152 Cancer ControlJuly 2011, Vol. 18, No. 3
Address correspondence to Marilyn M. Bui, MD, PhD, Department of
Anatomic Pathology, Moffi tt Cancer Center, 12902 Magnolia Drive,
Tampa, FL 33612. E-mail: Marilyn.Bui@moffi tt.org
No signifi cant relationship exists between the authors and the
companies/organizations whose products or services may be ref-
erenced in this article.
From the Departments of Pathology and Cell Biology (NDR, MMB) at
the University of South Florida College of Medicine, Tampa, Florida,
the Departments of Sarcoma (RJG, SA, MMB) and Anatomic Pathol-
ogy (MMB) at the H. Lee Moffi tt Cancer Center & Research Institute,
Tampa, Florida, and the Departments of Pathology/Microbiology,
Pediatrics and Orthopaedic Surgery at the University of Nebraska
Medical Center (JAB), Omaha, Nebraska.
Submitted June 14, 2010; accepted September 23, 2010.
Jeffrey Hessing. The West Bank. Oil on canvas, 18″ × 24″.
An accurate diagnosis of sarcoma
requires a correlation of clinical
and histological information with
A CD117 and CD34 Immunoreactive Sarcoma
Masquerading as a Gastrointestinal Stromal Tumor:
Diagnostic Pitfalls of Ancillary Studies in Sarcoma
Nicole D. Riddle, MD, Ricardo J. Gonzalez, MD, Julia A. Bridge, MD,
Scott Antonia, MD, and Marilyn M. Bui, MD, PhD
Background: The immunohistochemical hallmarks of gastrointestinal stromal tumors (GISTs) are positivity
for CD117 (c-kit) and CD34; however, CD117 is not specifi c for GISTs, and the list of CD117+ tumors/tissues is
increasing. Also, MDM2 is known to be amplifi ed in several types of mesenchymal tumors, including liposarcoma.
Methods: We report a spindle cell sarcoma arising in the mediastinum that morphologically and immunohis-
tochemically mimicked GIST to illustrate the potential diagnostic pitfalls of ancillary studies in sarcoma and
their appropriate use in conjunction with clinical content. Clinical information was obtained from electronic
medical databases. Cytological, histological, and ancillary studies were retrieved from the archives of the
Department of Anatomic Pathology at Moffi tt Cancer Center. Literature of the last 20 years was reviewed. The
role of biomarkers and their molecular testing in the prognosis and prediction of GIST is also discussed.
Results: A 75-year-old woman with a history of well-differentiated liposarcoma of the trunk/inguinal canal 5
years earlier developed a 5.5-cm heterogeneously enhancing mediastinal mass by computed tomography. Fine-needle
aspiration biopsy revealed spindle cells with moderate pleomorphism and immunohistochemically reactive to
CD117 and CD34 suggestive of GIST, but the clinical picture was unusual for GIST. Mutational analyses for KIT
and platelet-derived growth factor receptor alpha (PDGFRa) were negative; DOG1 was not immunoactive, and
this was believed to rule out GIST. An additional study of MDM2 by fl uorescent in situ hybridization was positive,
suggesting that this tumor was a dedifferentiated liposarcoma vs a spindle cell sarcoma not otherwise specifi ed.
Conclusions: CD117+/CD34+ sarcoma is not diagnostic for GIST. KIT and PDGFRa mutational analyses
are important in confi rming a diagnosis of GIST and predicting its response to imatinib therapy. MDM2+
sarcoma is not diagnostic for liposarcoma. Although MDM2 is almost always positive in well-differentiated
liposarcoma, which is useful in differentiating benign from atypical/well-differentiated lipomatous tumor, it
should not be used in differentiating liposarcoma from other sarcomas.
