Human Pathology (2018) 75, 34–40

www.elsevier.com/locate/humpath

Original contribution

Expression of MDM2 and p16 in angiomyolipoma☆

Xiaoqi Lin MD, PhD a,⁎, William B. Laskin MD b , Xinyan Lu MD a , Yaxia Zhang MD, PhD c
a
Department of Pathology, Northwestern University, Chicago, IL 60611
b
Department of Pathology, Yale School of Medicine, New Haven, CT 06520
c
Department of Pathology, Hospital for Special Surgery, New York, NY 10021

Received 2 October 2017; revised 19 January 2018; accepted 24 January 2018

Keywords:
Summary Angiomyolipoma (AML) arises primarily from the kidney but may grow into the retroperitoneal
Angiomyolipoma;
space mimicking a primary retroperitoneal tumor. Fine needle aspiration (FNA) and core needle biopsy of
Liposarcoma;
AML, particularly the fat-predominant variant, may be difficult to distinguish from retroperitoneal well-
Lipoma;
differentiated liposarcoma (WDLS) or lipoma. Commonly used immunomarkers, MDM2 and p16, have
Histomorphology;
proven useful in diagnosing WDLS and dedifferentiated liposarcoma (DDLS), while HMB45 and Melan-
Cytomorphology;
A are melanocyte-related markers characteristically expressed in AML. In this study, we investigated the
Immunohistochemistry;
utility of MDM2 and p16 along with HMB45 and Melan-A immunohistochemical analysis in distinguishing
Fluorescence in situ
AML from WDL/DDLS or lipoma. Immunohistochemically, AMLs demonstrated focal MDM2 expression
hybridization for MDM2
(40% of cases) and focal/diffuse expression of p16 (60%). AMLs marked focally or diffusely with HMB45
(76% of cases) and Melan-A (96%). These latter two immunomarkers were not expressed in any of the
WDLS/DDLSs or lipomas tested. WDLS/DDLSs showed focal/diffuse expression of MDM2 (91% of
cases) and p16 (97%). While focal expression of MDM2 and p16 was observed in 14% and 67% of lipomas,
respectively, no lipoma exhibited diffuse MDM2 positivity. In our hands, MDM2 expression by itself
cannot exclude the diagnosis of AML or lipoma, and p16 alone is not helpful in separating AML and
conventional lipoma from WDLS/DDLS. However, along with morphology, an immunohistochemical
battery including HMB45, Melan-A, MDM2 and p16 are useful in distinguishing AML from WDLS/DDLS
or lipoma. For equivocal cases, fluorescence in situ hybridization for MDM2 should be performed.
© 2018 Elsevier Inc. All rights reserved.

1. Introduction mesenchymal tumors that exhibit overlapping histologic fea-

Perivascular epithelioid cell tumors (PEComas), collective-
ly named after the enigmatic perivascular epithelioid cell from
which they presumably arise, constitute a family of
☆
Disclosures: All authors fully disclose no financial or ethical conflicts of
interest.
⁎ Corresponding author at: Northwestern Memorial Hospital, Northwest-
ern University, 251 East Huron Street, Galter Pavilion 7-132F, Chicago, IL,
60611.
E-mail address: xlin@northwestern.edu (X. Lin).
tures and a dual myogenic and melanocytic immunophenotype
[1,2]. Angiomyolipoma (AML), a prototypic PEComa, is a tri-
phasic neoplastic process composed of a variable admixture of
lipid-laden cells resembling adipocytes, spindled and epitheli-
oid myoid cells, and abnormal thick-walled blood vessels [2].
AML arises primarily within the kidney, but can grow into the
retroperitoneal space, or rarely present in retroperitoneum
without renal attachment [3], thereby mimicking a primary
retroperitoneal tumor.
Although well-differentiated liposarcoma (WDLS) repre-
sents the single most common primary sarcoma encountered

