Epilepsy surgery evaluation in patients with refractory epilepsy: prospective application of routine genetic testing

Epilepsy surgery is the only treatment that can fully cure patients from their
seizures without the need for lifelong medication. During surgery, the
epileptogenic zone is removed or disconnected with sparing of surrounding
normal functional tissue. The chance of postoperative seizure-freedom is
maximized when the epileptogenic lesion is completely removed/disconnected.
Whereas in the past only patients with visible structural MRI- lesions
(considered causative for the seizures) were considered surgical candidates, we
now increasingly evaluate people with refractory epilepsy and normal imaging.
Currently 60-70% of these MRI-negative (but presumed lesional/structural)
patients are rejected for surgery, often after extensive and invasive
source-localizing diagnostics. Operated MRI-negative patients have a lower
chance of reaching seizure freedom {Tellez 2010}. Furthermore, the absence of a
histopathological abnormality (~8% of all operated patients) is a major
predictor of poor outcome {Wang 2013}. When evaluating MRI-negative patients
with refractory focal seizures for surgical candidacy, the epilepsy is either
caused by an *invisible* microstructural lesion or malformation of cortical
development *probably implying a high chance of postoperative seizure-freedom*,
by a genetic syndrome without an identifiable lesion *probably reflecting a low
chance of surgical success *, or by a combination of the two (e.g. tuberous
sclerosis). The crucial differentiation between people with operable and
non-operable epilepsy (i.e. between a presumed lesional and non-lesional cause
of seizures) requires new and reliable biomarkers. Pathogenic variants in novel
and known epilepsy genes are such biomarkers that are currently not routinely
implemented in the presurgical evaluation. This project will apply preoperative
genetic testing and evaluate the ability of (new) genetic biomarkers to
distinguish between eligible and non-eligible surgical candidates, predict
outcome, and extend current genetic knowledge of focal epilepsies. Importantly,
the decision whether * and if so, how * individual patients* results will guide
decision making will be left to the discretion of the clinical team.
The most paradigm-shifting discoveries in epilepsy genetics are currently made
in focal epilepsy (FE) and malformations of cortical development (MCD), where
causal germline and somatic gene mutations prove that FE often has a genetic
basis {Pang 2008}. In particular, mutations in genes that regulate the mTOR
pathway are now recognized as the most frequent genetic cause of FE, detected
in up to 9% of patients. We recently reviewed the literature on surgical
outcome in different genetic causes of refractory epilepsy. The most important
finding was the difference between the low rate of seizure-freedom (14%) in
patients with mutations in genes associated with channel function or synaptic
transmission, versus a higher surgical success (71%) in patients with mutations
in genes involved in mTOR pathways. {Stevelink, submitted 2017} This further
supports the assumption that studying genetic variations in patients with FE
can improve decision-making and ultimately patient selection in the presurgical
evaluation.

Primary Objective:
Our overall aim is to improve selection of patients for epilepsy surgery and
predict surgical candidacy with use of genetic test results as a biomarker. The
specific objectives
are:

Cohort I
1. Apply genetic testing for epilepsy gene mutations in MRI-negative patients in
the presurgical evaluation program of three large European centers experienced
in epilepsy surgery. We aim to improve selection of surgical candidates, by
testing for mutations in known epilepsy genes, that can either point towards
structural focal and thus surgical remedial causes of epilepsy, or towards
primary genetic non-lesional causes of epilepsy. Identification of genes in
individual patients may thus indicate either good or poor surgical candidacy.
2. Evaluate the clinical utility of genetic biomarkers in epilepsy surgery
programs. We aim to measure the impact of genetic testing on presurgical
evaluation and decision making in MRI-negative patients, as well as
postsurgical AED treatment regimen. The genetic results from each individual
patient will be provided to the multidisciplinary epilepsy surgery teams (MDT)
at the participating institutes.

