REVIEW
www.advancedscience.com
Electrotherapies for Glioblastoma
Elise P. W. Jenkins, Alina Finch, Magda Gerigk, Iasonas F. Triantis, Colin Watts,
and George G. Malliaras*
Non-thermal, intermediate frequency (100–500 kHz) electrotherapies present
a unique therapeutic strategy to treat malignant neoplasms. Here, pulsed
electric fields (PEFs) which induce reversible or irreversible electroporation
(IRE) and tumour-treating fields (TTFs) are reviewed highlighting the
foundations, advances, and considerations of each method when applied to
glioblastoma (GBM). Several biological aspects of GBM that contribute to
treatment complexity (heterogeneity, recurrence, resistance, and blood-brain
barrier(BBB)) and electrophysiological traits which are suggested to promote
glioma progression are described. Particularly, the biological responses at the
cellular and molecular level to specific parameters of the electrical stimuli are
discussed offering ways to compare these parameters despite the lack of a
universally adopted physical description. Reviewing the literature, a
disconnect is found between electrotherapy techniques and how they target
the biological complexities of GBM that make treatment difficult in the first
place. An attempt is made to bridge the interdisciplinary gap by mapping
biological characteristics to different methods of electrotherapy, suggesting
important future research topics and directions in both understanding and
treating GBM. To the authors’ knowledge, this is the first paper that attempts
an in-tandem assessment of the biological effects of different aspects of
intermediate frequency electrotherapy methods, thus offering possible
strategies toward GBM treatment.
E. P. W. Jenkins, Dr. M. Gerigk, Prof. G. G. Malliaras
Division of Electrical Engineering
Department of Engineering
University of Cambridge
Cambridge CB3 0FA, UK
E-mail: gm603@cam.ac.uk
Dr. A. Finch, Prof. C. Watts
Institute of Cancer and Genomic Science
University of Birmingham
Birmingham B15 2TT, UK
Dr. I. F. Triantis
Department of Electrical and Electronic Engineering
City, University of London
London EC1V 0HB, UK
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/advs.202100978
© 2021 The Authors. Advanced Science published by Wiley-VCH GmbH.
This is an open access article under the terms of the Creative Commons
Attribution License, which permits use, distribution and reproduction in
any medium, provided the original work is properly cited.
DOI: 10.1002/advs.202100978
1. Introduction
Glioblastoma (GBM), a type of brain tu-
mour, is an aggressive, debilitating disease
with poor prognosis and limited changes
to the standard of care since 2005 despite
substantial research efforts.[1] Currently,
surgical resection where possible followed
by radiotherapy and adjuvant chemother-
apy,Temozolomide (TMZ)is often consid-
ered as the optimal treatment strategy. Ra-
diotherapy (RT), also known as radiation
therapy, uses high energy electromagnetic
fields or radioactive substances to directly
or indirectly damage DNA of tumour cells
to halt or slow tumour growth.[2] Temo-
zolomide (TMZ) is an alkylating agent ap-
proved for anti-cancer treatment of GBM
that is capable of crossing the blood-brain
barrier (BBB). Simply, it aims to dam-
age DNA and trigger cell suicide.[3] How-
ever, depending on the expression of the
MGMT (O-6-methylguanine-DNA methyl-
transferase) gene, GBM cells can repair
the DNA damage and become resistant to
treatment resulting in poor therapeutic ef-
ficacy and inevitable recurrence.[3] Recent
efforts that focus on cancer drug delivery
techniques to overcome the BBB include encapsulated nanoparti-
cles, drug delivery vehicles, convection enhanced delivery,[4] and
immunotherapy strategies to enhance T-cell infiltration.[5] De-
spite these efforts, the complexity of the disease results in treat-
ment failure, disease progression and fatality.
While pharmaceutical interventions, as well as, the study of
chemical and mechanical aspects of biology and medicine have
been the focus of therapeutic and diagnostic research and prac-
tice since antiquity, only glimpses of the therapeutic effects of
electricity were documented early on, notably through the use of
electric fish for pain relief by Scribonius Largus circa 47CE.[6] It is
mainly in the last few centuries that bioelectric aspects of diagno-
sis and therapy have been considered, the first through biology’s
electrical properties and signals and the latter through electrical
and electromagnetic interventions.[7] Electrical stimulation via
bespoke apparatus was demonstrated by Galvani over 200 years
ago, causing frog legs to twitch. The electrophysiology era con-
tinued well into the 1800s to date where scientific breakthroughs
led to auditory, visual, olfactory stimulation, cochlear implants,
cardiac pacemakers, functional electrical stimulation and deep
brain stimulation.[8] Akin to pharmaceutical interventions, be-
ing subject to dosage quantity, timing and frequency as well as
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (1 of 28)
www.advancedsciencenews.com www.advancedscience.com
Figure 1. Stages and potential biological modulation of electrotherapies. A) Pulsed electric fields (including nanosecond-pulsed electric fields (nsPEFs),
reversible electroporation and IRE techniques) targeting GBM are in pre-clinical validation withmost studies in vitro and others in canine and rat models.
These techniques typically involve mono- or bi-phasic waveforms. B) TTFs is FDA approved to treat new and recurrent GBM in combination with current
standard of care. It typically involves sinusoidal AC fields at 200 kHz. C) Potential mechanisms (from left to right): Neuron signaling contributes to GBM
growth which feeds back to increase neuron excitation. PEF techniques can target ion channels, lipid bilayers, blood-brain barrier opening, and other
intracellular mechanisms (not shown). TTFs target mitotic cells through dipoles and DEP forces. Created with BioRender.com
application route and location, the efficacy of electrical interven-
tions depends on the waveform, amplitude and frequency of the
applied signal as the topological aspects of the active electrodes.
Alternating current (AC) stimuli in the low frequency spectrum
<10 kHz when applied to living tissue can excite cells by depo-
larizing the cell membrane. Increasing the applied frequency be-
yond 10 kHz toward a few MHz (herein referred to as intermedi-
ate frequencies) were long believed to have insignificant cellular
effects. At higher frequencies, tissue heating phenomena are ob-
served, and this is exploited in many tissue ablation techniques,
most notably, radio-frequency tumour ablation.
For the purpose of establishing nomenclature, applied voltages
are herein denoted by V (in Volts) and electric fields by E (Volts
cm−1). In electrotherapy (see Figure 1), non-thermal intermedi-
ate frequencies treatments include pulsed-electric fields (PEFs)
which range from nanosecond to millisecond pulsing durations
and tumour-treating fields (TTFs) ranging from 100 to 500 kHz.
PEFs can be used to induce electroporation,[9] a technique
whereby electrodes are placedwithin target tissue to induce nano-
sized pores in cell membranes, increasing their permeability and
thus susceptibility of drug substances (e.g., for electrochemother-
apy (ECT)[10]) or delivery of DNA (electrogenetherapy[11]) allow-
ing for targeted pharmaceutical intervention. When the applied
field E is substantial (i.e., 1 kV cm−1), irreversible breakdown
of the cell membrane occurs leading to cell death (IRE).[12] Al-
ternatively, TTFs are non-invasive and use fields of E ≈ 1–3 V
cm−1 which are claimed to interfere with mitosis (although this
is disputed as indicated later) where dipole molecules like tubu-
lin dimers of microtubules align with the applied field leading
to improper polymerization and early metaphase exit. In addi-
tion, cells in the final stage of mitosis (telophase) are subject to
non-uniform E that create dielectrophoretic (DEP) forces moving
particles toward the furrow prompting DNA damage and cellu-
lar suicide.[13–15] In the treatment of GBM, electroporation tech-
niques are still in pre-clinical stages, while TTFs are Food and
Drug Administration (FDA) approved for newly diagnosed and
recurrent disease. Thesemethods have shown very promising re-
sults, with electroporation mostly demonstrated to affect cancer-
ous cells pre-clinically,[16] while TTFs has been approved for clini-
cal use to treat GBM (NCT00916409). Still, the first is not yet fully
proven for GBMwhilst the latter is not fully accepted by themed-
ical community, mainly due to the lack of understanding of its
underlying mechanisms and controversial clinical trial data.[17]
Both techniques target specific aspects of the tumour and thus
they are challenged by the multi-faceted nature of GBM much
like the challenges faced in bespoke pharmaceutical remedies.
Therefore, there is a need for better understanding of the effect
of each of the aforementionedmethods to biologicalmechanisms
that have a direct remedial impact on GBM. To be more precise,
there is a need for “mapping” the effects of each of the reported
stimulus parameters of these techniques to corresponding bio-
logical aspects of GBM. This is particularly challenging due to the
increased biological complexity related to the poorly understood
“multiforme” or heterogeneous nature of GBM tumors,[18,19] as
described later. As the authors have noticed, this is coupled with
an additional difficulty from the lack of consistency in literature
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (2 of 28)
www.advancedsciencenews.com www.advancedscience.com
when it comes to describing the physical characteristics of the ap-
plied electrotherapeutic signal. For example, some papers men-
tioning field strength of extracranial fields while others report in-
tracranial fields (i.e., volume conductors,[7] or applied voltage or
current amplitudes) without a clear “translation” from one to the
other, rendering certain studies incomparable. Additional incon-
sistencies come from differences in electrode topologies, as well
as their material or their proximity to the targeted tumour.
Finally, given the biophysical understanding of the cell mem-
brane potential and electrophysiology, it seems intuitive that elec-
trotherapy techniques could exploit the electrical characteristics
of glioma. While this may be the case, it appears there is an inter-
disciplinary disconnect at the interface of neuroscience, biology,
and applied physics/engineering that is limiting the pace of pro-
gression in this field, thus, we propose research questions that
require urgent consideration.
2. Addressing the Biological Challenge
2.1. Glioblastoma Multiforme
The central nervous system (CNS) is built up of neurons and
glial cells, with the latter primarily serving the needs of the first
by providing scaffolds during the development of the nervous
system.[20,21] Glia can also produce brief electric currents by open-
ing calcium (Ca2+) channels influencing many neurons almost
simultaneously and affecting neurotransmitter release, hence
contribute to the coordination of synaptic activity.[20,22,23,21] Glial
cells are usually divided into three categories: astrocytes, oligo-
dendrocytes and microglial cells. Astrocytes exhibit a variety of
shapes and appear to serve a homeostatic function. They also
surround blood capillaries and form very extensive tight junc-
tions between endothelial cells, thus decreasing the permeabil-
ity of brain capillaries and helping to establish a fully functional
BBB.[24–26]
Tumors that originate in glial cells are called gliomas and
they account for more than 70% of all brain tumors.[18] The
most frequent adult glioma (65%) is GBM, which develops from
astrocytes.[19,27,28] TheWHOclassifies GBMas a grade IV astrocy-
toma, accounting for ≈12–15% of all intracranial neoplasms and
60–75% of astrocytic tumors.[19,27,18] Combining the histopatho-
logical features of GBM—nuclear atypia, vascular thrombosis,
microvascular proliferation, mitotic activity, cellular pleomor-
phism, and necrosis[19,27]—with a putative cancer stem-likecell
subpopulations and a plethora of epigenetic and genetic lesions,
makes this type of cancer one of the most multifaceted human
tumour, and therefore it is incredibly difficult to treat.[19]
As implied by the moniker “multiforme,” GBM is character-
ized by a widespread inter (different between tumors) and intra-
tumoral (within the same tumour) heterogeneity.[18,19] Hetero-
geneity can present in multiple different ways within the same
tumour including molecular, metabolic, microenvironmental,
and vascular heterogeneity. Devastatingly, these factors can vary
within the same tumour leading to regional variations in ther-
apy response and cellular behaviors. This makes clinical man-
agement and long-term survival a much greater challenge and is
confounded by the lack of treatment options available. Overall,
there is extensive intra-tumoral heterogeneity displayed by GBM
and that this occurs at the genetic, metabolic, and microenviron-
mental levels which we later discuss (see Section 7.1.). Brief ex-
amples include: drastic variations in gene expression at spatially
distinct regions of the same tumour,[29] mosaicism in the expres-
sion of various receptor tyrosine kinases (RTKs) leading to mul-
tiple mechanisms of furthering their proliferation,[30] variation
in the vasculature leading to areas of high and low angiogenesis
with varying rates of perfusion,[31] and alterations to gene expres-
sion within the GBM microenvironment, potentially leading to
improved capacity for invasion.[32]
Inevitably, GBM recurs even following gross total resection
and the standard of care. This is due to a lack of specificity in the
treatments combined with the highly adaptable behavior of GBM
cells allowing them to become resistant to treatment. Recurrence
has been suggested to occur following the survival of brain tu-
mour initiating cells which are capable of re-populating tumour
mass and heterogeneity.[33] Taken together, we show that GBM
presents a unique challenge for researchers and clinicians in or-
der to understand how aspects of heterogeneity, BBB, resistance,
and recurrence can be targeted via electrotherapy techniques.
3. Electrophysiology
The nervous system has recently been implicated to play a critical
role in cancer progression for tumors occurring within the CNS.
Contrary to long standing belief, non-neuronal derived cancerous
cells including prostate,[34] glial and glia-derived glioma among
others retain the expression of voltage-gated ion channels[35]
and neurotransmitter receptors.[36] While these cells do no ex-
hibit true action potentials,[37] they do display oscillations at the
membrane potential (Vm) which, when depolarized, are func-
tionally relevant for glia-mediated neurotransmitter uptake and
release[38] and proliferation.[39] Hence, it is unsurprising that
the communication between neurons and cancer cells is a key
pathophysiological trait of gliomas.[40] Electrophysiological un-
derstanding of GBM is paramount for the progression of tar-
geted electrotherapies, therefore, we briefly discuss recent devel-
opments of their involvement in glioma.
3.1. Ion Channels
Ion channels are responsible for cellular homeostasis and regu-
late the influx and efflux of ions required for cell metabolism and
function. Glioma cells exhibit vast expression of ion channels in-
cluding sodium (Na+), potassium (K+), calcium (Ca2+), chloride
(Cl−), and transient receptor potential cation channels. Blockade
of a range of ion channel classes has been demonstrated to reduce
the viability of glioma stem-like cells (GSCs) in vitro thus provid-
ing more avenues to explore for novel, targeted therapies.[41]
K+ channels are responsible mostly for large efflux of K+ from
the intra-cellular space which aids the cell in maintaining a neg-
ative resting potential. They are classified as inward rectifying K+
channels (Kir), Ca
2+ activated K+ channels and voltage-gated K+
channels (Kv) which form the largest group of ion channels in the
membrane. Kv channel expression is altered across various can-
cers and is well known for roles in cell proliferation and neoplas-
tic progression.[42] Specific examples of ion channels involved in
GBM include human ether-a-go-go related gene (hERG), Kv3.3,
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (3 of 28)
www.advancedsciencenews.com www.advancedscience.com
and PIEZO1. hERG, encoded by the KCNH2 gene, is a delayed
rectifier K+ channel most commonly found in the heart where it
is critical to the rapid repolarization of the ventricular action po-
tential. Mutations in or pharmacological blockade of the hERG
channel can result in fatal arrhythmias stemming from altered
channel function.[43] In GBM, increased expression of hERG cor-
related with worse survival in a tissue microarray whilst patients
treated with inhibitors of hERG showed better survival than those
who did not. Interestingly, patients receiving multiple hERG
blockers showed greater survival than patients just receiving one
which indicates that there ismerit in increasing the level of hERG
inhibition.[44] Systemic administration of hERG blockers would
present a risk of patients developing potentially fatal arrhythmias,
therefore novel drug delivery mechanisms may be of benefit but
need to be carefully considered. Kv3.3, encoded by the KCNC3
gene, is also a delayed rectifier K+ channel which has a role in re-
polarizing rapid fire action potentials in the cerebellum.[45] It has
been shown that higher channel expression correlates with bet-
ter patient survival—note this is in opposition to hERG channels
indicating that simple K+ ion flux across the membrane is not
the driving force of malignancy. Studies on differentially methy-
lated regions found that the KCNC3 genewas hypermethylated in
GBM samples compared to controls which typically indicates si-
lencing of gene expression.[46] Interestingly, another study found
that KCNC3 expression was enriched in the GSC compartment
but not in the bulk GBM tissue.[41] Taken together, this informa-
tion suggests that overall tumour expression of KCNC3 is low,
but that it is specifically enriched in GSCs which are known to be
involved in insidious tumour processes such as local GBMmetas-
tasis, resistance and recurrence (see Section 7.4.3.). A final exam-
ple lies in PIEZO1, a mechnosensitive channel that permits the
movement of cations across the membrane in response to me-
chanical forces on the membrane.[47] PIEZO1 has been found
to have higher expression levels in high grade gliomas (WHO
grade III and IV) than in low grades (WHO grade II), further
to this the PIEZO1 gene was downregulated in isocitrate dehy-
drogenase (IDH) mutant gliomas irrespective of grade. This was
later found to be due to genetic hypermethylation in IDH mu-
tant gliomas as part of the wider epigenetic glioma CpG island
methylator phenotype (G-CIMP) signature. Further in vitro test-
ing showed that genetic knock-down of PIEZO1 in GBM cell
lines caused a reduction in growth whilst the same experiment
conducted in GSCs caused an inhibition in the sphere-forming
capabilities of the cells.[48]This data indicates that PIEZO1 plays
a role in the growth rate of glioma cells. Additionally, it was also
shown that high expression levels of PIEZO1 correlated with a
much shorter survival time and that this may serve as a novel
prognostic marker.[49] It should be noted that the positive corre-
lation between high PIEZO1 expression and growth rate may not
be strictly causative since the increase in PIEZO1 expression is
likely preceded by earlier genomic events such as IDHmutation.
However, PIEZO1 still holds an important role in controlling tu-
mour cell growth and has the advantage of being targeted either
pharmacologically to reduce ionic flux or mechanically to alter its
signaling capability.
Ion channels present a unique opportunity as drug targets, as
they are membrane-bound and therefore easily accessible from
the extracellular space. Previous studies have highlighted that tar-
geting specific K+ channels with inhibitors can increase suscep-
tibility of GBM cells to TMZ treatment, thus potentially helping
to overcome resistance to TMZ.[50]
Both Cl− and Ca2+ have been suggested to facilitate migra-
tion behavior of GBM cells. Additionally, oscillatory changes
in intracellular Ca2+ have been hypothesized to initiate GBM
invasion.[51,52] Mutations in Na+ channels of GBM samples
have shown to have shorter survival than mutations in K+ or
Ca2+. In particularly, epithelial Na+ channels which fall into
amiloride-sensitive Na+ channels are associated with prolifera-
tion and invasion in many cancers.[53] Glioma ion channels[35]
and inhibitors[39] used as a treatment modality have been re-
viewed. Combination of channel inhibitors with electrotherapy
could lead to greater understanding of the cellular mechanism
due to induced electric fields.
Unlike neuronal excitability which can be detected via multi-
ple techniques, the excitability of glioma cells is difficult to detect
with high selectivity due to theminute signals they present. Ultra-
sensitive platforms have been developed[37] and will be crucial for
understanding ion channel regulation thus electrophysiology of
glioma.
3.2. Neuron-Glioma Interactions
In high-grade gliomas, progression is vigorously regulated by
neuronal activity (Figure 1). Since neurons are crucial com-
ponents of glioma microenvironments and regulate activity-
dependant malignant growth, it was suggested that gliomas
may also engage in synaptic communication which could be
fundamental to its progression.[54] Validating this hypothesis,
Venkatesh et al.[38] show that neuron-glioma interactions in-
volve electrochemical communication via bona fide 𝛼-amino-
3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors
at the neuron-glioma synapses. They further show that the in-
terconnections form an electrically coupled network since neu-
ronal activity evoked non-synaptic activity-dependent K+ cur-
rents. Using in vivo optogenetics, the authors showed that de-
polarized gliomamembranes promoted proliferation while phar-
macological/genetic blockers of electrochemical signaling inhib-
ited growth. Ultimately, their findings postulate that synaptic and
electrical integration into neural circuits promote glioma pro-
gression.
As neuronal activity promotes glioma growth, gliomas too in-
crease neuronal activity in preclinical models of GBM.[55–57] Elec-
trocorticography recordings in awake human subjects with cor-
tical GBM confirm neuronal hyper excitability in the disease-
infiltrated brain compared to adjacent parenchyma.[38] Interest-
ingly, the primary excitatory neurotransmitter, glutamate, is im-
plicated in the bidirectional interactions between neurons and
glioma cells.[58] Thus, increased extracellular glutamate in and
around adult gliomas have been reported. It is therefore plausi-
ble, that the elevated extracellular glutamate could explain com-
mon clinical seizures associated with GBM.[55,56,59,60] The take-
away message is that neurons within the tumour microenviron-
ment and glioma cells are reciprocally engaged in a feedback loop
as the tumour grows. Thus far, the literature suggests that glioma
progression is regulated by electrochemical driving forces from
interconnected neurons, however, neuron’s role in glioma initia-
tion remains an unanswered question.[58]
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (4 of 28)
www.advancedsciencenews.com www.advancedscience.com
4. Pulsed Electric Fields
PEFs are used in several biological, medical, environmental,
and food processing applications. Short duration (1–10 µs pulse
width), mono- or biphasic waveforms have been used to kill
pathogens and spoilage microorganisms in vegetal or animal
tissue,[61,62] as well as, extraction of cellular compounds and
growth stimulation in fungi, soy, microalgae, and other cells.[63]
The biological effects induced by electric fields had been observed
and studied since the early 1700s,most notably Jean-AntioneNol-
lets’ “electric boy” experiment. In recent decades, PEFs have pro-
vided a non-thermal tissue ablation treatment technique for a
variety of malignant neoplasms. Broadly, this modality,termed
electroporation, can provide both reversible and irreversible ef-
fects to biological cells depending on the applied field parame-
ters. The terminology used in this field is typically described by
the intended use of electroporation[64] which is practiced across
multiple modalities as we discuss. For simplicity, in this section
we introduce electroporation generally, before describing applica-
tions of electroporation,ECT and nsPEFs, and how these are used
in glioma treatment. Finally, we introduce IRE protocols and how
they have been applied in gliomas to date.