https://doi.org/10.1016/j.humpath.2018.01.022
0046-8177/© 2018 Elsevier Inc. All rights reserved.
MDM2 and p16 in angiomyolipoma
in the retroperitoneum, lipoma and AML also occur in this
region [3], and may be difficult to separate on CT imaging
[4]. Compounding matters further, all three lesions are
composed at least in part of lipid-laden cells that have the
potential to exhibit cytologic atypia. In addition, the dedif-
ferentiated component of some examples of WDLS (DDLS)
mimics the spindle cell element of AML. These ambiguous
cytologic and histologic features are often magnified in the
setting of small tissue samples such as fine needle aspiration
(FNA) and core needle biopsies (CNB), and may result in
diagnostic confusion.
Important differences in clinical course and treatment of
these two entities further dictate correct diagnosis. Retroperito-
neal WDLS/DDLS requires surgery, and has a high rate of re-
currence due to difficulties in completely excising the tumor in
this region. Patients with uncomplicated AML can be initially
managed with initial active surveillance [5], whereas symp-
tomatic lesions are treated by radical and partial nephrectomy
[6,7], selective arterial embolization [7] or ablative therapies,
including cryoablation and radio frequency ablation [7]. In ad-
dition, patients with tuberous sclerosis complex–associated
angiomyolipomas have shown promising results with mTOR
inhibitors [7]; therefore, a correct diagnosis is imperative.
Immunohistochemistry is oftentimes used to assist in the
diagnosis of AML and WDLS/DDLS. The former characteris-
tically expresses melanocyte-related immunomarkers,
HMB45 and Melan-A, and less often, TFE3 and microphthal-
mia transcription factor, and myogenic markers, smooth mus-
cle actin, calponin and occasionally, h-caldesmon and desmin
[8-10]. The hallmark molecular event in the pathogenesis of
WDLS/DDLS is amplification of chromosome 12q13-15
resulting in increased copy numbers of MDM2 (considered
the “gold standard” of diagnosis) [11] and consequent overex-
pression of MDM2 [12,13]. Expression of p16 has proven a
sensitive and specific marker for separating WDLS (atypical
lipomatous tumor)/DDLS from other adipocytic tumors [14,15]
and has been touted as especially helpful in differentiating
WDLS (atypical lipomatous tumor) from deep lipoma [14].
To date, only a few studies have attempted to differentiate
WDLS/DDLS from AML and lipoma with immunohisto-
chemistry [13,16]. As this differential has some relevance in
the retroperitoneum [13,16], we decided to evaluate the ability
of commonly used immunoreagents MDM2, p16, HMB45,
and Melan-A to accomplish this task.
2. Materials and methods
2.1. Case selection
After approval by our institutional review board (STU
#82702), we retrieved 25 cases of AML (16 surgically resected
and 9 CNB cases), 33 retroperitoneal liposarcomas (27 WDLS
and 6 DDLS) of which 12 were initially diagnosed by CNB (9
WDLS and 3 DDLS), and 21 resected lipomas (4 with prior
35
CNB) from the Pathology Department of Northwestern
Memorial Hospital. Data collected included age and sex of
the patient, and the surgical and cytopathology diagnoses.
2.2. Core needle biopsy
Percutaneous CNBs were performed under guidance of ul-
trasound (US) or computed tomography (CT) imaging using a
20-gauge core biopsy device. One to four CNB passes with cor-
responding touch preparations (TP) were obtained. The touch
preparation slides were air-dried and stained with modified
Giemsa (Diff-Quik) stain for on-site evaluation for specimen ad-
equacy and interpretation by a board-certified cytopathologist.
