s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/survophthal

Major review

Orbital peripheral nerve sheath tumors

Adam R. Sweeney, MDa,*, Divakar Gupta, MDb, C. Dirk Keene, MD, PhDc,
Patrick J. Cimino, MD, PhDc, Christopher B. Chambers, MDa,
Shu-Hong Chang, MDa, Eissa Hanna, MDa
a
Department of Ophthalmology, University of Washington, Seattle, Washington, USA
b
Department of Ophthalmology, Duke University, Durham, North Carolina, USA
c
Department of Pathology, University of Washington, Seattle, Washington, USA

article info abstract

Article history:
Received 21 March 2016
Received in revised form 14 August
2016
Accepted 19 August 2016
Available online 26 August 2016
Peripheral nerve sheath tumors of the orbit and ocular adnexa are a rare group of neo-
plasms hallmarked by nonspecific clinical presentations, variable tumor locations, chal-
lenging therapeutic efforts, and occasional diagnostic dilemmas. We review these tumor
types and provide an updated summary on their clinical, histopathologic, radiological, and
emerging molecular features.
ª 2016 Elsevier Inc. All rights reserved.

Keywords:
peripheral nerve sheath tumor
schwannoma
neurilemmoma
neurinoma
neurofibroma
neurofibromatosis
malignant
magnetic resonance imaging
pathology
histology

1. Background differentiated from neurofibromas histologically by Ver-

In 1768, Akenside first described a patient with multiple
tumors involving the peripheral nerves. Over a century
later, von Recklinghausen reported on “neurofibroma-
tosis” and demonstrated numerous neurofibromas as
belonging to a single nerve. The neurinoma was
ocay in 1910 and later named schwannoma after Masson
confirmed that the cell of origin was the Schwann cell. The
term neurilemmoma was introduced by Stout, leaving
schwannoma and neurilemmoma synonymous through a
great span of medical literature until the advent of elec-
tron microscopy.