4.1. Electroporation and Electropermeabilization
Electroporation variations in cancer treatment are biophysical
tissue ablation techniques where rod/needle-like electrodes are
strategically placed directly in or surrounding the lesion of in-
terest. Electroporation can be an invasive modality and there-
fore limited by surgical intervention. Nonetheless, appropriate
placement of one or multiple electrodes which are selectively
stimulated with appropriate combinations of field strength and
pulse duration and can accordingly interfere with a cells Vm,
which sits anywhere from −50 to −90 mV in non-proliferating
glia.[65] When the strength of E is large enough to induce a
Vm ≈ 0.25–0.5 V, the membrane of the cell becomes unstable
and nanoscale pores are introduced. Here, the effect is consid-
ered reversible while an induced Vm ≈ 1 V results in irreversible
damage. The electroporation phenomenon was termed by Neu-
mann and Rosenheck.[9] Mechanisms here are discussed in brief
as detailed discussion on distinguishable mechanisms of electro-
poration and electropermeabilization are already available.[66]
4.1.1. Mechanisms
The extent of electroporation can be described according to the
applied electric field’s parameters used to induce changes in Vm.
The cell membrane is comprised of a tightly bound lipid bilayer
which allows for a semipermeable ion transfer of smallmolecules
between intra- and extracellular spaces. Larger molecules are
transferred via ion channels or transporters. As illustrated in
Figure 2, during low to medium field strength and/or respec-
tive long/short duration electroporation (e.g., 8 pulses of E ≈
1.3 kV cm−1, with pulse width ≈99 µs for reversible ECT[67]) wa-
ter molecules begin to penetrate the lipid bilayer inducing un-
stable hydrophobic pores (stage 1). During stage 2, the lipids re-
orientate their polar head groups causing metastable hydrophilic
pores throughout the cell membrane.[64] Eventually a state of
electropermeabilization is reached, allowing small molecules to
traverse the membrane, whose permeability increases due to a
range of physical or chemical mechanisms triggered by the ap-
plied electric field, such as modulation of membrane protein
functions.[66] The extent of electroporation and thus of electrop-
ermeabilization can be quantitatively estimated and predicted ac-
cording to changes induced inVm.When the strength of E is sub-
stantially increased with a relevant pulse width to induce a Vm ≈
1 V, the membrane breakdown reaches an irreparable state lead-
ing to cell death.
4.1.2. Quantitative Descriptions
Induced changes to a cellsVm due to an applied electric field were
first described mathematically by the Schwan equation which
considers a spheroidal cell shape.[68] In 1988, Glaser et al.[69] de-
scribed the formation of pores induced by electric breakdown
of the lipid bilayer as a function of pore radius and time. Fur-
ther, in 1999, Neu and Krassowska[70] introduced the asymptotic
model of membrane electroporation according to the Smolu-
chowski equation governing the distribution of pores to describe
hydrophilic pore formation occurring beyond a critical radius.
The required energy to form critical pores which lead to mem-
brane rupture decreases with increasing Vm.
[71]
Using similar approaches today, Kotnik et al.[66] describe that
even when the Vm exceeds 450 mV, if the pulse duration is short
enough, and Vm returns to 0 V prior to pores expanding beyond
20 nm, irreversible breakdown can be avoided. Overall, mem-
brane porosity is dynamic and numerical solutions suggest that
reversible field strength sub-microsecond pulses will induce mil-
lions of pores at ≈1 nm while longer pulses at similar strengths
produce up to tens of thousands of pores in a cell but at a much
greater pore radius ≥10 nm.[66,72] The prediction of these mod-
els strongly correlates with experimental results and this lay the
foundation of experimental planning.
Throughout the literature, there is poor consistency across
stimulation protocols which make it difficult to compare results.
Previous reports emphasize the effect of electroporation is pri-
marily dominated by two parameters: E (thus applied voltage and
distance between electrodes) and exposure duration (pulse width
in nanoseconds tomilliseconds), where strength-duration curves
were used as a metric to identify pulsing parameters (U.S. patent
no. US8282631B2). However, since different cell types have dis-
tinct electrical properties and sizes, these curves should be ap-
propriately calibrated. Additionally, responses have shown to vary
with temperature[73] and osmotic pressure.[74] More generally, the
waveform of use is particularly important for the target applica-
tion as shown in Figure 3. What is not clear in current literature
is how the variation of electrode materials, sizes, and shapes ef-
fects the field distribution. The resulting current flowing through
the electrodes would be a useful parameter to include in future
literature, as well as, the electrochemical impedance spectra of
the electrodes and effective (potential) electrochemical reactions
that may occur for the durations used in experiments much like
the expectation in neuromodulation studies. A description of the
electrical coupling (direct, capacitive, inductive or a combination)
should be included. Further details are provided in Section 8.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (5 of 28)
www.advancedsciencenews.com www.advancedscience.com
Figure 2. Stages of electroporation. Short but intense electric fields (induced Vm≈0.5 V) force water molecules to penetrate and disrupt the lipid bilayer
in the cell membrane leading to unstable hydrophobic pore formation (stage 1). The hydrophilic heads of the lipids begin to reorientate to form a
metastable pore of ≈2 nm wide allowing small molecules to enter the cell (stage 2). Depending on the strength and duration of the applied field,
modulation of ion channels/transporters have been reported. When the field strength is very intense (Vm ≥ 1 V), the cell membrane breaks down due
to lack of homeostasis and several cell death pathways have been suggested including pyroptosis, necrosis and apoptosis. Created with BioRender.com
4.2. Electrochemotherapy
ECT in oncology is an application in which electroporation is
used to introduce cytotoxic drugs to malignant cells. Although
the specific mechanisms in antitumor effectiveness in ECT treat-
ments are not fully understood, the basic understanding is that
the reversible permeabilization of the cell membrane enhances
the cytotoxic uptake of chemotherapeutics.[75,76] The first demon-
stration of ECT was performed in 1987,[10] and began routine use
in 2006.[77] ECT has been studied both pre-clinically and clini-
cally for a range of cutaneous, sub cutaneous and deep-seated
tumors including liver, lung, skin, ovary, prostate, pancreas, col-
orectal, head and neck, and bone.[64,78] In the brain, the chal-
lenges arise mainly due to the excitable tissue and intraopera-
tive difficulties. As wemention in our earlier sections, aggressive
brain tumors like GBM are heterogeneous and are either inher-
ently or progress to become resistant tomolecularly targeted drug
therapies. In addition, the BBB which is designed to limit toxic
perfusion presents an overwhelming task for poorly permeable
cytotoxic drugs to reach the tumour.
ECT presents an alternative mechanism for difficult-to-treat
tumour cells by making them more permeable and using elec-
trophoretic type driving forces. For clinical ECT, patients are
typically administered with an intra-tumoral or intravenous
chemotherapeutic agent that spreads throughout the tumour
vasculature.[78] Following this, the electrodes are placed in or
around the tumour according to the treatment planning soft-
ware of use (for example, Pulsar coupled with Cliniporator VI-
TAE) and deliver short, defined bursts to electrophoretically drive
molecules into the target permeabilized cells. Typical protocols
for ECT use between 2 and 24monopolar pulses of 100 µs–20ms
durations with respective therapeutic field strengths.[79] After a
short period of time (seconds to minutes depending on the pulse
parameters) the membrane begins to reseal and the drug can ex-
ert its cytotoxic effects. As mentioned earlier, TMZ is typically
the physician’s choice chemotherapy for GBM since it is able to
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (6 of 28)
www.advancedsciencenews.com www.advancedscience.com
Figure 3. Various electrical regimes used in electroporation techniques. a) ECT are typically monophasic, long in duration (up to 20 ms) with tolerable,
reversible field strengths. b) nsPEFs are biphasic and use the shortest duration of all electroporation techniques thus require substantial field strengths
to achieve electropermeablization or irreversible effects. However, in this regime, most effects are observed intracellularly. c) IRE traditionally involves a
short duration (≈100 µs), monophasic regime at lethal field strengths. d) High-frequency IRE (H-FIRE) are second generation IRE waveforms involving
biphasic short duration bursts at lethal thresholds to induce cell death. e) H-FIRE regimes with asymmetric inter-pulse and inter-phase delay.
penetrate the BBB. However, Bleomycin and Cisplatin are
charged drugs currently impermeable to the brain but have
demonstrated high therapeutic efficacy in alternate tumourmod-
els when used with ECT.[80] Thus, permeating the BBB creates
many explorative opportunities for drug therapies that are suc-
cessful in alternate tumour models. To date, the main methods
of ECT in the brain target BBB disruption rather than tumour in-
fused electroporation. We discuss the findings of BBB disruption
in Section 5.3.
Preclinical validation of intra-tumoral ECT in rat glioma mod-
els resulted in 9 of the 13 rats showing tumour regression (69%
complete response rate).[16] The ECT protocol in this study was
delivered first with 42 IU of bleomycin contained in a 14 µL pump
and the electric field was delivered at 4 repetitions of 8 pulses of
100 V, of 100 µs pulse width every 1 Hz. The authors explored the
safety of ECT in the brain on healthy rats (tumour free). Subse-
quent MRI data demonstrated that the treatment area was sub-
jected to necrosis and revealed fluid-filled cavities highlighting
that some risks may include oedema, infection or haemorrhage.
An additional consideration would be risk of exciting surround-
ing tissue prompting epileptic events. While this work provides
encouraging evidence for the use of ECT in the brain, future stud-
ies must mitigate and overcome these risks by optimizing elec-
trode designs, field distributions, and drug dosages.
Large electric field amplitudes at low frequencies would be
expected to cause hydrolysis or electrochemical reactions which
create additional complications in the brain. However, Bonakdar
et al.[81] showcased no chemical reaction or bubbling which they
suggest is likely due to the short pulse durations and low currents
(100 µs, ≈2 mA). Moreover, intracranial tumour electroporation
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (7 of 28)
www.advancedsciencenews.com www.advancedscience.com
electrode geometry and performance optimization have been
studied,[82] and neurosurgical techniques have been reviewed.[83]
4.3. Nanosecond Pulsed Electric Fields
Conventional electroporation-based techniques involve electric
fields delivered in the micro- to millisecond range with rel-
ative understanding of cell membrane disruption. The sub-
microsecond regime, nsPEF, has shown to induce intracellu-
lar effects (i.e., nucleic membrane disruption).[84] Although elec-
tropermeabilization is induced by nsPEF, the strength and du-
ration of the pulse can lead to nano-sized pores and electropo-
ration effects.[85–87] Given these parameters, cell electroporation
and permeabilization can be reversible and membrane recovery
can be in the range of minutes while intracellular repair may take
hours.[66] Electropermeabilization has previously been described
by Kotnik et al.[66] to occur in stages: Initiation, expansion, partial
recovery, membrane resealing, and memory.
Cell death mechanisms via sub-microsecond pulses initially
showed to be apoptotic in various cell lines and tumour tissue
including through caspase activation,[88] evidence of indicated
intracellular Ca2+ release,[89] loss of mitochondrial membrane
potential,[90] and DNA damage.[91] Interestingly, while excess
Ca2+ is known to causemicrotubule instability,[92] Carr et al.[93] re-
cently demonstrated in human U-87 GBM cell lines, that nsPEFs
affect the microtubule network by a mechanism that is indepen-
dent of intracellular Ca2+ concentrations or osmotic swelling. In
addition, nsPEF have shown to effect tumour growth[94] and vas-
cular perfusion[95] in vivo. For detailed discussion of nsPEF in-
duction of cell deathmechanismswe refer the reader to this study
by Beebe et al.[96]
Since the CNS combines both excitable and non-excitable tis-
sue, the safe use of electroporation-based protocols in the brain
requires a well-designed approach. Recently, a comparison by
Dermol-Cˇerne et al.[97] show for pulses of 10 ns and 10 ms du-
ration, plasma membrane depolarization thresholds in both ex-
citable and non-excitable cells (including GBM), indicated that
excitable cells require greater field strengths to depolarize than
non-excitable cells in vitro. Similarly, depolarizationwas achieved
at lesser field strength by the increasing pulse duration. Quies-
cent neuronal cells exhibit Vm ≈ 90 mV as mentioned, proliferat-
ing cancer cells display depolarised Vm compared to their healthy
counterparts.[65] It is therefore likely that various cancer cells, par-
ticularly, those expressing various voltage-gated ion channels are
susceptible to depolarization at lower field intensities and thus
appropriate stimulus parameters would allow for selectively ad-
dressing them without affecting surrounding healthy neuronal
tissue.
Rapid membrane depolarization of neurons was shown by
Pakhomov et al.[98] to occur within 1 ms of 200-nsPEF electropo-
ration and that voltage gated Ca2+ channels did not contribute to
depolarization. However, the opening of voltage gated Na+ chan-
nels which peak 4–5 ms after nsPEF application could lead to ac-
tion potential generation. Consistent with other studies, nsPEF
amplitudes above 1.5–3 kV cm−1 induces electroporation effects
and at these thresholds there is risk of evoking action potentials.
The suggestion in these cases is that longer duration pulses (100
µs–10 ms) allow time for voltage gated ion channels to respond
and ultimately reduces the risk of firing action potentials.
Burke et al.[99] recently observed in U-87 GBM cells that there
is a direct interaction between nsPEF (single 10 ns, 34 kV cm−1
pulse) and the activation of voltage-gated ion channels that in
turn have downstream effects on non-voltage dependent chan-
nels (see figure 12 in ref. [99]). Ion channel modulation studies by
nsPEF are ongoing in the field of neurostimulation and cancer
treatment. These results suggest that nsPEF protocols below elec-
troporation thresholds present opportunities that may be thera-
peutically beneficial for cancer treatment and should be consid-
ered in future studies.
Ultimately, electropermeabilization will also depend on cell
size, cell orientation (with reference to the applied electric field)
and cell density.[100,101] Comparing the literature, we find lethal
thresholds are inconsistent and vary across applications and ex-
periments. This could be attributed to the expression of ion chan-
nels, intracellular and extracellular electrical properties, the rest-
ing state of Vm or simply the electroporation protocol used.
4.4. Irreversible Electroporation
IRE is a predominately non-thermal tissue ablation technique
that uses lethal field strengths and pulse durations to induce per-
manent nano sized pores which destabilize cellmembranes caus-
ing lack of homeostasis and inevitable cell death.[12] It was pio-
neered by Davalos and Rubinsky (U.S. patent no. US8048067B2)
in 2003 and is now routinely used to treat various tumors.[102]
While traditional IRE protocols are relatively safe and demon-
strate efficacy for cancer treatment in a variety of organs, the
translation to human brain cancer still requires additional re-
search for safety and efficacy. We later discuss the “second gen-
eration” protocol, H-FIRE for treating brain cancers.
4.4.1. First Generation Irreversible Electroporation
There are two generations of IRE. The first-generation IRE,
Nanoknife, was FDA approved for human soft tissue tumors
(2008) and commercialized for research purposes in 2009. It
includes one or multiple electrode probes that are inserted di-
rectly in or near the site of a tumour where ablative pulses
are implemented to induce cell death while sparing critical
structures.[103–106] IRE traditionally uses short monophasic E de-
livered in the target tissue for 25–100 µs pulse width (see Fig-
ure 3) at lethal intensities typically greater than 1 kV cm−1 to in-
duce a Vm ≥ 1 V. These fields are typically delivered repetitively
every 1–4 Hz, to reach substantial irreversible tissue damage.
The mechanisms in which IRE induce cell death is suggested to
be apoptotic[107–110] and caused due to the breakdown of the cell
membrane. Further, numerical studies show that the size of ab-
lation areas can be estimated to inform adequate treatment plan-
ning. Real-time procedure monitoring can be performed via ul-
trasound and confirmed ablation with MRI.[111,103] IRE has been
considered a non-thermal treatmentmodality since the cell death
mechanism is not primarily caused by tissue heating, yet, that
does not exclude it from rising temperatures in tissue. Sano
et al.[112] show in vitro brain tumour models that for pulse du-
rations on the order of 1 µs, IRE is thermally mediated.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (8 of 28)
www.advancedsciencenews.com www.advancedscience.com
4.4.2. Pre-Clinical Validation of Irreversible Electroporation
While Nanoknife has been used in various in vivo experiments,
few relevant glioma models exist. In 2009, Garcia et al.[113]
produced the first feasibility results of focal ablation (using
Nanoknife) in vivo canine brain tissue using 10 × 50 µs pulses
between 500–2 kV cm−1 delivered every 4 Hz (a total of 9 times).
Although the data was preliminary, this work provided the prac-
ticality of non-thermal tissue ablation in the brain. The complica-
tions included oedema, seizures or bleeding (due to needle inser-
tion) all of which have suggested overcomingwith corticosteroids
and a hyperosmolar agent (oedema) or prophylactic anticonvul-
sant (seizures). Later, IRE procedures in the brain were further
corroborated by Ellis et al.[114] in 4 normal canine brains demon-
strating that the volume of ablation correlates with the strength
of E. The post-operative histopathologic and ultrastructural as-
sessments of this work detailed by Rossmeisl et al.[115] reveal that
at 72 h following treatment, cell death is primarily mediated by
necrosis. Further, the authors found that no significant activity of
caspase-3 or caspase-9 was evident, and therefore apoptosis was
not a significant contributor to cell death following IRE.
The first demonstration of using IRE treatment for an in-
operable, spontaneous malignant intracranial glioma in a ca-
nine patient was published in 2011 by Garcia et al.[116] Since
canine malignant gliomas present similarly to GBM in clini-
cal, biologic, pathologic, molecular, and genetic properties, they
are considered an excellent translation for human brain tumour
treatment.[117] The pre- (1.36 cm3) and 48 h post- (0.35 cm3) IRE
delivery resulted in an average 74.2% reduction of tumour vol-
umes. This work indicated tolerability and successful safe tu-
mour ablation with adjunctive radiotherapy, anti-oedema treat-
ment, and anticonvulsants with minimal exacerbating haemor-
rhage.
Most recently, traditional IRE protocols were further studied
by Rossmeisl et al.[118] utilizing the Nanoknife procedure in 7
dog glioma models. The mean pre-treatment tumour size was
1.9 ± 1.4 cm3 with dog 2 and 5 presenting with GBM. The re-
sults demonstrated that in 6 of the 7 dogs, IRE treatment was
achieved without inducing/exacerbating oedema or significant
haemorrhage and produced a median survival of 119 days post-
operative treatment.[118] While most adverse effects were mini-
mized or in line with typical post-operative surgery, one dog ex-
perienced severe cerebral oedema. The tumour location in this
case was close to the periventricular regions much like the com-
mon site of occurrence in human glioma and the oedema was
a result of excessive field strength in IRE. This emphasizes the
consideration of tumour location and potential effects in the pre-
treatment and planning of IRE in the brain.
4.5. High Frequency Irreversible Electroporation
H-FIRE[119] (second-generation IRE) delivers short, fast bursts
of substantialstrength, biphasic electric fields into/around the
region of interest. These short, microsecond (1–10 µs) pulsing
regimes delivered in a series of bursts achieve the same ener-
gized “ON” time as a single monopolar 100 µs pulse used in tra-
ditional IRE but requires much greater field strengthto achieve
the same lesion size.[120,121] H-FIRE was introduced to overcome
some of the existing challenges found in IRE. By utilizing the
same energized time as IRE, direct comparisons can bemade be-
tween protocols that are implemented in clinically relevant mod-
els. In addition, the waveform regime does not induce muscle
contractions[122,123] as previously shown in IRE treatments.[124]
Numerical work by Dermol-Cˇerne et al.[97] have also shown
through the Hodgkin-Huxley model why H-FIREmay permeabi-
lize but not excite tissue. Finally, the H-FIRE bursts are typically
delivered every 1 s corresponding to the clinical system delivery
rates that are synchronized with patient heart rates.[125]
An example of a symmetrical H-FIRE waveformwould involve
a positive phase pulse width of 2 µs, a 2 µs inter-phase delay,
followed by a 2 µs negative pulse repeated until the total aver-
age energized (non-zero) time would equal 100µs (see Figure 3).
This can be repeated for a desired number (5–200 times every
1 s) and is represented in the literature as 2-2-2 H-FIRE wave-
form. In some cases, asymmetric waveforms may be used where
the phase durations, inter-phase delays or inter-pulse delays vary.
While biphasic pulses achieve similar biological effect to IRE, the
inter-pulse delay is described to be an important aspect of the
delivery. A phenomenon termed the “cancellation effect” where
the opposing polarity phase of a pulse cancels the effect of the
first phase,[126] has been observed in both nano and microsecond
pulse ranges and is not fully understood.[127–129,120,130] Vinzinitin
et al.,[126] show that longer inter-phase and inter-pulse delay be-
tween biphasic pulses resulted in more significant cell death in
Chinese hamster ovary cells. However, the cell membrane per-
meabilization effect was not as obvious. It is possible to consider
that longer inter-pulse and inter-phase delays are more effective
as it would allow for dissipation of the formed electrical double
layer at the electrode-cellular interface. Sano et al.[79] also demon-
strated that asymmetric pulses reduced the lethal threshold to
induce cell death in both U-87 (human GBM) and MDA-MB-231
BR3 (human brain metastasis from breast) cells when compared
to equivalent energy symmetric waveforms.