2.3. Histology and immunohistochemistry
CNBs and representative sections from partial nephrecto-
my specimens and resected retroperitoneal tumors were
formalin-fixed, paraffin-embedded (FFPE), microtome proc-
essed, placed on glass slides, and stained with hematoxylin
and eosin (H&E) for histomorphology examination.
Immunohistochemical stains were performed on the sec-
tions of the FFPE tissue with appropriate positive and negative
controls (ie, positive controls expressed the immunoreagent
and negative controls showed no expression). Antibodies
against MDM2 (M41-113, Invitrogen, Carlson, CA), p16
(705-4713, Roche Diagnostics, Indianapolis, IN), HMB45
(M0634, DakoCytomation, Carpinteria, CA), and Melan-A
(M7196, DakoCytomation) were used. Positive result for
MDM2 and p16 required nuclear staining in viable tumor
cells. Immunohistochemical staining was graded in a semi-
quantitative manner and scored as “diffuse” if N50% of tumor
cells were positive, “focal” if between 5% and 50% of tumor
cells expressed the protein, and “rare” if b5% of tumor cells
were positive. When less than 5% of tumor cells were positive,
a negative score was assigned.
2.4. Fluorescence in situ hybridization
Four-micrometer slides were prepared from FFPE. Fluores-
cence in situ hybridization (FISH) was performed using a dual-
color MDM2 probe set with the chromosome 12 centromere
labeled as spectrum green serving as the control locus and
the MDM2 gene located at 12q15 labeled as spectrum orange
(Leica Biosystems, Buffalo Grove, IL). Standard laboratory
procedure and the instructions from the manufacturer’s proto-
col were followed in conducting FISH analysis.
The normal signal pattern consists of two MDM2 (orange)
and CEP12 (green) signals. Cells with gains in copy numbers
demonstrate a total of 3 to 5 copies of both CEP12 and MDM2
signals and are considered as exhibiting aneuploidy. The sig-
nal patterns are considered as positive for MDM2 amplifica-
tion when (1) the ratio of MDM2/CEP12 is ≥3.0, (2) there
are at least six or more MDM2 signals/per cell, or (3)
MDM2 signals are clustered [17].
36
2.5. Statistics
Student t test was used to test the statistical significance of
dichotomous parameters. The sensitivity, specificity, positive
predictive value and negative predictive value of the individual
immunohistochemical markers were calculated for AML,
WDLS/DDLS, and lipoma.
3. Results
3.1. Clinical results
AML patients included 2 males and 23 females ranging in
age from 28 to 71 years (mean ± standard deviation = 53.3 ±
11.6 years). Twenty-one men and 12 women ranging in age
from 30 to 81 years old (58.7 ± 12.0 years) were diagnosed
with WDLS or DDLS. Lipoma patients included 10 males
and 11 females ranging in age from 17 to 66 years old (49.9 ±
14.1 years). The 21 lipomas included 6 retroperitoneal, 7
intramuscular and 8 superficial lesions. AML showed a statis-
tically significant female predominance compared with
WDLS/DDLS and lipoma.
3.2. Cytomorphology of touch preparation of core
needle biopsy
On-site evaluation of the CNB TP for adequacy and inter-
pretation was performed by board-certified cytopathologists.
After 2 to 4 passes, 8 of 9 AML cases were cellular and consid-
ered “adequate,” while the cellularity of the remaining case
was deemed “inadequate.”
The TP of WDLS was characterized by the presence of ad-
ipocytes possessing variable-sized cytoplasmic vacuoles, and
nuclei exhibiting variation in size and chromatin content.
Examples of lipoma showed clusters of adipocytes with large
cytoplasmic vacuoles and small bland nuclei.
3.3. Histology
