Pediatric Neurology 63 (2016) 6e22

Contents lists available at ScienceDirect

Pediatric Neurology
journal homepage: www.elsevier.com/locate/pnu

Topical Review

Applying the Lessons of Tuberous Sclerosis: The 2015 Hower

Award Lecture
E. Steve Roach MD *
Division of Child Neurology, Ohio State University College of Medicine, Nationwide Childrens Hospital, Columbus, Ohio

abstract

Tuberous sclerosis complex is a dominantly inherited disorder that variably affects the brain, skin, kidneys, heart,
and other organs. Its neurological manifestations include epilepsy, autism, cognitive and behavioral dysfunction,
and giant cell tumors. A mutation of either TSC1 or TSC2 can cause tuberous sclerosis complex. Their two gene
products, hamartin and tuberin, form a physical complex which normally inhibits protein synthesis mediated
through the mechanistic target of rapamycin, so a TSC1 or TSC2 mutation results in overactivation of the mech-
anistic target of rapamycin cascade. In addition to their tumor suppressor roles, TSC1 and TSC2 help to regulate cell
size, neuronal migration, axon formation, and synaptic plasticity. Clinical trials of two different the mechanistic
target of rapamycin inhibitors have demonstrated substantial improvement of tuberous sclerosis complexerelated
tumors, and a recent trial also showed a benet from the mechanistic target of rapamycin inhibitor everolimus in
the treatment of refractory epilepsy due to tuberous sclerosis complex. Effective mechanism-based therapy is now
available for some manifestations of tuberous sclerosis complex.

Keywords: tuberous sclerosis complex, mTOR, rapamycin, everolimus, history, subependymal giant cell astrocytoma, epilepsy,
history
Pediatr Neurol 2016; 63: 6-22
2016 The Author. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).

Promise me youll always remember: youre braver than
you believe, and stronger than you seem, and smarter than
you think.
A. A. Milne
Winnie-the-Pooh
Introduction
Tuberous sclerosis complex (TSC) is a dominantly
inherited disorder of cellular differentiation and prolifera-
tion which variably affects the brain, skin, kidneys, heart,
and other organs. The neurological manifestations of TSC
include epilepsy, autism, cognitive and behavioral
dysfunction, and giant cell tumors. Up to three fourths of
Article History:
Received June 9, 2016; Accepted in nal form July 7, 2016
* Communications should be addressed to: Dr. Roach; Division of
Child Neurology; Nationwide Childrens Hospital; 700 Childrens Drive;
Columbus, Ohio 43245.
E-mail addresses: esroach@earthlink.net, roache@nationwidechildrens.org
the individuals with TSC have a spontaneous mutation.
Because of its striking variability of clinical expression, the
diagnosis of TSC is occasionally difcult, especially in
younger individuals or in those with subtle ndings.1 Two
genes are responsible for tuberous sclerosis: TSC1 on
chromosome 9q34 (which encodes the protein hamartin)
and TSC2 on chromosome 16p13.3 (which encodes the
protein tuberin). Effective treatment is now available for
some of the manifestations of TSC, therapy that specically
targets the molecular dysfunction caused by TSC mutations.
My own involvement with TSC began with a presenta-
tion at the 1986 annual Child Neurology Society meeting in
Boston. By the time of this presentation, TSC had already
been known to physicians for 124 years. Yet much of what
we now know about this often devastating disease has been
discovered in the three decades since 1986. That such
stunning progress can be made in the course of one gen-
eration of physicians provides hope that TSC can serve as a
prototype for the development of mechanism-based ther-
apy for other genetic diseases. Here I provide a historical
perspective on some of the things we have learned about

0887-8994/ 2016 The Author. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
http://dx.doi.org/10.1016/j.pediatrneurol.2016.07.003
 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
TSC in the last three decades. There are lessons from the
progress we have made with TSC that can be applied to
other serious genetic diseases.
The early history of TSC
The false sense of security afforded by misinformation or
misinterpretation of facts is far worse than simple
ignorance.
Tuberous sclerosis complex was rst described by von
Recklinghausen (Fig 1) at a March 25, 1862, meeting of the
Berlin Obstetrical Society.2 He presented a newborn baby
who died after a few breaths because of multiple cardiac
myomata. The babys brain contained numerous sclerotic
areas, although von Recklinghausen did not elaborate.
Although von Recklinghausens patient almost certainly
had TSC, his appreciation of its clinical manifestations was
clearly rudimentary.
Eighteen years later, in 1880, Bourneville3 (Fig 2) added
important clinical neurological details. He characterized the
brain pathology as sclrose tubreuse des circonvolution
crbrales, thus coining the name we still use for the dis-
order. Nevertheless, Bournevilles understanding of TSC was
initially limited. He noted his patients facial angiobromas
but concluded that they were coincidental. Bourneville and
Brissaud4 later would tie renal tumors to TSC.
The genetic nature of TSC was quickly recognized. In
1885, only ve years after the report by Bourneville, Balzer
and Menetrier5 described TSC in a mother and daughter.
They were also the rst to connect the characteristic facial
FIGURE 1.
Friedrich Daniel von Recklinghausen, 1833 to 1910.
7
angiobromas (Fig 3) with TSC. During the next few years,
Kothe described the ungual bromas of TSC, and Campbell7
described the retinal TSC lesions (Fig 4).
In 1908, Heinrich Vogt9 proposed quasi-diagnostic
criteria for TSC, a triad consisting of epilepsy, idiocy, and
adenoma sebaceum. Because facial angiobromas (ade-
noma sebaceum) have a high level of specicity for TSC,
most individuals who manifest all three features of Vogts
triad will in fact have TSC, but a clinician using Vogts triad
would likely fail to diagnose half the TSC patients that we
now recognize.
Hypomelanotic macules (ash leaf spots) occur in over
90% of the individuals with TSC, so it is surprising that the
association of these lesions with TSC went unrecognized
until a 1932 report by Critchley and Earl.10 Over three de-
cades later, Gold and Freeman11 along with Fitzpatrick
et al.12 nally established hypomelanotic macules (Fig 3) as
an important early sign of TSC.
Chasing the tuberous sclerosis phenotypes
No matter how well-accepted an idea it may be, without
evidence, dogma is still just dogma.
The clinical features of TSC are highly variable, and even
affected members of the same family often manifest widely
differing clinical ndings. The resulting clinical manifesta-
tions often represent a substantial nancial and psychoso-
cial burden to the patient and family.13 Not unexpectedly for
a complicated disorder such as TSC, over the years, there
have been points of controversy.
Cutaneous lesions
Skin ndings of TSC include hypomelanotic macules,
ungual or periungual bromas, a shagreen patch, and facial
angiobromas (Fig 3). Hypomelanotic macules occur in over
90% of individuals with TSC. They are often difcult to see in
light-skinned newborns without the aid of an ultraviolet
light. Facial angiobromas occur in about three fourths of
patients and typically start to appear during the preschool
years. A shagreen patch occurs in 20% to 30% of TSC patients,
typically on the lower back or ank area. It is an irregularly
shaped, slightly raised or textured skin lesion. Ungual -
bromas are eshy lesions that arise adjacent to or from
underneath the nails. Single bromas occasionally develop
after trauma. Sometimes the broma itself is not visible but
creates a prominent longitudinal groove in the ngernail, a
nding which also has diagnostic signicance (Fig 3).
Retinal lesions
Retinal lesions occur in up to 87% of people with TSC.14
Varied retinal abnormalities include the classic mulberry
lesion adjacent to the optic disc, abnormalities of the retinal
pigment layer, plaque-like hamartomas, and depigmented
areas in the retina (Fig 4). Most retinal lesions are clinically
inconsequential, but occasional patients have visual
impairment due to the occurrence of a large macular lesion,
a retinal detachment, or a vitreous hemorrhage.
8 E.S. Roach / Pediatric Neurology 63 (2016) 6e22

FIGURE 2.
(A) Dsir-Magloire Bourneville, 1840 to 1909. (B) Bournevilles illustration of his patients brain tubers.3