July 2011, Vol. 18, No. 3 Cancer Control 153
As in many areas of oncology, tissue diagnosis is es-
sential in planning the appropriate therapy in sarcoma
treatment. Ancillary studies are essential in arriving at
a defi nitive diagnosis and in subtyping sarcoma since
traditional hematoxylin and eosin (H&E) stain has limi-
tations.1,2 Ancillary studies include immunohistochem-
istry (IHC), electron microscopy, cytogenetics, and mo-
lecular analysis. IHC is currently the most commonly
used and well-established method that studies the im-
munoreactivity of archived tissue. IHC is specifi c in its
distribution in certain tissues that have characteristic
patterns such as cytoplasmic, nuclear, membranous, or a
combination thereof. The role of IHC in the workup of a
mesenchymal lesion is threefold: to elucidate the lineage
(eg, muscle, nerve, epithelial), to forecast the prognosis
(eg, p53 mutation and increased Ki-67 are associated
with poor prognosis), and to determine the response
to targeted therapy (eg, CD117 positivity for response
to imatinib therapy). To date, no IHC marker is available
to differentiate benign from malignant mesenchymal
tissue. Histopathology diagnosis by a pathologist using
traditional criteria such as cellularity, nuclear pleomor-
phism, mitosis, and necrosis remains the gold standard
in the diagnosis of a sarcoma. Electron microscopy
looks into the ultrastructure of tissues and requires a
specifi c fi xative and special processing. For the most
part, it is no longer used on a regular basis outside of
kidney disorders and has been replaced with IHC and
Cytogenetics requires fresh tissue and cell culture
to study the chromosomal structure of metaphased cells.
Certain structural and/or numerical abnormalities are
specifi c for some sarcomas. Most of the sarcomas have
complex but not specifi c chromosomal abnormalities.
The most commonly used molecular tests in sarcoma
include fl uorescent in situ hybridization (FISH) and mu-
tational analysis, although polymerase chain reaction
(PCR) and reverse transcription PCR (RT-PCR) are also
useful techniques for gene amplifi cation or translocation
detection. FISH can detect a specifi c translocation or
amplifi cation of interphased chromosomes on archived
or fresh tissue. Because these studies are probe-specifi c,
pathologists must be aware of the specifi c abnormality
to look for when ordering the study. Mutational analysis
studies mutation of genes using PCR and gene sequenc-
To illustrate the potential diagnostic pitfalls of ancil-
lary studies in sarcoma, we report a CD117+/CD34+/
MDM2+ sarcoma arising in the mediastinum that pre-
sented a diagnostic challenge. In addition, we review the
benefi ts and pitfalls of ancillary studies, focusing on IHC
and molecular analysis in the diagnosis of spindle cell
mesenchymal neoplasms (particularly GIST), as well as
the prognostic and predictive markers of GIST and the
role of MDM2 among mesenchymal neoplasms.
Materials and Methods
A 75-year-old woman with a history of well-differentiat-
ed, sclerosing variant liposarcoma of the trunk/inguinal
region 5 years prior was being followed for known 6-cm
low-grade residual disease in the inguinal canal and ex-
tending down to her right labia majora. Re-excision
with a right inguinal hernia repair 2 years after initial
resection was consistent with lipoma. She claimed that
the labial mass was unchanged since the time of her re-
section, which was concordant with cross-sectional im-
aging. Contrast-enhanced computed tomography (CT)
of the chest was obtained as part of a workup for gas-
troesophageal refl ux disease with associated dysphagia,
odynophagia, asthma, and chronic cough. CT identifi ed
a heterogeneously enhancing 5.5-cm mediastinal mass
(Fig 1A-B). Physical examination demonstrated no signs
of oral mucosal lesions, jugular venous distention, bruits,
abnormal breath sounds, or murmurs. Her inguinal ca-
nal lesion was soft and mobile, and it extended to the
right labia majora. An endoscopic, ultrasound-guided,
fi ne-needle aspiration biopsy of the mediastinal lesion
Fig 1A-B. — Computed tomography of the chest demonstrating a 5.5 × 4.6-cm heterogeneously enhancing mass involving the right paratracheal region.
(A) The lesion causes deviation of the left brachiocephalic vein (arrow) and (B) proximal superior vena cava (arrow).