* Corresponding author: Adam R. Sweeney, MD, Department of Ophthalmology, University of Washington, Box 359608, 325 Ninth
Avenue, Seattle, WA 98104, USA.
E-mail address: asweeney@uw.edu (A.R. Sweeney).
0039-6257/$ e see front matter ª 2016 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.survophthal.2016.08.002
44 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
Peripheral nerve sheath tumors (PNSTs) are derived from
the neuroectoderm and neural crest and may involve the
nerve fascicles or branches of cranial nerves IIIeVII in the
orbit.117 The PNSTs more commonly involved with these
nerves include the neurofibroma, schwannoma, and malig-
nant PNST (MPNST).
MPNSTs have historically been referred to as neurogenic
sarcoma, neurofibrosarcoma, malignant schwannoma, or
malignant neurilemmoma. Schwann cell origin has not been
demonstrated in all cases leaving MPNSTs as a distinct clini-
copathologic entity.43
Tumor subtypes differentiate the localized solitary neuro-
fibroma from the diffuse and plexiform neurofibromas.
Schwannomas are classified as conventional (solitary) or by
their distinguishing histologic features: melanotic, plexiform,
and neuroblastoma-like schwannoma. MPNSTs may also
have a histological variant referred to as the epithelioid sub-
type. PNSTs are associated with the neurofibromatoses;
however, 90% of solitary orbital PNSTs occur in the absence of
oculoneurocutaneous syndromes.128,134
Other exceptionally rare tumors affecting peripheral
nerves have been described but are out of the scope of this
review. These include traumatic neuroma, paraganglioma,
amputation neuroma, melanotic neuroectodermal tumor of
infancy, and primary orbital neuroblastoma.
2. Epidemiology
PNSTs often involve the head and neck, but are uncommon
in the orbit. Karcioglu summarized 5 large studies of biopsy-
proven orbital tumors and found the relative frequency of
PNSTs to all orbital tumors to be neurofibroma (all types)
0.4%e3.0%; schwannoma (subtypes not specified) 0.7%e
2.3%; and MPNST 0%e0.2%.62 Benign PNSTs most commonly
affect adults aged between 20 and 60 years,19,119 with the
exception of plexiform neurofibromas, in which 50% of le-
sions are diagnosed between 1 and 5 years of age,42,43
initiating debate for this tumor type to be considered a
congenital lesion.37 Previously, MPNSTs were considered
tumors of adulthood, with rare occurrences in children with
neurofibromatosis type 1 (NF-1)59; however, more recent
case reports demonstrate involvement in children with or
without NF-1.17,36,40 PNST involvement is approximately
equal between genders with some reporting a slightly
higher occurrence in women.27,119,141 No racial predilection
has been observed.
Localized orbital neurofibromas have an 11%e28% associ-
ation with systemic neurofibromatosis or a family history of
neurofibromatosis.76,119 This contrasts to extraorbital neuro-
fibromas, which are characteristic of NF-1 and infrequently
found outside of neurofibromatosis. An extraorbital neurofi-
broma fulfills 1 of 2 diagnostic criteria for NF-1. The plexiform
neurofibroma subtype is described by some as pathogno-
monic for NF-13; however, one group recently reported 3 in-
dividuals with plexiform neurofibroma without other
systemic features meeting the diagnostic criteria of NF-1, and
two of these individuals lacked the NF-1 mutation on pe-
ripheral blood DNA analysis.9 This highlights that no single
finding is diagnostic of NF-1.
Patients with NF-1 have a 2%e18% likelihood of developing
orbital solitary or diffuse neurofibromas and a 5% likelihood of
developing orbital plexiform neurofibroma.37 Patients with
neurofibromatosis are also at increased risk for other orbital
tumors, including optic nerve glioma (10%e15%),37 optic nerve
sheath meningioma (2%e8%),14 and less commonly, orbital
schwannoma (1.5%).117
Solitary schwannomas affecting the orbit are rarely asso-
ciated with systemic features of neurofibromatosis.118 Plexi-
form schwannomas have a much weaker association with the
neurofibromatoses, with only a few extraorbital cases asso-
ciated with NF-1 or neurofibromatosis type 2 (NF-2) reported
in the literature.149 Multiple schwannomas affecting the orbit
in the absence of vestibular schwannomas or other systemic
features of NF-2 may be classified into a different category
termed neurofibromatosis type 3, also referred to as
schwannomatosis.110
Rarely, benign neurofibromas and schwannomas may un-
dergo malignant transformation.32,47,50,117,123,139 Solitary neu-
rofibromas in the absence of NF-1 undergo malignant
transformation to MPNSTs exceedingly rarely with only one
reported.22 While only a few case series discuss the epidemi-
ology of orbital MPNSTs, more is known about extraorbital
MPNSTs, which have a prevalence of 4% in patients with NF-1
and 0.0001% in the general population.151 Radiotherapy for
prior malignancy may be associated with increased risk of
MPNSTs in patients with NF-1.34 To our knowledge, only 1 case
has been reported of a primary orbital epithelial MPNST,108 a
form of MPNST portending a dismal prognosis when found
outside the orbit.96
3. Etiology
The molecular etiology of PNSTs is only partially understood.
In patients with NF-1, neurofibromas are formed after biallelic
loss of the tumor suppressor gene NF-1 (17q11.2) in Schwann
cells. MPNSTs have been found to occur following both the
loss of NF-1 and overexpression of RAS coinciding with inac-
tivation of cell cycle regulators, such as p53.39 The NF-2 gene
(22q11.2) is also considered a tumor suppressor gene. Loss of
NF-2 or the gene’s encoded protein (merlin) in Schwann cells
results in Schwann cell hyperplasia and schwannomas.38 Less
is known about the molecular etiology of sporadic PNSTs. One
patient with sporadic plexiform neurofibroma without other
features of NF-1 was shown to have a biallelic loss of the NF-1
gene with a mosaic Schwann cell populationdsome cells
demonstrating a chromosomal rearrangement mutation in
one allele and a deletion in the other allele.10 This case sug-
gests a second hit phenomenon, which may be applicable to
other benign PNSTs as well.
Neurofibromas arise from nonmyelinated, neoplastic
Schwann cells and differentiate themselves from schwanno-
mas by their neoplastic incorporation of various other cell
types.29 These tumors primarily arise from the V1,62 rarely
from V243 or the nerves innervating the extraocular muscles.1
Schwannomas originate from myelin-producing Schwann
cells and principally grow via hyperplasia of this cell of origin.
Most orbital schwannomas arise from sensory nerves,
particularly, branches of V1; however, tumors of V2,25,118 the
 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
cranial nerves in the orbit,118,125 and the nerves innervating
the extraocular muscles20,70 have been reported. More rarely,
schwannomas can affect the globe, with reported origins
being the ciliary body78,129 choroid,29,131,159 iris,111 and
sclera49,95 with pathological demonstration of tumor origina-
tion from the posterior ciliary nerve.41,131 While Schwann cells
are specific to peripheral nerves, optic nerve schwannomas
occur,28,67,86,112,133 with the proposed mechanism stemming
from autonomic perivascular nerves surrounding the optic
nerve sheath.112
MPNSTs generally arise de novo62 but may also originate
from neurofibromas or schwannomas.3,62,149 An estimated
25%e50% of MPNSTs are found in patients with neurofibro-
matosis and a history of plexiform neurofibroma.88,158 The cell
type origin of many MPNSTs often remains unknown given
the diverse cellular proliferation; however, the Schwann cell is
considered to be the typical, but not the only known, cell type
of origin. Similar to benign PNSTs, MPNSTs have a predilection
for the superior orbit, specifically V1.34,90 MPNSTs are known
to metastasize to regional lymph nodes and lung,59,90 with a
higher risk of metastasis following postsurgical recurrence.62
4. Clinical features
In the orbit, benign PNSTs commonly present with an insid-
ious onset and are governed by the tumor location, their slow
rate of growth, and their noninvasive nature. Localized neu-
rofibromas and schwannomas present with lid swelling and
proptosis in 50%dwith ptosis, decreased vision, and diplopia
presenting features in 4%e20%.119 Late manifestations
including visual obscuration and pain are potentially ominous
manifestations of nerve root compression, globe indentation,
or transformation to invasive tumor types such as MPNST.
Approximately 1/3 of localized neurofibromas extend into the
superior orbital fissure.119 Similarly, the rate of extension into
the superior orbital fissure for schwannomas is 16%e
24%.32,119,148
Localized neurofibromas typically present with progressive
unilateral proptosis of less than 1-year duration. Visual acuity
is often preserved in the absence of optic nerve compression.
Extraorbital localized neurofibromas may undergo an accel-
erated rate of growth during pregnancy and puberty.149
Localized neurofibromas affecting the orbit are not typically
associated with NF-1; however, when this association is pre-
sent, the orbital lesions are generally preceded by café au lait
spots.149 Multiple localized neurofibromas in the same orbit in
the absence of NF-1 are rare.76,132 PNSTs affecting the orbits
bilaterally in the absence of NF-1 have been reported only in 2
patients: one with Charcot-Marie-Tooth disease (a disorder
with a defect on the same gene locus as NF-1)91 and one with
several features of multiple endocrine neoplasia type IIB.97
Orbital plexiform neurofibroma will often present in
childhood or adolescence with lid swelling in an S-shaped
curvature of the upper eyelid highly characteristic of NF-1.62
The strong association of postseptal neurofibroma with pre-
ceding eyelid involvement has prompted some to consider
orbital involvement a product of early posterior spread across
the septum.37 Careful palpation of the ocular adnexa may
demonstrate the classic “bag of worms” texture.149 Palpation
45
may also reveal pulsatile proptosis or enophthalmos, indi-
cating sphenoid wing dysplasia and herniation of orbital
contents into the craniumdwhich may be specific for plexi-
form neurofibroma in the setting of a benign, slow-growing
orbital tumor.
Neurofibromatosis may have additional ophthalmic man-
ifestations, notably iris melanocytic hamartomas (Lisch nod-
ules) occurring in 92% of patients
6 years of age with NF-1.87
The second most common ophthalmic manifestation in NF-1
is choroidal hamartoma.37 The most common orbital tumor in
patients with NF-1 is optic pathway glioma, which may arise
in the orbit or intracranially in 15%e20% of patients with NF-
1.62 Other ocular manifestations of NF-1 include retinal
astrocytic hamartoma, choroidal melanoma, choroidal
schwannoma, and retinal vascular occlusions.62 Ocular
manifestations of NF-2 are less common, but include juvenile
posterior subcapsular or peripheral cortical cataracts, epi-
retinal membrane, retinal hamartomas, and ocular motor
abnormalities, in that order of frequency.62,92,95
Orbital schwannomas present similarly to neurofibromas
with early manifestations of gradual nonpulsating proptosis
and lid swelling. Late manifestations include diplopia, re-
striction in ocular motility, mild impairment in visual acuity,
and symptoms of optic nerve compression including sco-
tomas, dyschromatopsia, and decreased contrast sensitivity.
When proptosis does occur, inferior displacement is most
common, as the superior quadrant is the location of primary
involvement in 40%-60%.19,135,148 Occasionally, orbital
schwannomas may present with a deep, dull pain or with
paresthesias in the distribution of the affected nerve. On
occasion, a mass in the orbit may be palpated. Only 1 case is
known with simultaneous bilateral orbital involvement.122
As with neurofibromas, case reports describe rapid growth
of orbital solitary schwannomas during pregnancy.13,21,142 In
one case, progesterone receptors were diffusely found on
immunohistochemistry of the tumor,21 and in another case,
estrogen and progesterone receptors were absent with his-
tology and serial imaging demonstrating intratumor hemor-
rhage as the likely etiology of rapid growth.142 This finding of
intratumor hemorrhage has also been observed in the rare
reports of rapid solitary schwannoma growth in nonpregnant
patients.135
MPNSTs affecting the orbit have increased incidences of
pain, redness, and rapidly progressing ptosis59,90,119 differen-
tiating them from their benign counterparts. Unfortunately,
MPNSTs may have early metastasis with spread along pe-
ripheral nerves to the cranium leading to the presentation of
neurological changes in addition to orbital complaints.59
Rapid recurrence of previously diagnosed benign PNSTs
following partial excision should raise clinical suspicion for
MPNSTs.59
5. Differential diagnosis
The clinical differential diagnosis for PNSTs includes other
slow-growing tumors such as meningioma, cavernous
hemangioma, lymphangioma, fibrous histiocytoma, lym-
phoma, dermoid cyst, hemangiopericytoma, and pleomor-
phic adenoma of the lacrimal gland. These lesions are very
46 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
difficult or impossible to differentiate on clinical
examination.
Imaging techniques have become extremely useful to aid-
ing the identification of orbital tumors; however, even with
modern, high-resolution imaging, many tumor types remain
difficult to distinguish. If consecutive imaging is performed,
particular attention should be focused on the tumors evolu-
tion in size, location of spread, and specific imaging contrast
techniques described in the following. An increase in the
growth rate or new development of pain is atypical for benign
PNST and should be concerning for high-grade neoplasm or
malignancy.
6. Diagnostics
6.1. Imaging
6.1.1. Computed tomography/magnetic resonance imaging
Radiographically, solitary lesions in the orbit include neuro-
fibroma, schwannoma, MPNST, cavernous hemangioma,
meningioma, lymphoma, solitary fibrous tumors, dermoid
cysts, and others. Both computed tomography (CT) and mag-
netic resonance imaging (MRI) assist in narrowing the differ-
ential of orbital masses. In relationship to PNSTs, CT is best
used for monitoring bony erosion and for surgical planning,
while MRI may help to assist in characterizing the structure of
the tumors and their involvement with adjacent soft-tissue
structures.105
6.1.1.1. Neurofibroma. Localized neurofibromas on CT
demonstrate homogenous density and may be smoothly
marginated, round, ovoid, or lobulated. The majority of
these tumors will be isodense to extraocular muscles32,43
and will variably enhance, with some demonstrating pe-
ripheral ring enhancement.43 Adjacent bony opacification
from tumor pressure may be seen, highlighting the tumors
slow growth.84 Neurofibromas typically lack the secondary
degenerative changes, including calcification, that may be
seen on CT in schwannomas, meningiomas, or more rarely
cavernous hemangiomas.149 Based on CT images, confusion
between cavernous hemangioma and neurofibroma is
common, as both demonstrate a well-outlined, enhancing
lesion with similar homogeneity. Neurofibromas, however,
may be more rounded and are more commonly extraconal
on CT.43
Diffuse neurofibromas on CT typically demonstrate an
irregular, poorly defined soft-tissue mass with variable
contrast enhancement. Typically, these tumors are isodense
with the extraocular muscles; however, there are cases of
nonorbital neurofibromas and 1 orbital case where these tu-
mors have appeared dark on CT imaging, similar to orbital
dermoid cyst or lipoma.154
Plexiform neurofibromas may be distinguished from soli-
tary neurofibromas, schwannomas, and other solitary tumors
by the degree of infiltration of adnexal structures seen on
imaging.65 These tumors typically appear as a multilobular,
oblong, diffuse mass. Absence of one or both wings of a
sphenoid bone is a characteristic feature of NF-1, particularly
in the presence of plexiform neurofibromas.42 Additionally,
adolescents with NF-1 may have remodeling and expansion of
the ipsilateral bony orbit.
On MRI, neurofibromas may appear similar to schwanno-
mas in that they may be hypointense to orbital fat and iso-
intense to gray matter on T1-weighted images and
intermediate to hyperintense on T2-weighted images
(Fig. 1).31,74 Commonly, a peripheral ring of hyperintensity on
T2-weighted images will create a “target sign,” further aiding
the differentiation of this tumor. MRI intensity and degree of
heterogeneity represents the histological features of the tu-
mors, with greater water content in myxoid regions of these
tumors shown to represent hyperintensity regions on T2-
weighted images and hypercellular and more collagenous
regions demonstrating hypointensity on T2-weighted im-
ages.31,74 In MRI studies, enhancement is similarly variable for
both schwannoma and neurofibroma.
6.1.1.2. Schwannoma. Orbital schwannomas tend to be less
round and more oval or spindle-shaped compared to other
PNSTs. Schwannomas are typically smooth, well-
circumscribed tumors molding to the shape of the cavity148
with characteristic growth along the axis of the orbit.32
Schwannomas imaged on CT appear homogenously dense,
typically isodense to extraocular muscles, with smooth round
or elongated morphologies.32 Schwannomas often demon-
strate enhancement with contrast on CT.118 Calcification
within the tumor has been reported for solitary primary
schwannomas.118,136 Schwannomas are more often extraco-
nal, whereas meningiomas and hemangiomas are often
intraconal.144 Extension through the superior orbital fissure is
more indicative of schwannoma than meningioma aiding in
differentiating these tumors.65,148 One proposal is that bone
expansion without erosion of the fissures is characteristic of
extraconal schwannomas,27 a finding demonstrated with
lesser frequency in cases of neurofibromas.19,32,130
Schwannomas affecting the orbit have various MRI fea-
tures; however, these lesions typically produce a hypointense
signal on T1-weighted images and a hyperintense signal on
T2-weighted images (Fig. 2). Orbital schwannomas may
demonstrate homogenous or heterogeneous enhancement.
Some MRI findings are a result of the underlying histology and
macroscopic morphology of these tumors.1,101,127 Shen and
colleagues demonstrated corresponding Antoni A regions to
have intermediate signal intensities with postcontrast
enhancement in both T1- and T2-weighted image, contrasting
to Antoni B regions demonstrating hypointensity on T1-
weighted images, hyperintensity in T2-weighted images, and
no contrast enhancement.127 Cystic degeneration may be seen
in as high as 41% of schwannomas, with these regions being
found to have similar radiologic features to Antoni B regions.63
MRI may be useful in differentiating schwannoma from
other lesions. On MRI, schwannomas and lymphomas may
have similarly round shapes, however, lymphomas have in-
termediate T2 signal and will classically mold around struc-
tures, in contrast to schwannoma which may abut and distort
anatomy of adnexal structures.65 Similarly, the shape of der-
moid cysts may correspond to that of schwannomas and
occur in similar locations; however, these round benign
masses are hyperintense on T1-weighted images and lack
gadolinium enhancement, differentiating them from
 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7 47

Fig. 1 e Axial MRI images of localized orbital neurofibroma histologically confirmed after excision. A: T1-weighted image, B:
T1-weighted image postcontrast, and C: T2-weighted image. MRI, magnetic resonance imaging.