Cohort II
3. Apply somatic mutation screening in resected tissue of patients who
underwent epilepsy surgery, to improve pathological characterization and
postoperative prognosis and treatment. We will also perform ultra-deep targeted
sequencing in a buccal sample of all patients who were operated to screen for
somatic mosaicism epilepsy gene mutations in order to improve the diagnostic
yield of genetic testing during the presurgical evaluation phase.

Hypothesis
We hypothesize that routine testing for germline epilepsy gene mutations in
MRI-negative patients with refractory epilepsy can improve the differentiation
between good surgical candidates (i.e. those with a focal structural cause of
seizures) and those who are less eligible for epilepsy surgery (particularly
those with non-lesional genetic epilepsy syndromes), ultimately improving the
rate of seizure-freedom after surgery and
preventing unnecessary invasive source localization techniques. In addition, we
hypothesize that the detection of somatic mutations in resected tissue can
improve post-surgery diagnosis, outcome prediction and postsurgical AED
regimens. Furthermore, we assume that detection of somatic mosaicism gene
mutations in saliva of operated patients with use of ultra-deep targeted
sequencing could improve the diagnostic yield of genetic testing during the
presurgical evaluation phase. This could ultimately improve the presurgical
differentiation between good ** presumed lesional focal epilepsy patients
** and poor surgical candidates.

Overall approach
We start this study with the presurgical phase in which we apply routine
genetic testing in MRI-negative patients and in resected tissue, demonstrate
proof-of-concept, and evaluate the effect of detecting genetic biomarkers on
surgical decision-making, lesion characterization, and outcome prediction. In a
parallel study, we will extend the scope of genetic biomarkers in a more
experimental phase to find novel biomarkers of focal epilepsy that will be
functionally validated and rapidly transferred to the clinical application
phase. For this study, a separate application for ethical approval will be
submitted.

General relevance
Seizure control is not reached in ~30% of all patients 12 months after epilepsy
surgery. Those with normal MRI scans have an even higher chance of poor
surgical outcome up to ~50%, because the preoperative differentiation between
patients with either focal lesional (structural) epilepsy, or with focal
genetic non-lesional epilepsy has proven to be difficult. It is of crucial
importance to improve the selection of suitable surgical candidates, and to
reduce the risk of unnecessary (deemed unsuccessful) invasive diagnostics and
resective surgery. We hypothesize that routine testing for the recently
discovered genetic causes of epilepsy in MRI-negative patients can improve the
recognition of poor surgical candidates and prevent further presurgical and
invasive diagnostics. Vice versa, demonstration of a genetic abnormality that
is likely to underlie a structural cause (such as FCD or mMCD) may improve the
recognition of patients who are good candidates for surgery, despite normal
MRI. Most importantly, in patients with refractory focal non-lesional epilepsy,
in whom semiology and ictal EEG findings are not concordant with other
functional imaging techniques to localize the epileptogenic zone (such as MEG,
PET, and SPECT), the demonstration of a genetic underlying cause of the
epilepsy syndrome should prevent invasive and expensive (further) presurgical
diagnostics, such as grid or SEEG monitoring. Currently, genetic studies have
presented the prelude for personalized medicine by demonstrating the potential
relevance of genetic mutations for focal epilepsy, yet these fundamental
discoveries have not been translated to improved care and clinical utility. Our
proposal will accelerate the implementation of genetic discoveries into the
epilepsy surgery program, by immediately applying genetic testing in
MRI-negative patents that are enrolled for surgery. By applying genetic
screening in clinical practice, and assessing prospectively and retrospectively
the impact on presurgical evaluation, we aim to provide the evidence needed for
long term application. The combination of basic and applied studies from at
least two different institutions, in collaboration with at least two renowned
foreign centers, will provide the clinical evidence needed to implement
molecular diagnostics in epilepsy surgery.