154 Cancer ControlJuly 2011, Vol. 18, No. 3
Tissue Diagnosis and Ancillary Studies
Cytology smears revealed spindle cells with moderate
pleomorphism and myxoid background consistent with a
spindle cell mesenchymal tumor (Fig 2A-B). A cell block
was prepared from the rinse of the needle. H&E prepa-
ration of the cell block revealed spindle cells with rare
mitosis and no necrosis (Fig 2C). These spindle cells
were IHC positive for CD117 and CD34 (Fig 2D-E). IHC
stains for additional markers were all negative: pankera-
tin and epithelial membrane antigen (EMA) to exclude
carcinoma; S-100 to exclude neural-based lesions and
melanoma; desmin, smooth muscle actin (SMA), and
myoglobin to exclude muscle-derived tumors; and me-
sothelin and calretinin to exclude mesothelioma. The
tumor was also positive for vimentin. These H&E and IHC
fi ndings were consistent with gastrointestinal stromal
tumor (GIST). However, this clinical picture was unusual
for GIST because greater than 95% of GISTs arise in the
gastrointestinal tract, with approximately 60% occurring
in the stomach, 30% in the small intestine, and 8% in the
esophagus, colon, and rectum. GISTs can occur in extra-
intestinal locations (< 5%); however, no mediastinal GIST
has ever been reported.
Mutational analyses for KIT (CD117) and platelet-
derived growth factor receptor alpha (PDGFRa), both
on chromosome 4, were sent to ARUP Laboratories in Salt
Lake City, Utah. These two molecular marker mutations
are mutually exclusive and together account for greater
than 95% of GISTs. These tests are performed by utiliz-
ing melting curves wherein the mutant DNA has a lower
Fig 2A-B. — Diff-Quik and Papanicolaou stains of the cytology smear cor-
responding to the cell block (400×).
Fig 2C. — Hematoxylin-eosin stain of the cell block showing pleomorphic
spindled cells (600×).
Fig 2D. — CD117 immunostain showing strong, diffuse positivity within
the cytoplasm of the tumor cells.
Fig 2E. — CD34 showing moderate cytoplasmic staining within the tumor
July 2011, Vol. 18, No. 3 Cancer Control 155
melting point than the normal (wild-type) DNA. In the
event of an atypical melting curve, direct sequence analy-
sis of the questionable exons is performed to confi rm or
refute a mutation. KIT mutations account for approxi-
mately 85% of cases with a constitutionally active tyro-
sine kinase. The most common mutations are in exons 9
and 11, with exons 13 and 17 also being reported. Most
of the remainder of GISTs shows a PDGFRa mutation
involving exons 12, 14, or 18. Cases have been described
that are KIT and PDGFRa wild-type, some studies claim-
ing as many as 10% to 15%, and it is postulated that these
cases harbor an as-of-yet unidentifi ed mutation.3 The
exon involved is important because approximately 10%
of GISTs are resistant to imatinib (Gleevec) therapy, and
these cases have been associated particularly with KIT
exon 9 or PDGFRa exon 18. These apparently imatinib-
resistant cases may respond to doses at twice the usual
level or may achieve a better response to a second-line
agent, sunitinib (Sutent). However, in our cases, these
studies were uniformly negative (not shown). An IHC
study of DOG1 was also negative. An additional analysis
of MDM2 by FISH was performed at the University of
Nebraska Molecular Diagnostic Laboratory and returned
positive (Fig 3). The fi nal pathologic differential diagnosis
was metastatic dedifferentiated liposarcoma vs a new
primary spindle cell sarcoma not otherwise specifi ed.
Multidisciplinary discussion regarding the imaging fi nd-
ings and pathology deemed the lesions unresectable
due to the proximity/involvement of major vascular
and pulmonary structures within the mediastinum. The
patient was treated with four cycles of doxorubicin
(Adriamycin) at a cumulative dose of 300 mg/m2 and
followed with repeat cross-sectional imaging. CT of
the chest at the conclusion of systemic therapy dem-
onstrated disease progression with worsening displace-
ment of the previously involved mediastinal structures.
She died of her disease 7 months after diagnosis of the
Spindle cell mesenchymal neoplasms are a common di-
agnostic challenge for even the most experienced pa-
thologist. The differential diagnosis is dependant on lo-
cation, but it may include smooth muscle tumors, neural
tumors, desmoids tumors, monophasic synovial sarcoma,
dermatofi brosarcoma protuberans, infl ammatory myofi -
broblastic tumor, fi brosarcoma, spindle cell sarcoma not
otherwise specifi ed, and GIST. In these cases, a well-
chosen panel of IHC will help elucidate the correct diag-
nosis (Fig 4). The IHC hallmark of GISTs is positivity for
CD117 (c-kit) and CD34. However, neither CD117 nor
Fig 3. — FISH for MDM2 showing an increase in expression.
