Polymorphism Microarray polymorphisms: a comprehensive review of the in-vitro in Patients with Epilepsy at a Tertiary Medical Center. J and human data. Pharmacogenetics 2002; 12: 251–263. Child Neurol 2015; 30: 1770–1777. 57. Lopez-garcia MA, Feria-romero IA, Serrano HF-, Escalante 40. Andrade DM, Bassett AS, Bercovici E, et al. Epilepsy: D, Suarez SO. Genetic polymorphisms associated with Transition from pediatric to adult care. Recommendations antiepileptic metabolism. Front Biosci 2014; E6: 377–386. of the Ontario epilepsy implementation task force. 58. Ottman R, Hirose S, Jain S, et al. Genetic testing in the Epilepsia 2017; 58: 1502–1517. epilepsies—Report of the ILAE Genetics Commission. 41. Scheffer IE, Berkovic SF. Generalized epilepsy with febrile Epilepsia 2010; 51: 655–670. seizures plus. A genetic disorder with heterogeneous 59. Nederlandse Vereniging voor Kindergeneeskunde. clinical phenotypes. Brain 1997; 120: 479–490. Concept richtlijn voor de initiële etiologische diagnostiek 42. Thomas RH, Berkovic SF. The hidden genetics of bij kinderen met een ontwikkelingsachterstand / epilepsy-a clinically important new paradigm. Nat Rev verstandelijke beperking. 2017. Not yet available online. Neurol 2014; 10: 283–292. 60. Miller DT, Adam MP, Aradhya S, et al. Consensus 43. Chen W-J, Lin Y, Xiong Z-Q, et al. Exome sequencing Statement: Chromosomal Microarray Is a First- identifies truncating mutations in PRRT2 that cause Tier Clinical Diagnostic Test for Individuals with paroxysmal kinesigenic dyskinesia. Nat Genet 2011; 43: Developmental Disabilities or Congenital Anomalies. 1252–1255. Am J Hum Genet. The American Society of Human 44. Suls A, Mullen SA, Weber YG, et al. Early-onset absence Genetics 2010; 86: 749–764. epilepsy caused by mutations in the glucose transporter 61. Resta R, Biesecker BB, Bennett RL, et al. A new definition GLUT1. Ann Neurol 2009; 66: 415–419. of genetic counseling: National Society of Genetic 45. Kalachikov S, Evgrafov O, Ross B, et al. Mutations in Counselors’ Task Force report. J Genet Couns 2006; 15: LGI1 cause autosomal-dominant partial epilepsy with 77–83. 10 auditory features. Am J Hum Genet 2008; 30: 335–341. 46. Bisulli F, Licchetta L, Baldassari S, Tommaso P, Tinuper P. DEPDC5 mutations in epilepsy with auditory features. Epilepsia 2016; 57: 335. 47. Steinlein OK, Mulley JC, Propping P, et al. A missense mutation in the neuronal nicotinic acetylcholine receptor alpha 4 subunit is associated with autosomal dominant nocturnal frontal lobe epilepsy. Nat Genet 1995; 11: 201–203.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 199 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 200 201

Chapter 11

Nederlandse samenvatting

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 201 202 CHAPTER 11

Epilepsie komt bij kinderen veel voor, namelijk bij 1 op de 50 tot 100 kinderen. Bij de aandoening epilepsie treden epileptische aanvallen op zonder dat daarbij sprake is van speciﬁeke uitlokkende factoren. Denk bijvoorbeeld aan koorts, ernstig slaaptekort of een lage bloedsuikerspiegel. Een epileptische aanval ontstaat door een verstoring in de signaaloverdracht in de hersenen, waardoor grote groepen hersencellen tegelijkertijd ontladen. De verschijnselen die als gevolg daarvan optreden hangen af van de plaats en uitgebreidheid van die ontladingen. Zo maken we onderscheid tussen focale (beperkt tot een hersengebied) en gegeneraliseerde (betrokkenheid van het gehele brein) aanvallen.