schwannomas.160 Solitary fibrous tumors and schwannomas 6.1.2. Ultrasound

have similar T1- and T2-weighted image intensities and lo-
cations. To differentiate between these tumors, dynamic
contrast-enhanced MRI may be of use. Studies using dynamic
contrast-enhanced MRI comparing contrast washout pro-
posed that tumors with high-cellularity stroma, such as soli-
tary fibrous tumors, have a smoother washout curve than
schwannomas, which have a persistent or plateau-shaped
washout curve presumably from their nonuniform, loose
cellular arrangement.160 Dynamic contrast-enhanced MRI has
also been shown to assist in differentiating cavernous hem-
angiomas from schwannomas with cavernous hemangiomas
demonstrating progressive enhancement in later images
which is atypical for schwannoma.152
Ultrasonography has a limited role in the evaluation of PNSTs;
however, it may still be used for expedited evaluation, routine
follow-up, or to monitor for progression. On B-scan, neurofi-
bromas are generally diffuse and irregular, with internal
vascularity sometimes appreciated.62 The characteristics of
schwannomas on B-scan have been more thoroughly
described. These tumors may appear as round, well-defined,
solid lesions with a highly reflective surface or sometimes as
a heterogeneous or cystic mass with several tissue interfaces
that attenuate the signal intensity.100,118,135 Additionally,
acoustic hollows have been suggested in schwannomas to
correlate with hemorrhage within the tumor.18 MPNSTs have
similar sonographic findings as schwannomas.100

6.1.1.3. MPNST. Imaging characteristics of MPNSTs are more
variable than the other PNSTs. On CT, MPNSTs often appear as
a well-defined, rounded, homogenously enhancing mass
commonly located in the superior orbit, indistinguishable
from other solitary orbital tumors, including benign
PNSTs.90,144 These tumors may be well-circumscribed and
found to erode through adjacent bony structures.100 On MRI,
orbital MPNSTs may have an isointense homogenous signal
on T1-weighted images, hyperintense on T2-weighted images
with diffuse enhancement to gadolinium with regions of
intense enhancement.17,52 Lesions with an irregular shape
and/or adjacent bone invasion should be highly concerning
for malignancy in patients with NF-1.
6.2. Histological features
6.2.1. Neurofibroma
On gross pathology, solitary neurofibromas appear as a poorly
vascularized, firm, rubbery, well-circumscribed gray mass. On
low power, neurofibromas can be distinguished from
schwannomas by the lack of a true capsule and lower cellular
density. Cytologically, there is variability in cell type density;
however, classically neurofibromas demonstrate an overall
low cellularity, with a loose myxoid background composed of
mucopolysaccharide matrix punctuated by circumscribed
proliferations of interlacing bundles of elongated, wavy,
spindle-shaped cells. This pattern is a result of the

Fig. 2 e Axial MRI images of orbital schwannoma histologically confirmed after excision. A: T1-weighted image, B: T1-
weighted image postcontrast, and C: T2-weighted image. MRI, magnetic resonance imaging.
48 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7

endoneurial growth pattern of neurofibromas in which the
nerve is expanded radially entrapping and separating cells.3,43
Characteristic wavy collagen bundles will be present, often
abundantly, throughout the tissue (Fig. 3A) and may be best
appreciated with a trichrome stain. With neurofilament im-
munostaining, neurofibromas can usually be shown to
contain scattered axons arranged individually or in small
clusters, a pattern that is typically absent in schwannomas
(Fig. 3B). In neurofibromas, spindle cells with darkly stained
nuclei resemble cells of endoneurial fibroblast origin, whereas
other spindle cells with comma shaped nuclei resemble those
of Schwann cell origin. Mast cells are often present
throughout the tumor (Fig. 3C).
Diffuse neurofibromas are morphologically appreciated
grossly or on low power as spreading diffusely, surrounding
structures of the ocular adnexa. These neoplasms have
increased vasculature compared to localized neurofibromas.
Histologically, diffuse neurofibromas have similar cell types
as localized neurofibromas; however, they may be distin-
guished from localized neurofibroma by a greater degree of
cellular density and by extensive infiltration to surround
structures of the ocular adnexa, corresponding to the gross
appearance. The mode of spread is via connective tissue septa
and intracellular spaces along the adnexa. Collections of
Schwann cell processes may make up ovoid bodies in this
subtype.
Plexiform neurofibromas may identify themselves on im-
aging or intraoperatively by their nodular, tortuous, “bag of
worms” appearance, owing to irregular hyperplasia that is
often referred to as “spilling out” into surrounding tissue.
Similarly to the diffuse subtype, plexiform neurofibromas may
contain greater vascularity than solitary neurofibromas
(Fig. 3D). Cytologically, this PNST contains the same cell types
as solitary and diffuse neurofibromas; however, plexiform
neurofibromas demonstrate different cell type densities with
a higher degree of endoneurium and perineurium hyperplasia
spacing out axons. Perineurial sheathing is characteristic.
Increased nuclear atypia and cellularity seen histologically
coincides with a greater risk of malignant transformation.149
All neurofibroma subtypes will stain positively for S-100;
however, neurofibromas generally only have a subset of
positively staining cells, grossly staining intermittently and
less intensely than schwannomas, demonstrating the greater
degree of mixed cell types in neurofibroma.3,43,149 The matrix
of a neurofibroma will stain positive for Alcian blue, whereas
the schwannoma will be negative.
The cellular composition of neurofibroma includes
neoplastic Schwann cells, fibroblasts, mast cells, and vascu-
lature.157 Both the neoplastic and microenvironment cellular
components host different molecular elements to neurofi-
broma tumorigenesis. Within the neoplastic cells, biallelic
inactivating mutations in the tumor suppressor NF-1, which
codes for the protein neurofibromin, are commonly associ-
ated with neurofibromas.30,114 The loss of NF-1 leads to over-
activation of Ras, which signals through 2 major effector
pathways, including PI3K/Akt/mTOR and Raf/MEK/
ERK.7,53,64,147 Recurrent gene mutations in NF-1 associated or
sporadic cases of neurofibromas outside of NF-1 have not been
described. In a case of an individual with NF-1, whole-exome
sequencing demonstrated a low variant allele frequency
missense mutation in MYB within a plexiform neurofi-
broma.55 On malignant transformation to MPNST, as well as