Poseidon MDM2 (12q15) DNA Probe
156 Cancer ControlJuly 2011, Vol. 18, No. 3
CD34 is specifi c for GISTs, and the list of CD117+ and/
or CD34+ tumors is increasing (melanoma, seminoma,
adenoid cystic carcinoma, small cell lung carcinoma,
follicular thyroid carcinoma, Ewing sarcoma, Kaposi’s
sarcoma, oncocytoma and chromophobe carcinoma of
the kidney, thymic lesions, angiosarcoma, and certain
GISTs are mesenchymal neoplasms arising primarily
in the gastrointestinal tract (95%) with approximately 60%
occurring in the stomach, 30% in the small intestine and
the remainder in the esophagus, colon, and rectum.4 Rare
cases have been reported in the omentum and mesen-
tery.5,6 As noted above, no mediastinal GIST has ever been
reported. They are believed to derive from the interstitial
cells of Cajal, and their pathogenesis involves activation
of the KIT signaling pathway, making a constitutionally
active tyrosine kinase.7-9 Most GISTs arise sporadically,
but a minority occur as part of a syndrome, either in a
familial setting with heritable germline KIT or PDGFRa
mutation, in patients with neurofi bromatosis type 1 syn-
drome (NF-1), or in the Carney triad (GISTs, pulmonary
chondroma, and paraganglioma, occurring more often in
young women).10,11 GISTs usually develop in the fi fth or
sixth decade and are rarely diagnosed in patients younger
than 40 years of age. They rarely have been reported in
children, the occurrence rate being under 1% and usually
developing in the second decade, with a predilection for
females, a gastric localization, and the epithelioid vari-
ant.12,13 No sex-related predilection has been reliably dem-
onstrated (although some data indicate a predominance in
males), nor have differences been reported based on race,
ethnicity, occupation, or specifi c geographic distribution.3
Histologically, GISTs vary from spindle cell tumors to
epithelioid and pleomorphic tumors. Most GISTs (95%)
express CD117 (c-kit), CD34 (70%), and heavy caldesmon
(80%), and 25% are positive for smooth muscle actin
Smooth Muscle Skeletal Muscle Neural, MelanomaVascular
(strong clinical suspicion)
Fig 4. — IHC approach to spindle cell mesenchymal neoplasms. SMA = smooth muscle actin, EMA = epithelial membrane antigen.
July 2011, Vol. 18, No. 3 Cancer Control 157
and < 5% for desmin.3,4,14,15 Despite histological overlap
with smooth muscle neoplasms, GISTs are usually readily
distinguishable from other mesenchymal tumors by their
typical IHC profi le. As described above, approximately
95% of GISTs express the kit receptor tyrosine kinase
CD117, the most specifi c marker for these lesions. CD34
is also a useful but less specifi c fi nding (described below).
C-kit, however, is not specifi c for GIST. Positivity is also
seen in mast cells and also in melanocytes, which may
be confusing in spindle cell melanomas that also have a
similar histology to GISTs.15-18 Other tumors that have
been shown to have some c-kit positivity include renal
cell carcinomas, seminoma, dedifferentiated liposarcoma,
Kaposi’s sarcoma, angiosarcoma, Ewing sarcoma and ex-
tramedullary myeloid tumors, and lymphoid tumors.3,19-21
CD117 positivity in GIST is indicative of a mutation in the
KIT gene on chromosome 4, a proto-oncogene encoding
a type III transmembrane tyrosine kinase receptor. The
most common exons involved are 9 and 11, with 13 and
17 also being reported.3 In GISTs, KIT gain-of-function
mutations lead to constitutive activation of KIT; this
overexpression is the target of the novel antityrosine
kinase therapy with imatinib. Exons 9 and 13 are associ-
ated with resistance to imatinib; in these cases, patients
may benefi t from double the usual dose of imatinib or
switching to a second-line agent such as sunitinib.22-25
As described above, approximately 15% of GISTs will
be c-kit negative. These cases almost always harbor a
mutation in PDGFR. PDGFR is a tyrosine kinase like c-kit
whose gene is also located on chromosome 4. Mutations
occur within exon 12 (1%), 14 (6%), or 18 (80%).3 Muta-
tions on exon 14 are associated with a worse prognosis.