De diagnose epilepsie mag worden gesteld als sprake is van: 1) twee niet uitgelokte epileptische aanvallen tenminste 24 uur na elkaar; of 2) één of meerdere niet uitgelokte epileptische aanvallen met een geschatte kans van minstens 60% op het optreden van nieuwe epileptische aanvallen; of 3) een epilepsie syndroom diagnose. Een epilepsie syndroom is een aandoening waarbij de epilepsie duidelijk de boventoon voert, maar er ook andere problemen kunnen zijn, zoals ontwikkelingsachterstand, gedragsproblemen of problemen met het bewegen. De epilepsie binnen dit syndroom heeft speciﬁeke kenmerken wat betreft aanvalstype(n), leeftijdsfase waarin de aanvallen optreden en epileptische functiestoornissen in het hersenﬁlmpje (EEG). De epilepsie syndromen verschillen in hun etiologie, therapie-gevoeligheid en prognose.

Epilepsie kent vele oorzaken, onderverdeeld in zes categorieën: structureel, infectieus, metabool, immunologisch, genetisch en onbekend. Het is belangrijk om vroegtijdig te proberen de etiologie te achterhalen. Dit heeft niet alleen behandelconsequenties, maar leidt ook tot meer kennis over de onderliggende aandoening en daarmee tot eerdere herkenning en voorkomen van co-morbiditeit. Voor het stellen van een juiste epilepsie(syndroom) diagnose inclusief de onderliggende etiologie is een gedetailleerde classiﬁcatie van de epilepsie (aanvalstypen en EEG kenmerken) en de co-morbiditeit onmisbaar.

In dit proefschrift richten wij ons op genetische vormen van epilepsie, die berusten op één of meerdere veranderingen (varianten) in het erfelijk materiaal (DNA). Het DNA bestaat uit ongeveer 10.000 genen, die ieder coderen voor een bepaalde erfelijke eigenschap. De genen zijn verdeeld over 23 chromosomen paren (een chromosoom van moeder en een chromosoom van vader). Gen varianten kunnen overgeërfd worden van één van de ouders, maar ook nieuw (de novo) ontstaan bij het kind.

Hoofdstuk 1 bevat een algemene inleiding waarin wordt uitgelegd hoe epilepsie het best kan worden geclassiﬁceerd op basis van de klinische kenmerken (fenotypering). Daarnaast worden verschillende moleculaire technieken beschreven om een genetische oorzaak voor epilepsie te vinden (genotypering). Het combineren van informatie over het fenotype en het genotype is een waardevolle manier om meer te leren over genetische epilepsie.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 202 NEDERLANDSE SAMENVATTING 203

In de laatste jaren zijn er veel nieuwe genen geïdentiﬁceerd die een rol spelen bij het ontstaan van epilepsie. Daarmee is het aantal patiënten met epilepsie bij wie genetisch onderzoek wordt verricht ook enorm toegenomen. Het is echter nog onvoldoende duidelijk wat de diagnostische opbrengsten en de psychologische eﬀecten van genetisch onderzoek zijn. Deel I van dit proefschrift richt zich hierop. Daarnaast is bij veel nieuwe epilepsie genen nog onbekend wat het volledige spectrum van klinische kenmerken is en waarom dit fenotype varieert tussen kinderen met varianten in hetzelfde gen. In deel II van dit proefschrift beschrijven wij een aantal genotype- fenotype correlatie studies. Daarnaast geven wij een voorbeeld van een therapie die is gestart op basis van het genotype, genaamd ‘precision medicine’, en de positieve uitkomst hiervan voor de patiënt.

DEEL I – DE OPBRENGST VAN GENETISCH ONDERZOEK

In hoofdstuk 2 beschrijven we de resultaten van een onderzoek naar de diagnostische opbrengst van een microarray onderzoek bij kinderen met epilepsie. Microarray is een geavanceerd chromosoom onderzoek waarbij kleine ontbrekende stukjes (deleties) of juist verdubbelde stukjes (duplicaties) van het chromosoom kunnen worden opgespoord. Het onderzoek betrof een groep kinderen met epilepsie uit het Universitair Medisch Centrum Groningen bij wie dit genetisch onderzoek is ingezet. Wij vonden met dit onderzoek bij 11% een variant die de epilepsie kon verklaren. Dit getal is zeer waarschijnlijk beïnvloed door de klinische selectie van patiënten. In ons ziekenhuis werden kinderen vaker geselecteerd voor een microarray onderzoek als zij ook ontwikkelings- of gedragsproblemen en/of bepaalde uiterlijke kenmerken hadden. Een microarray blijkt dus een waardevolle test voor het opsporen van de oorzaak bij kinderen met epilepsie, vooral bij de groep met de genoemde extra kenmerken.