Fig. 3 e Neurofibroma. A: Loose spindle cells interspersed with wavy collagen; B: longitudinal section of single, large,
myelinated axon in neurofibroma; and C: loose myxoid tissue with occasional mast cell and axon; and D: plexiform
neurofibroma demonstrating increased vascularity and sheet-like architecture.
 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
metastasis, the MYB variant allele frequency increases, indi-
cating more of the neoplastic cells harbored the mutation. The
importance of MYB gene mutation in the setting of neurofi-
bromas is yet to be determined. Recurrent copy number al-
terations of whole or partial chromosomes (outside of the 17q
region encoding NF-1) are not present in neurofibromas.116
Similar to their unique genetic signatures, recent epigenetic
profiling has shown that methylation profiles of neurofibroma
differ from that of MPNST and schwannoma.116 Furthermore,
methylation classification can discriminate between dermal,
localized intraneural, and plexiform subtypes.116 Methylation
classification has the potential for clinical diagnostic utility
when there is a diagnostic dilemma of plexiform neurofi-
broma, which may imply a possible association with NF-1. In
concert with the neoplastic component, the microenviron-
ment is important for neurofibroma tumorigenesis, mainly
contributed by hematopoetic-derived mast cells.94,140,155e157
The contribution by mast cells to neurofibroma development
and maintenance includes molecular effectors such as TGF-
beta and c-Kit.94,155,156
6.2.2. Schwannoma
Histologically, most schwannomas have features that justify
designation of “conventional” schwannoma; however, 4
additional histological variants have been described: cellular
schwannomas, melanotic schwannomas, plexiform schwan-
nomas, and neuroblastoma-like schwannomas.71,72,79 In the
orbit, cellular schwannomas are much less common than
conventional schwannomas. Melanotic, plexiform, and
neuroblastoma-like variants are exceptionally rare tumors in
the orbit.
Fig. 4 e Schwannoma. A: Verocay bodies classic for Antoni type A; B: Antoni type B with loose myxoid tissues; C: cellular
schwannomas with poorly formed Verocay bodies focally; and D: cystic schwannoma with degenerative features.
49
Macroscopically, conventional and cellular schwanno-
mas demonstrate a true fibrous smooth capsule formed
from the perineurium of the nerve of origin. Melanotic
schwannomas may have a thin fibrous membrane at most,
whereas a capsule may be less evident and lobulated in
plexiform variant.80,99 In cases where the neoplasm is small,
the parent nerve may be clearly identified with the tumor
growing eccentrically from it. This eccentric growth pattern
may help to distinguish the solitary schwannoma, espe-
cially from a neurofibroma, as the latter tend to grow within
the nerve expanding it radially.150 Larger schwannomas can
overgrow the nerve, but diffuse infiltration of the nerve, as
seen in neurofibroma, is not a typical feature of
schwannomas.
On low-power light microscopy, schwannomas distinguish
themselves by the strikingly greater concentration of
Schwann cells than seen in other tumor types. The charac-
teristic identifying feature is the distinct biphasic morpho-
logical pattern owing to variable, patchy amounts of Antoni A
and Antoni B tissue patterns.
Antoni A areas are hypercellular and composed of bundles
and fascicles of compact spindle cells with indistinct cyto-
plasmic borders arranged in parallel along their long axis. A
common feature of Antoni A areas are Verocay bodies, which
are strips or regions of nuclear areas bounded by clusters of
elongated spindle cell nuclei arranged in palisades (Fig. 4A). In
highly differentiated areas, whorled patterns of cells can
mimic morphology seen in meningioma.149 A periodic acid
Schiff stain will be strongly positive in schwannoma, as will
immunoperoxidase assay for Laminin, reflecting the fact that
each cell is producing a true basement membrane. At higher
50 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
magnification, mitotic activity may be seen but should not be
brisk.
Antoni B tissue pattern is less cellular and less organized.
The widely separated bland round often vacuolated cells
arranged in sheets in a myxoid or microcystic appearing
matrix. Scattered throughout this tissue pattern are irregu-
larly spaced small, thick-walled, hyalinized vessels and oc-
casional clusters of lipid rich (foamy) histiocytes,
inflammatory cells surrounding vessels, and delicate
collagen fibers (Fig. 4B).
These tissue patterns are each represented to a variable
degree within a given schwannoma, and occasionally one
pattern may predominate over the other or even be absent.
Schwannomas with less cellular density and neurofibromas
with greater density and a higher degree of organization can
appear remarkably similar. Distinguishing features described
previously, including infiltration of peripheral nerve, presence
of a capsule, and S-100 immunopositivity, should be consid-
ered. In most cases, however, the alternating pattern of
Antoni A and Antoni B regions, classic for schwannomas, will
not be present in neurofibromas.43
The cellular schwannoma subtype is distinguished on low-
power light microscopy by the paucity, or absence, of the
Antoni B tissue pattern. The cells typically take a storiform
appearance and are packed tightly together, arranged in fas-
cicles. Occasionally, they hint of nuclear palisades; however,
unlike the Antoni A tissue pattern of conventional counter-
type, cellular schwannomas typically lack well-formed Ver-
ocay bodies (Fig. 4C). Cytologic atypia may be increased in
cellular schwannomas, and mitotic activity may also be
increased, which often raise consideration for malignancy in
the differential diagnosis.124 Also in the histologic differential
diagnosis of cellular schwannomas are benign smooth muscle
neoplasms (leiomyomas), which can be distinguished by im-
munostaining for smooth muscle actin.
The plexiform schwannoma shares features with conven-
tional schwannoma, including a true capsule and cell type
population,149 but takes its origin from a nerve plexus or
ganglion forming bundles of palisading spindle cells sur-
rounding the involved nerve.80 This subtype may demonstrate
a higher degree of Antoni type A than Antoni type B tissue.80
Given this associated higher degree of cellularity, the plexi-
form subtype may appear similar to sarcoma, however, the
lack of mitotic activity, atypia, and geographic necrosis in the
presence of strong S-100 immunostaining should guide
pathologist to this benign entity.149
Similar to the plexiform subtype, melanotic schwannomas
are also exceedingly rare tumors with only 1 case reported
arising in the orbit.60 Melanotic schwannomas have variably
abundant pigment and are immunopositive for HMB-45 and
Melan-A. The greatest degree of pigmentation is found within
scattered melanophages. Differentiating between melanotic
schwannoma and melanocytic lesions including malignant
melanoma can be a diagnostic challenge in which the
pathologist must rely on histomorphology, cytoarchitecture,
and features of malignancy including high proliferative index
to differentiate the entities.
A rare histological variant of schwannoma,
neuroblastoma-like schwannoma, also has only a single case
involving the orbit.79 Microscopically, the hallmark finding of
these lesions is a “giant rosette” with a central eosinophilic
collagenous core surrounded by small round to oval cells
with hyperchromatic nuclei. Nuclear atypia and mitotic fig-
ures are low or absent. These neuroblastoma-like regions
may represent the majority of the neoplasm, or may be a
small portion in a tumor with otherwise conventional fea-
tures including Antoni A and Antoni B regions present.46
Immunohistochemical staining pattern is that of conven-
tional schwannomas, and negative for synaptophysin CD99
and neuron-specific enolase that permits relatively
straightforward differentiation of these tumors from
neuroblastoma.46
The term “cystic schwannoma” has also been used in
ophthalmology to describe a conventional schwannoma
where degenerative microcystic and myxoid areas have coa-
lesced to form a macrocystic form appreciated on low-power
light microscopy (Fig. 4D).81,141 Plump Schwann cells may
line microcystic spaces and produce a round or epithelioid
appearance that may be confused with true epithelial
differentiation.149
The ophthalmic literature has used the term “ancient
schwannoma” to describe longstanding tumors of the con-
ventional and cellular variants that have undergone degen-
erative changes over time, which may include microcystic
changes, hemorrhage, and focal calcification.66,98,122 Cytolog-
ically, ancient schwannomas can be a challenge owing to
frank cytologic atypia and nuclear pleomorphism, and may be
misconstrued as sarcomas. Indeed, so called “ancient change”
may be seen in individual cells in most schwannomas, but the
absence of mitotic figures, presence of Antoni A and B regions,
and positive S-100 staining should guide diagnosis in these
cases.66
All schwannoma variants stain strongly and diffusely
positive for S-100 with less-intense staining in Antoni B areas
as the Schwann cell population is less dense.149 Type-IV
collagen staining highlights pericellular collagen deposi-
tion.57,106 Schwannomas commonly show immunopositive
reactivity for SOX10, p16, and neurofibromin, while being
completely negative for epidermal growth factor receptor, and
the combination of these markers may aid in distinguishing
cellular schwannoma from MPNST.106
Schwannomas may arise in the context of a clinical syn-
drome related to germline mutations, including NF-2 and
schwannomatosis.89 NF-2 results from mutations in NF2,
located on chromosome 22q, and encodes for the protein
merlin which signals through the Hippo/YAP pathway.104,120
Schwannomatosis results from mutations in INI1/SMARCB1
or LZTR1 and possibly yet undiscovered genes on chromo-
some 22.16,58,107,137 In sporadic cases of schwannoma,
recurrent genetic mutations present often include the tumor
suppressor NF2, and to a lesser extent, INI1/SMARCB1.11,51,121
Mutations in the genes LATS1 and LATS2, which encode for
proteins that signal downstream from merlin, are found
infrequently in schwannomas.104 Loss of chromosome 22q is
the only frequently recurrent cytogenetic alteration found in
schwannoma.116 Like neurofibromas, methylation classifi-
cation can help distinguish schwannomas from other PNSTs
and may have clinical utility in cases where there is ques-
tionable morphologic and immunohistochemical
characteristics.116
 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
6.2.3. MPNST
On gross examination, MPNSTs are characterized as a fusi-
form, nodular mass with increased vascularity and perineural