Cardiac lesions of blood ow by an intraluminal tumor or to inadequate

Two thirds to three fourths of newborns with TSC have
one or more cardiac rhabdomyomas on echocardiography,
but most of these lesions remain asymptomatic.15-17 Rhab-
domyomas are often multiple (Fig 5) and are commonly
evident during prenatal ultrasounds.18 These lesions do not
grow and typically shrink during the rst few months after
birth. Most individuals with cardiac dysfunction present
soon after birth with heart failure due either to obstruction
normal myocardium to maintain perfusion. A few children
later develop cardiac arrhythmias.
Renal lesions
Renal angiomyolipomas (AMLs) are demonstrable in
75% to 80% of TSC patients by 10 years of age.19 Multiple
tumors are typical (Fig 6A), and lesions larger than 4 cm
are more likely to become symptomatic than smaller

FIGURE 3.
The cutaneous lesions of tuberous sclerosis complex include (A) facial angiobromas (adenoma sebaceum), (B) ungual broma, (C) hypomelanotic
macules (ash leaf spots), and (D) a shagreen patch. Parts C and D are reproduced with permission from Roach.6
 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
FIGURE 4.
(A) Retinal astrocytic hamartoma in an individual with tuberous sclerosis complex (TSC). (B) Retinal pigmentary lesions in an adult with TSC. Part A is
reproduced with permission from Roach.8
tumors. Most of these lesions represent histologically
benign tumors with varying amounts of vascular tissue,
fat, and smooth muscle (Fig 6B). Renal cell carcinoma or
other malignancies occur in a small number of TSC pa-
tients, and it can be difcult to noninvasively distinguish a
malignant kidney tumor from an AML that contains little
fat. Single or multiple renal cysts are also a feature of TSC,
and about 1% of the individuals with TSC also exhibit the
severe phenotype of polycystic kidney disease (PKD). Iso-
lated renal cysts may disappear on repeated imaging.
Pulmonary dysfunction
The classic pulmonary lesion of TSC is lymphangioleio-
myomatosis (LAM). Other patients exhibit multifocal
micronodular pneumocytic hyperplasia. Manifestations
include spontaneous pneumothorax, dyspnea on exertion,
and frank respiratory failure. Pulmonary disease is about
ve times more common in females than in males, and it is
uncommon to develop pulmonary symptoms before
adulthood.
FIGURE 5.
Prenatal ultrasound showing multiple echogenic cardiac rhabdomyomas in
a baby who later proved to have tuberous sclerosis complex.
9
Neurological dysfunction
The principal neurological manifestations of TSC are
intellectual disability, epilepsy, autism, and various
behavioral abnormalities. Various types of seizures occur
in over 90% of patients, and infantile spasms are particu-
larly common in younger individuals.21 Vigabatrin may be
more effective for infantile spasms due to TSC than those
arising from other causes, so it is typically administered
instead of corticosteroids or adrenocorticotrophic hor-
mone in these children.22-24 Although the seizures may be
refractory to therapy, seizure control with medication is
often possible, and medication can even be halted in some
individuals without the return of seizures.25 Surgical
removal of the epileptogenic cortical lesion is an option for
some individuals.26,27 Hemispherectomy is usually neces-
sary for the children with hemimegalencephaly related to
TSC,28 and corpus callosotomy can be helpful in children
with atonic seizures or frequent rapid secondary general-
ization of the seizures.
Almost all individuals with marked cognitive impair-
ment have epilepsy, but many patients who have epilepsy
are normally intelligent. Children with infantile spasms are
more likely to exhibit long-term cognitive impairment.29
Cognition ranges from normal to severe impairment, but
close to half of the patients are cognitively intact. Autism
spectrum disorder occurs in 25% to 50% TSC patients, and
attention decit disorder, aggressiveness, obsessive-
compulsive behavior, and even frank psychosis are com-
mon, often in conjunction with epilepsy or intellectual
decit.
The subependymal nodules which characterize TSC
typically arise adjacent to the lateral ventricular wall and
protrude into the ventricles (Fig 7).1 When calcied, these
lesions are easily conrmed with computed tomography
(Fig 8). Even uncalcied subependymal nodules are readily
apparent on magnetic resonance imaging (MRI), and the
cortical tubers are far more obvious with MRI than with
computed tomography. Evidence of abnormal neuronal
migration can be seen in some patients as a high signal
linear lesions running perpendicular to the cortex on T2-
weighted MRI (Fig 9).
10 E.S. Roach / Pediatric Neurology 63 (2016) 6e22

FIGURE 6.
(A) Longitudinally sectioned kidney reveals a large fat-laden angiomyolipoma (AML) in the lower pole. Several small AMLs are also evident in the renal
cortex. (B) Renal AMLs contain vascular, smooth muscle, and fat moieties (hematoxylin and eosin at
110). Part A was reproduced with permission from
Weiner et al.20

Depending on whether the diagnosis is rst made clini- lesions typically arise in the anterior portion of the lateral
cally or radiographically, subependymal giant cell tumors ventricles (Fig 9) and typically develop in children rather
(SEGAs) occur in 6% to 14% of individuals with TSC. These than in adults.31 Unlike the more common cortical tubers,

FIGURE 7.
(A) Brain section shows a cortical tuber in the left frontal region. (B) Subependymal nodules protruding into the ventricles (arrow). Reproduced with
permission from Roach.8
 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
FIGURE 8.
Computed tomography reveals typical calcied subependymal nodules
protruding into the lateral ventricles. Note also a large calcied cortical
lesion in the left posterior hemisphere (black arrow) and scattered cortical
hypodense cortical tubers (white arrows). Reproduced with permission
from Roach et al.30
SEGAs often enlarge and cause symptoms such as increased
intracranial pressure, new focal neurological decits, or
deterioration of seizure control.31,32 Surgical resection of an
expanding SEGA can be effective especially when the tumor
is not extensive, but increasingly, these lesions are managed
medically.31,33
The lesion burden conundrum
We suggested in 1987 that individuals with numerous
brain lesions on MRI were more likely to experience
intractable epilepsy and severe cognitive impairment.34
This seemingly common sense observation was unexpect-
edly controversial, because until then, severe intellectual
impairment in individuals with TSC had been attributed
solely to their uncontrolled seizures. A meta-analysis of this
and other studies conrmed that the cortical tuber count
constituted a biomarker that predicted an increased likeli-
hood of severe neurological impairment.35
The effect of the cortical lesion number on the severity of
neurological dysfunction is likely to be far more nuanced
than these early results suggest. The presence of numerous
brain lesions (Fig 10), for example, does not mean that
frequent seizures do not also contribute to a poor outcome.
Although the early analyses focused on the number of
cortical lesions, the size and location of the brain lesions are
likely to play a greater role than the number. Our 1987 study
used a prototype .15T MRI unit; it had limited image reso-
lution and did not identify the smaller lesions that are now
11
easily imaged, in effect ltering out all but the larger cortical
lesions.34 It is unlikely that a study using modern imaging
equipment that focused solely on the number of lesions
without regard to their size would show such a robust
correlation. Finally, recent studies suggest that the brain
lesions and the brains white matter connections may
evolve over time.36,37
Is surveillance testing worthwhile?
Another area of early debate centered on the need for
surveillance testing in an effort to identify TSC complications
at an earlier stage. Some physicians advocated an annual
complete assessment, whereas others preferred to deal with
any complications as they arose. Given the variety of
possible clinical manifestations, a complete annual evalua-
tion seemed burdensome to the families and often wasted
health care resources, but waiting on the complications to
occur could make them more difcult to address. At the
1998 consensus conference in Annapolis, Maryland, we
developed a systematic approach to periodic screening for
complications, which was based loosely on then existing
public health screening recommendations.38 To justify
screening, a complication should pose a signicant risk,
there should be a noninvasive means of evaluating it, and
earlier diagnosis should convey a substantial management
benet. We concluded that the lesions that met these three
requirements were renal AMLs, especially in postadolescent
patients, and cerebral giant cell tumors before adulthood.
Each of these lesions is treatable and is easier to manage if
identied when smaller and asymptomatic, so the surveil-
lance guidelines focused on periodic imaging of the brain
and kidneys.38 The latest revision of these guidelines
retained a similar philosophy but added periodic screening
for neuropsychiatric dysfunction,39,40 shifted the renal im-
aging modality from ultrasound to MRI, and added pulmo-
nary screening of adult women.41
Developing family partnerships
Recognize the power of an effective partnership between
families and clinicians and researchers.
Real heroes are ordinary people who do extraordinary
things. Adrianne Cohen, Verna Morris, Linda Hamm, and
Susan Diaz, each a mother of a child affected by TSC, foun-
ded the National Tuberous Sclerosis Association (NTSA) in
1974. Dissatised by the lack of understanding of their
childrens disease and the paucity of ongoing research,
these four women gathered for coffee one morning in one of
their kitchens, and from this meeting grew NTSA.
NTSA was renamed the Tuberous Sclerosis Alliance in
2000. The Tuberous Sclerosis Alliance has grown into a
particularly effective support organization, providing in-
formation for professionals and families, supporting clinical
and basic research, and supporting patients with TSC and
their families. The organization has provided start-up
funding for basic and clinical research and training grants
that launch new investigators. It supports research and
consensus conferences and press government ofcials for
more research funding for TSC.
12 E.S. Roach / Pediatric Neurology 63 (2016) 6e22