Net zo belangrijk als de diagnostische opbrengst, en door ons beschreven in hoofdstuk 3, is de uitkomst van genetisch onderzoek en raadgeving (counseling) vanuit het perspectief van patiënten met epilepsie of hun ouders. Het doel van genetische counseling is om patiënten hun ziekte beter te leren begrijpen en hen te helpen bij het aanpassen aan de medische, psychologische en familiaire gevolgen van het vinden van een genetische oorzaak van epilepsie. Wij onderzochten de impact van ons huidige genetische traject bestaande uit genetische counseling voorafgaand aan (pre-test counseling) en na aﬂoop van (post-test counseling) genetisch onderzoek in verband met epilepsie op patiënten of hun ouders met behulp van vragenlijsten. Wij gebruikten een belangrijke uitkomstmaat van counseling, namelijk empowerment. Empowerment geeft weer in hoeverre personen controle hebben over hun kennis, gedrag en besluitvorming ten aanzien 11 van hun epilepsie en in hoeverre zij hun emoties over de epilepsie kunnen regelen en hoop hebben voor de toekomst. Wij vonden dat patiënten of hun ouders meer empowerment hadden na het genetische counseling traject. Deze verandering in empowerment was niet verschillend tussen patiënten met een ziekte verklarende variant, een variant van onbekende betekenis en

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 203 204 CHAPTER 11

een normale genetische test uitslag. Een belangrijk deel van de toename in empowerment werd al gezien na pre-test counseling. Dit impliceert het belang van pre-test counseling voorafgaand aan genetisch onderzoek naar epilepsie.

DEEL II – GENOTYPE-FENOTYPE STUDIES

In de genotype-fenotype correlatie studies onderzochten wij de verschillende klinische kenmerken van patiënten met een variant in de genen SYNGAP1 (hoofdstuk 4) en GRIN2A (hoofdstuk 5). Deze studies helpen ons bij het herkennen van deze genetische vormen van epilepsie aan hun verschillende aanvalstypes, EEG bevindingen en co-morbiditeiten. Vroegtijdige herkenning en behandeling is belangrijk om de prognose van deze epilepsieën te verbeteren. Daarnaast vonden wij een duidelijk verband tussen het GRIN2A-genotype en -fenotype. Patiënten met varianten in twee specifieke domeinen van het gen hadden een veel ernstiger fenotype dan patiënten met varianten in de overige domeinen. Deze correlatie werd vervolgens functioneel onderzocht en bevestigd. Varianten in verschillende domeinen van GRIN2A leiden tot tegengestelde functionele effecten (toename versus verlies van functie). Deze bevinding is een eerste stap naar het ontwikkelen van nieuwe therapieën die rekening houden met deze verschillende onderliggende functionele effecten van GRIN2A varianten.

Comorbiditeiten kunnen net als epileptische aanvallen een grote impact hebben op het dagelijks leven van patiënten met epilepsie. Wij vonden dat bijna een kwart van de vrouwen met PCDH19-gerelateerde epilepsie een psychotische stoornis ontwikkelt op jong volwassen leeftijd (hoofdstuk 6). Wij benadrukken daarom het belang van het evalueren van psychiatrische problemen bij meisjes en vrouwen met PCDH19-gerelateerde epilepsie om psychotische stoornissen tijdig te herkennen en behandelen.

In hoofdstuk 7 presenteren wij de casus van een 18-jarige jongen met epilepsie met een speciﬁek aanvalstype (myoclone-atone aanvallen) op basis van een deletie op chromosoom 16 (chromosoom 16p11.2), waarin ook het STX1B gen is gelegen. Twee eerder gepubliceerde patiënten met een de novo variant in STX1B of een deletie van STX1B hadden hetzelfde epilepsie syndroom als onze patiënt. Wij concluderen dat het verlies van een kopie van het STX1B gen kan resulteren in epilepsie met myoclone-atone aanvallen.