extension. Invasion or envelopment of the ocular adnexal
structures is commonly seen.3,90 The absence of a capsule
surrounding MPNSTs is typical and will assist differentiation
of an early MPNST from a schwannoma.
The cell of origin of MPNSTs is difficult to surmise, as they
consist of spindle-shaped cells with heterochromic nuclei
with various nuclei arrangements including oval shaped,
comma shaped, and vesicular (Fig. 5A).3,149 Schwannomas and
MPNSTs similarly contain nuclear palisading regions; how-
ever, in MPNSTs these regions are less organized and more
sparse, only present focally, or entirely absent.3,149 Findings
that may be somewhat concerning for early malignant
changes in schwannoma are collections of positively staining
S-100 epithelioid cells, eosinophilic cytoplasm, vesicular
chromatin, and prominent nucleoli.96
A common challenge is distinguishing a low-grade MPNST
from a neurofibroma with atypical features. Histologically,
these lesions may demonstrate a continuum of cellular and
morphological changes with an admixture of neurofibroma or
schwannoma appearance as it recapitulates nerve sheath
cells in a disorganized manner.149 MPNSTs, however, will
classically demonstrate malignant features such as brisk
mitotic activity, increased cellularity, and a high degree of
pleomorphism with packed cellular fascicles, nodular aggre-
gates, and myoid zones (Fig. 5B).
Differentiation between a neurofibroma and an MPNST is
typically based purely on histomorphological features, but the
use of immunohistochemical techniques to evaluate PNST
suspicious for MPNST is growing. Myelin basic protein, Leu-7,
and PGP 9.5 are reportedly expressed in MPNST; p53 is
frequently positive.3,149 Some authors report the p16 antigen
to be present in neurofibromas, and absent in MPNSTs, aiding
the differentiation of these tumors in the possible continuum
of cellular change.149 At low magnification, MPNSTs have
weak, sparse staining to S-100 when compared to schwan-
nomas. More recently, antibodies to SOX10, a neural crest
transcription factor, has been used to stain cells of neural
crest origin including Schwann cells in both schwannomas
and neurofibromas, with more reliable staining of MPNST
tissues than S-100.103 The absence of SOX10, p16, and
Fig. 5 e Malignant peripheral nerve sheath tumor. A: High-power field of dense spindle-shaped cells with pleomorphic,
irregular hyperchromatic nuclei and B: low-power field demonstrating increased cellularity, myoid zone, and mitotic
activity.
51
neurofibromin immunostaining or the presence of epidermal
growth factor receptor immunopositivity is specific for MPNST
and can help differentiate MPNST from cellular
schwannoma.106
Epithelioid MPNSTs are a histologic variant of MPNSTs
accounting for 5% of extraorbital MPNSTs. Only 1 case has
been reported as a primary tumor of the orbit.108 This subtype
has a weaker association with NF-1 than the conventional
MPNST. Histologically, this subtype shares the malignant
features of conventional MPNST but may be distinguished by a
mixed morphology with rounded, polygon-shaped cells in
nest or chords. The appearance may resemble carcinoma or
malignant melanoma, however, will be negative for HMB 45
and cytokeratin, which are associated with melanoma and
carcinoma, respectively.82,108 Most epithelioid MPNSTs stain
positive for S-100, epithelial membrane antigen, and neuron-
specific enolase.82
Similar to their benign neurofibroma counterpart, biallelic
inactivation of NF-1 is well established to occur in the majority
of NF-1 associated and sporadic MPNST cases.15,113,115,145
There are specific mutations that occur as neurofibromas
transform into high-grade MPNST. CDKN2A, which encodes
for the cell cycle tumor suppressor proteins p14ARF, p15INK4b,
and p16INK4a, is frequently lost in MPNST and not lost in
neurofibroma.12,75,93,102 Loss of CDKN2A is independently
associated with a significant decrease, up to an 83.3%, in 5-
year overall survival rate.35 Loss of polycomb repressive
complex 2 components (EED or SUZ12) is frequent in NF-1
associated (70%) and sporadic (92%) cases of MPNST.85 Poly-
comb repressive complex 2 loss commonly cooccurs with loss
with NF-1 and CDKN2A loss.85 TP53 mutations are not
commonly found in MPNST.146 Genetic studies of 2 individual
cases of NF-1 associated MPNST arising from a neurofibroma
showed that transformation of neurofibroma in to high-grade
MPNST in each individual involved loss of TP53.55,138 In a se-
ries of 62 MPNST, the BRAF-V600E mutation was identified in 1
of 37 (2.7%) NF-1 associated and 5 of 25 (20.0%) sporadic cases
of MPNST.56 The prognostic and therapeutic relevance of
BRAF-V600E mutation in MPNST has yet to be determined. As
mentioned in previous sections, 450-k methylation profiling
can help distinguish MPNST from other PNSTs.116 High-grade
MPNST can be further divided into 2 methylation classifica-
tions, one of which that corresponds to loss of histone H3K27
52 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
trimethylation (H3K27me3).118 Loss of histone H3K27 trime-
thylation (H3K27me3) is found in a subset (34%e69%) of NF-1
associated and sporadic MPNST and is not present in neuro-
fibroma, which can help in the differential diagnosis of these
entities.26,109 Furthermore, within MPNST, loss of H3K27me3
is an independent predictor of a significantly worse overall
survival, with hazard ratio of 2.6 (confidence interval 1.2e5.7,
P ¼ 0.0017).26
7. Management
7.1. Surgical excision
Complete excision of benign neurofibromas and schwanno-
mas with full effort to maintain the capsular integrity is the
mainstay of treatment. Numerous surgical approaches have
been described to maximize access. Because most PNSTs are
in the superior orbit, the most common approach is anterior
orbitotomy through an eyelid crease incision.45 Lateral orbi-
totomy is commonly used for superolateral tumor locations.
An inferior transconjunctival incision may offer adequate
access for inferior or medial tumors. Medial access may be
best obtained via a transcaruncular incision45; however,
recently an endoscopic endonasal approach for medial wall
tumors near the apex has shown to be safe and effective for
those trained in this technique.143
Extension of benign tumors into the superior orbital fissure
considerably complicates the surgical approach. Rose and
colleagues reported 14/47 cases of benign PNSTs that had
superior orbital fissure extension limiting full extraction from
the orbital approach.119 When superior orbital fissure
extension is known, cosurgery with a neurosurgeon is rec-
ommended. Recent neurosurgical reports using a pterional-
extradural approach boast excellent exposure to the superior
posterior orbit, superior orbital fissure, and optic canal,126
which conceptually would improve the chances of complete
excision. One author advocates soft-tissue approaches for
benign intraconal tumors, reserving bony removal for tumors
located near the apex or within the superior orbital fissure.45
Regardless of the surgical approach, the surgeon should
make every effort to identify the involved nerve as sensory
rather than a motor nerve before sacrificing the nerve while
excising the tumor. Sacrificing the nerve may be avoided by
microsurgical techniques or careful capsular dissection when
there is confidence in the tumor identity before the procedure.
Tumors in locations that are difficult or contraindicated to
attempt full excision may be treated with a debulking pro-
cedure, with the consent of the patient to obtain postoperative
serial imaging for observation of tumor progression. Debulk-
ing procedures should be used as sparingly as possible for
plexiform neurofibromas with full attempt at complete exci-
sion, as these tumors often recur quickly with partial removal.
The treatment for plexiform neurofibroma is dependent on
the age of the patient, location of the tumor, and preferences
of the patient with no firm guidelines currently established.
For some patients, plexiform neurofibroma may never cause
visual deficits. In children, amblyopia secondary to orbital
plexiform neurofibroma in the absence of other optic pathway
tumors may be as high as 62%.6 Therefore, if the child is in the
critical stage of development, regular assessment for early
development of amblyopia is needed including testing for
stereopsis, color vision and visual acuity. Recently, 3D MRI has
been used to measure the volume of plexiform neurofibromas
of the orbit, associating a tumor volume of 10 cc to the
development of amblyopia, with a majority of cases arising
from anisometropia or ptosis.6 In children at risk of ambly-
opia, debulking procedures may be performed to maintain the
tumor size, improve cosmesis for facial hypertrophy, and treat
ptosis. Orbitotemporal plexiform neurofibroma should be
surgically treated through a multidisciplinary approach and
may include tumor reduction, tumor resection, grafting the
sphenoid wing to correct bony defects, eyelid surgery to cor-
rect ptosis, or in the setting of a painful, proptotic eye with
decreased visual acuity, enucleation.8
MPNSTs often present aggressively and with local,
regional, or even distant metastasis. Unfortunately, given the
aggressive nature of the tumor, excision with tumor-free
boarders is often impossible,48 with aggressive recurrence33
or metastasis with extension via the nerve sheath often
through bony fissures and into the intracranial vault.59 For
cases involving a well-circumscribed malignant lobe attached
to circumscribed lobe of benign tumor, outcomes may be
better with a simple intact removal as these tumors will likely
be found to harbor lower grade malignancy. In more advanced
cases with boney invasion, exenteration with adjacent bone
removal is recommended.43,62 Unfortunately, MPNSTs have
poor sensitivity to chemotherapy.22 Palliative measures may
include surgical debulking; however, the patient should be
warned before surgical debulking of likely rapid, aggressive
recurrence.
7.2. Radiation therapy
The role of radiation therapy in PNSTs is evolving. Early doc-
umentations of radiation therapy for orbital tumors reported
high incidences of optic neuropathy attributed to direct injury
or neurovascular damage.5,83 Retrospective studies have
suggested a perceivable increased risk of optic neuropathy
with radiation doses above 8e12 Gray.83,153 Fractionated
radiotherapy has been used to reduce individual doses of ra-
diation; however, despite advances in this modality, the optic
nerve is often exposed to dosages equal to that treating the
tumorddosages previously demonstrated to cause optic
neuropathydwhich has limited the use of this modality.61
Despite the known radiosensitivity of the optic apparatus,
however, the difficulty in surgically managing apical tumors