FIGURE 9.
(A) T1-weighted magnetic resonance imaging (MRI) shows subependymal nodules in a child with tuberous sclerosis complex. (B) Cranial MRI reveals radial
migration lines positioned between the subependymal area and the cortex. (C) MRI with gadolinium contrast shows the typical appearance and location
of a subependymal giant cell astrocytoma.

The Gomez textbook
A well-researched book or seminal paper can dene a eld
of study, introduce new lines of thought, and ignite
progress.
A turning point in our understanding of TSC occurred in
1979 with the publication of the rst edition of Tuberous
Sclerosis, edited by Manuel R. Manny Gomez (Fig 11).2 His
book was the rst comprehensive source of information
about TSC, with contributions from neurologists, cardiolo-
gists, nephrologists, radiologists, dermatologists, ophthal-
mologists, and pathologists. For the rst time, there was one
convenient source of information about all aspects of the
disease and a broadening recognition of TSC manifestations
beyond a single discipline.
The Gomez book documented the phenotypic variability
of TSC and demonstrated the existence of milder TSC phe-
notypes. The recognition of TSCs diverse and milder phe-
notypes heralded a several-fold increase in the prevalence
of TSC, now estimated to be about 1 in 6000 to 1 in 10,000
individuals.42,43
Diagnostic criteria for tuberous sclerosis
Diagnostic criteria represent only an operational denition
of a condition pending more information and insight.
Aside from the 1908 Vogt triad, the rst TSC diagnostic
criteria appeared in the Gomezs 1988 second edition of
Tuberous Sclerosis.44 The Gomez diagnostic criteria assumed
that a number of its more suggestive clinical manifestations
 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
FIGURE 10.
(A) T2-weighted magnetic resonance imaging (MRI) from a child with severe cognitive impairment shows numerous cortical and subcortical lesions. (B)
Note the paucity of hemispheric lesions on this MRI from another child with TSC who has normal cognitive function.
were pathognomonic of TSC, then tabulated the ndings to
arrive at a denitive or a presumptive diagnosis.
The Gomez diagnostic criteria provided a framework for
a clinician to consider less-specic ndings when making a
diagnosis, and for typical patients with multiple classic
features of TSC, the criteria were quite accurate. However,
there was no way to conrm the assumed pathognomonic
nature of the lesions, and indeed, some of the abnormalities
once considered specic for TSC are now known to occur in
other settings. In addition, it was possible to reach a pre-
sumptive diagnosis of TSC via combination relatively
nonspecic manifestations, such as the occurrence of a
gingival broma or dental pitting in an individual with an
immediate relative with TSC or the co-occurrence of in-
fantile spasms and myotonic, clonic, or atonic seizures.44
The 1988 Gomez diagnostic criteria appeared the year
after TSC was linked to chromosome 9 by Fryer et al.45 and
just before the effort to identify the second TSC gene. The
mistaken characterization of an unaffected individual as
affected can have a devastating effect on linkage analysis,
and, because of this concern, several of us were asked in
1992 to devise diagnostic criteria with a higher level of
specicity.46 Like Gomez, we had no way to ascertain the
true frequency or specicity of each TSC feature. To increase
the level of specicity, we eliminated the notion of patho-
gnomonic TSC features altogether and structured the
criteria to require at least two major features of TSC or one
major and two minor manifestations to characterize the
diagnosis as denitive. Somewhat controversially, we
excluded epilepsy and cognitive impairment from the
criteria altogether because both would almost always be
associated with brain imaging abnormalities and these
were already being tabulated.46 These criteria served their
initial purpose, but they were cumbersome for routine
clinical use and likely left some individuals with TSC with
only a probable diagnosis.
13
The criteria were revised at the 1998 consensus confer-
ence in Annapolis, Maryland.47 By this time, our under-
standing of the clinical features of TSC had improved, and
mutations could be identied at least in research settings,
making it easier to validate the diagnostic signicance of
some TSC manifestations.47 The most recent iteration of TSC
diagnostic criteria resulted from the 2012 consensus con-
ference in Washington, D.C.48 By then, commercially avail-
able mutational analysis was available, and the presence of
a disease-causing mutation of TSC1 or TSC2 was accepted as
the basis of a TSC diagnosis without regard to the clinical
manifestations. It was also possible to further simplify the
criteria. Members of this productive consensus group also
updated the recommended surveillance guidelines for TSC-
related complications and devised a standardized way to
assess individuals with TSC for the presence of neuropsy-
chological dysfunction.39-41
On the trail of the TSC genes
Clinical observations can expedite research and suggest
new lines of inquiry. Pay attention!
In 1987, Fryer et al.45 linked TSC to the distal long arm of
chromosome 9 by analyzing ABO blood groups in 19 fam-
ilies affected by TSC. This report was among the rst to link
a specic disease to a specic noneX chromosome. It soon
became apparent that other multigenerational families
with TSC did not exhibit linkage to chromosome 9, and the
search for a second gene locus began.49,50
At this time, it was not easy to establish gene linkage to a
particular chromosome or to pinpoint the gene itself.
Identifying multigenerational families with TSC who were
eager to participate in research was not easy. There were a
limited number of established markers for each chromo-
some, and it could take months of painstaking work just to
14 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
FIGURE 11.
Manuel Rodriguez Gomez, 1928 to 2006.
eliminate one chromosome. Raymond Kandt noted that
occasional individuals with TSC experience polycystic kid-
neys and that PKD had recently been linked to chromosome
16. This bit of serendipity paid off handsomely, for Kandt
et al.51 quickly found that TSC2 was not only linked to
chromosome 16, it was immediately adjacent to the PKD
gene on chromosome 16.
Subsequent progress was rapid. In 1993 the TSC2 gene
was identied at 16p13.3,52 and, two years later, tuberin
was identied as its gene product.53 In 1997, ten years after
its initial linkage to chromosome 9, TSC1 was nally iden-
tied at 9q34, and hamartin was promptly identied as its
gene product.54
Clinical gene testing for TSC1 and TSC2 became available
in 2002. Conrmatory mutational analysis can help to
establish the TSC diagnosis in individuals with milder or
atypical manifestations, and mutational analysis also aids
genetic counseling. However, several factors limit the
usefulness of mutational analysis for TSC.55 Most in-
dividuals acquire TSC via a spontaneous mutation,56 and
mutational analysis is ordinarily not considered until after
the diagnosis of the rst individual in the family.
Commercially available mutational analysis fails to identify
the disease-causing mutation in about 10% to 15% of the
individuals who meet the clinical diagnostic criteria for
TSC.57 Many of these individuals with unremarkable gene
testing probably represent somatic mosaicism,57 so it is
not surprising that individuals without an identied mu-
tation often exhibit a milder phenotype than those with a
dened mutation.58 Finally, germline mosaicism will not
be detected with standard mutational analysis. These in-
dividuals may have multiple offspring with TSC although
they have no clinical ndings to suggest TSC and despite
their lack of a disease-causing mutation on blood-based
testing.59
What do the TSC genes do?
The preservation of TSC1 and TSC2 across a broad range of
species was an early indication of their importance.
Plank et al. and others demonstrated in 1998 that tuberin
can be traced to the Golgi apparatus and that hamartin and
tuberin interact.60,61 Tuberin and hamartin form a protein
complex and work, probably explaining how a mutation on
either of the two genes could cause such a similar
phenotype.62,63
Rapamycin had been used clinically for several years
because of its broad antifungal and immunosuppressive
effects by the time it intersected with TSC research. Rapa-
mycin was named for Rapa Nui, the Easter Island whose soil
yielded rapamycins parent bacterium, Streptomyces hygro-
scopicus. Cafferkey et al.64 in 1993 identied two yeast
genes which, when mutated, eliminated rapamycin toxicity.
Subsequently, the proteins that arose from these two genes
were puried and characterized as the physical target of
rapamycin.65 Mammalian homologs of these proteins were
thus designated the mammalian targets of rapamycin
(mTOR1 and mTOR2).66-68 In recognition of mTORs pres-
ervation across a broad range of species and the entrenched
status of its abbreviation, mTOR was belatedly renamed the
mechanistic target of rapamycin.69,70
An early indication of the importance of the TSC genes
came in 2001 with the demonstration by Gao and Pan that
TSC1 mutations antagonize insulin signaling in
Drosophila.71 It was subsequently demonstrated that the
tuberinehamartin complex inhibits mTOR-mediated
signaling. The mTOR proteins are serine/threonine protein
kinases that combine with several other proteins to form
two large complexes (mTOR complex 1 and mTOR complex
2 or mTORC1 and mTORC2). Rapamycin inhibits mTOR by
interfering with the phosphorylation of the mTOR protein
complexes.72 The hamartinetuberin complex normally
dampens the distal mTOR pathway (Fig 12), so a TSC1 or
TSC2 mutation results in overactivation of the mTOR
cascade. mTORC1 phosphorylates and regulates several
downstream components of the cascade, including two
well-studied proteins, namely ribosomal protein S6 kinase
1 (S6K1) and eukaryotic initiation factor 4E-binding protein
1 (4E-BP1).73 These proteins, in turn, control several aspects
of translation and protein synthesis in response to changes
in nutrients, hormonal effects, and energy sufciency.74
Given the frequency and the variety of tumors that occur
in individuals with TSC, the tumor suppressor role of TSC1
and TSC2 is fairly obvious. However, overactivation of the
mTOR pathway can also exert a dramatic effect on cell size,
neuronal migration, axon formation, and synaptic plasticity.
TSC-related neurological lesions result from impaired
 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
FIGURE 12.
Schematic representation of the relationship of the tuberinehamartin complex to mTORC1 and mTORC2. Reproduced with permission from Tran and
Zupanc.21
neuronal migration along radial glial bers, abnormal
myelination, and from abnormal cell proliferation. Way
et al.75 created a mouse model with selective deletion of
TSC2 from radial glial progenitor cells in the cerebral cortex.
Reminiscent of human TSC, these mice developed postnatal
megalencephaly, abnormal cerebral cortical and hippo-
campal lamination, heterotopias, abnormal myelination,
and enlarged dysplastic neurons and glia. There is also ev-
idence that mTOR function inuences long-term synaptic
plasticity, which in turn could play a role in cognition and
behavior that is distinct from these migrational and pro-
liferational aberrations.72,76 Impaired local protein synthe-
sis in axonal growth cones and synapses could represent a
unifying mechanism for these seemingly different
processes.77
There are now several animal models of TSC,78 but mice
with prenatal loss of neuronal TSC1 by Meikle et al.79 was a
notable early contribution because the animals replicated
several of the clinical and neuropathological features of TSC
in humans. These mice failed to thrive and experienced
sometimes fatal seizures, typically dying between three
weeks and two months of age. They exhibited enlarged
dysplastic neurons, a larger than normal overall brain size,
and a striking lack of myelination. Remarkably, all these
manifestations were mitigated by the early administration
of an mTOR inhibitor. In the treated animals, seizures did
not occur, the median survival and weight gain normalized,
15
the brain size remained normal, and the myelination
remained close to normal (Fig 13).79
TSC1 and TSC2 are preserved across a broad range of
species and, via the mTOR pathway, inuence several
important cellular processes.70 The pathway may play a role
in the pathophysiology of such seemingly diverse condi-
tions as obesity, cancer, type 2 diabetes mellitus, aging, and
neurodegenerative disorders. These topics are well beyond
the scope of this presentation, but detailed reviews of the
protean effects of the mTOR pathway are available.70
The rise of mTOR inhibitors
Repurposing an approved drug can save years of work and
help patients much sooner than making a completely new
drug.
The TSC1 and TSC2 mutations result in mTOR activa-
tion, and rapamycin suppresses mTOR signaling. After TSC
was shown to affect the mTOR pathway, clinicians soon
began to study the already available mTOR inhibitor
rapamycin (sirolimus) in individuals with TSC. Clinical
trials were also begun with everolimus, a rapamycin de-
rivative that is said to act more selectively on mTOR
complex 1.21 For the rst time, targeted therapy based on
an understanding of the molecular pathophysiology of
TSC became a reality.80
16 E.S. Roach / Pediatric Neurology 63 (2016) 6e22