In de studie beschreven in hoofdstuk 8 vergeleken wij de PRRT2-gerelateerde fenotypes tussen patiënten met twee verschillende type genetische veranderingen, namelijk 1) een deletie van chromosoom 16p11.2 inclusief het PRRT2 gen en 2) een PRRT2 variant. Uit deze studie bleek dat slechts 3 van de 33 patiënten met een deletie PRRT2-gerelateerde symptomen had. Daarentegen hadden alle 12 patiënten met een variant een PRRT2 fenotype, waaronder epilepsie, een bewegingsstoornis of beide. Wij veronderstellen dat dit verschil kan worden verklaard door de

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 204 NEDERLANDSE SAMENVATTING 205

wijze waarop de patiënten voor PRRT2 analyse zijn geselecteerd. Bij patiënten met een PRRT2 variant gebeurde dat op basis van hun epilepsie of bewegingsstoornis, terwijl dat bij degenen met een deletie gebeurde vanwege andere problemen zoals een ontwikkelingsstoornis. Wij denken dat het percentage van de patiënten waarbij PRRT2 varianten aanleiding geeft tot klinische verschijnselen wordt overschat door deze wijze van selectie. Dit is een belangrijke bevinding aangezien hetzelfde kan gelden voor andere epilepsiegenen.

In hoofdstuk 9 geven wij een voorbeeld van de effectiviteit van een behandeling die werd gekozen op basis van het genotype van de patiënt. We presenteren de casus van een meisje met therapieresistente epilepsie op basis van een variant in het SCN2A gen. Deze variant leidt waarschijnlijk tot een toename van de functie van het betreffende natriumkanaal. Dit meisje werd aanvalsvrij na het starten van carbamazepine, een natriumkanaalblokker. Deze behandeling was mede begonnen omdat vader als kind een vergelijkbare epilepsie had gehad die, naar later bleek, ook goed reageerde op carbamazepine. Deze casus bevestigt nog eens dat het waardevol is om een gedetailleerde familieanamnese af te nemen inclusief het nagaan van de effectiviteit van anti-epileptica bij familieleden met epilepsie. Dit kan helpen bij het herkennen van een etiologische diagnose en starten van de meest geschikte therapie.

Concluderend heeft dit proefschrift geleid tot meer kennis over genetische vormen van epilepsie, zowel op diagnostisch, therapeutisch als psychologisch gebied. De verkregen onderzoeksresultaten en onze ervaringen in de klinische praktijk hebben wij in hoofdstuk 10 gebundeld tot een voorstel voor een nieuw diagnostisch stappenplan voor genetische diagnostiek bij patiënten met epilepsie. Dit kan artsen helpen in hun besluitvorming wanneer en hoe genetisch onderzoek kan worden ingezet bij hun patiënten met epilepsie. Wij benadrukken het belang van een nauwkeurig selectieproces van patiënten voor genetische diagnostiek bij wie de opbrengst waarschijnlijk het hoogst is. Daarnaast onderstrepen wij de waarde van goede counseling voorafgaand aan en volgend op genetisch onderzoek. Indien een genetische verklaring voor de epilepsie is gevonden, is het belangrijk om terug te gaan naar de klinische presentatie van de patiënt. Er kan worden gekeken of er comorbiditeiten, geassocieerd met de genetische epilepsie, kunnen worden geïdentiﬁceerd die monitoring en behandeling vereisen om de prognose voor de patiënt en zijn/haar familie te verbeteren.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 205 525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 206 Dankwoord Curriculum vitae

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 207 208

DANKWOORD

Graag maak ik hier gebruik van de gelegenheid om jullie – begeleiders, collega’s, vrienden en familie – te bedanken voor jullie fantastische bijdrage aan dit proefschrift.

Als eerst gaat mijn dank uit naar mijn begeleiders; ik had er maar liefst vier. Dit betekent vier keer zoveel aandacht, wijsheid en adviezen (uiteraard niet altijd dezelfde…), maar uiteindelijk ook artikelen die vier keer beter waren dan hun eerste versie. Jullie hebben mij doen beseﬀen hoe leuk en waardevol een goede samenwerking en communicatie tussen de verschillende vakgebieden is, niet alleen voor wetenschappelijk onderzoek maar ook voor de klinische praktijk. Ik kijk met veel plezier terug op onze samenwerking.