AML tissue from CNB or resection demonstrated a
haphazard arrangement of lipid-laden cells and spindled cells
(Fig. 1A). Some of the lipid-laden cells mimicked lipoblasts
by the presence of highly vacuolated cytoplasm. The spindled
cells had fibrillated eosinophilic cytoplasm indicating myoid
differentiation. Nuclei were generally round to oval, but occa-
sionally showed irregular contours. Thick-walled vessels from
which the spindled element appeared to emanate were also ob-
served. Sections of WDLS showed variable-sized adipocytes
with scattered enlarged and hyperchromatic nuclei including
lipoblasts possessing hyperchromatic nuclei whose contours
were scalloped by variable-sized cytoplasmic lipid vacuoles
(Fig. 1B). Atypical mesenchymal cells with enlarged, hyper-
chromatic nuclei were observed within fibrous septae travers-
ing the process. Lipomas consisted of adipocytes showing
X. Lin et al.
little size variation, a single large cytoplasmic vacuole, and
no hyperchromatism (Fig. 1C). No fat necrosis was identified
in any of the lipomas.
3.4. Immunohistochemistry
The immunohistochemical results for AML, WDLS/DDLS
and lipoma are available in the Table. AMLs showed focal ex-
pression of MDM2 (40%) (Fig. 1D), and diffuse or focal pos-
itivity for p16 (60%) (Fig. 1G), HMB45 (76% of cases)
(Fig. 1J) and Melan-A (96%) (Fig. 1M). One AML was posi-
tive for HMB45, while negative for Melan A. WDLS/DDLS
exhibited diffuse or focal positivity for MDM2 (91%) and
p16 (97%) (Fig. 1E and H). Focal expression for MDM2 and
p16 was observed in 14% and 64% of lipomas, respectively
(Fig. 1F and I). Among the 6 retroperitoneal lipomas, 2 were
focally positive for MDM2 and 4 were focally positive for
p16. No WDLS/DDLS or lipoma expressed melanocyte-
related immunomarkers, HMB45 and Melan-A (Fig. 1K, L,
N and -O).
AML showed a statistically significant lower frequency of
MDM2 expression than in WDLS/DDLS (P b .01), but
MDM2 positivity in AML was not statistically different from
that of lipoma (P N .05) (Table). Diffuse expression of
MDM2 was evident only in cases of WDLS/DDLS. The p16
positivity achieved statistical significance in separating
WDLS/DDLS from AML and lipoma (Table) (Fig. 1G-I) (P
b .05 and P b .01, respectively). An absence of Melan-A and
HMB45 immunoexpression in WDLS/DDLS (Fig. 1K and
N) and lipoma (Fig. 1L and O) was statistically significant in
differentiating these two lipogenic tumors from AML (P b
.01) (Table).
The sensitivity, specificity, positive predictive value and
negative predictive value of Melan-A for the diagnosis of
AML are 96%, 100%, 100%, and 98%, respectively. The sen-
sitivity, specificity, positive predictive value and negative pre-
dictive value of HMB45 for AML are 76%, 100%, 100%, and
90%, respectively. The sensitivity, specificity, positive predic-
tive value and negative predictive value of MDM2 for the di-
agnosis of WDLS/DDLS is 91%, 72%, 30%, and 92%,
respectively. The sensitivity, specificity, positive predictive
value and negative predictive value of p16 for WDLS/DDLS
is 97%, 37%, 52%, and 94%.
3.5. Fluorescence in situ hybridization
Seventeen adipocytic tumors with equivocal atypia and im-
munoreactivity to MDM2 were tested for amplification of
MDM2 by FISH to confirm a diagnosis of WDLS. Amplifica-
tion of MDM2 was identified in 7 tumors which were classi-
fied as WDLS/DDLS (Fig. 2B), while no amplification of
MDM2 was observed in 3 AMLs (Fig. 2A) and 7 lipomas
(Fig. 2C). The two retroperitoneal lipomas that were focally
immunoreactive to MDM2 and p16 did not show amplifica-
tion of MDM2.
MDM2 and p16 in angiomyolipoma 37