FIGURE 13.
Improved myelination in TSC1 null-neuron mice in response to rapamycin. The top row depicts myelin stained samples from the retrosplenial granular
region of the cerebral cortex, whereas the bottom images are myelin-stained samples from the CA3 region of anterior hippocampus. Note the normal myelin
staining in a control animal (A, F) and in a control animal treated with rapamycin (E, J). Myelination is severely reduced in the untreated TSC1-decient
animal (B, G). Administration of rapamycin to the mutant mice dramatically improved myelination (C, H), and halting rapamycin after 30 postnatal days
resulted in improved but patchy myelination (D, I). Images
60; both scale bars equal 100 mm. Reproduced with permission from Meikle et al.79

Franz et al.81 initially treated ve TSC patients with oral
rapamycin. The SEGA regressed in all ve of these in-
dividuals, exhibited transient regrowth after cessation of
therapy, and then shrank again upon reinstitution of rapa-
mycin (Fig 14). Interestingly, reduction of the tumor size in
response to rapamycin may not be associated with an
obvious histopathological change.82
Bissler et al.83 conducted a 24-month, nonrandomized,
open-label trial of rapamycin (sirolimus) in individuals with
TSC-related renal AMLs or LAM. Similar to the results of the
earlier SEGA study by Franz et al., the renal AMLs regressed
during rapamycin therapy (Fig 15A) but often grew again
after therapy was halted.83 The spirometry measurements
also improved in the LAM patients, and the improvement
persisted in some individuals (Fig 15B).83
Subsequently, Bissler et al.84 conducted a placebo-
controlled trial of the rapamycin derivative everolimus
(Fig 16) in 118 patients with either TSC or sporadic LAM who
had one or more renal AMLs that were more than 3 cm in
largest diameter. Thirty-three of the 79 (42%) patients who
received everolimus experienced a 50% or more reduction
of the target AML volume.84 None of the lesions regressed in
the 39 individuals who received placebo, and nine people in
the control group discontinued study participation because
of tumor growth.84 There was a similar robust response (i.e.,
more than 50% reduction of AML size in 80% of the patients)
among the younger patients receiving sirolimus in the SEGA
clinical trial.86
Krueger et al.87 conducted an open-label trial of ever-
olimus in 28 patients with TSC who exhibited increasing
SEGA size. All 28 individuals experienced a reduction in
SEGA size, and the tumor shrank by at least 30% in 21 pa-
tients (75%) and by 50% in nine individuals (32%). None of
the patients experienced new SEGA formation, worsening
hydrocephalus, or increased intracranial pressure.87 Over
half of the individuals who continued on therapy eventually
experienced a reduction in tumor size of more than 50%.88
Franz et al.89 conducted a randomized, placebo-
controlled trial of everolimus in 117 individuals with TSC
and at least one SEGA of more than one centimeter in
diameter. Twenty-seven of the 78 individuals (35%) who
received everolimus experienced a 50% or greater reduction
in SEGA size versus none of the control group.89 With
extended follow-up, about half of the patients achieved a
50% reduction in tumor size.90 Most patients tolerate the
extended use of mTOR inhibitors well, but measuring
trough drug concentrations may be useful in some
individuals.91
Double-blind clinical trials of mTOR inhibitors are dif-
cult to perform in individuals with TSC because of the drugs
often striking effect on the facial angiobromas. Not sur-
prisingly, a recently concluded multi-center clinical trial
demonstrated (Fig 17) that topically applied rapamycin
renders the facial angiobromas less prominent (Hope
Northrup, personal communication September, 2015).
Another report suggests that topical everolimus is also
effective for these facial lesions.92
Goyer et al.93 also successfully treated three neonates
with TSC-related tumors (two with a cardiac rhabdomyoma
and one with a SEGA) with everolimus. All three neonates
seemed to benet from the medication and all tolerated it
well.
What about pre-emptive therapy of TSC?
Most genetic disease therapies are aimed at secondary
complications. Mechanism-based therapy may change that
paradigm.
 E.S. Roach / Pediatric Neurology 63 (2016) 6e22 17