Since PDGFR is a tyrosine kinase, some of these cases
have also been shown to respond to imatinib. However,
a subset of cases, specifi cally those with the exon 18
mutation, will show resistance to imatinib and, similar to
the resistant c-kit mutations, these patients may benefi t
from higher doses or second-line treatment.26,27
CD34 is a monomeric transmembrane glycoprotein
that is expressed on hematopoietic progenitor cells and
endothelial cells. It is positive in early precursor leuke-
mias, vascular tumors, and some mesenchymal tumors.
CD34 has a role as an adhesion molecule in the bone
marrow, and it is postulated to have a role in present-
ing carbohydrate ligands to selectins and in regulating
adhesion to stromal cells. CD34 may react with a variety
of mesenchymal lesions, such as nerve sheath tumors,
dermatofi brosarcoma protuberans, and solitary fi brous
tumors.28-30 Extensive membrane and cytoplasmic stain-
ing for CD34 is also a consistent feature of spindle cell
lipomas.31-33 In addition, CD34 immunoreactivity identi-
fi es a sparse network of dendritic spindle cells in a variety
of lipomatous tumors, including angiolipomas, myxoid
liposarcomas, and atypical lipomatous tumor/well-differ-
entiated liposarcoma (ALT/WDLS).31-33 For ALT/WDLS,
staining of the scattered atypical stellate cells may be vari-
able, and positivity is usually widespread when present in
the malignant spindle cell population of dedifferentiated
liposarcoma.31-33 The utility of CD34 immunoreactivity
in distinguishing GIST from a lipomatous neoplasm has
not been established and requires other immunostains
or ancillary studies.
The DOG1 gene (discovered on GIST1) encodes
a hypothetical protein specifi cally expressed in GISTs
called FLJ10261. Immunoreactivity for DOG1 has been
shown in up to 98% of GISTs, regardless of KIT reactivity
or PDGFRα mutation status.34,35 Therefore, it is especially
useful in the diagnosis of KIT-negative GIST. DOG1.1 is a
recently described mouse monoclonal antibody reported
to have superior sensitivity and specifi city compared to
KIT (CD117) and CD34.36 However, emerging studies
have shown that other tumors on the differential for GIST
(desmoids fi brosis, schwannoma, and muscle tumors) are
immunonegative for DOG1.34 Other new markers such as
protein kinase C theta (PKCθ) and nestin are sensitive but
less specifi c for GIST.35 DOG1 is proposed for use with
CD117 as antibodies as choice for the diagnosis of GIST.34-36
Liposarcoma is the most common soft tissue sar-
coma, accounting for approximately 20% of all mesen-
chymal malignancies encountered by surgical patholo-
gists. Liposarcoma is classifi ed into four main groups:
(1) well-differentiated liposarcoma, including adipocytic,
sclerosing, infl ammatory, and spindle-cell, and dediffer-
entiated variants, characterized by ring or long marker
chromosomes derived primarily from the long arm of
chromosome 12, (2) myxoid/round cell liposarcoma char-
acterized in most cases by a reciprocal t(12;16)(q13;p11),
(3) dedifferentiated liposarcoma, and (4) pleomorphic
liposarcoma (no characteristic karyotype).37-41 Among
the different subtypes of liposarcoma, dedifferentiated
liposarcoma represents both a morphologically and bio-
logically challenging lesion in which transition from low-
grade to high-grade nonlipogenic morphology within a
well-differentiated liposarcoma is observed. The differ-
ential diagnosis of these tumors may be histologically dif-
fi cult, however, because they can mimic various different
tumors. In these cases, genetic analyses are particularly
useful to make or confi rm the diagnosis.
MDM2 (murine double-minute) oncogene is located
on chromosome 12 and has been shown to be positive
in numerous mesenchymal tumors, the most helpful of
which is liposarcoma. Like CD117 (c-kit), MDM2 can be
detected by IHC or FISH. However, like CD117/CD34
positivity in GIST, an MDM2-positive pleomorphic spindle
cell neoplasm is not automatically a liposarcoma. As
alluded to above, MDM2 is helpful in distinguishing a
well-differentiated liposarcoma from a benign lipoma.