4.5.1. Pre-Clinical Validation of High Frequency Irreversible
Electroporation
The cellular mechanism induced by H-FIRE is relatively unex-
plored. First, H-FIRE induced cell death dynamics were found
to result in both immediate and delayed cell death.[131] Mercadel
et al.[110] studied IRE and H-FIRE protocols in 3D models of ade-
nocarcinoma cell line BxPC-3 (pancreatic cancer) in collagen I
hydrogel-based scaffolds. Comparing 100 µs monopolar IRE and
variousH-FIRE protocols with different pulse length/ inter-pulse
delays (all of which had a total energized time of 100 µs per burst)
the authors indicate that cell death dynamics were consistent
with accidental cell death (ACD—instantaneous uncontrollable
cell death cause by extreme physical, chemical, or mechanical
disturbance) and regulated cell death (RCD—cell death as a re-
sult of active signaling transduction modules that are modulated
pharmacologically or genetically). For H-FIRE specifically, they
showed a reduced fraction of cells undergoing ACD compared to
IRE and suggested that H-FIRE has sufficient control over select-
ing the ACD or RCD mechanism via the electric field delivery.
A parametric analysis by Sano et al.[131] on the field strength
versus pulse width duration effects in H-FIRE with bursts from
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (9 of 28)
www.advancedsciencenews.com www.advancedscience.com
0.25 to 50 µs showed an inverse correlation between the pulse-
width and toxicity in vitro. Despite that equal energy quantities
were delivered in each burst the authors suggest that this could
be due to membrane charging times. In addition, they demon-
strate that the delay between positive and negative polarity pulses
(phase delay) has negligible effect on Vm but significantly affects
the nuclear envelope potential. Their numerical results indicate
that cells of similar size but with higher nucleus-to-cytoplasm
ratio (NCR) will achieve greater effects on the nuclear envelope
potential than those of a smaller NCR. Subsequently, the NCR
and cell membrane permittivity will determine the nuclear enve-
lope charging characteristics. Since malignant cells, particularly
infiltrative glioma cells, are often expressed with higher NCR,[132]
these findings suggest an element of selectivity when using H-
FIRE regimes.
IRE procedures have shown preservation of critical struc-
tures and major blood vessels in humans which is an advan-
tage of this technique over microwave or radiofrequency ablation
methods.[12,133] As H-FIRE protocols are known to produce rapid
and reproducible ablations,[134,106] Siddiqui et al.[135] evaluated
how the presence of critical vascular and biliary structures in vivo
porcine liver models were affected by H-FIRE protocols. While
histological examination indicated no visible collateral damage
to adjacent structures, Hematoxylin and Eosin (H&E) staining
revealed that endothelial cell damage/shedding was present par-
ticularly in vessels located proximal to the site of the electrode in-
sertion. For translation into the brain, this is an important remark
detailing the significance of predictable lesions and preservation
of critical tissue.
Tissue heating is another significant parameter for transla-
tion into the brain since temperature changes may cause pro-
tein denaturation, oedema and seizures. Fesmire et al.[136] eval-
uate the temperature dependence in 3D tumour models (both
U-118 MG human GBM and Panc-1 pancreatic cancer) compar-
ing Nanoknife IRE (NK-IRE) and H-FIRE pulses to aid clinical
treatment protocol development. The main observations were
that lethal thresholds for NK-IRE were consistent across the tem-
perature range (2–37 °C) and that the ablation zone increased by
7% across the ranging temperatures. Alternatively, H-FIRE was
strongly dependent on the treatment temperature where an in-
creasing temperature resulted in a decrease in lethal thresholds
and increase in ablation volumes.
In brain specificmodels performed in canines, Garcia et al.[201]
provides a detailed explanation of the therapeutic planning for
IRE and Latouche et al.[235] for meningioma treatment using H-
FIRE. In the H-FIRE case, patient specific plans were developed
by MRI tissue segmentation, volumetric meshing and finite el-
ement modelling. Each patient’s therapeutic procedure was cus-
tomized with a 3D generated patient and tumour specific output
depicting the expected electric field distribution, Joule heating
and electrode configuration to be used in treatment (for further
details see their Supporting Information).
As detailed in Section 4.4.2., Nanoknife was used to irre-
versibly electroporate spontaneous gliomas in caninemodels and
adverse effects were observed in 1 canine. The advancement ofH-
FIRE protocols has since presented an opportunity to overcome
these effects. Latouche et al.[137] used H-FIRE treatment in a fol-
low up experiment to selectively ablate intracranial meningioma
in 3 canines. MRI scans confirmed solitary mass lesions greater
than 1 cm in diameter with characteristics of meningioma. 6
month follow up revealed canine 1 alive, seizure free, and no ev-
idence of tumour, while canine 2 was alive, it required escalation
of anticonvulsants to control seizure activity, and was suspected
to have residual or recurrent tumour presenting inMRI 5months
post treatment. Due to recurrent status epilepticus, canine 3 died
in 76 days. In this feasibility study, no post-operative adverse ef-
fects attributed to H-FIRE were observed. Muscle/nerve excita-
tion or cardiac arrhythmia were not evident during treatment cor-
roborating the advantages of H-FIRE over traditional IRE, partic-
ularly in the brain.[122,137] This study provided the first evidence of
organ and indication specific feasibility of H-FIRE for brain tu-
mour ablation and presents an exciting translation opportunity
in the near future. Further H-FIRE studies have since been per-
formed in awake standing horses by Byron et al.[138] for superfi-
cial tumors. In this application, H-FIRE was delivered at 2 µs, up
to 3100 V, to horses that were treated without general anesthesia,
neuromuscular blockades or cardiac—impulse synchronization.
Tumour volumes evaluated by physical and ultra-sonographic ex-
amination indicated a reduction in the mean, 68.8%, SD± 31.39;
after 2–4 treatments.While sample size and post-operative follow
up where limited in this study, the clinical potential in soft tissue
was demonstrated. This work further supports the safety of H-
FIRE translation in the field of cancer treatment.
Continuing evidence supports the use of H-FIRE as a novel
and feasible treatment modality for difficult-to-treat cancers.
These efforts have led to the first in human trial of H-FIRE
for the treatment of prostate cancer (NCT03838432).[139] The
success of the canine ablation models due to the consider-
able efforts in H-FIRE research provides prospective treatment
for brain cancer and in particularly, GBM, where no cure ex-
ists. While significant work has led to the first in human trial,
there still exists many challenges to successful treatment out-
comes. Some of these challenges include: real-time field moni-
toring, demonstration of safety in human, examination of pro-
tocols invoking seizures, intraoperative difficulties in surgery.
Further understanding of therapeutic variations across different
GBM subtypes, microenvironmental changes, efficacy on cur-
rently resistant or recurrent cell types, immune response, and
combination therapy could greatly advance momentum in this
field.
5. Tumour-Treating Fields
TTFs is a relatively recent treatment for management of GBM
which uses low strength (E≈ 1–2V cm−1) intermediate frequency
(100–500 kHz) AC electric fields by extracranial application. Con-
trary to the mostly experimental aforementioned electroporation
techniques, this is a method that has entered the clinical domain.
Clinically, this treatment has been approved by the U.S. FDA un-
der the commercial name Optune (Novocure) for the treatment
of both newly diagnosed and recurrent GBM. Its developers and
those supportive of the method theorise that TTFs exhibit anti-
tumour effects by targeting dividing tumour cells whilst sparing
other cells in the brain that are not undergoing division. It is sur-
prising how little TTFs mechanism of action is explored beyond
physical mitotic disruption, especially given what is known about
electrophysiology in this field.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (10 of 28)
www.advancedsciencenews.com www.advancedscience.com
Figure 4. Single cell level mechanism of action of TTFs. TTFs targets mitotic cells inducing, DNA damage, replication stress, and mitotic arrest leading
to cell death. Small arrows indicate increase (upward) or decrease (downward). Created with BioRender.com.
5.1. Discovery
From the initial introduction of the concept in the publication by
Kirson et al. in 2004,[15] the research has focused on investigation
of the antimitotic effects of TTFs. Mitosis is known to be a pro-
cess which occurs in precisely choreographed stages (prophase,
prometaphase, metaphase, anaphase, and telophase). Aiming
to ensure that a single cell divides into two genetically identi-
cal daughter cells, the stages of mitosis must be executed with
exquisite fidelity. It has been shown that the antimitotic effect is
selective to proliferating cells, and has minimal impact on non-
proliferating cells.[15] TTFs perturb cells in mitosis resulting in
plasmamembrane contractions and instability and the formation
of plasma membrane blebbing.[15] Additionally, TTFs can affect
motility and assembly of intracellular macromolecules during
metaphase of the cell cycle, which, in homeostasis, those intracel-
lular macromolecules are required for mitotic spindle formation.
In consequence, the disruption leads tomitotic catastrophe, chro-
mosomal breakage forming micronuclei, and cell death.[15] This
is thought to be a consequence of the effect of the field on po-
lar macromolecules like tubulin dimers during polymerization-
depolymerization responsible for assembly and disassembly of
microtubules. Although questioned, the force moment acting
on tubulin dimers during this process is sufficient to interfere
with proper assembly. During the cleavage formation, the elec-
tric field distribution is no longer homogeneous creating DEP
force on intracellular charged and polar particles drawing them
toward the center of the furrow causing cell destruction and
suicide.
5.2. In Vitro Approaches to Investigating the Mechanisms of
Action
A series of in vitro studies reported that TTFs inhibited prolifera-
tion and killed tumour cells, including melanoma, glioma, lung,
ovary, prostate, and breast cancer cells.[14,15,140] These studies ob-
served that the electric field frequencies capable of inhibiting pro-
liferation are dependent of the sizes and shapes of cells.[15] Fo-
cusing strictly on publications concerning GBM, we notice that
most of the research has been conducted using the inovitro sys-
tem, which was developed by Novocure to aid investigation into
mechanisms of action of TTFs, as well as, finding therapeutical
combinations adjuvant with TTFs. There are only a few publica-
tions describing research where primary GBM cells were used in
in vitro setups, with themajority of authors relying on 2D culture
of GBM cell lines. Therefore, although these models can provide
information regarding general mechanism by which TTFs work,
it is difficult to predict the efficacy of the treatment in terms of its
impact on various cell types and even tumour subtypes within the
tumour mass and tumour microenvironment. Nonetheless, tak-
ing together the information from the available publications on
the effects of TTFs on GBM cells, we can group and summarize
the described mechanisms (Figure 4).
As mentioned, TTFs can disrupt the alignment of several cel-
lular structures, including the spindle structure and contrac-
tile ring. Those disruptions often happen during various cell
cycle phases—anaphase, telophase, and cytokinesis—and pre-
vent cytoplasm separation. In consequence, this leads to apop-
tosis. The programmed cell death can be induced via either
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (11 of 28)
www.advancedsciencenews.com www.advancedscience.com
p53-dependent or p53-independent pathway.[141,142] Considering
that TTFs present the ability to target several cell cycle phases,
therapies that disrupt Gap1/synthesis (G1/S) or Gap2/mitosis
(G2/M) phases could be combined to enhance treatment. The
type of cell death induced upon TTFs treatment may differ
between tumour entities and cell lines.[140] In U-87 and U-
373 human GBM cells, TTFs induced autophagy, and sup-
pressed autophagy may attenuate anticancer effects.[143] Further,
Akt2/mTOR/p70S6K axis are crucial pathways underlying TTF-
induced autophagy.[143]
Multiple studies have observed the correlation of cell doubling
time to TTFs efficacy both experimentally and numerically.[140,144]
This remark corroborates the results by Giladi et al.,[140] which ex-
perimentally demonstrates the correlation between TTFs efficacy
and cell doubling time, thus for any treatment duration, the effi-
cacy is greater on rapidly dividing cells than quiescent cells.
Interestingly, Neuhaus et al.[145] and Li et al.[146] have suggested
that TTFs have an effect on Vm. Since Vm of tumour cells are
typically depolarised and fluctuate during proliferation,[147] TTFs
could potentially induce considerable change to Vm resulting in
reduced cell counts. According to Li et al.[146] this may explain
why healthy glial cells, which feature resting Vm ≈ −90 mV are
unaffected by the relative change in Vm potentially induced by
TTFs. Experimentally, Neuhaus et al.[145] demonstrated that TTFs
were able to activate CaV1.2 channels in GBM cells. They also
demonstrated that clonogenic survival varied across GBM cell
lines and that Cav antagonist may augment the therapeutic ef-
fect.
If a tumour cells induced Vm is depolarised considerably, the
consequent changes in ion concentration due to downstream ac-
tivation of voltage-gated ion channels and transporters could ex-
plain the abnormalities expressed in mitotic cells from TTFs ex-
posure. Conversely, if TTFs exposure induces hyperpolarization,
this may explain why cells no longer divide and reduced cell
counts are observed since depolarization is believed to initiate
mitosis and DNA synthesis.[65] Moreover, the fluctuations of Vm
through-out phases of the cell cycle andmitosis could also explain
why different effects are observed. For example, in many cells,
depolarisation halts G1/S checkpoint, remains relatively hyper-
polarised during S phase, and finally the G2/M transition again
exhibits depolarization.[65]
5.3. In Silico Single-Cell Approaches to Understanding the
Mechanism of Tumour-Treating Fields
Single cells can be modelled using finite element software
like COMSOL Multiphysics to investigate potential underlying
physics of specific TTFs induced mechanisms. Regardless of
whether the model includes a single cell, tumour-like features or
full-scale head models, the same principle to estimate field dis-
tribution is considered. Electric field distributions can be approx-
imated according to a volume conductor model,[7] where electro
quasi-static approximations of Maxwell’ electrodynamics equa-
tions can be applied and wave propagation terms ignored. This
approximation is valid since the wavelength of the applied TTFs
frequency in tissue is much larger than the size of the human
head. From these assumptions, the electric potential can be com-
puted using Laplace’s equation, Equation (1).
∇ × 𝜎 ∇V = 0 (1)
where ?̃? = 𝜎 + i𝜔𝜀 is the complex conductivity, 𝜖 is the permit-
tivity and 𝜔 = 2𝜋f is the angular frequency.[142] The boundary
conditions typically assume that all interior boundaries have con-
tinuity of the normal component of the current density and ex-
ternal boundaries are electrically insulating.[148]
While TTFs is both field strength and frequency dependant,
many studies have also indicated that the direction of the field
changes the therapeutic effect for cells that are in telophase.[4]
Hence, for electric fields that are applied directionally parallel
to the cell division orientation (where the furrow is parallel to
field direction), the field strength at the furrow is greatest pro-
ducing maximum DEP force. Wenger et al.[142] indicates that the
optimal frequency (which they define by when the strength of E
is strongest) is cell cycle stage dependant. For example, for the
cell modelled to be in metaphase, the strength of E is maximum
at approximately 10 MHz, however, for a cell modelled in late
telophase, maximum E occurs at ≈100 kHz. This finding sug-
gest that the delivery of TTF should consider sweeping a range of
frequencies relative to cell cycle progression. While this is yet to
be validated experimentally, a consideration of cell cycle stage de-
pendency should also include the corresponding changes in Vm.
Wenger et al.[142] also show how variations in cell size will change
the exposure to TTFs, in particular, as the cell size increases, the
efficacy is reduced.
The mechanism that TTFs interrupts the mitotic spindle for-
mation in early stages of mitosis is said to be unlikely since the
forces exerted by the electric field strength are insufficient to
cause cytoskeleton disruption.[146,142,149] As explained in single
cell studies by Tuszynski et al.,[149] within the cytoplasm, electro-
static charges would be screened over the distances greater than
the Debye length and that the simplified force for unscreened
charged according to F = qE, (where E is static 1 V cm−1) would
result in 10 or 0.5 pN if Debye screening is accounted for. Sim-
ilarly, in the work presented by Li et al.[146] further show, taking
random thermalmotions into consideration, the torque and force
induced by TTFs at E≈ 2 V cm−1 are nearly four orders of magni-
tude smaller than thermal motion energy. This too suggests that
the forces are not comparable to that of Brownian motion and
would be ineffective to disrupt the tubulin dimer alignment dur-
ing mitosis.
While direct effects on microtubule polymerization are less
likely to be contributing to TTFs induced cell death during
prophase/metaphase, telophase and cytokinesis disruption is
said to be the main mechanism for cells that progress. The as-
sumptions of homogenous field distributions in tissue vary dras-
tically between electrotherapies making comparisons difficult.
Since the electric field distribution within a cell in late cytoki-
nesis is largely inhomogeneous, theoretically, significant DEP
forces will develop and cause polarizable particles to move as a
result of the induced non-uniform field acting on their dipole
moments.[149]
Li et al.[146] studied DEP forces arguing that previous theoret-
ical studies neglect the inclusion of cytoplasm viscosity and that
Stokes drag forces would strongly influence the movement of
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (12 of 28)
www.advancedsciencenews.com www.advancedscience.com
macromolecules toward the furrow. With this inclusion, the sim-
ulation shows at 1 and 10 min time points, no clear indication of
particle movement toward the furrow. Since telophase represents
a small percentage (2–5%) of the typical mitotic duration (12 h),
these results, although not experimentally validated, indicate that
macromolecule movement is not the cause of cell death in TTF.
5.4. Tumour-Treating Fields in Clinical Practice
Novocure’s Optune system consists of 4 patches containing an
array of 9 transducers (36 electrodes in total) that are placed on
a shaved head. Each pair of transducer array rests across the
left and right, temporal and parietal areas (LR array) and the
anterior-posterior pairs across the supraorbital and occipital re-
gions (AP array). These arrays are positioned optimally accord-
ing to the treatment planning software, NovaTAL (NovoCure ltd),
and deliver biphasic electric fields with a typical 200 kHz cen-
ter frequency at therapeutic thresholds estimating E ≥ 1 V cm−1
at the tumour location. The commutation time for each elec-
trode pair is 1 s on the left-right (LR) pair, followed by 1s on the
anterior-posterior (AP) pair repeatedly. The portable device is rec-
ommended to be worn for at least 18 h per day for optimal out-
comes.
5.4.1. Clinical Trials
The two pivotal clinical trials which had sizable patient popu-
lations and focused on outcomes directly related to TTFs were
EF-11 (237 patients) and EF-14 (695 patients),[150–152] and the pa-
tient registry dataset (PRiDe).[153] The EF-11 clinical trial was de-
signed to investigate TTFs as a monotherapy versus chemother-
apy for recurrent GBM patients, and the primary efficacy end-
point was overall survival (OS). The EF-14 studied TTFs for
newly diagnosed GBM patients, and the primary efficacy end-
point was progression-free survival (PFS). During the EF-11 trial,
chemotherapy was chosen for the control arm due to lack of an
established standard of care for recurrent GBM. The primary
endpoint, OS, was not superior in the TTFs arm compared to
chemotherapy (median 6.6 months vs 6.0 months).[150] After re-
ceiving FDA approval, the efficacy of TTFs was also assessed in
the PRiDe, a large post-market registry which included all recur-
rent GBM patients who began TTFs between October 2011 and
November 2013.[153] The study did not specify the start date for
measuring the overall survival from, though the report said that
the OS was 9.6 months which in comparison to the 6.6 months
reported for the TTFs arm of EF-11 was increased.
The demonstrated tolerability of TTFs from EF-11, provided
the basis for EF-14, a phase III randomized clinical trial in-
vestigating TTFs with maintenance TMZ versus maintenance
TMZmonotherapy.[152] In this trial, TTFs plusmaintenance TMZ
demonstrated significantly prolonged survival compared to TMZ
monotherapy—median PFS was 6.7 months versus 4.0 months,
and median OS was 20.9 months versus 16.0 months, respec-
tively. Efficacy of TTFs was similar across age, Karnofsky Perfor-
mance Status, MGMT methylation, extent of resection, or geo-
graphic location. Thus, no prognostic indicators were identified
for subpopulations of patients that may receive greater benefit
from the treatment. Overall, the significant increase in PFS and
OS demonstrated in EF-14 resulted in approval of TTF therapy
for newly diagnosed GBM patients.
5.4.2. Computational Head Models
Computational models offer a cheap and efficient way to esti-
mate the electric field distribution and informoptimal transducer
placement during treatment planning of TTF. Asmentioned, No-
vaTAL is the proprietary treatment planning software for Optune,
and it creates personalized electrode placements based on patient
specific data. It was not included as part of the EF-14 clinical trial
but was approved following a clinical system user study.[154,155]In
bespoke systems, properties for the scalp, skull, cerebrospinal
fluid (CSF), grey matter (GM), and white matter (WM) can be
obtained according to segmentation of MRI (isotropic) or diffuse
tensor MRI (anisotropic). In addition to MRI based techniques,
water-content electrical property tomography has also been used
for mapping brain tissue conductivities in the intermediate fre-
quency range 200–1000 kHz.[156] GM and WM surface meshes
(most typically using anisotropic conductivity tensors[157]) are cre-
ated in software’s like the SimNibs pipeline while other tissue
meshes can be modelled using the Brainsuite package. Mim-
ics are an additional software package used to create virtual tu-
mors and to correct bumps or holes in the total mesh. Complete
meshes are imported into common finite element analysis pack-
ages to solve the electric field distribution . These models are de-
veloped to be reflective of the NovaTAL system.[148]
Recently, Lok et al.[158] compared NovaTAL and generated be-
spoke transducer array positions for cerebellar GBM. Shifting lat-
eral arrays backward and posteroanterior arrays to the lower oc-
cipital and upper cervical regions revealed superior field cover-
age at the tumour location to the NovaTAL-generated position-
ing. Similarly, Korshoej et al.[159] showed that oblique orienta-
tions where the operational electrodes are separated by 45° rel-
ative to the sagittal plane, produced greater field strength across
tumour locations than standard LP-AP orientation.