FIGURE 14.
These magnetic resonance images illustrate one patients subependymal giant cell tumor (SEGA) response to rapamycin. (A) Before starting therapy. (B)
After 2.5 months of therapy. (C) Follow-up imaging after stopping therapy for 4 months. (D) Eight months after resuming therapy, the SEGA is again smaller.
Images reproduced with permission from Franz, et al.81

Most children with TSC initially appear neurologically
normal, offering a potential opportunity to minimize or
prevent some of the disorders complications. Would early
therapy prevent TSC neurological manifestations from
developing? The evidence is intriguing but far from
complete.80
Jozwiak et al.94 rst suggested that prophylactic admin-
istration of an antiepileptic agent to infants at high risk for
epilepsy due to TSC before the onset of seizures greatly
improves their developmental outcome and reduces their
likelihood of drug-resistant seizures. More recently, Wu
et al.95 assessed serial electroencephalography in 40
seizure-free children with TSC before seven months of age.
Two thirds of the children in this cohort subsequently
developed seizures, and just over half of them developed
infantile spasms. Seventy-four percent of the children who
experienced seizures developed epileptiform discharges on
electroencephalography at an average of 1.9 months before
the onset of seizures, demonstrating the feasibility of
identifying individuals at high risk for seizures before the
seizures actually begin.95
Even more intriguing is the possibility of mitigating ep-
ilepsy or cognitive dysfunction before their onset or at least
soon after their appearance.94 The administration of rapa-
mycin to TSC-decient mice appears to diminish seizures
and behavioral abnormalities.80 Anecdotal reports from TSC
patient families suggest improvement of epilepsy or
behavior after the administration of an mTOR inhibitor, and
some individuals in the sirolimus clinical trials reported a
substantial reduction in their seizure frequency.88,96
We had long assumed that the intractable epilepsy,
cognitive impairment, and behavioral changes in in-
dividuals with TSC resulted from long-standing multifocal
cortical dysplasia, aberrations of neuronal migration, and
abnormal neuronal connections. This conjecture was
bolstered by the apparent correlation between the brains
lesion burden and the severity of the individuals epilepsy
and cognitive impairment.34,35 However, substantial
18 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
FIGURE 15.
(A) All 18 individuals with tuberous sclerosis complexerelated renal
angiomyolipomas experienced a substantial reduction in tumor size during
12 months of rapamycin (sirolimus) therapy, but only ve of them main-
tained a tumor reduction of 30% or more during the following year after
discontinuation of therapy. (B) The individuals with lymphangioleiomyo-
matosis exhibited improved spirometry measurements during sirolimus
therapy. Note the improved performance in the forced vital capacity and
forced expiratory volume during the 12-month treatment period and the
subsequent deterioration of the readings once therapy was discontinued.
FEV1 forced expiratory volume 1; FVC forced vital capacity. Repro-
duced with permission from Bissler, et al.83
improvement of long-established neurological dysfunction
in response to an mTOR inhibitor would suggest that at least
some aspect of the dysfunction represents an ongoing
cellular effect of the mutation, not just the effects of an
already abnormal brain structure.
During the clinical trials of mTOR inhibitors for TSC-
related tumors, families often documented a reduction in
the number of their childs seizures or described improved
cognition or behavior after starting the medication.
Following these observations, clinical trials began to
determine if the mTOR inhibitor everolimus could improve
epilepsy or autistic symptoms related to TSC.
French and colleagues97 recently reported the initial re-
sults of the EXIST-3 trial, a randomized, multicenter,
placebo-controlled phase 3 of everolimus in 366 individuals
with refractory epilepsy due to TSC. The patients were
randomly assigned to receive placebo, lower-dose ever-
olimus (3 to 7 ng/mL), or higher-dose everolimus (9 to
15 ng/mL) after an initial eight-week observation period.
They determined that 15.1% of the placebo group, 28.2% of
the lower-dose everolimus group (P < 0.008 versus pla-
cebo), and 40% of the higher-dose everolimus group
(P < 0.001 versus placebo) reached the primary endpoint of
a 50% seizure reduction. They also analyzed the percentage
reduction in seizure frequency: 14.9% for placebo, 29.3% for
lower-dose everolimus (P < 0.003 versus placebo), and
39.6% for higher-dose everolimus (P < 0.001 versus pla-
cebo). These are remarkable results, especially given the
severity of the epilepsy in these individuals, but other
questions remain. Will earlier therapy minimize the en-
cephalopathy that often accompanies years of poorly
controlled epilepsy? Have we discovered the optimal
medication dose? Would an mTOR inhibitor affect seizures
due to other conditions that affect the mTOR pathway or
some forms of focal cortical dysplasia? Would there be a
benet to starting therapy after the TSC diagnosis but before
the onset of seizures?
Also left unanswered is whether mTOR inhibitors will
improve cognition and behavior in some individuals with
TSC. Clinical trials now in progress should provide answers
to these questions. The availability of mechanism-based
therapy might make it feasible to treat the disease rather
than waiting on its symptoms to develop.98
What causes TSC phenotypic variation?
Not everything is easily explained by simple Mendelian
genetics.
Phenotypic variability is striking in individuals with TSC.
Affected individuals within the same family, who typically
share the same mutation of the same TSC gene, a large
number of other common genes, and similar environmental
circumstances, often exhibit strikingly different TSC mani-
festations. Even identical twins often differ dramatically in
regard to their TSC manifestations.99 Our deepening un-
derstanding of the genetic and molecular basis of TSC has
explained some parts of the phenotypic variability of TSC,
but other aspects of the variability remain mysterious.
Clearly, some manifestations of TSC are age or gender
related, a factor that must be considered when assessing
phenotypic variation.1 Infants are unlikely to exhibit facial
angiobromas or ungual bromas, and cardiac rhabdo-
myomas are unlikely to become symptomatic after the
neonatal period. Renal AMLs are evident in 80% of patients
by age ten years on imaging but seldom become symp-
tomatic until adolescence or adulthood.19 Pulmonary TSC
seldom becomes symptomatic until adulthood and is far
more likely to occur among women than among men. Ce-
rebral giant cell tumors, in contrast, are relatively unlikely
to develop de novo in adults.32
 E.S. Roach / Pediatric Neurology 63 (2016) 6e22 19

FIGURE 16.
The chemical structure of rapamycin and everolimus. Reproduced with permission from Curatolo, et al.85

That up to three fourths of affected individuals develop
TSC via a spontaneous mutation has long been appreciated,
but an estimated 15% or more of the individuals with TSC
exhibit somatic mosaicism and another 1% or so have
germline mosaicism.59 Somatic mosaicism explains at least
some of the cross-generational variability (a mildly affected
mosaic parent may have severely affected children),
whereas germline mosaicism explains instances in which
seemingly unaffected couples have more than one child
with identical TSC mutations.
The variability of tumor formation from individual to
individual can be explained at least in part by Knudsons
second hit phenomenon. Tumors arise after the tumor-
forming tissue sustains one or more additional mutations
that affect the function of the previously normal second
allele. This phenomenon may explain both the occur-
rence of a tumor and the variability in the rate of tumor
growth.
The variable phenotype of TSC is not fully explained by
the affected gene or the specic mutation. Not surprisingly,
individuals with a TSC2 mutation are more likely to exhibit
multiple renal cysts. Genotypeephenotype analyses
consistently show that TSC2 mutations tend to cause more
severe manifestations than TSC1 mutations.58,100 In familial
cases, TSC1 and TSC2 mutations occur with roughly equal
frequency, whereas TSC2 mutations are identied more
frequently among individuals with a new mutation.101 On
the whole, individuals with familial TSC tend to have milder
disease manifestations than individuals with a new muta-
tion, possibly reecting the increased likelihood of TSC2
mutations among people with spontaneous mutations. In-
dividuals who meet the clinical diagnostic criteria but have
no mutation identied on standard testing tend to have a
milder phenotype.58 Both their mild phenotype and their
unremarkable test results may be explained by somatic
mosaicism.