often strongly adherent to nearby structures in a compact
area with critical neurovascular structures has warranted
ongoing nonsurgical alternative therapies including targeted
radiation applications.Stereotactic radiotherapy using
gamma knife surgery in orbital tumors has been increasingly
reported, with encouraging outcomes in treating benign
schwannomas and neurofibromas. One large study demon-
strated tumor size reduction or stability in 16/23 patients with
schwannomas and 3/3 with neurofibromas at a median
follow-up of 34.5 months.153
Single-session gamma knife surgery in the treatment of
orbital tumors near the apex has been deemed unsafe owing to
frequently associated posttreatment optic neuropathy
 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
manifesting as visual field deficits and vision acuity deteriora-
tion.54,68 Multisession gamma knife surgery for orbital tumors
offers decreased morbidity from radiation exposure with pre-
liminary outcomes demonstrating tumor control, with vision
improved or unchanged in 46/49 patients.2 Multisession
gamma knife surgery has been used in orbital schwannomas,
with 1 study reporting tumor control (reduced or unchanged
tumor size) in 6/7 patients with vision deterioration seen in 2/
7.69 Another smaller study included one patient with known
partially resected schwannoma and another with presumed
schwannoma. Both underwent multisession gamma knife
surgery with 70%e87% tumor volume reduction and improve-
ment in visual acuity with no adverse side effects.44
Multisession gamma knife surgery may be an appropriate
modality in small, unresectable, inaccessible, or postsurgical
benign schwannomas and localized neurofibromas. The in-
dications for radiation therapy as an alternative to extraction,
however, are not well established, and practices vary across
specialties.
Plexiform neurofibromas recur frequently following
extraction resulting in multiple surgeries and considerable
morbidity; thus, noninvasive strategies may have greater
impact in this population. Low-frequency radiotherapy has
been used in young patients with these tumors in effort to
decrease the natural progression of these tumors8; however,
in these studies and other similar studies analyzing adults,
results are mixed.24
The role of radiotherapy and chemotherapy in MPNSTs is
controversial.59,90 Some authors describe tumor radio-
resistance,48 and others have reported improvement in local
control of extraorbital MPNSTs with adjuvant external beam
radiation treatment with doses of
60 Gy.151 This dose is
clearly associated with optic neuropathy in the eye and may
best be used as a palliative measure with full disclosure of
limited efficacy and possible visual harm.
7.3. Decompression
In patients with documented serial imaging suggesting a sol-
itary neurofibroma or schwannoma affecting the apex or
where resection is not feasible, orbital decompression alone
may be beneficial. Two groups have demonstrated orbital
decompression efficacy in alleviation of optic nerve
compression symptoms, while avoiding risks of excision in
well-documented, slow-growing tumors affecting the
apex.4,73 Clearly, this strategy is most appropriate for cases
with a high degree of suspicion for benign, slow-growing tu-
mors. Additional criteria for the consideration of decompres-
sion may include intact vision, rapidly developing clinical
decline, no history of systemic malignancy, advanced patient
age, and patient agreement to serial imaging following
decompression. Orbital decompression is not without adverse
effects, notably diplopia, with hypoglobus, enophthalmos,
and cerebrospinal fluid leak more rare adverse events.
8. Prognosis
Rose and colleagues described postoperative results for pa-
tients with benign neurofibroma and benign schwannoma
53
and reported nonworsened visual acuity in 19/22 and 22/25,
respectively, with approximately 45% in both groups demon-
strating restricted eye movement postoperatively that
improved considerably with time.119 Chronic ptosis and fixed
mydriasis were found to be postoperative complications in a
minority of patients.119 Rose and colleagues also reported
incomplete excision in 13/25 cases of solitary neurofibroma,
secondary to difficult to reach tumors, with no patients found
to have recurrence at a mean time of 6.8 years following
excision.119 On the contrary, diffuse and plexiform neurofi-
bromas associated with NF-1 have high recurrence rates
following incomplete excision, and incomplete excision
should be considered a palliative treatment in these sce-
narios.42,62 When complete excision is achieved, there is a
good prognosis for solitary benign schwannomas.63 Few cases
of schwannoma recurrence have been described in the
absence of systemic neurofibromatosis after complete exci-
sion.23,77,116,141 Kron and colleagues reported on 2 such oc-
currences with 1 man developing 2 distinct tumors 6 years
after complete excision, and 1 child with multiple plexiform
schwannoma recurrences following incomplete extraction.77
Given the 2 simultaneous tumors, the former patient was
considered to have schwannomatosis.77
Long-term follow-up is needed to monitor for recurrence
and malignant transformation. This is rare in isolated tumors,
but may be higher in isolated neurofibroma than isolated
schwannomas, particularly in patients with NF-1.
MPNSTs have been shown to recur, particularly with
increased odds of metastasis at recurrence.59 While it is pru-
dent to consider syndromic association with any PNST
recurrence, it is particularly relevant to recurrence of MPNSTs,
given the increased risk of NF-1 with MPNSTs. Unfortunately,
in the largest reported case series of orbital MPNSTs, 9/13
patients died within 5 years of diagnosis.90
9. Conclusion
PNSTs are a rare group of neoplasms to affect the orbit with
substantial variability in tumor activity predicated on histo-
logical, genetic or syndromic, and positional factors. Consid-
eration of these tumor types in the differential diagnosis for
orbital lesions will assist in early detection and appropriate
counseling. Advances in CT and MRI imaging may offer
improved ability to diagnose these lesions and guide treat-
ment. Surgical excision remains the gold standard for defini-
tive therapy; however, new modalities including radiation
therapy demonstrate promise. Further studies to assist in
noninvasive diagnostics are needed, as are randomized
controlled trials comparing modern treatment modalities.
10. Method of literature search
This review was prepared using Medline with PubMed from
1900 until August 2015 using the following Medline subject
headings: orbit, orbital disease, or orbital neoplasms in com-
bination with neurilemmoma, neurofibroma, neurofibroma-
tosis, neurofibromatosis 1, or neurofibromatosis 2. The search
included non-English language articles. Abstracts were
54 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
reviewed for relevance and excluded for the following criteria:
invasion or extension of PNSTs from other locations, orbital
symptoms or diseases from PNSTs located outside the orbit,
mention of PNSTs as 1 case in a group of cases studying a
different disease, report on unrelated diseases with a patient
history of PNSTs elsewhere, ocular PNSTs, surgical procedures
unless specifically related to PNSTs removal, PubMed search
results without abstracts, articles not pertaining to humans.
Additionally, selected articles cited in the reference list of
other articles were reviewed.
11. Disclosures
The authors have no financial interests and nothing to
disclose.
references
1. Abe T, Kawamura N, Homma H, et al. MRI of orbital
schwannomas. Neuroradiology. 2000;42:466e8
2. Adler JR, Gibbs IC, Puataweepong P, Chang SD. Visual field
preservation after multisession cyberknife radiosurgery for
perioptic lesions. Neurosurgery. 2008;62(Suppl 2):733e43
3. Albert DM, Miller JW, Azar DT. Albert & Jakobiec’s principles
and practice of ophthalmology Philadelphia, Saunders/
Elsevier; 3rd ed., 2008, p 1, online resource (4 v).
4. Almond MC, Cheng AG, Schiedler V, et al. Decompression of
the orbital apex: an alternate approach to surgical excision
for radiographically benign orbital apex tumors. Arch
Otolaryngol Head Neck Surg. 2009;135:1015e8
5. Anegawa S, Hayashi T, Torigoe R, et al. [Orbital neurinoma
presenting orbital apex syndrome]. No Shinkei Geka.
1997;25:473e7
6. Avery RA, Dombi E, Hutcheson KA, et al. Visual outcomes in
children with neurofibromatosis type 1 and orbitotemporal
plexiform neurofibromas. Am J Ophthalmol.
2013;155:1089e94.e1
7. Banerjee S, Crouse NR, Emnett RJ, et al. Neurofibromatosis-1
regulates mTOR-mediated astrocyte growth and glioma
formation in a TSC/Rheb-independent manner. Proc Natl
Acad Sci U S A. 2011;108:15996e6001
8. Baujat B, Krastinova-Lolov D, Blumen M, et al.
Radiofrequency in the treatment of craniofacial plexiform
neurofibromatosis: a pilot study. Plast Reconstr Surg.
2006;117:1261e8
9. Bechtold D, Hove HD, Prause JU, et al. Plexiform
neurofibroma of the eye region occurring in patients without
neurofibromatosis type 1. Ophthal Plast Reconstr Surg.
2012;28:413e5
10. Beert E, Brems H, Renard M, et al. Biallelic inactivation of NF1
in a sporadic plexiform neurofibroma. Genes Chromosomes
Cancer. 2012;51:852e7
11. Begnami MD, Palau M, Rushing EJ, et al. Evaluation of NF2
gene deletion in sporadic schwannomas, meningiomas, and
ependymomas by chromogenic in situ hybridization. Hum
Pathol. 2007;38:1345e50
12. Berner JM, Sørlie T, Mertens F, et al. Chromosome band 9p21
is frequently altered in malignant peripheral nerve sheath
tumors: studies of CDKN2A and other genes of the pRB
pathway. Genes Chromosomes Cancer. 1999;26:151e60
13. Birkholz ES, Lee AG, Nerad JA, et al. A pregnant pause. Surv
Ophthalmol. 2010;55:162e8
14. Bosch MM, Wichmann WW, Boltshauser E, Landau K. Optic
nerve sheath meningiomas in patients with
neurofibromatosis type 2. Arch Ophthalmol.
2006;124:379e85
15. Bottillo I, Ahlquist T, Brekke H, et al. Germline and somatic
NF1 mutations in sporadic and NF1-associated malignant
peripheral nerve sheath tumours. J Pathol. 2009;217:693e701
16. Boyd C, Smith MJ, Kluwe L, et al. Alterations in the SMARCB1
(INI1) tumor suppressor gene in familial schwannomatosis.
Clin Genet. 2008;74:358e66
17. Briscoe D, Mahmood S, O’Donovan DG, et al. Malignant
peripheral nerve sheath tumor in the orbit of a child with
acute proptosis. Arch Ophthalmol. 2002;120:653e5
18. Byrne BM, van Heuven WA, Lawton AW. Echographic
characteristics of benign orbital schwannomas
(neurilemomas). Am J Ophthalmol. 1988;106:194e8
19. Cantore G, Ciappetta P, Raco A, Lunardi P. Orbital
schwannomas: report of nine cases and review of the
literature. Neurosurgery. 1986;19:583e8
20. Capps DH, Brodsky MC, Rice CD, et al. Orbital intramuscular