Beste prof. van Ravenswaaij, beste Conny, wat ben ik blij dat jij mij hebt ondergedompeld in de revolutionaire wereld van de genetica. Ik wil je bedanken voor al je geduld om mij dit vreselijk ingewikkelde vak beetje bij beetje te leren. Want wat wist ik nu eigenlijk toen ik begon? Ik bewonder jouw toewijding aan het onderzoek, waarbij je altijd kritisch, innoverend en tegelijkertijd zo ﬁjn pragmatisch blijft. Dit, plus dat ik niet té hard moet werken (niet jou achterna gaan zoals je zelf altijd zegt), ga ik proberen mee te nemen in mijn verdere carrière.

Beste prof. Brouwer, beste Oebo, als ik de laatste maanden uw kamer in liep om even bij te kletsen memoreerde u vaak hoe ik mij de laatste jaren heb ontwikkeld als persoon, arts en onderzoeker. Dergelijke “speeches” – wijze adviezen altijd inbegrepen – kenmerken u ten voeten uit en heb ik altijd enorm gewaardeerd. Als ik het spoor even bijster was, wist u voor mij de hoofdzaken er feilloos uit te halen zodat ik weer verder kon. Ik denk met heel veel plezier terug aan de maandagochtend poli’s, de ﬁetstochtjes naar het Martini, de congresbezoeken en de overleggen bij de genetica. Uw enthousiasme over het vak kinderneurologie heeft u zeker op mij overgebracht; de rust en het vertrouwen wat u uitstraalt naar ouders en kinderen hoop ik mee te nemen.

Beste Petra, dat wij een zelfde manier van werken hebben werd al vroeg tijdens de stage wetenschap duidelijk. Onze overleggen verliepen daarom altijd verrassend productief en waren bovendien gezellig. Met het invullen van onze grote database heb jij mij geleerd kritisch naar epilepsie te kijken en hoe dit te classiﬁceren. Daarnaast wil ik je bedanken voor je betrokkenheid, ook bij wat mij persoonlijk bezig hield. Het feit dat we in Istanbul een kamer konden delen, ook toen ik ziek, zwak en misselijk was, typeert onze ﬁjne band.

Beste Patrick, van jou heb ik geleerd om nog kritischer te kijken naar de resultaten van mijn en andermans onderzoek. Soms is het daarvoor goed om even de tijd te nemen en even ‘uit de rijdende trein te stappen’, zoals jij mij eens zo mooi vertelde. Ik wil je bedanken voor dit waardevolle inzicht. Met jouw genetische ‘bril’ heb je mij veel geleerd over de interpretatie van genetische data. Ik zal de valkuilen nu niet meer vergeten; althans dat ga ik proberen. Dank voor al je hulp!

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 208 DANKWOORD 209

Dear Prof. Scheffer and Prof. Berkovic, thank you for welcoming me in the Melbourne Brain Centre in Australia and for letting me participate in your fantastic research clinic. My stay in Australia has been a wonderful experience. Dear Ingrid, I admire your passion for research and clinical work and how you integrate them in your research center; I really think you do make a difference for the patients and families. Thank you for teaching me about epilepsy classification and syndromology and writing good papers. I will not forget our great time in the clinic, discussions about research, finding exciting novel things and I will definitely not forget our long Skype-sessions in the night (Dutch time of course). It is a great honor that you will be present at my thesis defense in December. I am hoping to further collaborate in the future.

Beste prof. dr. H.P.H. Kremer, prof. dr. N.V.A.M. Knoers en prof. dr. K.P.J. Braun, hartelijk dank voor het lezen en beoordelen van dit proefschrift. Ik zie uit naar de verdediging op 5 december 2018.

Klinisch onderzoek is alleen mogelijk met het enthousiasme en de medewerking van patiënten en hun ouders. Mijn dank gaat daarom uit naar ieder van hen, die hebben deelgenomen aan mijn studies. A special thanks goes to all the parents from the SYNGAP1 facebook group. I totally agree with you that we should bridge the gap between clinicians and patients and I believe that your help in this is wonderful.

Beste Yvonne Vos en Trijnie Dijkhuizen, ik besef mij nu pas hoe beperkt mijn basale kennis over de genetica was toen ik mijn promotieonderzoek startte. Dank voor jullie geduld en uitleg om de basis van dit interessante vak beter te begrijpen.

Beste co-auteurs, hartelijk dank voor jullie bijdrage aan de artikelen. Lucia en Chania, het was erg leuk om jullie als student te begeleiden. Dank jullie wel voor jullie enthousiasme en inzet.