Following the phase 3 EF-14 trial, Urman et al.[160] used com-
putational modelling to correlate the PFS and overall (OS) sur-
vival of 119 patients to the relative field distributions. Their mod-
els show that when at least 95% (E95) of the combined volume of
the gross tumour volume and proximal boundary zone achieved
E ≥ 1.3 V cm−1, PFS was 11.9 months and OS was 33 months
compared to PFS 7.5 months and OS 21.9 months for E ≤ 1.3 V
cm−1. Ultimately, these results suggest that greater E strength at
the tumour location corresponds to improved patient outcomes.
For further reviews specific to computational headmodelling, we
refer the readers to Bomzon et al.[161] and Wenger et al.[142]
Several conclusions can be drawn from computational mod-
els, but not without limitations. Most commonly, relating diffuse
MRI data to stereotactic in vivo conductivity measurements with
specific emphasis on the heterogeneous properties would create
more accurate and predictable results. Reducing the time con-
straint and complexity of creating patient specific models would
greatly assist physicians involved in treatment planning and per-
haps improve adoption in the clinic.
While TTFs have demonstrated efficacy in vitro, in vivo and
in clinical trials, and is an approved treatment protocol for both
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (13 of 28)
www.advancedsciencenews.com www.advancedscience.com
newly diagnosed and recurrent GBM, there is skepticism and
apprehension from both researchers and physicians. Skepticism
arises due to incoherent results, with certain clinicians taking a
wait it out approach.[162] As discussed by Wick’s review,[163] the
main source of skepticism is the lack of a clear mechanism of ac-
tion in complexmodels and localized tissue. As discussed in both
single cell and more complicated head models, there are contra-
dicting theories which create enough uncertainty to warrant con-
tinued research into the mechanisms behind the observed out-
comes.
6. Intra-Tumoral Modulation Therapy
Optune as the approved treatment modality for GBM patients
has changed the procedures inmanagement of the disease. How-
ever, there are various reasons for compliance difficulties for this
system. Those are related to operational aspects (e.g., requiring
a shaved scalp, dermatological complications, perpetual applica-
tion) and stigma of using an external treatment system. Treat-
ment efficacy may also be limited by an inability to conform
field dimensions tomaximize stimulation strength and avoid off-
target injury.
GBM progression most commonly occurs as an extension
from the site of the original lesion. The extension has been recog-
nized as aggressive and incessant.[164–166] Due to this characteris-
tic of GBM, it has been endeavored to design locoregional strate-
gies which could restraint growth of unresectable tumors, thus in
consequence preventing recurrences.[167–170] Consequently, there
are research groups currently challenging the method of external
placement of transducer arrays. If effective, those alternative in-
ventions may prompt development of implantable technology to
deliver low-strength stimulation within tumour-affected brain re-
gions.
Intra-tumoral modulation therapy (IMT) has been introduced
and tested in vitro and in vivo for the treatment of GBM. The
delivery of IMT has been developed in a form of an in-dwelling
device which delivers electric charge to those brain regions which
have been affected by the tumour. This method explores how the
known electro-sensitivity of GBM cells can be used as a strategy
for treatment in a localized, targetedmanner. The device has been
designed to allow for sustained and titratable therapy, offering
very low maintenance, as well as, hidden away hardware which
can have an impact on patients’ quality of life.
The first evidence of IMT was described by Xu et al.[171] The
study sought proof-of-concept evidence for the in vitro anti-
tumour efficacy of pulsed (90 µs pulse width at 130 Hz) small
amplitude (V = 4 V) electric stimulation continuously delivered
to the epicenter of tumour cell preparation via an indwelling elec-
trode. The results showed that the viability of patient-derived pri-
mary GBM cells was reduced under the influence of IMT treat-
ment with negligible impact on primary post-mitotic rat neu-
rons. Additionally, the study has shown apoptosis and enhanced
chemotherapeutic effect in GBM cells treated with IMT. How-
ever, it currently is not knownwhether this treatment shares com-
mon mechanisms of action with other electrotherapeutic modal-
ities.
Di Sebastiano et al.[172] tested a new profile of parameters, us-
ing intermediate frequency at 200 kHz and a sinusoidal wave-
form to deliver continuously at V ≈ 2 V. Their experiments fur-
ther confirmed in vitro that primary human and, F98 GBM rat
cells but not primary post-mitotic rat neurons were exquisitely
sensitive to low amplitude, sinusoidal pulses at a frequency out
of range for neuronal entrainment or thermal injury. Moreover,
they presented the use of special purpose, MRI-compatible bio-
electrodes strategically positioned within, or adjacent to, tumour-
affected regions. A 1-week course of continuous IMT monother-
apy produced a significant reduction (19.7+/− 24.3%with outlier
excluded) of mean GBM volume in the living rat brains.
The in vivo test showed that the key feature of IMT is the abil-
ity to reach any aspect of the CNS to provide focused, titratable
therapy directly within areas of disease. Bioelectrodes could be
designed for personalized and comprehensive treatment cover-
age of GBM resection beds of non-operated lesions within elo-
quent or deep-seated CNS regions. The proximity of the IMTfield
source to GBM pathology will permit a broad, versatile spectrum
of stimulation parameters custom optimized to tumour location
and treatment response. Such a concealed, indwelling system is
expected to support patient quality of life providing sustained,
low maintenance therapy that potently complements radiation,
and ongoing chemotherapeutic options.
7. Mapping Electrotherapies to Glioblastoma
Characteristics
We present in this review thus far that GBM is of complex na-
ture, strongly correlated to neuronal innervations which are sub-
ject to reciprocal feedback in glioma growth. We discuss the ways
in which PEFs which cause various forms of reversible and IRE
have shown pre-clinical efficacy in gliomas and some alternate
tumour models thus far. We describe TTFs in terms of the pro-
posed mechanisms affecting mitotic spindle formation and duc-
tile ring destabilization in GBM and the role of TTF-induced Vm
changes as a topic of future research. Here, we highlight how
each modality of the electrotherapies mentioned show evidence
(or lack) of treating or addressing different aspects of GBM intri-
cacy (Figure 5) and questions which remain to be addressed in
order to progress the field.
7.1. Heterogeneity
7.1.1. Molecular and Cellular Heterogeneity
At the molecular level, differences in transcription programmes
and DNA methylation levels contribute to overall heterogeneity.
The identification of the four transcriptional subtypes; proneu-
ral, neural, classical, and mesenchymal, was a ground-breaking
discovery that has great potential to change the classification, di-
agnosis and treatment of patients in the future.[173] The neural
subtype has since been shown to likely be an artefact of contami-
nation and has since been discounted as a subtype.[174] Different
fragments of the same tumour could be classified into different
subtypes, with some patients exhibiting up to three different sub-
types. Since the different subtypes respond differently to thera-
pies and have differential survival times, the discovery of mul-
tiple subtypes existing within the same tumour complicates pa-
tient diagnosis and treatment (Figure 6).[175,176] Spatially distinct
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (14 of 28)
www.advancedsciencenews.com www.advancedscience.com
Figure 5. How can electrotherapy be improved in order to better treat GBM? Here we depict both the complex nature of GBM and what strategies of
electrotherapy have attempted to overcome these challenges. The prospect of electrotherapies ought to address numerous challenges which are currently
poorly understood including various aspects of heterogeneity, BBB, treatment resistance, and recurrence and how innate immunity can be harnessed.
Created with BioRender.com
tumour fragments showed that heterogeneity increaseswith time
resulting in different areas of the tumour being classified into dif-
ferent transcriptional subtypes. Further to this, multiple biopsies
taken from the same tumour were analyzed for their transcrip-
tion subtype. The proneural subtype was predominantly found at
the leading edge and in regions of invasion whilst the mesenchy-
mal subtype was mostly found in more hypoxic regions such as
pseudopalisades or the necrotic core.[29] This data showed that
not only could more than one subtype be found within the same
tumour, but that they occupy spatially distinct regions.
5-aminolevulinic acid (5-ALA) is an oral compound used as
a tool to aid maximal surgical resection as it produces red fluo-
rescence under blue light selectively within cancer cells. Studies
have indicated the fluorescence is not uniform throughout the
tumour volume[177] and that this spatial heterogeneity of fluores-
cence intensity could be down to cell-cell differences inmetabolic
programmes.[178] Spatial heterogeneity has also been indicated
through findings demonstrating that the metabolism of tyro-
sine is higher at the core of the tumour than at its edge. An in-
crease in tyrosine metabolism was found within tumour cells of
the MES-like transcriptional subtype indicating that heterogene-
ity within cellular metabolism and transcriptional heterogeneity
may be linked and contribute to overall poor prognosis.[179] Dif-
ferences in the metabolism of tyrosine in these GBM cell popula-
tions may have implications on dopamine and glutamate signal-
ing within the GBMmicroenvironment. Both neurotransmitters
are derived from tyrosine and glioma-initiating cells have been
shown to expressed dopamine receptor 2. Activation of these re-
ceptors by dopamine caused a hyperpolarization reminiscent of
neurons and long-term stimulation resulted in a higher sphere-
forming ability in vitro.[180] Additionally, tyrosine can be metab-
olized to produce glutamate, which has been shown to induce
glioma cell invasion due to action on glioma cell AMPA receptors
across neuron-glioma cell synapses.[181] This research brings into
question whether heterogeneously altered tyrosine metabolism
in GBM cells has the potential to produce both dopamine and
glutamate which can act in an autocrine manner to promote tu-
morigenesis of surrounding tumour cells. These distinctions are
important for electrotherapy since the application of an electric
field will not perturb each region with equal strength. Current
models using electrotherapy assume homogeneity of the tissue
when in fact, understanding the variation in the electrical prop-
erties of the different subtypes could provide more realistic data
for computational models, thus informing appropriate stimula-
tion parameters and electrode geometries for future bioelectronic
devices.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (15 of 28)
www.advancedsciencenews.com www.advancedscience.com
Figure 6. A cartoon representation of the current pathway newly diagnosed GBM patients follow and the most common progression of events (Created
with BioRender.com). A) Patient presents to clinic with a brain tumour and receives gross total resection (GTR) surgery. B) Despite GTR and margins,
some tumour cells remain close to the resection cavity, and some may have begun spreading further through the parenchyma. C) A representation
of the complex heterogeneity of cell types surrounding GBM cells in the brain microenvironment. D) Following GTR, the patient receives TMZ, and
radiotherapy which aims to kill remaining sensitive tumour cells; however some resistant cells persist following the cessation of therapies (shown in
purple). E) These persisting cells begin the process of re-populating the tumour and re-establishing the original heterogeneity. F) Despite standard of
care, the tumour recurs either at sites close to or further afield from the original tumour. Recurrences may present on the ipsilateral or contralateral side.
7.1.2. Heterogeneity within the Microenvironment and Immune Cell
Population
Glioma cells themselves contribute extensively to the heterogene-
ity of the disease, however the interaction with non-tumour cells
within the brain as well as the immune system further compli-
cate the heterogeneity. Some of these cell types include; neurons,
vascular cells, immune cells, and glial cells.[182,183] The role of
the tumour microenvironment has raised the possibility of har-
nessing the immune system to treat gliomas. The role of dif-
ferent types of immune cell in the GBM microenvironment as
well as recent advances of immunotherapeutics have been ex-
tensively reviewed[184] and revealed that alterations in gene ex-
pression influence the microenvironment as well as the effect
of infiltrating cells on glioma physiology. Briefly, the expres-
sion of heparanase in GBM samples following chemoradiation
was found to be heterogeneous.[32] This is a potential mecha-
nism by which GBM cells can physically alter their environment
by breaking down ECM components to aid invasion. Addition-
ally, neuroinflammation may be a driving factor in the acquisi-
tion of a more aggressive transcriptional programme and has
been demonstrated using exogenous IL𝛽1 on proneural GBM
cells. These cells slowly gained a more aggressive, mesenchy-
mal transcription programme which shows that the transcrip-
tional programmes of GBM cells are sensitive to input from the
microenvironment.[185] Concerning the application of PEFs, IRE
techniques have shown to induce severe cerebral oedema.[118]
While this was suggested to be overcome with appropriate phar-
maceuticals and electric field parameters, studies should con-
sider the risk of induced inflammation not only from a patient
safety perspective, but additionally, for the potential of triggering
aggressive transcription and invasion.
7.1.3. High Frequency Irreversible Electroporation in 3D Scaffolds
In 3D cultured GBM microenvironment mimics, Ivey et al.[186]
investigated if IRE or H-FIRE protocols could specifically tar-
get morphological characteristics of cells in heterogeneous mi-
croenvironments. They illustrate that malignant cells (human
U-87 GBM, human DBTRG GBM, rat C6 glioma) had signifi-
cantly reduce lethal thresholds E ≈ 530–810 V cm−1 compared
to healthy counterparts E ≈ 930–1200 V cm−1 (non-malignant
astrocytes, D1TNC1 and PC12 cells). The authors suggest that
treatment regimens may leverage these observations by tuning
the applied fields between these thresholds such the ablation of
tumour cells are achieve while sparing healthy astrocytes. Fur-
ther, Ivey et al.[187] later showed that molecularly enhanced nu-
clear to cytoplasm ratio with eA1 adjuvant further enhances the
selectivity of PEFs to induce cell death in tumors and that overex-
pressed EphA2 receptors could be exploited in combination treat-
ment regimes.
Cell-to-cell interactions in the GBM tumour microenviron-
ment under the influence of PEFs are not well studied and hence
not understood. Additionally, specific responses to cellular and
molecular variations between GBM samples are yet to be re-
ported in detail. However, there is evidence that PEFs are effective
against GSCs as discussed in Section 7.4.4.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (16 of 28)
www.advancedsciencenews.com www.advancedscience.com
7.1.4. Tumour-Treating Fields and Heterogeneity
Various in vitro studies of GBM cell lines have been used to study
the effect and mechanisms of TTFs with some cell lines more re-
sponsive than others. The cell size and corresponding dielectric
properties will impact the field distribution.[142] It is therefore rea-
sonable that the response of electrical stimulus is not unanimous
across cell variations. On this notion, TTFs do not show the same
therapeutic efficacy across different GBM cells lines. For exam-
ple, Neuhaus et al.[188] showed that U251 and T98G (both ma-
lignant human GBM derived) stimulated by TTF combined with
Ca2+ channel blocker benidipine increased cell death effects in
T98G which was not observed in U251 despite that Ca2+ entry
could be completely blocked in both cells with benidipine. Con-
trastingly, TTFs decreased aneuploidy and increased clonogenic
survival in U251 compared to T98G illustrating selective differ-
ences to TTFs. Thus, TTFs alone (U-251), or in combination with
benidipine (T98G) where shown to potentially trigger intrinsic
apoptosis.[188]
For applications in human, patients presentingwithGBMs dif-
fered in location and size, thus were at different stages of disease.
TTFs have exhibited positive outcomes in clinical trials were pa-
tients displayed varying heterogeneous characteristics between
samples. While interpretation of these remarks are to be taken
with caution, the study key endpoints of the EF-14 trial com-
paring TTF combination with TMZ (6.7 months OS) and TMZ
monotherapy (4 months OS) suggest that TTFs have therpeutic
effect for heterogeneous tumours.[152]
7.2. Immune Response
7.2.1. The Effect of Immune Cell Infiltration on Gliomas
Cells of the immune system including microglia and
macrophages also show heterogeneity in GBM tissue and
their involvement in GBM has been recently reviewed.[189] Dys-
functional T cells are involved in GBM and exhaustion of these
T cells is due to long term overstimulation. The co-expression of
multiple immune checkpoints is a brain specific T cell signature
with triple positive T cells being less able to produce cytokines
representing hypo-responsiveness and T cell exhaustion.[190]
This renders them ineffective at targeting tumor cells and makes
immunotherapies less effective. Thus, methods or strategies that
favor a “re-programming” of T cells into non-exhausted states
prior to immunotherapy may be of therapeutic interest. Studies
have found that the mesenchymal subtype has the highest
levels of T cell and macrophage infiltration among the different
subtypes.[191] Given that there is a variation in GBM subtypes
within a single tumour, a high level of intra-tumoral heterogene-
ity in immune signatures is therefore expected. This indicates
that some transcriptional subtypes may be more immunogenic
and therefore respond differently to immunotherapies. This
also proves a potential need for clinical trials to classify patients
based on their majority subtype, as the more immunologically
“cold” tumours would respond less and therefore confound
trial results.[192] Primary GBM samples can be clustered into
2 infiltration profiles,[193] one with a significantly increased-
and another with much lower expression of genes relating to
immune suppression compared to normal brain. Overall, data
indicates the importance of using immune cell infiltration data
when stratifying patients for therapies as patients belonging to
the first profile could be better candidates for immune based
therapies.
With respect to electrotherapies, neither PEFs’ or TTFs have
shown evidence of harnessing or addressing an immune re-
sponse in GBMmodels. There is however evidence that IRE pro-
tocols in alternate models induced T cell activation and enhanced
immune memory.[194] Additionally, sub-cutaneous tumours in
immunocompetent mice models showed substantial response to
IRE treatment than immunodeficient mice.[195] H-FIRE regimes
in breast cancer models show up-regulation of genes/pathways
related to necrosis and pyroptosis (lytic programmed cell death)
in vitro.[196] More conclusively, gene expression profiling data
indicated down-regulation of genes associated with immuno-
suppression and pro-inflammatory genes were increased after H-
FIRE treatment. These results suggest that PEFs treatment has
the potential to induce an immune response.
Similarly, mechanisms of immunogenic engagements have
been considered for TTFs outside the CNS.[197] TTFs can inter-
fere with DNA replication fork by repositioning fragments of
DNA created during replication, which leads to severe disrup-
tion of DNA damage repair and breast cancer 1-mediated ho-
mologous recombination pathways.[198–200] Moreover, TTFs may
cause endoplasmic reticulum stress which triggers adenosine
monophosphate-activated protein kinase (AMPK)-dependent
autophagy[201] and/or immunogenic cell death.[197] Immuno-
genic cell death can additionally be initiated by cytoplasmic
double-stranded DNA which leads to activation of pyroptosis via
the STING pathway.[202] Again, these studies imply that TTFs can
induce programmed immunogenic cell death when applied to
tissue outside of the brain.
While these results are promising, the microenvironments of
these models and their interactions with the immune system di-
verge from what could be expected in the brain. Consequently,
conclusions cannot be drawn from this research without limita-
tion.
Macrophages have become a recent topic of therapeutic in-
terest given that their polarization (M1/M2) is involved in
both pathological conditions (including tissue repair, cancer,
allergy, and chronic inflammation) and various physiological
conditions.[203] The complexity of the tumour microenvironment
demonstrates the ability of neoplastic brain cells to strongly in-
fluence the polarization of tumour associated macrophages.[204]
Since the membrane potential has been recognized as a key
regulator in many biological transitions/functions, macrophage
polarization was explored through bioelectric targeting of ATP
sensitive potassium channels (KATP).
[205] The outcome of these
experiments suggests that Vm acts as an instructive signal for
macrophage polarization and that KATP channels are potential
targets for immunomodulation. This is an important consid-
eration for uncovering the role of macrophages in electrother-
apy and whether bioelectric modulation can be tuned to induce
macrophage polarization toward tumour-killing phenotype. Cur-
rently, the role of the immune system in response to electrother-
apies in GBM remains an open question.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (17 of 28)
www.advancedsciencenews.com www.advancedscience.com
Figure 7. The “vascular lock effect” in reversible conditions. Electroporation induces vasoconstriction, reduce blood flow, and structural changes to
tight endothelial cell junctions between 0 and ≈2 h after application. The gaps allow paracellular permeability of ions (i.e., cytotoxic agents) to reach
the tumour and extracellular space. ≈8–12 h post application, the vessel is able to gradually recover its shape and function and full recovery is typically
evident within 24 h. Created with BioRender.com.
7.3. Blood-Brain Barrier
Due to the BBB presence, systemic drug delivery to brain tu-
mours remains problematic. BBB acts as a protection for neu-
ral tissues by hindering the inflow of potentially harmful com-
pounds from the blood to the brain. Thus, it shields the brain
pathogens. Additionally, its function involves regulating ions and
nutrients entering and exiting the brain.[206]
Since the connection formed between endothelial cells lin-
ing cerebral blood vessel—adherent and endothelial junctions—
regulate trans- and paracellular transport across the BBB, those
cells are the core anatomical unit of the BBB.[207] Research has
shown that in humanGBM, the connections between endothelial
cells are less stable, therefore creating leakymicro-vessels.[208–210]
Those openings between endothelial cells can result in cerebral
oedema.[211] Interestingly,multiple groups have observed that the
BBB within high-grade gliomas are disrupted at the center of the
tumour mass but intact at the peripheries, thus impeding drug
delivery.[211–213]
Electrotherapies present a unique therapeutic advantage to
overcome BBB since their modalities are typically physical. ECT
aside, electroporation techniques, and IMT protrude the tumour
environment, and TTFs are applied extra-cranially (though they
rely on the therapeutic combination of TMZ for maximum ef-
ficacy). Hence, the electrotherapy discussion in this section de-
scribes how these techniques have permeated BBB.