Fig. 1 Histomorphology and immunostains of angiomyolipoma (A, D, G, J, and M), WDLS (B, E, H, K, and N) and lipoma (C, F, I, L, and O).
A-C, Histology, H&E stain, original magnification ×400. D-F, Immunostain for MDM2, original magnification ×400. G-I, Immunostain for p16.
J-L, Immunostain for HMB45, original magnification ×400. M-O, Immunostain for Melan A, original magnification ×400.

4. Discussion markers, HMB45 and Melan-A, identified in 76% and 96%

In our study, we found that MDM2 and p16 were not only
expressed in WDLS (91% and 97% of cases, respectively), but
also in AMLs (40% and 60%, respectively) and lipoma (14%
and 64%, respectively). However, only WDLS/DDLS showed
diffuse expression of MDM2, and melanocytic-related
of AML cases tested, respectively, were not detected in any
examples of WDLS/DDLS and lipoma. Indeed, we found that
presence of HMB45 and Melan-A immunoexpression exhibit-
ed greater specificity and a higher positive predictive value for
diagnosing AML than MDM2 or p16 for WDLS/DDLS. Our
findings indicate that HMB45 and Melan-A expression should
38 X. Lin et al.

Table Immunohistochemical results of AML, WDLS and lipoma

Tumor No. Pattern MDM2 p16 HMB45 Melan-A
AML 25 Diffuse 0 3 8 16
Focal 10 12 11 8
Total (%) 10 (40%) 15 (60%) 19 (76%) 24 (96%)
WDLS 33 Diffuse 11 25 0 0
Focal 19 7 0 0
Total (%) 30 (91%) 32 (97%) 0 (0%) 0 (0%)
Lipoma 21 Diffuse 0 0 0 0
Focal 3 14 0 0
Total % 3 (14%) 14 (67%) 0 (0%) 0 (0%)
Pa b.01 b.05 b.01 b.01
Pb b.01 b.01
Pc b.01 b.01
NOTE. All statistical analyses were χ2 tests. Abbreviations: AML, angiomyolipoma; WDLS, well-differentiated liposarcoma.
a
Comparison of AML with WDLS.
b
Comparison of AML with lipoma.
c
Comparison of WDLS with lipoma.

be evaluated in a problematic retroperitoneal fatty tumor
where AML is within the differential diagnosis.
MDM2 amplification is considered highly specific for the
WDLS family of tumors and, to date, has not been detected
in AML or lipoma [18,19]. However, not all laboratories have
the ability to perform molecular testing, or the cost of testing
may be prohibitive. Therefore, MDM2 immunohistochemistry
oftentimes serves as a surrogate. Initially reported as having a
sensitivity of at least 80% for WDLS/DDLS in large series
[13,15,20], a more recent study evaluating only histologically
challenging differentiated fatty tumors requiring FISH confir-
mation for a diagnosis of WDLS found a sensitivity of only
45% [21]. p16 has been found a highly sensitive and specific
immunomarker for separating WDLS/DDLS from other adi-
pocytic tumors [15], and has been found particularly useful
in differentiating WDLS from deep lipoma [14]. In our study,
the frequency of immunoexpression of MDM2 and p16 in
WDLS/DDLS (91% and 97%, respectively) correlates well
with that documented in the literature [14,21-24], and both
immunomarkers proved highly sensitive for WDLS/DDLS.
Interestingly, we found that MDM2 was also expressed in
AML (40%) and lipoma (14%) in a focal manner. Only a
few studies examining the utility of immunohistochemistry
for distinguishing AML from WDLS/DDLS and lipoma
appear in the literature. One study failed to detect expression
of MDM2 in all 4 AML cases and in 27 examples of both su-
perficial and deep lipoma tested [13]. The authors concluded
that the combination of MDM2, CDK4, and p16 is a helpful
adjunct in differentiating WDLS from a variety of benign fatty
tumors, including AML and lipoma. However, the small num-
ber of AML cases tested in this study precludes any meaningful
conclusion regarding MDM2 expression in this entity. A more
recent paper reported MDM2 expression in 14% of AML in-
cluding 23% of fat-predominant variants [16]. Although their
figure of 14% is lower than our 60% expression rate in
AML, both studies demonstrate that MDM2 cannot be used
alone for separating WDLS from AML. In our study, the fre-
quency of MDM2 expression in lipoma (14%) was within
the range of 1.6% to 16% reported in the literature [21,22].
One caution to heed regarding MDM2 expression is that
lipid-laden histiocytes in foci of fat necrosis frequently are im-
munoreactive [25]. However, no fat necrosis was seen in the sec-
tions used for IHC studies in our study. We contend that diffuse
expression of MDM2 highly favors WDLS over AML or lipoma,
as only examples of WDLS/DDLS diffusely expressed this
protein. On the other hand, focal MDM2 expression plays only
a limited role in distinguishing WDLS from AML or lipoma.
Although MDM2 is a sensitive immunomarker for WDLS