FIGURE 17.
Facial angiobromas shown before (A) and after (B) administration of topical rapamycin. Photographs courtesy of Drs. Mary Beth Koenig and Hope
Northrup.
20 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
None of these factors, however, fully explains the
phenotypic variability among affected family members,
who should have the same mutation of the same TSC gene.
Family members are susceptible to the effects of modier
genes and the inuences of other metabolic pathways, but
even these factors cannot explain the occasional reports of
identical twins with widely differing phenotypes.99
Conclusion
TSC is a dominantly inherited disorder of cellular differ-
entiation and proliferation which variably affects the brain,
skin, kidneys, heart, and other organs. The predominant
neurological manifestations are epilepsy, autism, cognitive
and behavioral dysfunction, and SEGAs. TSC1 and TSC2
mutations cause an overlapping phenotype because their
protein products tuberin and hamartin act together as a
functional complex. Although the TSC2 phenotype is often
more severe than that of TSC1, the variable phenotype of
TSC is not fully explained by the gene that is affected nor by
the specic mutation. Mechanism-based therapy with
mTOR inhibitors is now available for some TSC complica-
tions, and pre-symptomatic therapy may prove to be
feasible in some situations. Although our understanding of
TSC and our ability to prevent or treat its complications
have grown tremendously in the last generation, ongoing
research is necessary if we are to solve the remaining rid-
dles. Nevertheless, the progress so far suggests that TSC can
serve as a prototype for the development of mechanism-
based therapy for other genetic diseases.
This review article is derived from the Hower Award Lecture at the annual Child
Neurology Society meeting in Washington, D.C., on October 10, 2015.
References
1. Roach ES, Sparagana SP. Diagnosis of tuberous sclerosis complex.
J Child Neurol. 2004;19:643-649.
2. Gomez MR. Tuberous Sclerosis. 1st ed. New York: Raven Press;
1979.
3. Bourneville DM. Sclrose tubreuse des circonvolution crbrales:
Idiotie et pilepsie hemiplgique. Arch de Neurologie Paris. 1880;1:
81-91.
4. Bourneville DM, Brissaud E. Idiotie et epilepsie symptomatiques de
sclroses tubreuse ou hypertrophique. Arch Neurol (Paris). 1890;
10:29-39.
5. Balzer F, Menetrier P. Etude sur un cas dadnomes sbacs de la
face et du cuir chevelu. Arch Physiol Norm Pathol (srie III). 1885;6:
564-576.
6. Roach ES. Diagnosis and management of neurocutaneous syn-
dromes. Semin Neurol. 1988;8:83-96.
7. Campbell AW. Cerebral sclerosis. Brain. 1906;28:382-396.
8. Roach ES. Neurocutaneous syndromes. Pediatr Clin North Am. 1992;
39:591-620.
9. Vogt H. Diagnostik der tubersen Sklerose. Z fr die Erforschung
Behandlung des jugendlichen Schwachsinns auf Wissenschaftlicher
Grundlage. 1908;2:1-16.
10. Critchley M, Earl CJC. Tuberose sclerosis and allied conditions.
Brain. 1932;55:311-346.
11. Gold AP, Freeman JM. Depigmented nevi: the earliest sign of tu-
berous sclerosis. Pediatrics. 1965;35:1003-1005.
12. Fitzpatrick TB, Szabo G, Hori Y, Simone AA, Reed WB,
Greenberg MH. White leaf-shaped macules. Earliest visible sign of
tuberous sclerosis. Arch Dermatol. 1968;98:1-6.
13. Rentz AM, Skalicky AM, Liu Z, et al. Tuberous sclerosis complex: a
survey of health care resource use and health burden. Pediatr
Neurol. 2015;52:435-441.
14. Kiribuchi K, Uchida Y, Fukuyama Y, Maruyama H. High incidence of
fundus hamartomas and clinical signicance of a fundus score in
tuberous sclerosis. Brain Dev. 1986;8:509-517.
15. Gibbs JL. The heart and tuberous sclerosis. An echocardiographic
and electrocardiographic study. Br Heart J. 1985;54:596-599.
16. DiMario FJ, Diana D, Leopold H, Chameides L. Evolution of cardiac
rhabdomyoma in tuberous sclerosis complex. Clin Pediatr. 1996;
12:615-619.
17. Jozwiak S, Kawalec W, Dluzewska J, Daszkowska J, Mirko-
wicz-Malek M, Michalowicz R. Cardiac tumors in tuberous
sclerosis: their incidence and course. Eur J Pediatr. 1994;153:
155-157.
18. Crawford DC, Garrett C, Tynan M. Cardiac rhabdomyomata as a
marker for the antenatal detection of tuberous sclerosis. J Med
Genet. 1983;20:303-304.
19. Ewalt DH, Shefeld E, Sparagana SP, Delgado MR, Roach ES. Renal
lesion growth in children with tuberous sclerosis complex. J Urol.
1998;160:141-145.
20. Weiner DM, Ewalt DE, Roach ES, Hensle TW. The tuberous sclerosis
complex: a comprehensive review. J Am Coll Surg. 1998;187:
548-561.
21. Tran LH, Zupanc ML. Long-term everolimus treatment in in-
dividuals with tuberous sclerosis complex: a review of the current
literature. Pediatr Neurol. 2015;53:23-30.
22. Jones K, Go C, Boyd J, et al. Vigabatrin as rst-line treatment for
infantile spasms not related to tuberous sclerosis complex. Pediatr
Neurol. 2015;53:141-145.
23. Wanigasinghe J, Arambepola C, Sri RS, Sumanasena S,
Attanapola G. Randomized, single-blind, parallel clinical trial on
efcacy of oral prednisolone versus intramuscular corticotropin on
immediate and continued spasm control in West syndrome.
Pediatr Neurol. 2015;53:193-199.
24. Wanigasinghe J, Arambepola C, Sri RS, Sumanasena S,
Muhandiram EC. The efcacy of moderate-to-high dose oral
prednisolone versus low-to-moderate dose intramuscular corti-
cotropin for improvement of hypsarrhythmia in West syndrome: a
randomized, single-blind, parallel clinical trial. Pediatr Neurol.
2014;51:24-30.
25. Sparagana SP, Delgado MR, Batchelor LL, Roach ES. Seizure
remission and antiepileptic drug discontinuation in children with
tuberous sclerosis complex. Arch Neurol. 2003;60:1286-1289.
26. Bebin EM, Kelly PJ, Gomez MR. Surgical treatment for epilepsy in
cerebral tuberous sclerosis. Epilepsia. 1993;34:651-657.
27. Fallah A, Rodgers SD, Weil AG, et al. Resective epilepsy surgery for
tuberous sclerosis in children: determining predictors of seizure
outcomes in a multicenter retrospective cohort study. Neurosur-
gery. 2015;77:517-524.
28. Cuddapah VA, Thompson M, Blount J, Li R, Guleria S, Goyal M.
Hemispherectomy for hemimegalencephaly due to tuberous scle-
rosis and a review of the literature. Pediatr Neurol. 2015;53:
452-455.
29. Shepherd CW, Houser OW, Gomez MR. MR ndings in tuberous
sclerosis complex and correlation with seizure development and
mental impairment. AJNR Am J Neuroradiol. 1995;16:149-155.
30. Roach ES, Kerr J, Mendelsohn D, Laster DW, Raeside C. Diagnosis of
symptomatic and asymptomatic gene carriers of tuberous sclerosis
by CCT and MRI. Ann N Y Acad Sci. 1991;615:112-122.