However, it cannot reliably be used to distinguish be-
tween dedifferentiated liposarcoma and other pleomor-
phic sarcomas. In our case, if the tumor is located in the
region where the well-differentiated liposarcoma was
located, it is reasonable to suspect that it is a recurrent
158 Cancer ControlJuly 2011, Vol. 18, No. 3
liposarcoma that is now dedifferentiated. However, due
to its mediastinum location, the possibility of a new pri-
mary sarcoma cannot be excluded.
MDM2 codes for the MDM2 protein. The key tar-
get of MDM2 is the p53 tumor suppressor gene where
it represses p53 transcriptional activity by blocking
the N-terminal transactivation domain. MDM2 is also
upregulated by p53, wherein its transcription can be
activated by p53. Thus, when p53 is stabilized, the tran-
scription of MDM2 is also induced, resulting in higher
MDM2 protein levels. In addition, MDM2 acts as a li-
gase, targeting both itself and p53 for degradation by
proteosomes. Its link to p53 suppression substantiates
its oncogenic potential. Further supporting this role is
its increased level in several tumors, including soft tis-
sue sarcomas, osteosarcomas, and some breast tumors.
Well-differentiated and dedifferentiated liposarcomas
(WDLPS/DDLPS) contain amplifi ed sequences from the
12q13–15 region, including the MDM2 gene.42,43 The
detection of MDM2 overexpression (by IHC) or ampli-
fi cation (by FISH) was originally proposed as a reliable
tool for distinguishing benign adipose tissue tumors from
WDLPS and distinguishing DDLPS from other poorly dif-
ferentiated tumors.44 However, recent studies, including
this article, have shown that other tumors can be posi-
tive for MDM2 including chordomas, leiomyosarcomas,
intimal sarcomas of the pulmonary artery, and focally,
chondrosarcomas.45,46 MDM2 positivity is still a reliable
marker to differentiate lipoma from an atypical lipoma-
tous tumor/well-differentiated liposarcoma.
Ancillary studies are an integral part of pathologic di-
agnostics; however, these studies need to be used in
conjunction with clinical and histological information.
Pathologists must be aware that though several tumors
have hallmark IHC profi les, much overlap exists between
tumors. We used the example of a sarcoma initially
thought to be a GIST to elucidate this as it pertains to
CD117, CD34, and MDM2. It is important to note that
a CD117+/CD34+ spindle cell neoplasm is not neces-
sarily a GIST, and a MDM2+ sarcoma is not necessarily
a liposarcoma. Awareness of the potential diagnostic
pitfalls of ancillary studies, as well as their appropriate
applications, is critical.
Rosai and Ackerman’s Surgical Pathology Review. Vol 1. 9th ed. New York,
NY: Mosby; 2004.
2. Bui MM, Smith P, Agresta SV, et al. Practical issues of intraoperative
frozen section diagnosis of bone and soft tissue lesions. Cancer Control. 2008;
3. Fülöp E, Marcu S, Milutin D, et al. Gastrointestinal stromal tumors:
review on morphology, diagnosis and management. Rom J Morphol Embryol.
4. Fletcher CDM, Unni K, Mertens F, eds. World Health Organization
Classifi cation of Tumours Pathology and Genetics of Tumours of Soft Tissue
and Bone. Lyon, France: IARC Press; 2002.
5. Llenas-García J, Guerra-Vales JM, Moreno A, et al. Primary extragas-
1. Rosai J. Special techniques in surgical pathology. In: Rosai J, ed.
trointestinal stromal tumors in the omentum and mesentery: a clinicopatho-
logical and immunohistochemical study. Hepatogastroenterology. 2008;55
6. Reith JD, Goldblum JR, Lyles RH, et al. Extragastrointestinal (soft
tissue) stromal tumors: an analysis of 48 cases with emphasis on histologic
predictors of outcome. Mod Pathol. 2000;13(5):577-585.
7. Hirota S, Nishida T, Isozaki K, et al. Gain-of-function mutation at the
extracellular domain of KIT in gastrointestinal stromal tumours. J Pathol.
8. Nakahara M, Isozaki K, Hirota S, et al. A novel gain-of-function muta-
tion of c-kit gene in gastrointestinal stromal tumors. Gastroenterology. 1998;
9. Sircar K, Hewlett BR, Huizinga JD, et al. Interstitial cells of Cajal as
precursors of gastrointestinal stromal tumors. Am J Surg Pathol. 1999;23(4):
10. Tortella BJ, Matthews JB, Antonioli DA, et al. Gastric autonomic
nerve (GAN) tumor and extra-adrenal paraganglioma in Carney’s triad: a
common origin. Ann Surg. 1987;205(3):221-225.