Fig. 2 Fluorescence in situ hybridization (FISH) for MDM2. A, Angiomyolipoma. B, WDLS. C, Lipoma.
MDM2 and p16 in angiomyolipoma
(91%) with a high negative predictive value (92%), the specificity
and positive predictive value of MDM2 for the diagnosis of
WDLS is only 72% and 30%, respectively. Therefore, we agree
with the recent recommendation that molecular testing for
MDM2 amplification should be performed in problematic cases
showing equivocal cytologic atypia [21].
p16, the protein product of a suppressor gene that inhibits
cyclin-dependent kinases 4 and 6 from activating retinoblastoma
protein, has found relevance in combination with MDM2 and
CDK4 for diagnosing WDLS/DDLS [15] and alone in distin-
guishing WDLS from deep lipoma [14]. In addition, one paper
concluded that p16 immunohistochemistry can be valuable in
differentiating the uncommon fat-predominant AML variant
from atypical lipomatous tumor (ALT)/WDLS [26]. The fre-
quency of expression of p16 (97%) in WDLS in our study is
similar to the 83% to 91% reported in previous studies
[14,23]. Similar to the caveat regarding MDM2 expression in
fatty tumors, p16 can be expressed in areas of fat necrosis
[27]. We detected focal p16 expression in 67% of lipomas,
and diffuse or focal p16 expression in 60% of AMLs studied.
We found that, like MDM2, only diffuse expression of p16 as-
sists in distinguishing atypical lipomatous tumor/WDLS from
lipoma, but is not useful in differentiating WDLS/DDLS from
AML.
In our study, HMB45 and Melan-A expression were ob-
served in 88% and 96% of AML, respectively, but these
markers were not detected in any case of WDLS/DDLS or li-
poma. The high sensitivity, specificity, positive predictive val-
ue of HMB45 and Melan-A for the diagnosis of AML confirm
their utility in its differentiation from other fat-forming retro-
peritoneal lesions.
In rare instances, differentiating WDLS/DDLS from AML
may be problematic, especially with small biopsies from the
retroperitoneal region. In many pathology laboratories, immu-
nohistochemistry remains the main ancillary study that serves to
separate these two entities. We found that MDM2 and p16,
which are commonly expressed in WDLS/DDLS, are also found
in AML and, therefore, HMB45 and Melan-A immunostaining
should also be performed in questionable cases to distinguish
AML from benign and malignant adipocytic tumors. Our study
demonstrates that a panel of commonly used immunomarkers,
MDM2, p16, HMB45 and Melan-A, is sufficient to differentiate
AML from WDL/DDL. However, detection of MDM2 amplifi-
cation by FISH, multiplex ligation-dependent probe amplifica-
tion or real-time polymerase chain reaction is recommended in
truly problematic cases where histology and immunohistochem-
istry fail to differentiate WDL/DDLS from AML or lipoma
[21,28,29].
References
[1] Folpe AL. Neoplasms with perivascular epithelioid cell differentiation
(PEComas). In: Fletcher DM, Krishnan Unni K, Fletcher DM, editors.
World Health Organization Classification of Tumours: Pathology & Ge-