31. Roth J, Roach ES, Bartels U, et al. Subependymal giant cell astro-
cytoma: diagnosis, screening, and treatment. Recommendations
from the International Tuberous Sclerosis Complex Consensus
Conference 2012. Pediatr Neurol. 2013;49:439-444.
32. Torres OA, Roach ES, Delgado MR, et al. Early diagnosis of sub-
ependymal giant cell astrocytoma in patients with tuberous scle-
rosis. J Child Neurol. 1998;13:173-177.
33. Kotulska K, Borkowska J, Roszkowski M, et al. Surgical treatment of
subependymal giant cell astrocytoma in tuberous sclerosis com-
plex patients. Pediatr Neurol. 2014;50:307-312.
34. Roach ES, Williams DP, Laster DW. Magnetic resonance imaging in
tuberous sclerosis. Arch Neurol. 1987;44:301-303.
35. Goodman M, Lamm SH, Engel A, Shepherd CW, Gomez MR. Cortical
tuber count: a biomarker indicating cerebral severity of tuberous
sclerosis complex. J Child Neurol. 1997;11:85-90.
 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
36. Vaughn J, Hagiwara M, Katz J, et al. MRI characterization and
longitudinal study of focal cerebellar lesions in a young tu-
berous sclerosis cohort. AJNR Am J Neuroradiol. 2013;34:
655-659.
37. Baumer FM, Song JW, Mitchell PD, et al. Longitudinal changes in
diffusion properties in white matter pathways of children with
tuberous sclerosis complex. Pediatr Neurol. 2015;52:615-623.
38. Roach ES, DiMario FJ, Kandt RS, Northrup H. Tuberous sclerosis
consensus conference: recommendations for diagnostic evalua-
tion. J Child Neurol. 1999;14:401-407.
39. de Vries PJ, Whittemore VH, Leclezio L, et al. Tuberous sclerosis
associated neuropsychiatric disorders (TAND) and the TAND
Checklist. Pediatr Neurol. 2015;52:25-35.
40. Leclezio L, Jansen A, Whittemore VH, de Vries PJ. Pilot validation of
the tuberous sclerosis-associated neuropsychiatric disorders
(TAND) checklist. Pediatr Neurol. 2015;52:16-24.
41. Krueger DA, Northrup H. Tuberous sclerosis complex surveillance
and management: recommendations of the 2012 International
Tuberous Sclerosis Complex Consensus Conference. Pediatr Neurol.
2013;49:255-265.
42. Osborne JP, Fryer A, Webb D. Epidemiology of tuberous sclerosis.
Ann N Y Acad Sci. 1991;615:125-127.
43. OCallaghan FJK, Shiell AW, Osborne JP, Martyn CN. Prevalence of
tuberous sclerosis estimated by capture-recapture analysis. Lancet.
1998;351:1490.
44. Gomez MR. Tuberous Sclerosis. 2nd ed. New York: Raven Press;
1988.
45. Fryer AE, Chalmers A, Connor JM, et al. Evidence that the gene for
tuberous sclerosis is on chromosome 9. Lancet. 1987;1:659-661.
46. Roach ES, Smith M, Huttenlocher P, Bhat M, Alcorn D, Hawley L.
Diagnostic criteria: tuberous sclerosis complex. Report of the
Diagnostic Criteria Committee of the National Tuberous Sclerosis
Association. J Child Neurol. 1992;7:221-224.
47. Roach ES, Gomez MR, Northrup H. Tuberous sclerosis complex
consensus conference: revised clinical diagnostic criteria. J Child
Neurol. 1998;13:624-628.
48. Northrup H, Krueger DA. Tuberous sclerosis complex diagnostic
criteria update: recommendations of the 2012 International Tu-
berous Sclerosis Complex Consensus Conference. Pediatr Neurol.
2013;49:243-254.
49. Kandt RS, Pericak-Vance MA, Hung WY, et al. Absence of linkage of
ABO blood group locus to familial tuberous sclerosis. Exp Neurol.
1989;104:223-228.
50. Northrup H, Kwiatkowski DJ, Roach ES, et al. Evidence for genetic
heterogeneity in tuberous sclerosis: one locus on chromosome 9
and at least one locus elsewhere. Am J Hum Genet. 1992;51:
709-720.
51. Kandt RS, Haines JL, Smith M, et al. Linkage of an important gene
locus for tuberous sclerosis to a chromosome 16 marker for
polycystic kidney disease. Nat Genet. 1992;2:37-41.
52. The European Chromosome 16 Tuberous Sclerosis Consortium.
Identication and characterization of the tuberous sclerosis gene
on chromosome 16. Cell. 1993;75:1305-1315.
53. Wienecke R, Konig A, DeClue JE. Identication of tuberin, the tu-
berous sclerosis-2 product. J Biol Chem. 1995;270:16409-16414.
54. van Slegtnenhorst M, de Hoogt R, Hermans C, Nellist M, Janssen B,
Verhoef S. Identication of the tuberous sclerosis gene TSC1 on
chromosome 9q34. Science. 1997;277:805-808.
55. Roach ES. Are diagnostic criteria for tuberous sclerosis still rele-
vant? Pediatr Neurol. 2013;49:223-224.
56. Roach ES, Kerr J, Mendelsohn D, Laster DW, Raeside C. Detection of
tuberous sclerosis in parents by magnetic resonance imaging.
Neurology. 1991;41:262-265.
57. Tyburczy ME, Dies KA, Glass J, et al. Mosaic and intronic mutations
in TSC1/TSC2 explain the majority of TSC patients with no muta-
tion identied by conventional testing. PLoS Genet. 2015;11:
e1005637.
58. Au KS, Williams AT, Roach ES, et al. Genotype/phenotype corre-
lation in 325 individuals referred for a diagnosis of tuberous
sclerosis complex in the United States. Genet Med. 2007;9:88-100.
59. Rose VM, Au K-S, Pollom G, Roach ES, Prashner HR, Northrup H.
Germ-line mosaicism in tuberous sclerosis: how common? Am J
Hum Genet. 1999;64:986-992.
60. Plank TL, Yeung RS, Henske EP. Hamartin, the product of the tu-
berous sclerosis 1 (TSC1) gene, interacts with tuberin and appears
21
to be localized to cytoplasmic vesicles. Cancer Res. 1998;58:
4766-4770.
61. van Slegtenhorst M, Nellist M, Nagelkerken B, et al. Interaction
between hamartin and tuberin, the TSC1 and TSC2 gene products.
Hum Mol Genet. 1998;7:1053-1057.
62. Johnson MW, Emelin JK, Park S-H, Vinters HV. Co-localization of
TSC1 and TSC2 gene products in tubers of patients with tuberous
sclerosis. Brain Pathol. 1999;9:45-54.
63. Nellist M, Verhaaf B, Goedbloed MA, Reuser AJ, van den
Ouweland AM, Halley DJ. TSC2 missense mutations inhibit tuberin
phosphorylation and prevent formation of the tuberin-hamartin
complex. Hum Mol Genet. 2001;10:2889-2898.
64. Cafferkey R, Young PR, McLaughlin MM, et al. Dominant missense
mutations in a novel yeast protein related to mammalian phos-
phatidylinositol 3-kinase and VPS34 abrogate rapamycin cytotox-
icity. Mol Cell Biol. 1993;13:6012-6023.
65. Kunz J, Henriquez R, Schneider U, Deuter-Reinhard M, Movva NR,
Hall MN. Target of rapamycin in yeast, TOR2, is an essential
phosphatidylinositol kinase homolog required for G1 progression.
Cell. 1993;73:585-596.
66. Brown EJ, Albers MW, Shin TB, et al. A mammalian protein targeted
by G1-arresting rapamycin-receptor complex. Nature. 1994;369:
756-758.
67. Sabers CJ, Martin MM, Brunn GJ, et al. Isolation of a protein target
of the FKBP12-rapamycin complex in mammalian cells. J Biol Chem.
1995;270:815-822.