7.3.1. Electrochemotherapy—The Vascular Lock Effect
The application of high-voltage PEF’s in tissue induces an imme-
diate constriction but transient reduction in blood flow, a termi-
nology in the field called the “vascular lock” effect (Figure 7). In
ECT, this reduced blood flow can be advantageous since it is less
likely the combined drug will washout[64] or that invasive elec-
trodes will cause bleeding.[214]
Using intravital multiphoton microscopy techniques, Markelc
et al.[215] demonstrated in vivo that the main mechanism respon-
sible for increased electroporation induced permeability on nor-
mal blood vessels is due to the alterations in endothelial cell-to-
cell junctions. In addition, the authors elucidate that this may
not be the same case throughout the tumour vasculature. Since
leaky tumour vasculature and improper organization of endothe-
lial cell-to-cell junctions are inherent hallmarks, electroporation
effects on their organization may be less profound.
7.3.2. Electroporation-Induced Blood-Brain Barrier Disruption for
Increased Drug Transport
While several techniques exist to target the BBB temporarily per-
meabilizing it (ultrasound, osmotic disruption, drug delivery ve-
hicles), many are not able to reach the necessary concentrations
in the target tumour tissue.[81] Localized reversible electropora-
tion has been used to permeate the BBB to allow systemic cyto-
toxic drugs to cross and enter the tumour environment in vivo.
Recently, Sharabi et al.[216] were able to demonstrate in rat
gliomas models that under specific treatment conditions, the
BBB could be reversibly disrupted. The experiment included
two therapeutics, Cisplatin, which typically has limited perfu-
sion across the BBB but has demonstrated high cytotoxicity in
other tumours, andMethotrexate (MTX), which can partially pen-
etrate BBB, is not considered highly toxic and is used in treat-
ment of CNSmetastases. The average pre-treatment tumour vol-
umes were 22.8, 27.6, and 25.2 mm3 for control, MTX and elec-
troporation +MTX groups and 17.96 and 24.6 mm3 for the Cis-
platin and electroporation + Cisplatin groups. It was found that
while electroporation induced BBB disruption in combination
with MTX was able to slow tumour growth rates (1.02 mm3)
compared to control group (5.2 mm3), there was no significant
difference to rats treated with MTX only group (1.7mm3). How-
ever, the combination of electroporation and Cisplatin was able
to slow the growth rates (0.98 and 1.2mm3 for 90 and 180 pulses)
compared to 6.4 mm3 in the control group, while Cisplatin
delivery alone showed no significant effect on tumour growth
rates (3.8 mm3). This evidence supports that the combination of
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (18 of 28)
www.advancedsciencenews.com www.advancedscience.com
electroporation + Cisplatin has the greatest therapeutic benefit.
However, these results are to be taken with caution since large
portions of this group suffered immediate weight loss, porphyrin
discharge around eyes and nose and some rats showing transient
partial paralysis of the lower limbs.[216]
Using voltages from 250 to 650 V, with 4 Hz bursts of 50–
90 pulses of 50–70 µs widths a correlation was found in Hjouj
et al.[217] between the applied voltage and tissue damage and sim-
ilarly the corresponding volume of BBB disruption. Since the ef-
fects of reversible and IRE are transient, the authors also show
that BBB disruption was strongly correlated with delayed tissue
damage as found by later contrast enhanced T1W MRI data. In
similar work, diffusion-weighted MRI was considered by Mah-
mood et al.[218] as a quantitative method of assessing the electro-
poration inducedmembrane permeabilization of rat brain tissue.
They found that transient and permanent membrane permeabi-
lization could be distinguished as early as 5 min and 24 h post
electroporation treatment. In addition to in situ treatment mon-
itoring, statistical models based on the Peleg-Fermi model previ-
ously used to predict cell death in IRE have also been adapted by
Sharabi et al.[219] to show predictable BBB disruption.
While animal studies are typically more translatable, in
vitro models give less complicated exploration freedom to re-
searchers and still provided quantitative data.[220] For BBB
modelling to study electropermeabilization, several platforms
have been recently employed such as transwells,[221–223] and
microfluidics.[81,224]
Microfluidic models present alternative in vitro platforms to
study and mimic BBB. In the work by Mohammad et al.,[81] the
reversible and IRE thresholds were investigated focusing on both
magnitude and number of pulses using an in vitro model incor-
porating a “brain endothelial cells on chip” setup ofmicrofluidics
to simulate BBB. The experiments conclude that the majority of
cells electroporated with 10 pulses were able to recover while 30
and 90 pulses induced irreversible damage for E ≈646–714 V
cm−1. The least cell death was observed for E <400 V cm−1. Fur-
thermore, Bonakdar et al.[224] present a microfluidic model that
enables real-timemeasurements of BBB permeability in pre- and
post- treatment conditions of PEFs.Using human cerebralmicro-
capillary endothelial cells cultured on a permeable membrane,
they suggest that sub-electroporation pulsed BBB permeabiliza-
tion can occur via para-cellular pathways due to cellular deforma-
tion and opening of the endothelial tight junctions.
Similarly, Sharabi et al.[223] showed transient BBB disruption
utilizing 10 pulses of V≈ 5–100 V in low voltage experiments and
V≈ 200–2000 V in high voltage experiments. The durations in
each case were 50 µs applied at a frequency of 1 Hz. In the cell vi-
ability assay, the results demonstrated that for amplitudes<100 V
there was no statistically significant differences when compared
to control wells suggesting reversible damage only. The experi-
ments indicated that BBB disruption could not be attributed to
electroporation for PEF’s <100 V. For amplitudes >1400 V, cell
viability was reduced indicating cell death and lethal thresholds.
7.3.3. Tumour-Treating Fields Targeting Blood-Brain Barrier
In vitro studies using GBM cell lines have shown that TTFs
can increase the plasma membrane’s permeability by causing
mis-localisation of specific tight junction proteins (claudin-5
and ZO-1).[225] Images of the human GBM cell line U-87and
murine astrocytoma cell line KR158B treated with TTFs for 1
h revealed many perforations scattered throughout the plasma
membrane.[226] The results showed that the perforations allowed
the uptake of 20 kDa fluorescently labelled dextran particles. Ad-
ditionally, in vivo experiments performed in the same study re-
vealed that TTFs could induce BBB disruption for 48–96 h.[226]
Thus, TTFs application could enhance pharmacological com-
pounds’ delivery to the brain. Moreover, it could be used to in-
crease the uptake of intraoperative agents, for instance, 5-ALA,
to improve intraoperative marking of brain tumour margins.[227]
Finally, as TTFs can also weaken the mechanism of angiogen-
esis, there is an opportunity to use TTFs combined with anti-
angiogenic agents like bevacizumab.[228,229]
7.4. Resistance and Recurrence
7.4.1. Recurrence
The average life expectancy of a patient with GBM lies between
several weeks and several months after postoperative radiother-
apy, with a protracted course when treated with TMZ in addi-
tion to radiotherapy alone. Median survival is generally less than
one year from the time of diagnosis and most patients die within
two years, with most long-term survivors being given the wrong
histological diagnosis at first.[230,1] Unless the neoplasm has de-
veloped from a lower grade astrocytoma, in more than 50% of
cases the clinical history in less than 3 months. Although infil-
trative spread is a common feature of all diffuse astrocytic tu-
mours, GBM is particularly notorious for its rapid invasion of
the neighboring brain structures.[19] Invading cells reside outside
the contrast-enhancing rim of the tumour thereby escaping sur-
gical resection and evading radiotherapy. Thus, recurrence is typ-
ically inevitable. However, the generation of metastases outside
of the CNS remains very rare since the subarachnoid space and
CSF tend to remain unaffected.[27] PEF studies are yet to include
long term in vivo data that determine the rate of recurrence fol-
lowing electrical intervention. Additionally, excluding ECT, com-
bination treatment with pharmaceutical strategies are yet to be
compared within current in vivo or in vitro models. In the cur-
rent pre-clinical stage, there is no way to compare the techniques
to the current standard of care. This data is crucial to inform the
clinical translation roadmap for future interventions.
7.4.2. Patterns of DNA Methylation Show Intra-Tumoral Variation
Methylation of gene promoters is an epigenetic mechanism for
regulating the expression of that gene, thus unwarranted changes
to DNA methylation have the potential to severely disrupt gene
expression. Low grade gliomas (LGGs) and secondary GBMs are
excellent examples of this seeing as they commonly harbor mu-
tations in the IDH1 gene which leads to a widespread hyperme-
thylated phenotype termed G-CIMP.[231] A recent study of LGGs
(WHO grade II and III) was able to identify six methylation
subtypes M1–M6 which they were able to categorize based on
IDH status and tumour type.[232] Mutant IDH1 is a precursor for
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (19 of 28)
www.advancedsciencenews.com www.advancedscience.com
global methylation changes in gliomas, however WT IDH1 tu-
mours also show methylation changes. WT IDH1 primary GBM
shows significant inter-tumoral heterogeneity in the pattern of
DNA methylation. Methylation status was accurately correlated
with the three transcriptional subtypes, suggesting that DNA
methylation is also heterogeneous within GBMs. Patients with
shorter time periods between primary and recurrent tumours
had greater differences in their promoter methylation which in-
dicates that aggressive tumours are capable of changing methy-
lation status more quickly.[233] Spatially separate biopsies from
GBM patients showed that different subclasses of methylation
status existed within the same tumour. Analysis of the CpG is-
lands showed that samples fromwithin the same tumour, instead
of clustering together, had a high degree of cross-over with sam-
ples from other patients.[234] This suggests a high level of both
inter- and intra-tumoral heterogeneity at the DNA methylation
level.
Understanding epigenetic mechanisms and their role in GBM
initiation and progression is essential as it is so tightly linkedwith
gene expression and as has been discussed, can be correlated
with the transcriptional subtypes of GBM. Being able to pharma-
cologically manipulate the epigenetics and transcriptomic sub-
type of GBM would provide routes to systematically eliminate bi-
ologically distinct regions of the tumour. Electrotherapies such
as ECT would allow pharmacological agents access to their intra-
cellular targets and ECT has the benefit of being bespoke to the
location and size of a targetable area. For example, drugs target-
ing the tyrosine metabolism pathway would be more effective in
regions of differential tyrosinemetabolism,which corresponds to
the visual information provided by 5-ALA (discussed previously
in Section 7.1.1.). Thus, surgeons would be able to visually in-
spect and identify a candidate region and utilize ECT methods to
directly apply the drug of choice.
7.4.3. Treatment Resistance and Glioma Stem-like Cells
Metabolic phenotypes existing within patient-derived GSC lines
of GBMcan be categorized into fast cycling cells (FCCs), and slow
cycling cells (SCCs). SCCs generated highly invasive tumours in
mouse models extending long processes out into healthy brain
tissue whilst FCCs formed more contained tumours with less
invasive behavior. SCCs accumulate fatty acids as lipid droplets
within cells and switch to lipid metabolism upon glucose restric-
tion or pharmacological inhibition of glycolysis. This suggests
that the SCC population of GBMhas the capacity to switch energy
sources depending on external conditions. SCCs were demon-
strated to be inherentlymore TMZ resistant that FCCs. This com-
bined with their ability to switch to fatty acid oxidation means
they are likely the surviving population of standard of care proce-
dures and repopulate the tumour over time.[235] This also shows
thatmetabolic plasticity is an ability possessed by only certain cell
populations in GBM.
GSCs express unique surface markers, are capable of modu-
lating characteristic signaling pathways which promote tumori-
genesis, play essential roles in vascular formation, are less pro-
liferative and often confer treatment resistance.[236,237] Further,
GSCs contribute to both recurrence and resistance by alter-
ation of DNA damage response pathways,[238] creating hypoxic
microenvironments[239] and through a number of signaling path-
ways (Notch, RTK and sonic hedgehog).[236] While several stud-
ies have targeted the stemness signaling of GSC, some studies
have looked at inducing differentiation in GSCs via bone mor-
phogenetic proteins signaling activation which markedly attenu-
ated the frequency of GSC sphere formation in vivo.[240] Pulsed-
current electrical stimulation has also shown to include differ-
entiation in cultured neural stem cells impacting cell fate and
neurite extension.[241] This raises opportunity for functional and
therapeutic targeting of GSCs pharmacologically, electrically or
in combination.
GSCs have been shown to possess an incredibly dynamic abil-
ity to adapt to the environmental changes and re-establish hetero-
geneity following treatment cessation. This population seems to
not be enriched by TMZ, indicating that selection of pre-existing
resistant clones is not a mechanism of resistance.[242] This
suggests that therapies targeting membrane epitopes of GSCs
should be avoided as these cells are able to very quickly adapt
their membrane protein expression. A study of the metabolism
and efflux of drug therapies showed there was an increase in 23
genes involved in drug metabolism[243] which suggests GBM can
gain resistance via upregulation of drug metabolism and efflux
pathways.
Resistance to TMZ can be acquired via changes to the methy-
lation status of the MGMT gene,[3] however alternative mecha-
nisms of TMZ resistance have been recently reviewed.[244] Other
studies have shown that GBM cells exhibit a heterogeneous
expression of MGMT at the cellular level[245] and that TMZ-
resistant cells are capable of detoxifying mitochondrial ROS
thereby enhancing their survival following TMZ treatment.[246]
In vitro studies have shown that a loss of inhibition of the JAK-
STAT pathway leads to cells with increased radio-resistance[247]
indicating that mechanisms of reinstating this inhibition may be
viable treatment avenues. Increased levels of the cytokine CXCL1
in GBM samples have also been linked to radio-resistance via in-
creased NF-𝜅B signaling.[248] Taken together, these studies serve
to show that GBM cells possess a unique ability to endure and
survive current SOC procedures.
While physical modalities allow an element of selectivity to re-
sistant cells, combination treatments are typically required for
significant therapeutic benefit. Both TTFs and PEFs have demon-
strated that they can be efficacious to previously identified resis-
tant GBM cells and GSCs. TTFs combined modality have been
presented and compared extensively in vitro and in human while
PEFs are still pre-clinical and require extensive research for clin-
ical translation. The current state of research is however, encour-
aging.
7.4.4. High Frequency Irreversible Electroporation Targeting
Tumour-Initiating and Glioma Stem-like Cells
Structural characterization of GSCs have shown evidence of in-
crease nucleus atypia,[249] hence, Ivey et al.[250] hypothesize that
GSCs will be highly susceptible to H-FIRE over healthy astro-
cytes with regular nuclear size. They study GSCs in 3D type 1
collagen hydrogel scaffolds and first show that the nuclear sizes
of GSCs (GBM10, VTC-061, and VTC-064) and U-251 differ-
entiated GBM cells are significantly larger than non-malignant
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (20 of 28)
www.advancedsciencenews.com www.advancedscience.com
healthy astrocytes. Using a 0.5-2-0.5 H-FIRE protocol, the au-
thors demonstrate ablation areas are also significantly greater
in GSCs and U-251 cells when compared to healthy astrocytes
and non-malignant neural stem cells. A similar trend was ob-
served using hyaluronic acid scaffolds which have shown to have
more realistic in vivo biological and clinical behavior of the GBM
microenvironment.[251] The ability for H-FIRE to selectively ab-
late GSC that previously were resistant to chemotherapy presents
deeply encouraging potential for future treatment modalities.
7.4.5. Tumour-Treating Fields for Resistant and Recurrent
Glioblastoma
Since the anti-tumour activity of TTF is correlated to cell size,
which impacts the optimal frequency of use, GSCs may be less
responsive than other tumour cells since their sizes are more
heterogeneous and proliferate at a slow rate.[237] Cells treated
with TTFs have shown to have an increase in both cell volume
and granularity[141] which have been typically associated with
senescence[252] and autophagy.[253] Since senescence has not yet
been observed in TTFs, Shteingauz et al.[201] hypothesize that
granularity may be associated with autophagosome vesicle accu-
mulation and show that TTFs upregulate proteotoxic stress re-
sponse triggering activation of AMPK and increased autophagic
flux in treated cells. Since autophagy facilitates resistance of tu-
mour cells to antineoplastic agents,[254] these findings suggest
that greater autophagy functions as a resistant mechanism to
TTFs, which could be evaded by targeting autophagy.[201]
TTFs have been shown in vitro to inhibit cell migration and
invasion by inducing a more adhesive cell phenotype. This is
achieved through dysregulation of cytoskeletal structures and
proteins related to the epithelial-mesenchymal transition (e.g.,
actins, vimentin, and cadherin), potentially reducing the likeli-
hood of recurrence.[255,256] Clinically, analysis of the TTF phase
3 trial (EF-15) show that 15% of patients with recurrent GBM
responded with complete or partial radiological response.[257] In
comparison, only 9.6% of patients who received TMZ only ex-
hibited objective radiological response. Interestingly, the analy-
sis also indicated that 44% of GBM tumours in this trial ini-
tially display growth before reversing and shrinking 2–7 months
later.[257] This suggests that TTFs provide a better radiological
response.[225]
Smaller group studies on TTFwith combined treatment strate-
gies have been shown. Lu et al.[258] compared TTFs with a triple-
drug regimen of TMZ, bevacizumab, and irinotecan versus TTFs
with bevacizumab-based chemotherapies for recurrent GBM pa-
tients. While the former group had significantly prolonged me-
dian PFS compared to the latter from time of recurrence (10.7
months vs 4.7 months, respectively), there was no significant dif-
ference in OS from time of recurrence. Finally, Ansstas et al.[259]
reported a small case series of 8 patients who received TTFs for
recurrent GBM refractory to bevacizumab, of which 5 were sub-
sequently re-challenged with bevacizumab after progression on
TTF. Median OS was 216 days (7.2 months) following initiation
of TTFs. Since these studies were primarily focused on subgroup
analyses or combination therapies and had much smaller sam-
ples sizes than the studies previously mentioned, they add lim-
ited insight into the effects of TTFs.
7.5. Summary of Known Electrotherapy Risks
The application of electric fields in biological tissue carries in-
herent risks some of which can be mitigated, reduced or suit-
ably maintained. MRI data has indicated that treatment areas
subject to intra-tumoral ECT showed necrosis and fluid-filled
cavities suggesting oedema, infection, or haemorrhage all of
which should be approached with caution and better under-
stood for clinical translation.[16] Where electroporation was used
to induce BBB disruption in vivo, there were no signs of dis-
tress, neurological defects one day post-treatment.[216] However,
H&E results revealed haemorrhages along electrode path possi-
bly caused by needle insertion or ratmovement. Bleeding was ob-
served in tumour core at treatment location and muscle contrac-
tions were also evident and overcome with muscle relaxants.[216]
Drug choice for BBB disruption requires careful deliberation
since highly toxic agents delivered intravenously will have wider
access to penetrate and damage healthy brain tissue.
NsPEF can rapidly depolarize neurons 1 ms after application
and are at risk of evoking action potentials.[98] Careful consider-
ation of protocols in particularly, pulsing durations, to allow for
the response of voltage-gated channels and reduce the likelihood
of neuronal firing are encouraged.
While IRE is considered non-thermally mediated, in vitro
models have shown that for pulse durations on the order of
1µs are thermally mediated.[112] Real-time temperature monitor-
ing during IRE applications will be a necessity to ensure that
protein denaturation, oedema and seizures are avoided. Early
pre-clinical validation of IRE in tumour free canine models in-
dicated complications of oedema, seizures and bleeding (due
to needle insertion) which were overcome with corticosteroids
and a hyperosmolar agent (oedema) or prophylactic anticonvul-
sant (seizures).[113] Furthermore, in canine models with sponta-
neous occurring gliomas, IRE treatment was well tolerated and
tumours were successfully ablated with adjunctive radiotherapy,
anti-oedema, and anti-seizure treatment.[116] In a later study in-
volving 7 canines with glioma, one canine experienced severe
cerebral oedema due to excessive field strengths.[118] This out-
come stresses the importance of pre-operation planning and the
demand for real-time field monitoring in tissue to ensure safe
delivery.
Treatment with H-FIRE improved the outcomes observed in
some IRE models. First, the protocol does not induce muscle
contractions,[122,123] it has theoretically been suggested to perme-
abilize but not excite cells,[97] and is synchronized with cardiac
rhythm.[125] Pre-clinical studies of H-FIRE indicated no adverse
post-operative effects attributed to the therapy.[137]
The external application of TTFs have indicated dermatologic
adverse events in 16% of those enrolled in the phase III trial.
This due to the contact electrodes with the skin causing irritation
and rash which can be overcome with topical or oral antibiotics,
topical corticosteroids.[260]
Finally, in pre-clinical models of IMT, one animal died 3 days
after IMT initiation, however, histological examination excluded
the complication to be related to IMT treatment since there was
no evidence of electrolysis or hematoma.[172]
It is evident that the most serious risks include oedema and
seizures which have improved as techniques have been devel-
oped and can be maintained with appropriate supporting drugs.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (21 of 28)
www.advancedsciencenews.com www.advancedscience.com
It is important that these techniques continue to minimize any
associated risks as they transition from benchtop to clinic.
7.6. Multi-Modality Integration—A Promising Approach
Due to the multi-faceted nature of GBM, it seems obvious to ap-
proach treatment with a multi-modal solution. In the presented
findings, following surgical resection and radiotherapy, the ad-
dition of TTFs with maintenance TMZ resulted in greatest pa-
tient outcomes compared TMZ alone.[152] ECT when used in
the context of BBB disruption would allow for substantial trials
of drug repurposing and targeting in GBM. Several approaches
to drug repurposing for GBM have been recently reviewed,[261]
some of which include Chloroquine, an antimalarial agent found
to increase chemosensitivity to TMZ,[262] Naringenin, a bioac-
tive flavonoid which has shown to eliminate migration and in-
vasion in GBM,[263] and Phloroglucinol a phenolic derivative
and natural phlorotannin component of brown algae has sug-
gested to inhibit proliferation.[261] Additionally, Bevacizumab,
Disulfiram and platinum-based therapeutics are all considered
promising for drug re-purposing to treat gliomas and for poten-
tial administration alongside the electrotherapy techniques dis-
cussed. Combined electroporation and local drug delivery devices
with adjacent or integrated ion-pumps (i.e., convection enhanced
delivery,[264]iontronic pumps[265]) to the working electrode ad-
ministered under appropriate conditions, localized delivery fol-
lowed by electroporation or TTFs, could also allow for delivery
of chemotherapies that have limited BBB perfusion to be trialed.