schwannoma. Am J Ophthalmol. 1990;110:535e9
21. Chang BY, Moriarty P, Cunniffe G, et al. Accelerated growth
of a primary orbital schwannoma during pregnancy. Eye
(Lond). 2003;17:839e41
22. Cheng SF, Chen YI, Chang CY, et al. Malignant peripheral
nerve sheath tumor of the orbit: malignant transformation
from neurofibroma without neurofibromatosis. Ophthal
Plast Reconstr Surg. 2008;24:413e5
23. Chisholm IA, Polyzoidis K. Recurrence of benign orbital
neurilemmoma (schwannoma) after 22 years. Can J
Ophthalmol. 1982;17:271e3
24. Chopra R, Morris CG, Friedman WA, Mendenhall WM.
Radiotherapy and radiosurgery for benign neurofibromas.
Am J Clin Oncol. 2005;28:317e20
25. Clarençon F, Jafari A, Lefevre M, et al. Infraorbital nerve
schwannoma. J Neuroradiol. 2009;36:301e3
26. Cleven AH, Sannaa GA, Briaire-de Bruijn I, et al. Loss of
H3K27 tri-methylation is a diagnostic marker for
malignant peripheral nerve sheath tumors and an
indicator for an inferior survival. Mod Pathol.
2016;29:582e90
27. Cockerham KP, Cockerham GC, Stutzman R, et al. The
clinical spectrum of schwannomas presenting with visual
dysfunction: a clinicopathologic study of three cases. Surv
Ophthalmol. 1999;44:226e34
28. Cunliffe IA, Moffat DA, Hardy DG, Moore AT. Bilateral optic
nerve sheath meningiomas in a patient with
neurofibromatosis type 2. Br J Ophthalmol. 1992;76:310e2
29. Damato B, Damato EM, Konstantinidis L, et al. Choroidal
schwannoma: a case series of five patients. Br J Ophthalmol.
2014;98:1096e100
30. DeClue JE, Cohen BD, Lowy DR. Identification and
characterization of the neurofibromatosis type 1 protein
product. Proc Natl Acad Sci U S A. 1991;88:9914e8
31. De Potter P, Shields CL, Shields JA, et al. The CT and MRI
features of an unusual case of isolated orbital neurofibroma.
Ophthal Plast Reconstr Surg. 1992;8:221e7
32. Dervin JE, Beaconsfield M, Wright JE, Moseley IF. CT findings
in orbital tumours of nerve sheath origin. Clin Radiol.
1989;40:475e9
33. de Silva DJ, Tay E, Rose GE. Schwannomas of the lacrimal
gland fossa. Orbit. 2009;28:433e5
34. Dutton JJ, Tawfik HA, DeBacker CM, et al. Multiple
recurrences in malignant peripheral nerve sheath tumor of
the orbit: a case report and a review of the literature.
Ophthal Plast Reconstr Surg. 2001;17:293e9
35. Endo M, Kobayashi C, Setsu N, et al. Prognostic significance
of p14ARF, p15INK4b, and p16INK4a inactivation in
 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
malignant peripheral nerve sheath tumors. Clin Cancer Res.
2011;17:3771e82
36. Eviatar JA, Hornblass A, Herschorn B, Jakobiec FA. Malignant
peripheral nerve sheath tumor of the orbit in a 15-month-
old child. Nine-year survival after local excision.
Ophthalmology. 1992;99:1595e9
37. Farris SR, Grove AS. Orbital and eyelid manifestations of
neurofibromatosis: a clinical study and literature review.
Ophthal Plast Reconstr Surg. 1996;12:245e59
38. Ferner RE. Neurofibromatosis 1 and neurofibromatosis 2: a
twenty first century perspective. Lancet Neurol.
2007;6:340e51
39. Ferner RE, O’Doherty MJ. Neurofibroma and schwannoma.
Curr Opin Neurol. 2002;15:679e84
40. Fezza JP, Wolfley DE, Flynn SD. Malignant peripheral nerve
sheath tumor of the orbit in a newborn: a case report and
review. J Pediatr Ophthalmol Strabismus. 1997;34:128e31
41. Freedman SF, Elner VM, Donev I, et al. Intraocular
neurilemmoma arising from the posterior ciliary nerve in
neurofibromatosis. Pathologic findings. Ophthalmology.
1988;95:1559e64
42. Friedrich RE, Stelljes C, Hagel C, et al. Dysplasia of the orbit
and adjacent bone associated with plexiform neurofibroma
and ocular disease in 42 NF-1 patients. Anticancer Res.
2010;30:1751e64
43. Garrity JA, Henderson JW, Cameron JD. Henderson’s orbital
tumors. Philadelphia, Lippincott Williams & Wilkins; 2007
44. Goh AS, Kim YD, Woo KI, Lee JI. Benign orbital apex tumors
treated with multisession gamma knife radiosurgery.
Ophthalmology. 2013;120:635e41
45. Goldberg RA, Rootman DB, Nassiri N, et al. Orbital tumors
excision without bony marginotomy under local and general
anesthesia. J Ophthalmol. 2014;2014:424852
46. Goldblum JR, Beals TF, Weiss SW. Neuroblastoma-like
neurilemoma. Am J Surg Pathol. 1994;18:266e73
47. Gonzalez LF, Lekovic GP, Eschbacher J, et al. A true
malignant schwannoma of the eighth cranial nerve: case
report. Neurosurgery. 2007;61:E421e2, discussion E2.
48. González-Orús Álvarez-Morujo R, Garcı́a Leal R, Lasso
Vázquez JM, Scola Yurrita B. [Malignant peripheral nerve
sheath tumour of the infra-orbital nerve]. Neurocirugia
(Astur). 2014;25:240e3
49. Graham CM, McCartney AC, Buckley RJ. Intrascleral
neurilemmoma. Br J Ophthalmol. 1989;73:378e81
50. Grinberg MA, Levy NS. Malignant neurilemoma of the
supraorbital nerve. Am J Ophthalmol. 1974;78:489e92
51. Hadfield KD, Smith MJ, Urquhart JE, et al. Rates of loss of
heterozygosity and mitotic recombination in NF2
schwannomas, sporadic vestibular schwannomas and
schwannomatosis schwannomas. Oncogene.
2010;29:6216e21
52. Han JC, Kim YD, Suh YL, et al. Fibroblastic low-grade
malignant peripheral nerve sheath tumor in the orbit.
Ophthal Plast Reconstr Surg. 2012;28:e97e8
53. Harrisingh MC, Lloyd AC. Ras/Raf/ERK signalling and NF1.
Cell Cycle. 2004;3:1255e8
54. Hayashi M, Chernov M, Tamura N, et al. Gamma Knife
surgery for abducent nerve schwannoma. Report of 4 cases.
J Neurosurg. 2010;113(Suppl):136e43
55. Hirbe AC, Dahiya S, Miller CA, et al. Whole exome
sequencing reveals the order of genetic changes during
malignant transformation and metastasis in a single patient
with NF1-plexiform neurofibroma. Clin Cancer Res.
2015;21:4201e11
56. Hirbe AC, Pekmezci M, Dahiya S, et al. BRAFV600E mutation
in sporadic and neurofibromatosis type 1-related malignant
peripheral nerve sheath tumors. Neuro Oncol. 2014;16:466e7
55
57. Huang HY, Park N, Erlandson RA, Antonescu CR.
Immunohistochemical and ultrastructural comparative
study of external lamina structure in 31 cases of cellular,
classical, and melanotic schwannomas. Appl
Immunohistochem Mol Morphol. 2004;12:50e8
58. Hulsebos TJ, Plomp AS, Wolterman RA, et al. Germline
mutation of INI1/SMARCB1 in familial schwannomatosis.
Am J Hum Genet. 2007;80:805e10
59. Jakobiec FA, Font RL, Zimmerman LE. Malignant peripheral
nerve sheath tumors of the orbit: a clinicopathologic study
of eight cases. Trans Am Ophthalmol Soc. 1985;83:332e66
60. Jensen OA, Bretlau P. Melanotic schwannoma of the orbit.
Immunohistochemical and ultrastructural study of a case
and survey of the literature. APMIS. 1990;98:713e23
61. Jo KI, Im YS, Kong DS, et al. Multisession Gamma Knife
surgery for benign orbital tumors. J Neurosurg.
2012;117(Suppl):102e7
62. Karcioglu ZA. Orbital tumors: diagnosis and treatment. New
York, Springer; 2004
63. Kashyap S, Pushker N, Meel R, et al. Orbital schwannoma
with cystic degeneration. Clin Experiment Ophthalmol.
2009;37:293e8
64. Kaul A, Toonen JA, Cimino PJ, et al. Akt- or MEK-mediated
mTOR inhibition suppresses Nf1 optic glioma growth. Neuro
Oncol. 2015;17:843e53
65. Khan SN, Sepahdari AR. Orbital masses: CT and MRI of
common vascular lesions, benign tumors, and malignancies.
Saudi J Ophthalmol. 2012;26:373e83
66. Khwarg SI, Lucarelli MJ, Lemke BN, et al. Ancient
schwannoma of the orbit. Arch Ophthalmol. 1999;117:262e4
67. Kim DS, Choi JU, Yang KH, Jung JM. Optic sheath
schwannomas: report of two cases. Childs Nerv Syst.
2002;18:684e9
68. Kim MS, Park K, Kim JH, et al. Gamma knife radiosurgery for
orbital tumors. Clin Neurol Neurosurg. 2008;110:1003e7
69. Kim BS, Woo KI, Kim YD, Lee JI. Multisession Gamma Knife
Radiosurgery for Orbital Apex Tumors. World Neurosurg.
2015;84:1005e13
70. Kiratli H, Yildiz S, Soylemezog lu F. Neurofibromatosis type 2:
optic nerve sheath meningioma in one orbit, intramuscular
schwannoma in the other. Orbit. 2008;27:451e4
71. Kleihues P, Louis DN, Scheithauer BW, et al. The WHO
classification of tumors of the nervous system.
J Neuropathol Exp Neurol. 2002;61:215e25, discussion 26e29.
72. Kleihues P, Sobin LH. World Health Organization
classification of tumors. Cancer. 2000;88:2887
73. Kloek CE, Bilyk JR, Pribitkin EA, Rubin PA. Orbital
decompression as an alternative management strategy for
patients with benign tumors located at the orbital apex.
Ophthalmology. 2006;113:1214e9
74. Kottler UB, Conway RM, Schlötzer-Schrehardt U,
Holbach LM. Isolated neurofibroma of the orbit with
extensive myxoid changes: a clinicopathologic study
including MRI and electron microscopic findings. Orbit.
2004;23:59e64
75. Kourea HP, Orlow I, Scheithauer BW, et al. Deletions of the
INK4A gene occur in malignant peripheral nerve sheath
tumors but not in neurofibromas. Am J Pathol.
1999;155:1855e60
76. Krohel GB, Rosenberg PN, Wright JE, Smith RS. Localized
orbital neurofibromas. Am J Ophthalmol. 1985;100:458e64
77. Kron M, Bohnsack BL, Archer SM, et al. Recurrent orbital
schwannomas: clinical course and histopathologic
correlation. BMC Ophthalmol. 2012;12:44
78. Küchle M, Holbach L, Schlötzer-Schrehardt U, Naumann GO.
Schwannoma of the ciliary body treated by block excision. Br
J Ophthalmol. 1994;78:397e400
56 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
79. Kukreja M, Gupta R, Julka A, Sharma MC. Neuroblastoma-
like schwannoma as a rare cause of proptosis: a case report.
Can J Ophthalmol. 2007;42:624e5
80. Kurtkaya-Yapicier O, Scheithauer B, Woodruff JM. The
pathobiologic spectrum of Schwannomas. Histol
Histopathol. 2003;18:925e34
81. Lam DS, NG JS, To KF, et al. Cystic schwannoma of the orbit.
Eye (Lond). 1997;11(Pt 6):798e800
82. Laskin WB, Weiss SW, Bratthauer GL. Epithelioid variant of
malignant peripheral nerve sheath tumor (malignant
epithelioid schwannoma). Am J Surg Pathol. 1991;15:1136e45
83. Leber KA, Berglöff J, Langmann G, et al. Radiation sensitivity
of visual and oculomotor pathways. Stereotact Funct
Neurosurg. 1995;64(Suppl 1):233e8
84. Lee LR, Gigantelli JW, Kincaid MC. Localized neurofibroma of
the orbit: a radiographic and histopathologic study. Ophthal
Plast Reconstr Surg. 2000;16:241e6
85. Lee W, Teckie S, Wiesner T, et al. PRC2 is recurrently
inactivated through EED or SUZ12 loss in malignant
peripheral nerve sheath tumors. Nat Genet. 2014;46:1227e32
86. Leruez S, Gohier P, Menei P, Milea D. [Optic nerve
schwannoma]. J Fr Ophtalmol. 2013;36:e49e53
87. Lewis RA, Riccardi VM. Von Recklinghausen
neurofibromatosis. Incidence of iris hamartomata.
Ophthalmology. 1981;88:348e54
88. Lindsay RA, Gupta D, Keene CD, et al. Malignant peripheral
nerve sheath tumor of the lower eyelid: case presentation
and literature review. Ophthal Plast Reconstr Surg
2015;http://dx.doi.org/10.1097/IOP.0000000000000496
89. Lu-Emerson C, Plotkin SR. The neurofibromatoses. Part 2:
NF2 and schwannomatosis. Rev Neurol Dis. 2009;6:E81e6
90. Lyons CJ, McNab AA, Garner A, Wright JE. Orbital malignant