68. Sabatini DM, Erdjument-Bromage H, Lui M, Tempst P, Snyder SH.
RAFT1: a mammalian protein that binds to FKBP12 in a rapamycin-
dependent fashion and is homologous to yeast TORs. Cell. 1994;78:
35-43.
69. Curatolo P. Mechanistic target of rapamycin (mTOR) in tuberous
sclerosis complex-associated epilepsy. Pediatr Neurol. 2015;52:
281-289.
70. Laplante M, Sabatini DM. mTOR signaling in growth control and
disease. Cell. 2012;149:274-293.
71. Gao X, Pan D. TSC1 and TSC2 tumor suppressors antagonize
insulin signaling in cell growth. Genes Dev. 2001;15:
1383-1392.
72. Hoeffer CA, Klann E. mTOR signaling: at the crossroads of plasticity,
memory and disease. Trends Neurosci. 2010;33:67-75.
73. Tee AR, Fingar DC, Manning BD, Kwiatkowski DJ, Cantley LC,
Blenis J. Tuberous sclerosis complex-1 and -2 gene products
function together to inhibit mammalian target of rapamycin
(mTOR)-mediated downstream signaling. Proc Natl Acad Sci U S A.
2002;99:13571-13576.
74. Holz MK, Ballif BA, Gygi SP, Blenis J. mTOR and S6K1 mediate as-
sembly of the translation preinitiation complex through dynamic
protein interchange and ordered phosphorylation events. Cell.
2005;123:569-580.
75. Way SW, McKenna III J, Mietzsch U, Reith RM, Wu HC,
Gambello MJ. Loss of Tsc2 in radial glia models the brain pathology
of tuberous sclerosis complex in the mouse. Hum Mol Genet. 2009;
18:1252-1265.
76. von der BC, Waltereit R, Zhang L, Beck H, Kirschstein T. Impaired
synaptic plasticity in a rat model of tuberous sclerosis. Eur J Neu-
rosci. 2006;23:686-692.
77. Martin KC. Local protein synthesis during axon guidance and
synaptic plasticity. Curr Opin Neurobiol. 2004;14:305-310.
78. Sahin M, Henske E, Manning B, et al. Advances and future di-
rections for tuberous sclerosis complex research: recommenda-
tions from the 2015 strategic planning conference. Pediatr Neurol.
2016;60:1-12.
79. Meikle L, Pollizzi K, Egnor A, et al. Response of a neuronal
model of tuberous sclerosis to mammalian target of rapamycin
(mTOR) inhibitors: effects on mTORC1 and Akt signaling lead
to improved survival and function. J Neurosci. 2008;28:
5422-5432.
80. Julich K, Sahin M. Mechanism-based treatment in tuberous scle-
rosis complex. Pediatr Neurol. 2014;50:290-296.
81. Franz DN, Leonard J, Tudor C, et al. Rapamycin causes regression of
astrocytomas in tuberous sclerosis complex. Ann Neurol. 2006;59:
490-498.
82. Cheng S, Hawkins C, Taylor MD, Bartels U. Pathological ndings of
a subependymal giant cell astrocytoma following treatment with
rapamycin. Pediatr Neurol. 2015;53:238-242.
22 E.S. Roach / Pediatric Neurology 63 (2016) 6e22
83. Bissler JJ, McCormack FX, Young LR, et al. Sirolimus for angio-
myolipoma in tuberous sclerosis complex or lymphangioleio-
myomatosis. N Engl J Med. 2008;358:140-151.
84. Bissler JJ, Kingswood JC, Radzikowska E, et al. Everolimus for
angiomyolipoma associated with tuberous sclerosis complex or
sporadic lymphangioleiomyomatosis (EXIST-2): a multicentre,
randomised, double-blind, placebo-controlled trial. Lancet. 2013;
381:817-824.
85. Curatolo P, Moavero R. mTOR inhibitors in tuberous sclerosis
complex. Curr Neuropharmacol. 2012;10:404-415.
86. Kingswood JC, Jozwiak S, Belousova ED, et al. The effect of ever-
olimus on renal angiomyolipoma in patients with tuberous scle-
rosis complex being treated for subependymal giant cell
astrocytoma: subgroup results from the randomized, placebo-
controlled, Phase 3 trial EXIST-1. Nephrol Dial Transplant. 2014;
29:1203-1210.
87. Krueger DA, Care MM, Holland K, et al. Everolimus for sub-
ependymal giant-cell astrocytomas in tuberous sclerosis. N Engl J
Med. 2010;363:1801-1811.
88. Franz DN, Agricola K, Mays M, et al. Everolimus for subependymal
giant cell astrocytoma: 5-year nal analysis. Ann Neurol. 2015;78:
929-938.
89. Franz DN, Belousova E, Sparagana S, et al. Efcacy and safety of
everolimus for subependymal giant cell astrocytomas associated
with tuberous sclerosis complex (EXIST-1): a multicentre, rando-
mised, placebo-controlled phase 3 trial. Lancet. 2013;381:125-132.
90. Franz DN, Belousova E, Sparagana S, et al. Everolimus for sub-
ependymal giant cell astrocytoma in patients with tuberous scle-
rosis complex: 2-year open-label extension of the randomised
EXIST-1 study. Lancet Oncol. 2014;15:1513-1520.
91. Landecho MF, Alegre F, Lucena JF, Huerta A, Fernandez-Ros N,
Aldaz A. Monitoring of everolimus trough concentrations for
achieving optimal tolerance in the management of tuberous scle-
rosis complex. Pediatr Neurol. 2014;51:e9-e10.
92. Dill PE, DeBernardis G, Weber P, Losch U. Topical everolimus for
facial angiobromas in the tuberous sclerosis complex. A rst case
report. Pediatr Neurol. 2014;51:109-113.
93. Goyer I, Dahdah N, Major P. Use of mTOR inhibitor everolimus in
three neonates for treatment of tumors associated with tuberous
sclerosis complex. Pediatr Neurol. 2015;52:450-453.
94. Jozwiak S, Kotulska K, Domanska-Pakiela D, et al. Antiepileptic
treatment before the onset of seizures reduces epilepsy severity
and risk of mental retardation in infants with tuberous sclerosis
complex. Eur J Paediatr Neurol. 2011;15:424-431.
95. Wu JY, Peters JM, Goyal M, et al. Clinical electroencephalographic
biomarker for impending epilepsy in asymptomatic tuberous
sclerosis complex infants. Pediatr Neurol. 2016;54:29-34.
96. Wiegand G, May TW, Ostertag P, Boor R, Stephani U, Franz DN.
Everolimus in tuberous sclerosis patients with intractable epi-
lepsy: a treatment option? Eur J Paediatr Neurol. 2013;17:631-638.
97. French JA, Lawson JA, Yapici Z, et al. Adjunctive everolimus
therapy for treatment-resistant focal-onset seizures associated with
tuberous sclerosis (EXIST-3): A phase 3, randomised, double-blind,
placebo-controlled study. Lancet. Published online September 6,
2016 http://dx.doi.org/10.1016/S0140-6736(16)31419-2.
98. Roach ES. Mechanism-based therapy of genetic neurological dis-
ease. Pediatr Neurol. 2014;50:285-286.
99. Gomez MR, Kuntz NL, Westmoreland BF. Tuberous sclerosis, early
onset of seizures, and mental subnormality: study of discordant
homozygous twins. Neurology. 1982;32:604-611.
100. Dabora SL, Jozwiak S, Franz DN, et al. Mutational analysis in a
cohort of 224 tuberous sclerosis patients indicates increased
severity of TSC2, compared with TSC1, disease in multiple organs.
Am J Hum Genet. 2001;68:64-80.
101. Jones AC, Shyamsundar MM, Thomas MW, et al. Comprehensive
mutation analysis of TSC1 and TSC2- and phenotypic correlations
in 150 families with tuberous sclerosis. Am J Hum Genet. 1999;64:
1305-1315.