Integrating a small, separately controlled light source (i.e., Light
emitting diode or optical fiber) to an implantable electrotherapy
device could provide options for high-resolution photodynamic
therapy.[266] These device modifications would provide a multi-
functional treatment regime that could substantially impact the
current standard of care.
8. Final Remarks
Glioblastoma is exceptionally complex and currently incurable.
Several therapeutic options are yet to significantly advance the
standard of care. Electrotherapy presents a unique therapeutic
strategy but must look at addressing the complex biological chal-
lenges that currently make the disease so difficult to treat in
the first place. In order to progress interdisciplinary research in
this field we must first address the following: 1) The vast het-
erogeneity displayed by GBM is in itself a major problem when
treating the disease. The presence of cells capable of repopulat-
ing the tumour heterogeneity post-treatment and aggressively
invasive cells means that current treatment cannot prevent the
inevitable recurrence. Treatments cannot be expected to rise to
this challenge if the challenge itself is not understood, thus fur-
ther research into the mechanisms of heterogeneity are vital if
novel treatments are going to be pursued. 2) While electrother-
apy techniques show promise, there remains a gap of addressing
the biological complexities. How do PEFs affect recurrence? Is
there a neuro-immune response? What happens to neurons in
the micro-environment and surrounding parenchyma? What is
the efficacy if it is combined with TMZ, or other chemothera-
pies? How does it compare to the current clinical management?
How does it differ between varying samples of GBM and in vivo?
For TTFs, how does this technique alter the membrane poten-
tial, and thus is there a better way to apply the stimulus? Does
this better explain the target mechanism, and could this reduce
apprehension in the field? How does it affect the immune re-
sponse?More generally, how do electromagnetic fields affect neo-
plastic progression and migration? We need to understand how
we can use one or multiple techniques in a combined manner
to both understand and treat GBM. 3) A universal approach to
describe the applied electrical stimulus in electrotherapy. Cur-
rently, there is no generic ways of explaining what parameters are
used in an experiment (whether it be in vitro, in vivo or in silico)
whichmakes comparing data and reproducibility exhausting. We
suggest in future a better characterized setup where all methods
should include the current density of the applied stimulus (tak-
ing the integral of the active phase) such that any applied stimu-
lus can be scaled accordingly thus compared accurately. Authors
should be clear whether or not their method uses direct, capaci-
tive, or inductive coupling (or combinations of), In addition, elec-
trode characterization including impedance frequency response
and would be advantageous. 4) Safety, TTFs excluded, all other
electrotherapies reviewed here are in vitro or vivo validation in
the brain. Electroporation techniques particularly need sufficient
evidence that demonstrates the safety of high field strengths in
the brain where current concerns raised in Section 7.5. are ad-
dressed.
It is clear that the techniques described in this review present
valuable strategies for advanced treatment of glioblastoma. We
encourage all the scientists working on electrotherapies to ac-
quire an interdisciplinary approach, marrying the technological
input with the biological knowledge to further advance PEFs- and
TTFs-based therapies to create efficient and robust methods for
tackling the most aggressive type of primary brain tumours.
Acknowledgements
This work was supported by the EPSRC IRC in Targeted Delivery for Hard-
to-Treat Cancers (EP/S009000/1). E.P.W.J. was supported by a Trinity Hall
Research Studentship and the Cambridge Trust. A.F. was supported by
the University Hospitals Birmingham Charity. C.W. was supported by The
Brain Tumour Charity & Minderoo Foundation.
Conflict of Interest
The authors declare no conflict of interest.
Author Contributions
E.P.W.J., A.F., and M.G. contributed to the content and figures. E.P.W.J.
and A.F. contributed to final editing and preparing the manuscript for sub-
mission. All authors contributed to drafting the outline and editing the
manuscript.
Keywords
bioelectronics, electroporation, electrotherapy, glioblastoma, tumour-
treating fields
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (22 of 28)
www.advancedsciencenews.com www.advancedscience.com
Received: March 11, 2021
Revised: May 20, 2021
Published online:
[1] R. Stupp, W. P. Mason, M. J. van den Bent, M. Weller, B. Fisher, M.
J. B. Taphoorn, K. Belanger, A. A. Brandes, C. Marosi, U. Bogdahn,
J. Curschmann, R. C. Janzer, S. K. Ludwin, T. Gorlia, A. Allgeier, D.
Lacombe, J. G. Cairncross, E. Eisenhauer, R. O. Mirimanoff,N. Engl.
J. Med. 2005, 352, 987.
[2] S. Gianfaldoni, R. Gianfaldoni, U. Wollina, J. Lotti, G. Tchernev, T.
Lotti, Open Access Macedonian J. Med. Sci. 2017, 5, 521.
[3] J. Zhang, M. F. G. Stevens, T. D. Bradshaw, Curr. Mol. Pharmacol.
2011, 5, 102.
[4] M. Fakhoury, Artif. Cells, Nanomed., Biotechnol. 2015, 44, 1365.
[5] T. McGranahan, K. E. Therkelsen, S. Ahmad, S. Nagpal, Curr. Treat.
Options Oncol. 2019, 20, 24.
[6] E. Ottestad, D. S. Orlovich, Pain Med. 2020, 21, S3.
[7] J. Malmivuo, R. Plonsey, Bioelectromagnetism – Principles and Appli-
cations of Bioelectric and Biomagnetic Fields, Oxford University Press,
New York 1995.
[8] R. Feiner, T. Dvir, Nat. Rev. Mater. 2017, 3, 17076.
[9] E. Neumann, K. Rosenheck, J. Membr. Biol. 1972, 10, 279.
[10] M. Okino, H. Mohri, Jpn. J. Cancer Res. 1987, 78, 1319.
[11] E. Neumann, M. Schaefer-Ridder, Y. Wang, P. H. Hofschneider,
EMBO J. 1982, 1, 841.
[12] R. V. Davalos, L. M. Mir, B. Rubinsky, Ann. Biomed. Eng. 2005, 33,
223.
[13] C. F. Gonzalez, V. T. Remcho, J. Chromatogr. A 2005, 1079, 59.
[14] E. D. Kirson, V. Dbalý, F. Tovaryš, J. Vymazal, J. F. Soustiel, A. Itzhaki,
D. Mordechovich, S. Steinberg-Shapira, Z. Gurvich, R. Schneider-
man, Y. Wasserman, M. Salzberg, B. Ryffel, D. Goldsher, E. Dekel, Y.
Palti, Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 10152.
[15] E. D. Kirson, Z. Gurvich, R. Schneiderman, E. Dekel, A. Itzhaki, Y.
Wasserman, R. Schatzberger, Y. Palti, Cancer Res. 2004, 64, 3288.
[16] B. Agerholm-Larsen, H. K. Iversen, P. Ibsen, J. M. Moller, F. Mah-
mood, K. S. Jensen, J. Gehl, Cancer Res. 2011, 71, 3753.
[17] A. B. Lassman, A. E. Joanta-Gomez, P. C. Pan,W.Wick,Neuro-Oncol.
Adv. 2020, 2, vdaa069.
[18] D. W. Parsons, S. Jones, X. Zhang, J. C. H. Lin, R. J. Leary, P. An-
genendt, P. Mankoo, H. Carter, I. M. Siu, G. L. Gallia, A. Olivi, R.
McLendon, B. A. Rasheed, S. Keir, T. Nikolskaya, Y. Nikolsky, D. A.
Busam, H. Tekleab, L. A. Diaz Jr., J. Hartigan, D. R. Smith, R. L.
Strausberg, S. K. N. Marie, S. M. O. Shinjo, H. Yan, G. J. Riggins,
D. D. Bigner, R. Karchin, N. Papadopoulos, G. Parmigiani, Science
2008, 321, 1807.
[19] F. B. Furnari, T. Fenton, R. M. Bachoo, A. Mukasa, J. M. Stommel, A.
Stegh, W. C. Hahn, K. L. Ligon, D. N. Louis, C. Brennan, L. Chin, R.
A. DePinho, W. K. Cavenee, Genes Dev. 2007, 21, 2683.
[20] C. Agulhon, J. Petravicz, A. B. McMullen, E. J. Sweger, S. K. Minton,
S. R. Taves, K. B. Casper, T. A. Fiacco, K. D. McCarthy, Neuron 2008,
59, 932.
[21] P. Brodal, The Central Nervous System: Structure and Function, 4th
ed., Oxford University Press, New York 2010.
[22] A. Volterra, J. Meldolesi, Nat. Rev. Neurosci. 2005, 6, 626.
[23] N. J. Allen, B. A. Barres, Curr. Opin. Neurobiol. 2005, 15, 542.
[24] A. V. Gourine, S. Kasparov, Exp. Physiol. 2011, 96, 411.
[25] S. Liebner, C. J. Czupalla, H. Wolburg, Int. J. Dev. Biol. 2011, 55, 467.
[26] T. Saito, K. Shibasaki, M. Kurachi, S. Puentes, M. Mikuni, Y. Ishizaki,
Neurosci. Lett. 2011, 497, 116.
[27] D. N. Louis, H. Ohgaki, O. D. Wiestler, W. K. Cavenee, P. C. Burger,
A. Jouvet, B. W. Scheithauer, P. Kleihues, Acta Neuropathol. 2007,
114, 97.
[28] H. Ohgaki, P. Kleihues, Acta Neuropathol. 2005, 109, 93.
[29] D. H. Heiland, A. Gaebelein, M. Börries, J. Wörner, N. Pompe, P.
Franco, S. Heynckes, M. Bartholomae, D. HAilín, M. S. Carro, M.
Prinz, S. Weber, I. Mader, D. Delev, O. Schnell, Mol. Cancer Res.
2018, 16, 655.
[30] A. P. Patel, I. Tirosh, J. J. Trombetta, A. K. Shalek, S. M. Gillespie, H.
Wakimoto, D. P. Cahill, B. V. Nahed, W. T. Curry, R. L. Martuza, D. N.
Louis, O. Rozenblatt-Rosen, M. L. Suvà, A. Regev, B. E. Bernstein,
Science 2014, 344, 1396.
[31] J. Juan-Albarracín, E. Fuster-Garcia, A. Pérez-Girbés, F. Aparici-
Robles, Á. Alberich-Bayarri, A. Revert-Ventura, L. Martí-Bonmatí, J.
M. García-Gómez, Radiology 2018, 287, 944.
[32] A. V. Suhovskih, G. M. Kazanskaya, A. M. Volkov, A. Y. Tsidulko, S.
V. Aidagulova, E. V. Grigorieva, Int. J. Mol. Sci. 2020, 21, 1301.
[33] M. A. Qazi, P. Vora, C. Venugopal, S. S. Sidhu, J. Moffat, C. Swanton,
S. K. Singh, S. K. Singh, Ann. Oncol. 2017, 28, 1448.
[34] M. Cabello, H. Ge, C. Aracil, D. Moschou, P. Estrela, J. Manuel
Quero, S. I. Pascu, P. R. F. Rocha, Sensors 2019, 19, 139.
[35] R. J. Molenaar, A. Gambardella, J. A. Rey, R. Yamanaka, Int. Scholarly
Res. Not. 2011, 2011, 590249.
[36] T. Brismar, GLIA 1995, 15, 231.
[37] P. R. F. Rocha, A. Elghajiji, D. Tosh, Bioelectricity 2019, 1, 131.
[38] H. S. Venkatesh, W. Morishita, A. C. Geraghty, D. Silverbush, S. M.
Gillespie, M. Arzt, L. T. Tam, C. Espenel, A. Ponnuswami, L. Ni, P.
J. Woo, K. R. Taylor, A. Agarwal, A. Regev, D. Brang, H. Vogel, S.
Hervey-Jumper, D. E. Bergles, M. L. Suvà, R. C. Malenka, M. Monje,
Nature 2019, 573, 539.
[39] M. Griffin, R. Khan, S. Basu, S. Smith, Cancers 2020, 12, 3068.
[40] M. Monje, Cancer Res. 2020, 80, 2979.
[41] J. Pollak, K. G. Rai, C. C. Funk, S. Arora, E. Lee, J. Zhu, N. D. Price,
P. J. Paddison, J. M. Ramirez, R. C. Rostomily, PLoS One 2017, 12,
e0172884.
[42] C. Serrano-Novillo, J. Capera, M. Colomer-Molera, E. Condom, J. C.
Ferreres, A. Felipe, Cancers 2019, 11, 287.
[43] M. C. Sanguinetti, M. Tristani-Firouzi, Nature 2006, 440, 463.
[44] K. B. Pointer, P. A. Clark, K. W. Eliceiri, M. S. Salamat, G. A. Robert-
son, J. S. Kuo, Clin. Cancer Res. 2017, 23, 73.
[45] Y. Zhang, L. K. Kaczmarek, J. Physiol. 2016, 594, 4677.
[46] S. Kan, S. Chai, W. Chen, B. Yu, Oncol. Lett. 2019, 18, 1679.
[47] L. Soattin, M. Fiore, P. Gavazzo, F. Viti, P. Facci, R. Raiteri, F. Difato,
M. Pusch, M. Vassalli, Biophys. Chem. 2016, 208, 26.
[48] X. Chen, S. Wanggou, A. Bodalia, M. Zhu, W. Dong, J. J. Fan, W. C.
Yin, H. K. Min, M. Hu, D. Draghici, W. Dou, F. Li, F. J. Coutinho, H.
Whetstone, M. M. Kushida, P. B. Dirks, Y. Song, C.-c. Hui, Y. Sun,
L.-Y. Wang, X. Li, X. Huang, Neuron 2018, 100, 799.
[49] W. Zhou, X. Liu, J. W. M. van Wijnbergen, L. Yuan, Y. Liu, C. Zhang,
W. Jia, Sci. Rep. 2020, 10, 16121.
[50] T. T. Sales, F. F. B. Resende, N. L. Chaves, S. S. Titze-De-Almeida,
S. N. Báo, M. L. Brettas, R. Titze-De-Almeida, Oncol. Lett. 2016, 12,
2581.
[51] V. A. Cuddapah, H. Sontheimer, J. Biol. Chem. 2010, 285, 11188.
[52] A. Bordey, H. Sontheimer, J. Trouslard, J. Membr. Biol. 2000, 176, 31.
[53] S. Xu, C. Liu, Y. Ma, H. L. Ji, X. Li, Biomed Res. Int. 2016, 2016,
2190216.
[54] H. S. Venkatesh, L. T. Tam, P. J. Woo, J. Lennon, S. Nagaraja, S. M.
Gillespie, J. Ni, D. Y. Duveau, P. J. Morris, J. J. Zhao, C. J. Thomas,
M. Monje, Nature 2017, 549, 533.
[55] S. C. Buckingham, S. L. Campbell, B. R. Haas, V. Montana, S. Robel,
T. Ogunrinu, H. Sontheimer, Nat. Med. 2011, 17, 1269.
[56] S. L. Campbell, S. C. Buckingham, H. Sontheimer, Epilepsia 2012,
53, 1360.
[57] C. C. J. Lin, K. Yu, A. Hatcher, T. W. Huang, H. K. Lee, J. Carlson, M.
C. Weston, F. Chen, Y. Zhang, W. Zhu, C. A. Mohila, N. Ahmed, A. J.
Patel, B. R. Arenkiel, J. L. Noebels, C. J. Creighton, B. Deneen, Nat.
Neurosci. 2017, 20, 396.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (23 of 28)
www.advancedsciencenews.com www.advancedscience.com
[58] S. Gillespie, M. Monje, Annu. Rev. Cancer Biol. 2020, 4, 371.
[59] N. J. Ullrich, S. L. Pomeroy, K. Kapur, P. E. Manley, L. C. Goum-
nerova, T. Loddenkemper, Epilepsia 2015, 56, 1599.
[60] Z. Ye, H. Sontheimer, Cancer Res. 1999, 59, 4383.
[61] H. Vega-Mercado, O. Martín-Belloso, B. L. Qin, F. J. Chang, M. M.
Góngora-Nieto, G. V. Barbosa-Cánovas, B. G. Swanson, Trends Food
Sci. Technol. 1997, 8, 151.
[62] S. Min, G. A. Evrendilek, H. Q. Zhang, IEEE Trans. Plasma Sci. 2007,
35, 59.
[63] L. Buchmann, A. Mathys, Front. Bioeng. Biotechnol. 2019, 7, 265.
[64] B. Geboers, H. J. Scheffer, P. M. Graybill, A. H. Ruarus, S. Nieuwen-
huizen, R. S. Puijk, P. M. van den Tol, R. V. Davalos, B. Rubinsky,
T. D. de Gruijl, D. Miklavcˇicˇ, M. R. Meijerink, Radiology 2020, 295,
192190.
[65] M. Yang, W. J. Brackenbury, Front. Physiol. 2013, 4, 185.
[66] T. Kotnik, L. Rems, M. Tarek, D. Miklavcˇicˇ, Annu. Rev. Biophys. 2019,
48, 63.
[67] R. Heller, M. J. Jaroszeski, L. F. Glass, J. L. Messina, D. P. Rapaport,
R. C. DeConti, N. A. Fenske, R. A. Gilbert, L. M. Mir, D. S. Reintgen,
Cancer 1996, 77, 964.
[68] H. P. Schwan, Adv. Biol. Med. Phys. 1957, 5, 147.
[69] R. W. Glaser, S. L. Leikin, L. V. Chernomordik, V. F. Pastushenko, A.
I. Sokirko, Biochim. Biophys. Acta, Biomembr. 1988, 940, 275.
[70] J. C Neu, W Krassowska, Phys. Rev. E 1999, 59, 3471.
[71] K. N. Aycock, R. V. Davalos, Bioelectricity 2019, 1, 214.
[72] R. S. Son, K. C. Smith, T. R. Gowrishankar, P. T. Vernier, J. C. Weaver,
J. Membr. Biol. 2014, 247, 1209.
[73] M. Kandušer, M. Šentjurc, D. Miklavcˇicˇ, Bioelectrochemistry 2008,
74, 52.
[74] M. Golzio, M. P. Mora, C. Raynaud, C. Delteil, J. Teissié, M. P. Rols,
Biophys. J. 1998, 74, 3015.
[75] G. Serša, M. Cˇemažar, D. Miklavcˇicˇ, Cancer Res. 1995, 55, 3450.
[76] S. Orlowski, J. Belehradek, C. Paoletti, L. M. Mir, Biochem. Pharma-
col. 1988, 37, 4727.
[77] L. G. Campana, I. Edhemovic, D. Soden, A. M. Perrone, M. Scarpa,
L. Campanacci, M. Cemazar, S. Valpione, D. Miklavcˇicˇ, S. Mocellin,
E. Sieni, G. Sersa, Eur. J. Surg. Oncol. 2019, 45, 92.
[78] U. Probst, I. Fuhrmann, L. Beyer, P. Wiggermann, Technol. Cancer
Res. Treat. 2018, 17, 153303381878532.
[79] M. B. Sano, R. E. Fan, L. Xing, Sci. Rep. 2017, 7, 40747.
[80] M. Wichtowski, D. Murawa, K. Kulcenty, K. Zaleska, Breast Care
2017, 12, 409.
[81] M. Bonakdar, E.M.Wasson, Y.W. Lee, R. V. Davalos,Biophys. J. 2016,
110, 503.
[82] F. Mahmood, J. Gehl, Bioelectrochemistry 2011, 81, 10.
[83] A. Rodriguez, S. B. Tatter, W. Debinski, Pharmaceutics 2015, 7, 175.
[84] K. H. Schoenbach, S. J. Beebe, E. S. Buescher, Bioelectromagnetics
2001, 22, 440.
[85] P. T. Vernier, Y. Sun, L. Marcu, C. M. Craft, M. A. Gundersen, Biophys.
J. 2004, 86, 4040.
[86] A. G. Pakhomov, R. Shevin, J. A. White, J. F. Kolb, O. N. Pakhomova,
R. P. Joshi, K. H. Schoenbach, Arch. Biochem. Biophys. 2007, 465,
109.
[87] A. G. Pakhomov, A.M. Bowman, B. L. Ibey, F.M. Andre, O.N. Pakho-
mova, K. H. Schoenbach, Biochem. Biophys. Res. Commun. 2009,
385, 181.
[88] P. T Vernier, A. Li, L. Marcu, C. M. Craft, M. A. Gundersen, IEEE
Trans. Dielectr. Electr. Insul. 2003, 10, 795.
[89] J. A. White, P. F. Blackmore, K. H. Schoenbach, S. J. Beebe, J. Biol.
Chem. 2004, 279, 22964.
[90] W. Ren, S. J. Beebe, Apoptosis 2011, 16, 382.
[91] M. Stacey, J. Stickley, P. Fox, V. Statler, K. Schoenbach, S. J. Beebe, S.
Buescher, Mutat. Res., Genet. Toxicol. Environ. Mutagen. 2003, 542,
65.
[92] E. T. O’Brien, E. D. Salmon, H. P. Erickson, Cell Motil. Cytoskeleton
1997, 36, 125.
[93] L. Carr, S. M. Bardet, R. C. Burke, D. Arnaud-Cormos, P. Leveque, R.
P. O’Connor, Sci. Rep. 2017, 7, 41267.
[94] S. J. Beebe, P. M. Fox, L. J. Rec, K. Somers, R. H. Stark, K. H. Schoen-
bach, IEEE Trans. Plasma Sci. 2002, 30, 286.