peripheral nerve sheath tumours. Br J Ophthalmol.
1989;73:731e8
91. Ma’luf RN, Noureddin BN, Ghazi NG, et al. Bilateral, localized
orbital neurofibromas and Charcot-Marie-Tooth disease.
Arch Ophthalmol. 2005;123:1443e5
92. MacCollin M, Chiocca EA, Evans DG, et al. Diagnostic criteria
for schwannomatosis. Neurology. 2005;64:1838e45
93. Mantripragada KK, Dı́az de Ståhl T, Patridge C, et al.
Genome-wide high-resolution analysis of DNA copy number
alterations in NF1-associated malignant peripheral nerve
sheath tumors using 32K BAC array. Genes Chromosomes
Cancer. 2009;48:897e907
94. McDaniel AS, Allen JD, Park SJ, et al. Pak1 regulates multiple
c-Kit mediated Ras-MAPK gain-in-function phenotypes in
Nf1þ/ mast cells. Blood. 2008;112:4646e54
95. McLaughlin ME, Pepin SM, Maccollin M, et al. Ocular
pathologic findings of neurofibromatosis type 2. Arch
Ophthalmol. 2007;125:389e94
96. McMenamin ME, Fletcher CD. Expanding the spectrum of
malignant change in schwannomas: epithelioid malignant
change, epithelioid malignant peripheral nerve sheath
tumor, and epithelioid angiosarcoma: a study of 17 cases.
Am J Surg Pathol. 2001;25:13e25
97. Meyer DR, Wobig JL. Bilateral localized orbital
neurofibromas. Ophthalmology. 1992;99:1313e7
98. Moloney G, Brewer J, O’Donnell BA. ‘Ancient’
schwannoma of the orbit. Clin Experiment Ophthalmol.
2004;32:637e8
99. Morgenstern KE, Vadysirisack DD, Zhang Z, et al. Expression
of sodium iodide symporter in the lacrimal drainage system:
implication for the mechanism underlying nasolacrimal
duct obstruction in I(131)-treated patients. Ophthal Plast
Reconstr Surg. 2005;21:337e44
100. Morton AD, Elner VM, Frueh B. Recurrent orbital malignant
peripheral nerve sheath tumor 18 years after initial
resection. Ophthal Plast Reconstr Surg. 1997;13:239e43
101. Mulkens TH, Parizel PM, De Schepper AM, et al. MRI of
acoustic schwannoma: a retrospective study of 89 tumours.
Rofo. 1993;158:362e7
102. Nielsen GP, Stemmer-Rachamimov AO, Ino Y, et al.
Malignant transformation of neurofibromas in
neurofibromatosis 1 is associated with CDKN2A/p16
inactivation. Am J Pathol. 1999;155:1879e84
103. Nonaka D, Chiriboga L, Rubin BP. Sox10: a pan-schwannian
and melanocytic marker. Am J Surg Pathol. 2008;32:1291e8
104. Oh JE, Ohta T, Satomi K, et al. Alterations in the NF2/LATS1/
LATS2/YAP pathway in schwannomas. J Neuropathol Exp
Neurol. 2015;74:952e9
105. Park WC, White WA, Woog JJ, et al. The role of high-
resolution computed tomography and magnetic resonance
imaging in the evaluation of isolated orbital neurofibromas.
Am J Ophthalmol. 2006;142:456e63
106. Pekmezci M, Reuss DE, Hirbe AC, et al. Morphologic and
immunohistochemical features of malignant peripheral
nerve sheath tumors and cellular schwannomas. Mod
Pathol. 2015;28:187e200
107. Piotrowski A, Xie J, Liu YF, et al. Germline loss-of-function
mutations in LZTR1 predispose to an inherited disorder of
multiple schwannomas. Nat Genet. 2014;46:182e7
108. Prescott DK, Racz MM, Ng JD. Epithelioid malignant
peripheral nerve sheath tumor in the infraorbital nerve.
Ophthal Plast Reconstr Surg. 2006;22:150e1
109. Prieto-Granada CN, Wiesner T, Messina JL, et al. Loss of
H3K27me3 expression is a highly sensitive marker for
sporadic and radiation-induced MPNST. Am J Surg Pathol.
2016;40:479e89
110. Purcell SM, Dixon SL. Schwannomatosis. An unusual variant
of neurofibromatosis or a distinct clinical entity? Arch
Dermatol. 1989;125:390e3
111. Rácz M, Szabó J. [Two cases of neurilemmoma of the
anterior uvea (author’s transl)]. Klin Monbl Augenheilkd.
1973;163:605e9
112. Ramey WL, Arnold SJ, Chiu A, Lemole M. A rare case of optic
nerve schwannoma: case report and review of the literature.
Cureus. 2015;7:e265
113. Rasmussen SA, Overman J, Thomson SA, et al. Chromosome
17 loss-of-heterozygosity studies in benign and malignant
tumors in neurofibromatosis type 1. Genes Chromosomes
Cancer. 2000;28:425e31
114. Ratner N, Miller SJ. A RASopathy gene commonly mutated in
cancer: the neurofibromatosis type 1 tumour suppressor.
Nat Rev Cancer. 2015;15:290e301
115. Reuss DE, Habel A, Hagenlocher C, et al. Neurofibromin
specific antibody differentiates malignant peripheral nerve
sheath tumors (MPNST) from other spindle cell neoplasms.
Acta Neuropathol. 2014;127:565e72
116. Röhrich M, Koelsche C, Schrimpf D, et al. Methylation-based
classification of benign and malignant peripheral nerve
sheath tumors. Acta Neuropathol. 2016;131:877e87
117. Rootman J. Diseases of the orbit: a multidisciplinary
approach. Philadelphia, Lippincott Williams & Wilkins; 2003
118. Rootman J, Goldberg C, Robertson W. Primary orbital
schwannomas. Br J Ophthalmol. 1982;66:194e204
119. Rose GE, Wright JE. Isolated peripheral nerve sheath
tumours of the orbit. Eye (Lond). 1991;5(Pt 6):668e73
120. Rouleau GA, Merel P, Lutchman M, et al. Alteration in a new
gene encoding a putative membrane-organizing protein
causes neuro-fibromatosis type 2. Nature. 1993;363:515e21
121. Rousseau G, Noguchi T, Bourdon V, et al. SMARCB1/INI1
germline mutations contribute to 10% of sporadic
schwannomatosis. BMC Neurol. 2011;11:9
122. Sales-Sanz M, Sanz-Lopez A, Romero JA. Bilateral
simultaneous ancient schwannomas of the orbit. Ophthal
Plast Reconstr Surg. 2007;23:68e9
 s u r v e y o f o p h t h a l m o l o g y 6 2 ( 2 0 1 7 ) 4 3 e5 7
123. Schatz H. Benign orbital neurilemoma. Sarcomatous
transformation in von Recklinghausen’s disease. Arch
Ophthalmol. 1971;86:268e73
124. Scheithauer BW, Erdogan S, Rodriguez FJ, et al. Malignant
peripheral nerve sheath tumors of cranial nerves and
intracranial contents: a clinicopathologic study of 17 cases.
Am J Surg Pathol. 2009;33:325e38
125. Scheller C, Rachinger JC, Prell J, et al. Intraorbital
oculomotor nerve schwannoma affecting only the
parasympathetic fibers. J Neurol Surg A Cent Eur
Neurosurg. 2013;74:120e3
126. Schick U, Bleyen J, Hassler W. Treatment of orbital
schwannomas and neurofibromas. Br J Neurosurg.
2003;17:541e5
127. Shen WC, Yang DY, Ho WL, et al. Neurilemmoma of the
oculomotor nerve presenting as an orbital mass: MR
findings. AJNR Am J Neuroradiol. 1993;14:1253e4
128. Shields JA. Diagnosis and management of orbital tumors.
Philadelphia, Saunders; 1989
129. Shields JA, Hamada A, Shields CL, et al. Ciliochoroidal nerve
sheath tumor simulating a malignant melanoma. Retina.
1997;17:459e60
130. Shields JA, Kapustiak J, Arbizo V, et al. Orbital neurilemoma
with extension through the superior orbital fissure. Arch
Ophthalmol. 1986;104:871e3
131. Shields JA, Sanborn GE, Kurz GH, Augsburger JJ. Benign
peripheral nerve tumor of the choroid: a clinicopathologic
correlation and review of the literature. Ophthalmology.
1981;88:1322e9
132. Shields JA, Shields CL, Lieb WE, Eagle RC. Multiple orbital
neurofibromas unassociated with von Recklinghausen’s
disease. Arch Ophthalmol. 1990;108:80e3
133. Simpson RK, Harper RL, Kirkpatrick JB, Cooper B.
Schwannoma of the optic sheath. J Clin Neuroophthalmol.
1987;7:219e22
134. Singh AD. Clinical ophthalmic oncology. Edinburgh, Elsevier
Saunders; 2007, p 1, online resource (xv, 611 p).
135. Singh M, Singh U, Zadeng Z, et al. Clinico-radiological
spectrum and management of orbital schwannomas: A
tertiary care institute study. Orbit 2013;http://dx.doi.org/10.
3109/01676830.2013.788661
136. Singh U, Sukhija J, Raj S, et al. Calcification in schwannoma
of the lacrimal gland region. Eye (Lond). 2004;18:218,
discussion 218e219.
137. Smith MJ, Isidor B, Beetz C, et al. Mutations in LZTR1 add to
the complex heterogeneity of schwannomatosis. Neurology.
2015;84:141e7
138. Spurlock G, Knight SJ, Thomas N, et al. Molecular evolution
of a neurofibroma to malignant peripheral nerve sheath
tumor (MPNST) in an NF1 patient: correlation between
histopathological, clinical and molecular findings. J Cancer
Res Clin Oncol. 2010;136:1869e80
139. Standal B. Neurinoma of the orbit. Acta Ophthalmol
(Copenh). 1950;28:49e70
140. Staser K, Yang FC, Clapp DW. Mast cells and the
neurofibroma microenvironment. Blood. 2010;116:157e64
57
141. Subramanian N, Rambhatia S, Mahesh L, et al. Cystic
schwannoma of the orbit-a case series. Orbit. 2005;24:125e9
142. Sugo N, Yokota K, Nemoto M, et al. Accelerated growth of an
orbital schwannoma during pregnancy. J Neuroophthalmol.
2007;27:45e7
143. Sun MT, Wu W, Yan W, et al. Endoscopic endonasal-assisted
resection of orbital schwannoma. Ophthal Plast Reconstr
Surg 2015;http://dx.doi.org/10.1097/IOP.0000000000000528
144. Tailor TD, Gupta D, Dalley RW, et al. Orbital neoplasms in
adults: clinical, radiologic, and pathologic review.
Radiographics. 2013;33:1739e58
145. Upadhyaya M, Spurlock G, Monem B, et al. Germline and
somatic NF1 gene mutations in plexiform neurofibromas.
Hum Mutat. 2008;29:E103e11
146. Verdijk RM, den Bakker MA, Dubbink HJ, et al. TP53 mutation
analysis of malignant peripheral nerve sheath tumors.
J Neuropathol Exp Neurol. 2010;69:16e26
147. Wang Y, Kim E, Wang X, et al. ERK inhibition rescues defects
in fate specification of Nf1-deficient neural progenitors and
brain abnormalities. Cell. 2012;150:816e30
148. Wang Y, Xiao LH. Orbital schwannomas: findings from
magnetic resonance imaging in 62 cases. Eye (Lond).
2008;22:1034e9
149. Weiss SW, Goldblum JR, Enzinger FM. Enzinger and Weiss’s
soft tissue tumors. Philadelphia, PA, Mosby Elsevier; 2008
150. Wippold FJ, Lubner M, Perrin RJ, et al. Neuropathology for the
neuroradiologist: Antoni A and Antoni B tissue patterns.
AJNR Am J Neuroradiol. 2007;28:1633e8
151. Wong WW, Hirose T, Scheithauer BW, et al. Malignant
peripheral nerve sheath tumor: analysis of treatment
outcome. Int J Radiat Oncol Biol Phys. 1998;42:351e60
152. Xian J, Zhang Z, Wang Z, et al. Evaluation of MR imaging
findings differentiating cavernous haemangiomas from
schwannomas in the orbit. Eur Radiol. 2010;20:2221e8
153. Xu D, Liu D, Zhang Z, et al. Gamma Knife surgery in the
management of orbital tumors. J Neurosurg.
2010;113(Suppl):34e8
154. Yan J, Li Y, Wu Z. Orbital neurofibroma presenting with a
negative Hounsfield unit on computerized tomography.
Orbit. 2006;25:239e41
155. Yang FC, Chen S, Clegg T, et al. Nf1þ/ mast cells induce
neurofibroma like phenotypes through secreted TGF-beta
signaling. Hum Mol Genet. 2006;15:2421e37
156. Yang FC, Ingram DA, Chen S, et al. Nf1-dependent tumors
require a microenvironment containing Nf1þ/ and c-kit-
dependent bone marrow. Cell. 2008;135:437e48
157. Yang FC, Staser K, Clapp DW. The plexiform neurofibroma
microenvironment. Cancer Microenviron. 2012;5:307e10
158. Yanoff M, Sassani JW. Ocular pathology. Edinburgh, Mosby/
Elsevier; 2009
159. You JY, Finger PT, Iacob C, et al. Intraocular schwannoma.
Surv Ophthalmol. 2013;58:77e85
160. Zhang Z, Shi J, Guo J, et al. Value of MR imaging in
differentiation between solitary fibrous tumor and
schwannoma in the orbit. AJNR Am J Neuroradiol.
2013;34:1067e71