11. Carney JA, Swee RG. Carney complex. Am J Surg Pathol. 2002;26
12. Miettinen M, Lasota J, Sobin LH. Gastrointestinal stromal tumors of
the stomach in children and young adults: a clinicopathologic, immunohisto-
chemical, and molecular genetic study of 44 cases with long-term follow-up
and review of the literature. Am J Surg Pathol. 2005;29(10):1373-1381.
13. Prakash S, Sarran L, Socci N, et al. Gastrointestinal stromal tumors
in children and young adults: a clinicopathologic, molecular, and genomic
study of 15 cases and review of the literature. J Pediatr Hematol Oncol. 2005;
14. Bates AW, Feakins RM, Scheimberg I. Congenital gastrointestinal
stromal tumour is morphologically indistinguishable from the adult form, but
does not express CD117 and carries a favourable prognosis. Histopathology.
15. Graadt van Roggen JF, van Velthuysen ML, Hogendoorn PC. The
histopathological differential diagnosis of gastrointestinal stromal tumours. J
Clin Pathol. 2001;54(2):96-102.
16. Arber DA, Tamayo R, Weiss LM. Paraffi n section detection of the
c-kit gene product (CD117) in human tissues: value in the diagnosis of mast
cell disorders. Hum Pathol. 1998;29(5):498-504.
17. Gibson PC, Cooper K. CD117 (KIT): a diverse protein with selective
applications in surgical pathology. Adv Anat Pathol. 2002;9(1):65-69.
18. Pelle MT, Parc Y, Mourra N. c-Kit expression in spindle cell melanoma:
a diagnostic pitfall in anorectal biopsy. Arch Pathol Lab Med. 2009;133(8):
19. Berman J, O’Leary TJ. Gastrointestinal stromal tumor workshop.
Hum Pathol. 2001;32(6):578-582.
20. Miliaras D, Karasavvidou F, Papanikolaou A, et al. KIT expression in
fetal, normal adult, and neoplastic renal tissues. J Clin Pathol. 2004;57(5):
21. Espinosa I, Lee CH, Kim MK, et al. A novel monoclonal antibody
against DOG1 is a sensitive and specifi c marker for gastrointestinal stromal
tumors. Am J Surg Pathol. 2008;32(2):210-218.
22. Koay MH, Goh YW, Iacopetta B, et al. Gastrointestinal stromal tu-
mours (GISTs): a clinicopathological and molecular study of 66 cases. Pa-
23. Miettinen M, Lasota J. Gastrointestinal stromal tumors: review on
morphology, molecular pathology, prognosis, and differential diagnosis.
Arch Pathol Lab Med. 2006;130(10):1466-1478.
24. Lasota J, Corless CL, Heinrich MC, et al. Clinicopathologic profi le of
gastrointestinal stromal tumors (GISTs) with primary KIT exon 13 or exon 17
mutations: a multicenter study on 54 cases. Mod Pathol. 2008;21(4):476-484.
25. Longley BJ, Reguera MJ, Ma Y. Classes of c-KIT activating muta-
tions: proposed mechanisms of action and implications for disease clas-
sifi cation and therapy. Leuk Res. 2001;25(7):571-576.
26. Corless CL, Schroeder A, Griffi th D, et al. PDGFRA mutations in
gastrointestinal stromal tumors: frequency, spectrum and in vitro sensitivity
to imatinib. J Clin Oncol. 2005;23(23):5357-5364.
27. Lasota J, Stachura J, Miettinen M. GISTs with PDGFRA exon 14 mu-
tations represent subset of clinically favorable gastric tumors with epithelioid
morphology. Lab Invest. 2006;86(1):94-100.
28. van de Rijn M, Rouse RV. CD34: a review. Appl Immunohistochem.
29. Weiss SW, Nickoloff BJ. CD-34 is expressed by a distinctive cell
population in peripheral nerve, nerve sheath tumors, and related lesions.
Am J Surg Pathol. 1993;17(10):1039-1045.