[95] S. M. Bardet, L. Carr, M. Soueid, D. Arnaud-Cormos, P. Leveque, R.
P. O’Connor, Sci. Rep. 2016, 6, 34443.
[96] S. Beebe, N. Sain, W. Ren, Cells 2013, 2, 136.
[97] J. Dermol-Cˇerne, D. Miklavcˇicˇ, M. Reberšek, P. Mekucˇ, S. M. Bardet,
R. Burke, D. Arnaud-Cormos, P. Leveque, R. O’Connor, Bioelectro-
chemistry 2018, 122, 103.
[98] A. G. Pakhomov, I. Semenov, M. Casciola, S. Xiao, Biochim. Biophys.
Acta, Biomembr. 2017, 1859, 1273.
[99] R. C. Burke, S.M. Bardet, L. Carr, S. Romanenko, D. Arnaud-Cormos,
P. Leveque, R. P. O’Connor, Biochim. Biophys. Acta, Biomembr. 2017,
1859, 2040.
[100] B. Valicˇ, M. Golzio, M. Pavlin, A. Schatz, C. Faurie, B. Gabriel, J.
Teissié, M. P. Rols, D. Miklavcˇicˇ, Eur. Biophys. J. 2003, 32, 519.
[101] R. Susil, D. Šemrov, D. Miklavcˇicˇ, Electro- Magnetobiol. 1998, 17,
391.
[102] B. Rubinsky, Technol. Cancer Res. Treat. 2007, 6, 255.
[103] B. Rubinsky, G. Onik, P. Mikus, Technol. Cancer Res. Treat. 2007, 6,
37.
[104] E. Maor, A. Ivorra, J. Leor, B. Rubinsky, Technol. Cancer Res. Treat.
2007, 6, 307.
[105] G. Onik, P. Mikus, B. Rubinsky, Technol. Cancer Res. Treat. 2007, 6,
295.
[106] T. Miklovic, E. L. Latouche, M. R. DeWitt, R. V. Davalos, M. B. Sano,
Ann. Biomed. Eng. 2017, 45, 2524.
[107] W. Zhou, Z. Xiong, Y. Liu, C. Yao, C. Li, J. Cancer Res. Ther. 2012, 8,
80.
[108] Z. Zhang, W. Li, D. Procissi, P. Tyler, R. A. Omary, A. C. Larson,
Nanomedicine 2014, 9, 1181.
[109] H. B. Kim, C. K. Sung, K. Y. Baik, K. W. Moon, H. S. Kim, J. H. Yi, J.
H. Jung, M. H. Moon, O. K. Choi, Biochem. Biophys. Res. Commun.
2013, 435, 651.
[110] B. Mercadal, N. Beitel-White, K. N. Aycock, Q. Castellví, R. V. Dava-
los, A. Ivorra, Ann. Biomed. Eng. 2020, 48, 1451.
[111] E. W. Lee, C. T. Loh, S. T. Kee, Technol. Cancer Res. Treat. 2007, 6,
287.
[112] C. C. Fesmire, R. A. Petrella, J. D. Kaufman, N. Topasna, M. B. Sano,
Bioelectrochemistry 2020, 135, 107544.
[113] P. A. Garcia, J. H. Rossmeisl, J. Robertson, T. L. Ellis, R. V. Davalos,
in Proc. 2009 Annual International Conference of the IEEE Engineering
in Medicine and Biology Society, 2009, pp. 6513.
[114] T. L. Ellis, P. A. Garcia, J. H. Rossmeisl, N. Henao-Guerrero, J.
Robertson, R. V. Davalos, J. Neurosurg. 2011, 114, 681.
[115] J. H. Rossmeisl, P. A. Garcia, J. L. Roberston, T. L. Ellis, R. V. Davalos,
J. Vet. Sci. 2013, 14, 433.
[116] P. A. Garcia, T. Pancotto, J. H. Rossmeisl, N. Henao-Guerrero, N. R.
Gustafson, G. B. Daniel, J. L. Robertson, T. L. Ellis, R. V. Davalos,
Technol. Cancer Res. Treat. 2011, 10, 73.
[117] C. Herranz, F. Fernández, R. Martín-Ibáñez, E. Blasco, E. Crespo, C.
De la Fuente, S. Añor, R. M. Rabanal, J. M. Canals, M. Pumarola, J.
Comp. Pathol. 2016, 154, 169.
[118] J. H. Rossmeisl, P. A. Garcia, T. E. Pancotto, J. L. Robertson, N.
Henao-Guerrero, R. E. Neal, T. L. Ellis, R. V. Davalos, J. Neurosurg.
2015, 123, 1008.
[119] C. B. Arena, M. B. Sano, M. N. Rylander, R. V. Davalos, IEEE Trans.
Biomed. Eng. 2011, 58, 1474.
[120] M. B. Sano, C. B. Arena, K. R. Bittleman, M. R. Dewitt, H. J. Cho, C.
S. Szot, D. Saur, J. M. Cissell, J. Robertson, Y. W. Lee, R. V. Davalos,
Sci. Rep. 2015, 5, 14999.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (24 of 28)
www.advancedsciencenews.com www.advancedscience.com
[121] D. C. Sweeney, M. Reberšek, J. Dermol, L. Rems, D. Miklavcˇicˇ, R. V.
Davalos, Biochim. Biophys. Acta, Biomembr. 2016, 1858, 2689.
[122] C. B. Arena, M. B. Sano, J. H. Rossmeisl, J. L. Caldwell, P. A. Garcia,
M. N. Rylander, R. V. Davalos, Biomed. Eng. Online 2011, 10, 102.
[123] M. B. Sano, R. E. Fan, K. Cheng, Y. Saenz, G. A. Sonn, G. L. Hwang,
L. Xing, J. Vasc. Interventional Radiol. 2018, 29, 893.
[124] K. Nielsen, H. J. Scheffer, J. M. Vieveen, A. A. J. M. van Tilborg, S.
Meijer, C. van Kuijk, M. P. van den Tol, M. R. Meijerink, R. A. Bouw-
man, Br. J. Anaesth. 2014, 113, 985.
[125] A. Deodhar, T. Dickfeld, G. W. Single, W. C. Hamilton, R. H. Thorn-
ton, C. T. Sofocleous, M. Maybody, M. Gónen, B. Rubinsky, S. B.
Solomon, Am. J. Roentgenol. 2011, 196, W330.
[126] A. Vižintin, J. Vidmar, J. Šcˇancˇar, D. Miklavcˇicˇ, Bioelectrochemistry
2020, 134, 107523.
[127] E. C. Gianulis, M. Casciola, S. Xiao, O. N. Pakhomova, A. G. Pakho-
mov, Bioelectrochemistry 2018, 119, 10.
[128] A. G. Pakhomov, I. Semenov, S. Xiao, O. N. Pakhomova, B. Gregory,
K. H. Schoenbach, J. C. Ullery, H. T. Beier, S. R. Rajulapati, B. L. Ibey,
Cell. Mol. Life Sci. 2014, 71, 4431.
[129] B. L. Ibey, J. C. Ullery, O. N. Pakhomova, C. C. Roth, I. Semenov, H.
T. Beier, M. Tarango, S. Xiao, K. H. Schoenbach, A. G. Pakhomov,
Biochem. Biophys. Res. Commun. 2014, 443, 568.
[130] A. G. Pakhomov, S. Grigoryev, I. Semenov, M. Casciola, C. Jiang, S.
Xiao, Bioelectrochemistry 2018, 122, 123.
[131] M. B. Sano, C. B. Arena, M. R. DeWitt, D. Saur, R. V. Davalos, Bio-
electrochemistry 2014, 100, 69.
[132] F. Burel-Vandenbos, L. Turchi, M. Benchetrit, E. Fontas, Z. Pedeu-
tour, V. Rigau, F. Almairac, D. Ambrosetti, J.-F. Michiels, T. Virolle,
Neuro-Oncology 2013, 15, 1278.
[133] K. F. Chu, D. E. Dupuy, Nat. Rev. Cancer 2014, 14, 199.
[134] W. Cheung, H. Kavnoudias, S. Roberts, B. Szkandera, W. Kemp, K.
R. Thomson, Technol. Cancer Res. Treat. 2013, 12, 233.
[135] I. A. Siddiqui, R. C. Kirks, E. L. Latouche, M. R. DeWitt, J. H. Swet, E.
H. Baker, D. Vrochides, D. A. Iannitti, R. V Davalos, I. H. McKillop,
Surg. Innovation 2017, 24, 276.
[136] C. C. Fesmire, R. A. Petrella, C. A. Fogle, D. A. Gerber, L. Xing, M. B.
Sano, Ann. Biomed. Eng. 2020, 48, 2233.
[137] E. L. Latouche, C. B. Arena, J. W. Ivey, P. A. Garcia, T. E. Pancotto,
N. Pavlisko, S. S. Verbridge, R. V. Davalos, J. H. Rossmeisl, Technol.
Cancer Res. Treat. 2018, 17, 153303381878528.
[138] C. R. Byron, M. R. DeWitt, E. L. Latouche, R. V. Davalos, J. L. Robert-
son, Front. Vet. Sci. 2019, 6, 265.
[139] S. Dong, H. Wang, Y. Zhao, Y. Sun, C. Yao, Technol. Cancer Res. Treat.
2018, 17, 153303381878969.
[140] M. Giladi, R. S. Schneiderman, T. Voloshin, et al., Sci Rep 2015, 5,
https://doi.org/10.1038/srep18046.
[141] N. Gera, A. Yang, T. S. Holtzman, S. X. Lee, E. T. Wong, K. D. Swan-
son, PLoS One 2015, 10, e0125269.
[142] C. Wenger, P. C. Miranda, R. Salvador, A. Thielscher, Z. Bomzon, M.
Giladi, M. M. Mrugala, A. R. Korshoej, IEEE Rev. Biomed. Eng. 2018,
11, 195.
[143] E. H. Kim, Y. Jo, S. Sai, M. J. Park, J. Y. Kim, J. S. Kim, Y. J. Lee, J. M.
Cho, S. Y. Kwak, J. H. Baek, Y. K. Jeong, J.-Y. Song, M. Yoon, S.-G.
Hwang, Oncogene 2019, 38, 6630.
[144] L. Berkelmann, A. Bader, S. Meshksar, A. Dierks, G. H. Majernik, J.
K. Krauss, K. Schwabe, D. Manteuffel, A. Ngezahayo, Sci. Rep. 2019,
9, 7362.
[145] E. Neuhaus, L. Zirjacks, K. Ganser, L. Klumpp, U. Schüler, D. Zips,
F. Eckert, S. M. Huber, Cancers 2019, 11, 110.
[146] X. Li, F. Yang, B. Rubinsky, IEEE Trans. Biomed. Eng. 2020, 67, 2594.
[147] M. A. Persinger, R. M. Lafrenie, Int. Lett. Chem., Phys. Astron. 2014,
36, 67.
[148] P. C. Miranda, A. Mekonnen, R. Salvador, P. J. Basser, Phys. Med.
Biol. 2014, 59, 4137.
[149] J. A. Tuszynski, C. Wenger, D. E. Friesen, J. Preto, Int. J. Environ. Res.
Public Health 2016, 13, 1128.
[150] R. Stupp, E. T. Wong, A. A. Kanner, D. Steinberg, H. Engelhard, V.
Heidecke, E. D. Kirson, S. Taillibert, F. Liebermann, V. Dbalý, J. Hon-
norat, A. Sloan, M. Malkin, J. C. Landolfi, F. Payer, M. Mehdorn, R. J.
Weil, S. C. Pannullo, M. Westphal, M. Smrcka, L. Chin, H. Kostron,
S. Hofer, Eur. J. Cancer 2012, 48, 2192.
[151] R. Stupp, S. Taillibert, A. A. Kanner, S. Kesari, D. M. Steinberg, S. A.
Toms, L. P. Taylor, F. Lieberman, A. Silvani, K. L. Fink, G. H. Barnett,
J.-J. Zhu, J. W. Henson, H. H. Engelhard, T. C. Chen, D. D. Tran, J.
Sroubek, N. D. Tran, A. F. Hottinger, J. Landolfi, R. Desai, M. Caroli,
Y. Kew, J. Honnorat, A. Idbaih, E. D. Kirson, U. Weinberg, Y. Palti, M.
E. Hegi, Z. Ram, JAMA, J. Am. Med. Assoc. 2015, 314, 2535.
[152] R. Stupp, S. Taillibert, A. Kanner, W. Read, D. M. Steinberg, B. Lher-
mitte, S. Toms, A. Idbaih, M. S. Ahluwalia, K. Fink, F. D. Meco, F.
Lieberman, J.-J. Zhu, G. Stragliotto, D. D. Tran, S. Brem, A. F. Hot-
tinger, E. D. Kirson, G. Lavy-Shahaf, U. Weinberg, C.-Y. Kim, S.-H.
Paek, G. Nicholas, J. Bruna, H. Hirte, M. Weller, Y. Palti, M. E. Hegi,
Z. Ram, JAMA, J. Am. Med. Assoc. 2017, 318, 2306.
[153] M. M. Mrugala, H. H. Engelhard, D. D. Tran, Y. Kew, R. Cavaliere, J.
L. Villano, D. A. Bota, J. Rudnick, A. L. Sumrall, J. J. Zhu, N. Butowski,
Semin. Oncol. 2014, 41, S4.
[154] S. H. Burri, V. Gondi, P. D. Brown, M. P. Mehta, Am. J. Clin. Oncol.
2018, 41, 191.
[155] A. Chaudhry, L. Benson, M. Varshaver, O. Farber, U. Weinberg, E.
Kirson, Y. Palti,World J. Surg. Oncol. 2015, 13, 316.
[156] C. Wenger, H. S. Hershkovich, C. Tempel-Brami, M. Giladi, Z. Bom-
zon, in Brain and Human Body Modeling (Eds: S. Makarov, M. Horn-
der, G. Noetscher), Springer International Publishing, Berlin 2019,
pp. 367–393.
[157] P. M. Biol, P. C. Miranda, A. Mekonnen, R. Salvador, Phys. Med. Biol.
2015, 60, 7339.
[158] E. Lok, P. San, O. Liang, V. White, E. T. Wong, J. Neuro-Oncol. 2020,
147, 125.
[159] A. R. Korshoej, F. L. Hansen, N. Mikic, G. von Oettingen, J. C. H.
Sørensen, A. Thielscher, PLoS One 2018, 13, e0201957.
[160] N. Urman, S. Levy, A. Frenkel, D. Manzur, H. S. Hershkovich, A.
Naveh, O. Yesharim, C. Wenger, G. Lavy-Shahaf, E. Kirson, Z. Bom-
zon, in Brain and Human Body Modeling (Eds: S. Makarov, M. Horn-
der, G. Noetscher), Springer International Publishing, Berlin 2019,
pp. 139–154.
[161] Z. Bomzon, H. S. Hershkovich, N. Urman, A. Chaudhry, D. Garcia-
Carracedo, A. R. Korshoej, U. Weinberg, C. Wenger, P. Miranda, Y.
Wasserman, Kirson ED, Yoram. in Proc. 2016Annual International
Conference of the IEEE Engineering in Medicine and Biology Society,
2016, pp. 6461.
[162] C. Printz, Cancer 2015, 121, 969.
[163] W. Wick, Neuro-Oncology 2016, 18, 303.
[164] L. E. Gaspar, B. J. Fisher, D. R. Macdonald, D. V. Leber, E. C.
Halperin, S. C. Schold, J. G. Cairncross, Int. J. Radiat. Oncol., Biol.,
Phys. 1992, 24, 55.
[165] E. C. Halperin, P. C. Burger, D. E. Bullard, Int. J. Radiat. Oncol., Biol.,
Phys. 1988, 15, 505.
[166] S. W. Lee, B. A. Fraass, L. H. Marsh, K. Herbort, S. S. Gebarski, M.
K. Martel, E. H. Radany, A. S. Lichter, H. M. Sandler, Int. J. Radiat.
Oncol., Biol., Phys. 1999, 43, 79.
[167] L. S. Ashby, K. A. Smith, B. Stea,World J. Surg. Oncol. 2016, 14, 225.
[168] E. Barbarite, J. T. Sick, E. Berchmans, A. Bregy, A. H. Shah, N. El-
sayyad, R. J. Komotar, Neurosurg. Rev. 2017, 40, 195.
[169] L. B. de Paula, F. L. Primo, A. C. Tedesco, Biophys. Rev. 2017, 9, 761.
[170] D. S. Hersh, A. J. Kim, J. A. Winkles, H. M. Eisenberg, G. F. Wood-
worth, V. Frenkel, Neurosurgery 2016, 79, 643.
[171] H. Xu, F. Bihari, S. Whitehead, E. Wong, S. Schmid, M. O. Hebb,
Anticancer Res. 2016, 36, 71.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (25 of 28)
www.advancedsciencenews.com www.advancedscience.com
[172] A. R. Di Sebastiano, A. Deweyert, S. Benoit, E. Iredale, H. Xu, C. De
Oliveira, E. Wong, S. Schmid, M. O. Hebb, Sci. Rep. 2018, 8, 7301.
[173] R. G. W. Verhaak, K. A. Hoadley, E. Purdom, V. Wang, Y. Qi, M. D.
Wilkerson, C. R. Miller, L. Ding, T. Golub, J. P. Mesirov, G. Alexe, M.
Lawrence, M. O’Kelly, P. Tamayo, B. A. Weir, S. Gabriel, W. Winckler,
S. Gupta, L. Jakkula, H. S. Feiler, J. G. Hodgson, C. D. James, J. N.
Sarkaria, C. Brennan, A. Kahn, P. T. Spellman, R. K. Wilson, T. P.
Speed, J. W. Gray, M. Meyerson, Cancer Cell 2010, 17, 98.
[174] Q. Wang, B. Hu, X. Hu, H. Kim, M. Squatrito, L. Scarpace, A. C. de-
Carvalho, S. Lyu, P. Li, Y. Li, F. Barthel, H. J. Cho, Y.-H. Lin, N. Satani,
E. Martinez-Ledesma, S. Zheng, E. Chang, C.-E. G. Sauvé, A. Olar, Z.
D. Lan, G. Finocchiaro, J. J. Phillips,M. S. Berger, K. R. Gabrusiewicz,
G. Wang, E. Eskilsson, J. Hu, T. Mikkelsen, R. A. DePinho, F. Muller,
Cancer Cell 2017, 32, 42.
[175] A. Sottoriva, I. Spiteri, S. G. M. Piccirillo, A. Touloumis, V. P. Collins,
J. C. Marioni, C. Curtis, C. Watts, S. Tavaré, Proc. Natl. Acad. Sci. U.
S. A. 2013, 110, 4009.
[176] M. Guo, M. Van Vliet, J. Zhao, T. D. D. Ståhl, M. S. Lindström,
H. Cheng, S. Heller, M. Nistér, D. Hägerstrand, Neuro-Oncol. Adv.
2020, 2, vdaa061.
[177] C. G. Hadjipanayis, G. Widhalm, W. Stummer, Neurosurgery 2015,
77, 663.
[178] S. Kim, J. E. Kim, Y. H. Kim, T. Hwang, S. K. Kim, W. J. Xu, J. Y. Shin,
J. Il Kim, H. Choi, H. C. Kim, H. R. Cho, A. Choi, T. Chowdhury, Y.
Seo, Y.-S. Dho, J. W. Kim, D. G. Kim, S.-H. Park, H. Kim, S. H. Choi,
S. Park, S.-H. Lee, C.-K. Park, Sci. Rep. 2017, 7, 12221.
[179] D. Yamashita, J. D. Bernstock, G. Elsayed, H. Sadahiro, A. Mo-
hyeldin, G. Chagoya, A. Ilyas, J. Mooney, D. Estevez-Ordonez, S.
Yamaguchi, et al., J. Neurosurg. 2020, 134, 721.
[180] S. P. Caragher, J. M. Shireman, M. Huang, J. Miska, F. Atashi, S.
Baisiwala, C. H. Park, M. R. Saathoff, L. Warnke, T. Xiao, M. S. Les-
niak, C. D. James, H. Meltzer, A. K. Tryba, A. U. Ahmed, J. Neurosci.
2019, 39, 1982.
[181] V. Venkataramani, D. I. Tanev, C. Strahle, A. Studier-Fischer, L.
Fankhauser, T. Kessler, C. Körber, M. Kardorff, M. Ratliff, R. Xie, H.
Horstmann, M. Messer, S. P. Paik, J. Knabbe, F. Sahm, F. T. Kurz,
A. A. Acikgöz, F. Herrmannsdörfer, A. Agarwal, D. E. Bergles, A.
Chalmers, H. Miletic, S. Turcan, C. Mawrin, D. Hänggi, H.-K. Liu,
W. Wick, F. Winkler, T. Kuner, Nature 2019, 573, 532.
[182] S. Darmanis, S. A. Sloan, D. Croote, M. Mignardi, S. Chernikova, P.
Samghababi, Y. Zhang, N. Neff, M. Kowarsky, C. Caneda, G. Li, S.
D. Chang, I. D. Connolly, Y. Li, B. A. Barres, M. H. Gephart, S. R.
Quake, Cell Rep. 2017, 21, 1399.
[183] L. J. M. Perus, L. A. Walsh, Front. Immunol. 2019, 10, 2294.
[184] A. R. P. Antunes, I. Scheyltjens, J. Duerinck, B. Neyns, K. Movahedi,
J. A. Van Ginderachter, Elife 2020, 9, e52176.
[185] M. Niklasson, T. Bergström, M. Jarvius, A. Sundström, F. Nyberg, C.