Dear Jackie Senior and Kate McIntyre, I am thankful for your careful reading and your well-taken suggestions that improved the readability of my work.

Lieve Nienke, waar zal ik beginnen? Je gezelschap tijdens college, de avondjes in de Tango en de Negende Cirkel, de berendokter ochtenden, de city tripjes, onze avonturen aan de Gorechtkade, het aan een blik of woord genoeg hebben, het delen van onze onderzoek successen en miserie, kortom; je vriendschap. Ik kan me 5 december niet anders voorstellen dan dat jij naast me staat als paranimf.

Lieve Wieke, ons wetenschappelijk avontuur begon samen in het aquarium nu toch al enkele jaren geleden. Naast onze gedeelde passie voor de kinderneurologie was het vooral ook heel gezellig om met jou te werken en zijn we inmiddels goede vrienden geworden, ook buiten de muren van het ziekenhuis. Ik ben blij dat jij mijn paranimf bent 5 december!

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 209 210

Allerleukste (oud)collega’s van V4, Anouk, Arnoud, Dan, Didin, Elze, Esther, Hans, Harmen, Jeannette, Jeﬀrey, Jonathan, Gerrit, Madelein, Marenka, Maraike, Marja, Marieke, Marlous, Marouska, Mayra, Myrthe de K., Myrthe S., Roald, Robbert, Sanne, Sterre, Sygrid, Tinka, Wieke en Zeus, kunnen we nog eens een dagje terug met zijn allen? De iets te rumoerige meetings voor de lunch om stipt 12 uur, de neuroklaas, de noodstroomborrels, de NRC puzzels in de middag, ons zwemclubje ‘de dolﬁjn’ (we kletsten soms meer dan we zwommen), de jaarlijks retreats (volgend jaar lustrum!); door jullie voelde werk eigenlijk nooit als werk aan.

Met een gecombineerde PhD op verschillende afdelingen, had ik ook het geluk van extra collega PhD’ers, waar ik altijd op terug kon vallen voor adviezen en tips. Dank jullie wel, Aafke, Barbara, Christa, Monica, Nicole en Renée (in liefdevolle herinnering). Also to my colleagues in Melbourne, many thanks for making me feel at home immediately. I had a great time in the Melbourne brain center, but especially during karaoke, the curling event, home cinema, Friday morning runs and winter markets. I am looking forward to coming back to Melbourne one day.

Lieve collega’s in het Martini Ziekenhuis, wat ﬁjn dat mijn nieuwe werkplek net zo leuk is! Nynke, ik wil jou speciaal bedanken voor het meedoen tijdens de aanvalspoli in het Martini Ziekenhuis. Dit was niet alleen een welkome afwisseling naast het (soms redelijk eentonige) onderzoekswerk; ik heb als beginnend dokter ook enorm veel geleerd. Ik hoop dat je, hoewel niet meer oﬃcieel, toch nog een beetje mijn mentor blijft.

Tot slot wil ik graag mijn lieve vrienden en familie bedanken. Gelukkig valt er ook naast de muren van het ziekenhuis (die je soms best kunnen opslokken), heel veel te beleven. En dat maakt het werken weer leuker. Dank dat jullie mij dit helpen herinneren. Mijn band met jullie is mij heel dierbaar.

Lieve Corinna, Elsa, Irene, Janna, Lian, Lisa en Susan, hoe bijzonder is het dat we al vrienden zijn vanaf de middelbare schooltijd. Ik ben trots hoe ieder van ons zichzelf ontwikkelt en tegelijkertijd blij dat we elkaar blijven vinden. Lieve Irene, als Groningen-maatje richt ik me graag speciaal tot jou; mijn PhD-tijd was nog veel leuker door onze onderzoek successen en frustraties samen te delen. Ik kijk uit naar jouw promotie!

Lieve Mare, Mirthe en Lisanne, na een avondje met jullie heb ik meestal drie dagen spierpijn van het lachen. Dat zijn de beste avondjes. We zijn al op veel plekken in de wereld samen geweest (Australië, Sri Lanka, Uganda, Tanzania, Malawi); ik kijk er naar uit dat we straks weer samen in Nederland wonen.