30. van de Rijn M, Lombard CM, Rouse RV. Expression of CD34 by soli-
tary fi brous tumors of the pleura, mediastinum, and lung. Am J Surg Pathol.
31. Suster S, Fisher C. Immunoreactivity for the human hematopoietic
progenitor cell antigen (CD34) in lipomatous tumors. Am J Surg Pathol. 1997;
32. Templeton SF, Solomon AR Jr. Spindle cell lipoma is strongly CD34
positive: an immunohistochemical study. J Cutan Pathol. 1996;23(6):546-550.
33. Silverman JS, Tamsen A. Fibrohistiocytic differentiation in subcuta-
July 2011, Vol. 18, No. 3 Cancer Control 159
neous fatty tumors: study of spindle cell, pleomorphic, myxoid, and atypical
lipoma and dedifferentiated liposarcoma cases composed in part of CD34+
fi broblasts and FXIIIa+ histiocytes. J Cutan Pathol. 1997;24(8):484-493.
34. West RB, Corless CL, Chen X, et al. The novel marker, DOG1, is
expressed ubiquitously in gastrointestinal stromal tumors irrespective of KIT
or PDGFRA mutation status. Am J Pathol. 2004;165(1):107-113.
35. Novelli M, Rossi S, Rodriguez-Justo M, et al. DOG1 and CD117 are
the antibodies of choice in the diagnosis of gastrointestinal stromal tumours.
36. Liegl B, Hornick JL, Corless CL, et al. Monoclonal antibody DOG1.1
shows higher sensitivity than KIT in the diagnosis of gastrointestinal stromal
tumors, including unusual subtypes. Am J Surg Pathol. 2009;33(3):437-446.
37. Weiss SW, Rao VK. Well-differentiated liposarcoma (atypical lipoma)
of deep soft tissue of the extremities, retroperitoneum, and miscellaneous
sites: a follow-up study of 92 cases with analysis of the incidence of “dedif-
ferentiation”. Am J Surg Pathol. 1992;16(11):1051-1058.
38. Dei Tos AP, Mentzel T, Newman PL, et al. Spindle cell liposarcoma,
a hitherto unrecognized variant of liposarcoma: analysis of six cases. Am J
Surg Pathol. 1994;18(9):913-921.
39. Evans HL. Atypical lipomatous tumor, its variants, and its combined
forms: a study of 61 cases, with a minimum follow-up of 10 years. Am J
Surg Pathol. 2007;31(1):1-14.
40. Meis-Kindblom JM, Sjögren H, Kindblom LG, et al. Cytogenetic
and molecular genetic analyses of liposarcoma and its soft tissue simula-
tors: recognition of new variants and differential diagnosis. Virchows Arch.
41. Sandberg AA. Updates on the cytogenetics and molecular genetics of
bone and soft tissue tumors: liposarcoma. Cancer Genet Cytogenet. 2004;155
42. Pedeutour F, Forus A, Coindre JM, et al. Structure of the supernu-
merary ring and giant rod chromosomes in adipose tissue tumors. Genes
Chromosomes Cancer. 1999;24(1):30-41.
43. Italiano A, Bianchini L, Keslair F, et al. HMGA2 is the partner of MDM2
in well-differentiated and dedifferentiated liposarcomas whereas CDK4 be-
longs to a distinct inconsistent amplicon. Int J Cancer. 2008;122(10):2233-
44. Binh MB, Sastre-Garau X, Guillou L, et al. MDM2 and CDK4 immu-
nostainings are useful adjuncts in diagnosing well-differentiated and dedif-
ferentiated liposarcoma subtypes: a comparative analysis of 559 soft tissue
neoplasms with genetic data. Am J Surg Pathol. 2005;29(10):1340-1347.
45. Daugaard S, Christensen LH, Høgdall E. Markers aiding the diagnosis
of chondroid tumors: an immunohistochemical study including osteonectin,
bcl-2, cox-2, actin, calponin, D2-40 (podoplanin), mdm-2, CD117 (c-kit), and
YKL-40. APMIS. 2009;117(7):518-525.
46. Bode-Lesniewska B, Zhao J, Speel EJ, et al. Gains of 12q13-14 and
overexpression of mdm2 are frequent fi ndings in intimal sarcomas of the
pulmonary artery. Virchows Arch. 2001;438(1):57-65.