Haglund, R. Larsson, B. Westermark, B. Segerman, A. Segerman, J.
Pathol. 2019, 249, 295.
[186] J. W. Ivey, E. L. Latouche, M. B. Sano, J. H. Rossmeisl, R. V. Davalos,
S. S. Verbridge, Sci. Rep. 2015, 5, 17157.
[187] J. W. Ivey, E. L. Latouche, M. L. Richards, G. J. Lesser, W. Debinski,
R. V. Davalos, S. S. Verbridge, Biophys. J. 2017, 113, 472.
[188] E. Neuhaus, L. Zirjacks, K. Ganser, L. Klumpp, U. Schüler, D. Zips,
F. Eckert, S. Huber, Cancers 2019, 11, 110.
[189] Y. Pires-Afonso, S. P. Niclou, A. Michelucci, Int. J. Mol. Sci. 2020, 21,
689.
[190] K. Woroniecka, P. Chongsathidkiet, K. Rhodin, H. Kemeny, C.
Dechant, S. H Farber, A. A. Elsamadicy, X. Cui, S. Koyama, C. Jack-
son, L. J. Hansen, T. M. Johanns, L. Sanchez-Perez, V. Chandramo-
han, Y.-R. A. Yu, D. D. Bigner, A. Giles, P. Healy, G. Dranoff, K. J.
Weinhold, G. P. Dunn, P. E. Fecci, Clin. Cancer Res. 2018, 24, 4175.
[191] M. Martinez-Lage, T. M. Lynch, Y. Bi, C. Cocito, G. P. Way, S. Pal, J.
Haller, R. E. Yan, A. Ziober, A. Nguyen, M. Kandpal, D. M. O’Rourke,
J. P. Greenfield, C. S. Greene, R. V. Davuluri, N. Dahmane, Acta Neu-
ropathol. Commun. 2019, 7, 203.
[192] M. Lim, Y. Xia, C. Bettegowda, M. Weller,Nat. Rev. Clin. Oncol. 2018,
15, 422.
[193] M. Jeanmougin, A. B. Håvik, L. Cekaite, P. Brandal, A. Sveen, T. R.
Meling, T. H. Ågesen, D. Scheie, S. Heim, R. A. Lothe, G. E. Lind,
Mol. Oncol. 2020, 14, 1016.
[194] C. Bastianpillai, N. Petrides, T. Shah, S. Guillaumier, H. U. Ahmed,
M. Arya, Tumor Biol. 2015, 36, 9137.
[195] R. E. Neal, J. H. Rossmeisl, J. L. Robertson, C. B. Arena, E. M.
Davis, R. N. Singh, J. Stallings, R. V. Davalos, PLoS One 2013, 8,
e64559.
[196] V. M. Ringel-Scaia, N. Beitel-White, M. F. Lorenzo, R. M. Brock, K.
E. Huie, S. Coutermarsh-Ott, K. Eden, D. K. McDaniel, S. S. Ver-
bridge, J. H. Rossmeisl, K. J. Oestreich, R. V. Davalos, I. C. Allen,
EBioMedicine 2019, 44, 112.
[197] T. Voloshin, N. Kaynan, S. Davidi, Y. Porat, A. Shteingauz, R. S.
Schneiderman, E. Zeevi, M. Munster, R. Blat, C. T. Brami, et al., Can-
cer Immunol. Immunother. 2020, 69, 1191.
[198] N. K. Karanam, K. Srinivasan, L. Ding, B. Sishc, D. Saha,M. D. Story,
Cell Death Dis. 2017, 8, e2711.
[199] N. K. Karanam, L. Ding, A. Aroumougame, M. D. Story, Transl. Res.
2020, 217, 33.
[200] M. Giladi, M. Munster, R. S. Schneiderman, T. Voloshin, Y. Porat,
R. Blat, K. Zielinska-Chomej, P. Hååg, Z. Bomzon, E. D. Kirson, U.
Weinberg, K. Viktorsson, R. Lewensohn, Y. Palti, Radiat. Oncol. 2017,
12, 206.
[201] A. Shteingauz, Y. Porat, T. Voloshin, R. S. Schneiderman, M. Mun-
ster, E. Zeevi, N. Kaynan, K. Gotlib, M. Giladi, E. D. Kirson, U. Wein-
berg, A. Kinzel, Y. Palti, Cell Death Dis. 2018, 9, 1074.
[202] D. Chen, N. Thomas, J. Ren, S. Le, C. Deng, D. Jin, M. Sebastian, D.
Nguyen, D. Tran, Neuro-Oncology 2018, 20, vi130.
[203] A. Sica, M. Erreni, P. Allavena, C. Porta, Cell. Mol. Life Sci. 2015, 72,
4111.
[204] E. Guadagno, I. Presta, D. Maisano, A. Donato, C. K. Pirrone, G.
Cardillo, S. D. Corrado, C. Mignogna, T. Mancuso, G. Donato, M.
D. B. D. Caro, N. Malara, Int. J. Mol. Sci. 2018, 19, 1005.
[205] C. Li, M. Levin, D. L. Kaplan, Sci. Rep. 2016, 6, 1.
[206] B. T. Hawkins, T. P. Davis, Pharmacol. Rev. 2005, 57, 173.
[207] H. L. McConnell, C. N. Kersch, R. L. Woltjer, E. A. Neuwelt, J. Biol.
Chem. 2017, 292, 762.
[208] S. Liebner, A. Fischmann, G. Rascher, F. Duffner, E. H. Grote, H.
Kalbacher, H. Wolburg, Acta Neuropathol. 2000, 100, 323.
[209] S. Liebner, U. Kniesel, H. Kalbacher, H. Wolburg, Eur. J. Cell Biol.
2000, 79, 707.
[210] M. C. Papadopoulos, S. Saadoun, C. J. Woodrow, D. C. Davies, P.
Costa-Martins, R. F. Moss, S. Krishna, B. A. Bell, Neuropathol. Appl.
Neurobiol. 2001, 27, 384.
[211] M. W. Pitz, A. Desai, S. A. Grossman, J. O. Blakeley, J. Neuro-Oncol.
2011, 104, 629.
[212] L. Rosso, C. S. Brock, J. M. Gallo, A. Saleem, P. M. Price, F. E.
Turkheimer, E. O. Aboagye, Cancer Res. 2009, 69, 120.
[213] R. L. Fine, J. Chen, C. Balmaceda, J. N. Bruce, M. Huang, M. Desai,
M. B. Sisti, G. M. McKhann, R. R. Goodman, J. S. Bertino, A. N.
Nafziger, M. R. Fetell, Clin. Cancer Res. 2006, 12, 5770.
[214] T. Jarm, M. Cemazar, D. Miklavcic, G. Sersa, Expert Rev. Anticancer
Ther. 2010, 10, 729.
[215] B. Markelc, E. Bellard, G. Sersa, T. Jesenko, S. Pelofy, J. Teissié, R.
Frangez, M. P. Rols, M. Cemazar, M. Golzio, J. Controlled Release
2018, 276, 30.
[216] S. Sharabi, D. Guez, D. Daniels, I. Cooper, D. Atrakchi, S. Liraz-
Zaltsman, D. Last, Y. Mardor, Sci. Rep. 2020, 10, 2178.
[217] M. Hjouj, D. Last, D. Guez, D. Daniels, S. Sharabi, J. Lavee, B. Ru-
binsky, Y. Mardor, PLoS One 2012, 7, e42817.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (26 of 28)
www.advancedsciencenews.com www.advancedscience.com
[218] F.Mahmood, R.H.Hansen, B. Agerholm-Larsen,H. Gissel, P. Ibsen,
J. Gehl, Acta Oncol. 2015, 54, 289.
[219] S. Sharabi, B. Kos, D. Last, D. Guez, D. Daniels, S. Harnof, Y. Mar-
dor, D. Miklavcic, Radiol. Oncol. 2016, 50, 28.
[220] J. Bicker, G. Alves, A. Fortuna, A. Falcão, Eur. J. Pharm. Biopharm.
2014, 87, 409.
[221] M. P. L. Chan, S. Morisawa, A. Nakayama, Y. Kawamoto, M. Yoneda,
Environ. Toxicol. 2006, 21, 223.
[222] S. Nakagawa,M. A. Deli, H. Kawaguchi, T. Shimizudani, T. Shimono,
Á. Kittel, K. Tanaka, M. Niwa, Neurochem. Int. 2009, 54, 253.
[223] S. Sharabi, Y. Bresler, O. Ravid, C. Shemesh, D. Atrakchi, M.
Schnaider-Beeri, F. Gosselet, L. Dehouck, D. Last, D. Guez, D.
Daniels, Y. Mardor, I. Cooper, Drug Delivery 2019, 26, 459.
[224] M. Bonakdar, P. M. Graybill, R. V. Davalos, RSC Adv. 2017, 7,
42811.
[225] A. F. Kessler, T. Linsenmann, T. Westermaier, W. Wolber, J. Weiland,
C. M. Monoranu, M. Breun, C. Hagemann, R. I. Ernestus, M. Löhr,
Oncol. Lett. 2020, 19, 557.
[226] A. F. Kessler, E. Salvador, D. Domröse, M. Burek, C. Schaeffer, C. T.
Brami, T. Voloshin, M. Giladi, R. I. Ernestus, M. Löhr, C. Förster, C.
Hagemann, Int. J. Radiat. Oncol., Biol., Phys. 2019, 105, S162.
[227] E. Chang, C. B. Patel, C. Pohling, C. Young, J. Song, T. A. Flores, Y.
Zeng, L.-M. Joubert, H. Arami, A. Natarajan, R. Sinclair, S. S. Gamb-
hir, Cell Death Discovery 2018, 4, 113.
[228] E. H. Kim,H. S. Song, S. H. Yoo,M. Yoon,Oncotarget 2016, 7, 65125.
[229] F. Winkler, M. Osswald, W. Wick, Oncol. Res. Treat. 2018, 41, 181.
[230] D. Doyle, Oxford Textbook of Palliative Medicine, Oxford University
Press, Oxford 2004.
[231] C. F. de Souza, T. S. Sabedot, T. M. Malta, L. Stetson, O. Morozova,
A. Sokolov, P.W. Laird,M.Wiznerowicz, A. Iavarone, J. Snyder, A. de-
Carvalho, Z. Sanborn, K. L. McDonald, W. A. Friedman, D. Tirapelli,
L. Poisson, T. Mikkelsen, C. G. Carlotti Jr., S. Kalkanis, J. Zenklusen,
S. R. Salama, J. S. Barnholtz-Sloan, H. Noushmehr, Cell Rep. 2018,
23, 637.
[232] H. Binder, E. Willscher, H. Loeffler-Wirth, L. Hopp, D. T. W. Jones, S.
M. Pfister, M. Kreuz, D. Gramatzki, E. Fortenbacher, B. Hentschel,
M. Tatagiba, U. Herrlinger, H. Vatter, J. Matschke, M. Westphal, D.
Krex, G. Schackert, J. C. Tonn, U. Schlegel, H.-J. Steiger, W. Wick, R.
G. Weber, M. Weller, M. Loeffler, Acta Neuropathol. Commun. 2019,
7, 59.
[233] J. Klughammer, B. Kiesel, T. Roetzer, N. Fortelny, A. Nemc, K.
H. Nenning, J. Furtner, N. C. Sheffield, P. Datlinger, N. Peter,
M. Nowosielski, M. Augustin, M. Mischkulnig, T. Ströbel, D. Al-
par, B. Ergüner, M. Senekowitsch, P. Moser, C. F. Freyschlag,
J. Kerschbaumer, C. Thomé, A. E. Grams, G. Stockhammer, M.
Kitzwoegerer, S. Oberndorfer, F. Marhold, S. Weis, J. Trenkler, J.
Buchroithner, J. Pichler, Nat. Med. 2018, 24, 1611.
[234] A. Wenger, S. F. Vega, T. Kling, T. O. Bontell, A. S. Jakola, H. Caren,
Neuro-Oncology 2019, 21, 616.
[235] L. B. Hoang-Minh, F. A. Siebzehnrubl, C. Yang, S. Suzuki-Hatano,
K. Dajac, T. Loche, N. Andrews, M. S. Massari, J. Patel, K. Amin, A.
Vuong, A. Jimenez-Pascual, P. Kubilis, T. J. Garrett, C. Moneypenny,
C. A. Pacak, J. Huang, E. J. Sayour, D. A. Mitchell, M. R. Sarkisian,
B. A. Reynolds, L. P. Deleyrolle, EMBO J. 2018, 37, e98772.
[236] B. D. Liebelt, T. Shingu, X. Zhou, J. Ren, S. A. Shin, J. Hu, Stem Cells
Int. 2016, 2016, 7849890.
[237] S. Zheng, L. Xin, A. Liang, Y. Fu, Cytotechnology 2013, 65, 505.
[238] S. Bao, Q. Wu, R. E. McLendon, Y. Hao, Q. Shi, A. B. Hjelmeland,
M. W. Dewhirst, D. D. Bigner, J. N. Rich, Nature 2006, 444, 756.
[239] E. R. Blazek, J. L. Foutch, G. Maki, Int. J. Radiat. Oncol., Biol., Phys.
2007, 67, 1.
[240] S. G. M. Piccirillo, B. A. Reynolds, N. Zanetti, G. Lamorte, E. Binda,
G. Broggi, H. Brem, A. Olivi, F. Dimeco, A. L. Vescovi, Nature 2006,
444, 761.
[241] F. Pires, Q. Ferreira, C. A. V. Rodrigues, J. Morgado, F. C. Ferreira,
Biochim. Biophys. Acta, Gen. Subj. 2015, 1850, 1158.
[242] A. Dirkse, A. Golebiewska, T. Buder, P. V. Nazarov, A. Muller, S.
Poovathingal, N. H. C. Brons, S. Leite, N. Sauvageot, D. Sarkis-
jan, M. Seyfrid, S. Fritah, D. Stieber, A. Michelucci, F. Hertel, C.
Herold-Mende, F. Azuaje, A. Skupin, R. Bjerkvig, A. Deutsch, A.
Voss-Böhme, S. P. Niclou, Nat. Commun. 2019, 10, 1787.
[243] N. Chaicharoenaudomrung, P. Kunhorm, W. Promjantuek, N. Ru-
janapun, N. Heebkaew, N. Soraksa, P. Noisa, In Vivo 2020, 34, 199.
[244] A. A. Stepanenko, V. P. Chekhonin, Biomedicines 2019, 7, 92.
[245] M. Rabé, S. Dumont, A. Álvarez-Arenas, H. Janati, J. Belmonte-
Beitia, G. F. Calvo, C. Thibault-Carpentier, Q. Séry, C. Chauvin, N.
Joalland, F. Briand, S. Blandin, E. Scotet, C. Pecqueur, J. Clairam-
bault, L. Oliver, V. Perez-Garcia, A. Nadaradjane, P.-F. Cartron, C.
Gratas, F. M. Vallette, Cell Death Dis. 2020, 11, 19.
[246] A. Lo Dico, D. Salvatore, C. Martelli, D. Ronchi, C. Diceglie, G. Lu-
cignani, L. Ottobrini, Cells 2019, 8, 1315.
[247] M. P. Ventero, M. Fuentes-Baile, C. Quereda, E. Perez-Valeciano, C.
Alenda, P. Garcia-Morales, D. Esposito, P. Dorado, V. M. Barbera,
M. Saceda, PLoS One 2019, 14, e0212581.
[248] W. Alafate, X. Li, J. Zuo, H. Zhang, J. Xiang, W. Wu, W. Xie, X. Bai, M.
Wang, J. Wang, CNS Neurosci. Ther. 2020, 26, 475.
[249] S. Yamamuro, Y. Okamoto, E. Sano, Y. Ochiai, A. Ogino, T. Ohta, H.
Hara, T. Ueda, T. Nakayama, A. Yoshino, Y. Katayama, Int. J. Oncol.
2015, 47, 91.
[250] R. V. D., S. S. V. J. W. Ivey, E. M. Wasson, N. Alinezhadbalalami, A.
Kanitkar, W. Debinski, Z. Sheng, AAAS Res. 2019, 2019, 14.
[251] S. J. Florczyk, K. Wang, S. Jana, D. L. Wood, S. K. Sytsma, J. G. Sham,
F. M. Kievit, M. Zhang, Biomaterials 2013, 34, 10143.
[252] K. Gosselin, E. Deruy, S.Martien, C. Vercamer, F. Bouali, T. Dujardin,
C. Slomianny, L. Houel-Renault, F. Chelli, Y. De Launoit, C. Abbadie,
Am. J. Pathol. 2009, 174, 423.
[253] M. Ogata, S. Hino, A. Saito, K. Morikawa, S. Kondo, S. Kanemoto,
T. Murakami, M. Taniguchi, I. Tanii, K. Yoshinaga, S. Shiosaka, J. A.
Hammarback, F. Urano, K. Imaizumi,Mol. Cell. Biol. 2006, 26, 9220.
[254] X. Sui, R. Chen, Z. Wang, Z. Huang, N. Kong, M. Zhang, W. Han,
F. Lou, J. Yang, Q. Zhang, X. Wang, C. He, H. Pan, Cell Death Dis.
2013, 4, e838.
[255] E. D. Kirson, R. S. Schneiderman, V. Dbal, F. Tovary, J. Vymazal, A.
Itzhaki, D. Mordechovich, Z. Gurvich, E. Shmueli, D. Goldsher, Y.
Wasserman, Y. Palti, BMC Med. Phys. 2009, 9, 1.
[256] M. Silginer, M. Weller, R. Stupp, P. Roth, Cell Death Dis.. 2017, 8,
e2753.
[257] J. Vymazal, E. T. Wong, Semin. Oncol. 2014, 41, S14.
[258] G. Lu, M. Rao, P. Zhu, B. Liang, R. T. El-Nazer, E. Fonkem, M. B.
Bhattacharjee, J. J. Zhu, Front. Neurol. 2019, 10, 42.
[259] G. Ansstas, D. D. Tran, Case Rep. Neurol. 2016, 8, 1.
[260] M. E. Lacouture, M. E Davis, G. Elzinga, N. Butowski, D. Tran, J. L.
Villano, L. Dimeglio, A. M. Davies, E. T. Wong, Semin. Oncol. 2014,
41, S1.
[261] S. D Precilla, S. S. Kuduvalli, A. T. Sivasubramanian, Cell Biol. Int.
2021, 45, 18.
[262] E. B. Golden, H. Y. Cho, A. Jahanian, F. M. Hofman, S. G. Louie, A.
H. Schönthal, T. C. Chen, J. Neurosurg. 2014, 37, E12.
[263] Y.-Y. Chen, Y.-M. Chang, K.-Y. Wang, P.-N. Chen, Y.-C. Hseu, K.-M.
Chen, K.-T. Yeh, C.-J. Chen, L.-S. Hsu, Environ. Toxicol. 2019, 34, 233.
[264] T. H. Ung, H. Malone, P. Canoll, J. N. Bruce, CNS Oncol. 2015, 4,
225.
[265] L. Waldherr, M. Seitanidou, M. Jakešová, V. Handl, S. Honeder, M.
Nowakowska, T. Tomin, M. K. Rad, T. Schmidt, J. Distl, R. Birner-
Gruenberger, G. von Campe, U. Schäfer, M. Berggren, B. Rinner, M.
Asslaber, N. Ghaffari-Tabrizi-Wizsy, S. Patz, D. T. Simon, R. Schindl,
Adv. Mater. Technol. 2021, 6, 2001302.
[266] S. W. Cramer, C. C. Chen, Front. Surg. 2020, 6, 81.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (27 of 28)
www.advancedsciencenews.com www.advancedscience.com
Elise Jenkins receivedherB.Eng. (Hons) in Electrical andElectronic Engineering in 2017 fromGriffith
University, Australia. In 2017 she studiedmethodsof vibration energy harvestingusingpiezoelectric,
micro-electromechanical systemsat theNanoscienceCentre,University of Cambridge. In 2019 she
startedher Ph.D. in bioelectronics at theDepartment of Engineering,University of Cambridge, devel-
oping an in vitro platform to study the biophysical effects of electrical stimulation in brain cancer. She
is particularly interested in proliferation and cell signalingunder the influenceof electric fields and is
exploringopportunities for therapeutic translation.
Alina Finch receivedherBSc (Hons) inMedical Physiology (2014) andPh.D. (2018) from theUniver-
sity of Leicester in ion channel physiologywhere she studied the biophysics and functionof hEAG1,
anoncogenic potassiumchannel.Her current research focuses ongeneratingpatient-specific in vitro
and in vivomodels of humanbrain tumours to improvepre-clinical drugs testing and thus increase
the chances of successful clinical trials. She is interested in further studyinghowneurons and tumour
cells interactwith one another andhow tumour cells co-opt healthy surrounding cells to further their
growth and invade into other brain regions.
MagdaGerigk receivedherMSc inPharmaceutical Biotechnology in 2013 fromGdanskUniversity of
Technology, Poland. From2014 to 2015 sheworkedwithin at theUniversity of Alabama,Birmingham,
USA,where sheusedmousemodels of brain tumours to study gliomagenesis. From2015 to 2019
sheundertook aPh.D. projectwithin theDepartment of Engineering at theUniversity of Cambridge,
where shedeveloped amicrofluidic-basedplatform for investigating the interactions betweenbrain
tumour stemcells andnormal brain vasculature.Her current focus is ondeveloping an implantable
drugdelivery device that could beused for the treatment of brain cancer.
Adv. Sci. 2021, 2100978 © 2021 The Authors. Advanced Science published by Wiley-VCH GmbH2100978 (28 of 28)