De vijf musketiers - Lieke, Thijs, Tommy en Nienke - mijn studententijd was niet half zo leuk geweest zonder jullie! Onze intellectuele (toegegeven; dokters praten toch wel veel over het ziekenhuis), maar vooral ook de niet-intellectuele gesprekken zijn erg ﬁjn.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 210 DANKWOORD 211

Lieve Christien, MD/PhD en Uganda maatje, dank voor je creativiteit dat heeft geresulteerd in de prachtige voorkant, die nu extra persoonlijk is geworden. Ik had hem me niet mooier voor kunnen stellen.

Lindsey en Anne, lieve en gekke (ex)huisgenootjes, dank voor heerlijke aanwezigheid thuis, zodat ik altijd mijn stoom kon afblazen.

Lieve Rony, Christine en Emma, dank voor jullie interesse in mijn onderzoek. Ik ben benieuwd wat jullie van een Nederlandse PhD-verdediging vinden.

Lieve Martijn, ook al leerde ik jou pas kennen halverwege mijn promotietraject vlak nadat ik besloot naar Australië te vertrekken; niets hield je tegen om mij achterna te vliegen. Het Australië avontuur werd hierdoor nog leuker. Dank je wel voor je spontaniteit, je liefde, je humor en je steun. Dit heeft me zeker geholpen bij de laatste loodjes van mijn promotie. Ik kijk uit naar veel nieuwe avonturen samen.

Lieve pap, mam en Lyan, dank jullie wel voor jullie onvoorwaardelijke liefde, vertrouwen en steun. Zonder dit ben ik nergens. Lieve Lyan, wat ﬁjn dat we alles met elkaar kunnen delen. Het hebben van een zus als jij is van onschatbare waarde. Ik wil je speciaal bedanken voor je humoristische peptalks als ik weer eens doordraai; ik zal ze nog eens nodig hebben (vrees ik). En toegegeven: de ﬁguren zien er nu heel mooi kleurrijk uit! Lieve Robin, dank je wel voor je betrokkenheid, je gevatte opmerkingen en je vrolijke aanwezigheid. Lieve pap en mam, dank jullie wel dat jullie mij aanmoedigen om mijn dromen achterna en dat ik altijd op jullie terug kan vallen. Ik ben trots op waar dit me nu al heeft gebracht. Pap, bedankt dat je me leert relativeren; ik besef steeds meer hoe belangrijk dit is. Mam, dank je wel voor je luisterend oor en je wijsheid; vaak ken jij mij beter dan ik mezelf.

Ik zie er naar uit om mijn promotie-traject met jullie te vieren!

Danique Vlaskamp.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 211 212

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 212 CURRICULUM VITAE 213

CURRICULUM VITAE

Danique Rienke Maria Vlaskamp werd geboren op 15 mei 1991 te Raalte. In 2009 behaalde zij cum laude haar gymnasium diploma aan het Carmel College Salland te Raalte en begon zij haar studie geneeskunde aan de Rijksuniversiteit Groningen. Tijdens de drie jaar durende bachelor werd Danique haar interesse gewekt voor kindergeneeskunde, in het bijzonder de kinderneurologie.

In 2012 begon Danique haar master geneeskunde met haar wetenschappelijk stage op de afdeling kinderneurologie, onder begeleiding van Prof. O.F. Brouwer en Dr. P.M.C. Callenbach. Uit deze stage vloeide de aanvraag voor een MD/PhD traject voort, waarvoor zij in 2013 werd aangenomen. Tijdens dit traject combineerde Danique haar coschappen in Groningen, Enschede en Uganda met het doen van promotieonderzoek. Zij deed haar semi-arts stage op de kinderafdeling in het Martini Ziekenhuis en haar verdiepingsstage op de kinderneurologie in het Universitair Medisch Centrum Groningen. In 2016 behaalde Danique cum laude haar master in de geneeskunde. Hierna resteerde nog anderhalf jaar onderzoek tijd, waarbij zij in 2017 zes maanden onderzoek heeft gedaan in Melbourne, Australië, onder begeleiding van Prof. I.E. Scheﬀer.

Momenteel werkt Danique als arts-assistent bij de kindergeneeskunde in het Martini Ziekenhuis te Groningen en hoopt zij in de toekomst klinisch werk in de kindergeneeskunde dan wel kinderneurologie te combineren met doen van onderzoek.

525699-L-sub01-bw-Vlaskamp Processed on: 29-10-2018 PDF page: 213