This study is a non-intervention study to explore etiology and prognosis and
will be conducted in two stages. The setting of the study is the Department of
Child Neurology and the Department of Genetics (Center for Molecular Medicine)
at the University Medical Center Utrecht. DNA will be collected from eligible
patients the moment they are enrolled in a presurgical evaluation program at
any of the participating European centers (UCL Great Ormond Street Institute of
Child Health & Great Ormond Street Hospital for Children, London, UK; Hospices
Civils de Lyon, Lyon, France), including the University Medical Center Utrecht
(Utrecht, NL). The duration of this study is estimated to be four years. In the
past years, the participating centers evaluated ~100 MRI-negative patients per
year for surgical candidacy, of whom ~25 underwent surgery. We therefore
expect that in four years 400 MRI-negative patients will be eligible. We
estimate that more than 75% of patients will participate in this study based on
previous experience, and thus we envision that 75 patients will be enrolled per
year (300 in four years).

Cohort I
During the first stage of this prospective study, DNA (blood sample by
venepuncture) will be collected from eligible patients the moment they are
enrolled in a presurgical evaluation program at any of the participating
centers. WES-based gene panel sequencing and array technology will be performed
at one of the contracted sequencing facilities (UMC Utrecht or BGI Genomics).
If similar genetic analysis is already performed as part of routine clinical
care, genetic diagnostics as part of the study will not be repeatedly
performed. The sequencing results will be analysed through the standard
diagnostic procedures, and a clinical report of detected significant genetic
findings will be discussed in the UMCU multidisciplinary work meeting between
geneticists, clinicians, and researchers.
The genetic results from each individual patient will be provided to the
multidisciplinary epilepsy surgery teams (MDT) at the participating institutes.
Importantly, the decision whether * and if so, how * individual patients* test
results will guide decision making will be left to the discretion of the MDT.
With a descriptive analysis, we will investigate how the genetic findings
influence the presurgical diagnostic trajectory and decision-making. We will
compare patients with and without genetic findings on: 1) process indicators
(duration of presurgical evaluation until decision, number of ancillary
investigations performed, need for/yield of/complications of invasive
procedures, number proceeding to invasive diagnostics and to resective
surgery), and 2) postoperative seizure outcome (according to Engel
classification, at 1 year after surgery).

Cohort II
We will prospectively perform DNA diagnostics * using a NGS DNA panel analysis
* in resected tissue of all patients who underwent surgery for drug-resistant
focal epilepsy (FE) and had MCD, GNT, or negative pathology in the UMCU. The
aim is to improve the pathological characterization and improve outcome
prediction and patient care.
Based on current practise, we expect to include 50 tissue samples per year,
totalling to 200 samples in four years. Furthermore, we will perform ultra-deep
targeted sequencing in a buccal sample of the same patients who underwent
epilepsy surgery to screen for somatic mosaicism epilepsy gene mutations.
Finally, for selected samples that showed no putative disease variant through
panel sequencing, we will perform WES sequencing to allow follow up for novel
epilepsy genes that are discovered during the course of this project, or
through collaboration with international consortia.
We will share results with our international partners through several genetic
consortia that are aimed at identifying epilepsy genes. By sharing our results
we will increase the likelihood of finding similar patients with mutations in
the same gene, which is necessary as evidence for causality.

The current proposal aims to describe the implications of genetic testing in
patients with refractory epilepsy who receive a clinical assessment for
epilepsy surgery, and to confirm the as well as elucidate the association
between different subtypes of gene mutations and surgical candidacy. The burden
of patients in both cohort studies is minimal: standard clinical care in these
patients commonly includes a blood draw by venepuncture. For those patients
participation in the study, advantage is taken of the blood draw for clinical
purposes by drawing one extra tube of blood for the purpose of the study. In
patients who do not acquire blood draw for clinical purposes, the additional
burden of the study is a blood draw by venepuncture solely for the purpose of
the study. For the cohort II study, the burden of participation consists of
collection of a buccal swab.
All in all, the overall burden of participation in this study is low, while the
associated risk is negligible (vene puncture for DNA blood draw or buccal
swab). There is no clear benefit associated with this study, however knowledge
obtained from this study will contribute to the decision making and outcome
prediction in patients with focal refractory epilepsy who are evaluated for
epilepsy surgery. The inclusion of minors in the current study is required and
justified; meaningful results can only be acquired through this group-related
design.