netics of Tumours of Soft Tissue and Bone. Lyon: IARC Press; 2002.
p. 221-2.
39
[2] Martignoni G, Amin MB. Angiomyolipoma. In: Eble JN, Sauter G,
Epstein JI, Sesterhenn IA, editors. World Health Organization Classifica-
tion of Tumours: Pathology & Genetics of Tumours of the Urinary
System and Male Genital Organs. Lyon: IARC Press; 2004. p. 65-7.
[3] Venyo AK. A review of the literature on extrarenal retroperitoneal angio-
myolipoma. Int J Surg Oncol 2016;2016:6347136.
[4] Israel GM, Bosniak MA, Slywotzky CM, Rosen RJ. CT differentiation
of large exophytic renal angiomyolipomas and perirenal liposarcomas.
AJR Am J Roentgenol 2002;179:769-73.
[5] Bhatt JR, Richard PO, Kim NS, et al. Natural history of renal angiomyoli-
poma (AML): most patients with large AMLs N4cm can be offered active
surveillance as an initial management strategy. Eur Urol 2016;70:85-90.
[6] Nelson CP, Sanda MG. Contemporary diagnosis and management of re-
nal angiomyolipoma. J Urol 2002;168:1315-25.
[7] Flum AS, Hamoui N, Said MA, et al. Update on the diagnosis and man-
agement of renal angiomyolipoma. J Urol 2016;195:834-46.
[8] Folpe AL, Mentzel T, Lehr HA, Fisher C, Balzer BL, Weiss SW. Peri-
vascular epithelioid cell neoplasms of soft tissue and gynecologic origin:
a clinicopathologic study of 26 cases and review of the literature. Am J
Surg Pathol 2005;29:1558-75.
[9] Hornick JL, Fletcher CD. PEComa: what do we know so far?
Histopathology 2006;48:75-82.
[10] Folpe AL, Kwiatkowski DJ. Perivascular epithelioid cell neoplasms: pa-
thology and pathogenesis. HUM PATHOL 2010;41:1-15.
[11] Pedeutour F, Suijkerbuijk RF, Forus A, et al. Complex composition
and co-amplification of SAS and MDM2 in ring and giant rod marker
chromosomes in well-differentiated liposarcoma. Genes Chromosomes
Cancer 1994;10:85-94.
[12] Pilotti S, Della Torre G, Mezzelani A, et al. The expression of MDM2/
CDK4 gene product in the differential diagnosis of well differentiated
liposarcoma and large deep-seated lipoma. Br J Cancer 2000;82:1271-5.
[13] Binh MB, Sastre-Garau X, Guillou L, et al. MDM2 and CDK4
immunostainings are useful adjuncts in diagnosing well-differentiated and
dedifferentiated liposarcoma subtypes: a comparative analysis of 559 soft tis-
sue neoplasms with genetic data. Am J Surg Pathol 2005;29:1340-7.
[14] He M, Aisner S, Benevenia J, Patterson F, Aviv H, Hameed M. p16 im-
munohistochemistry as an alternative marker to distinguish atypical lipo-
matous tumor from deep-seated lipoma. Appl Immunohistochem Mol
Morphol 2009;17:51-6.
[15] Thway K, Flora R, Shah C, Olmos D, Fisher C. Diagnostic utility of p16,
CDK4, and MDM2 as an immunohistochemical panel in distinguishing
well-differentiated and dedifferentiated liposarcomas from other adipo-
cytic tumors. Am J Surg Pathol 2012;36:462-9.
[16] Asch-Kendrick RJ, Shetty S, Goldblum JR, et al. A subset of fat-
predominant angiomyolipomas label for MDM2: a potential diagnostic
pitfall. HUM PATHOL 2016;57:7-12.
[17] Erickson-Johnson MR, Seys AR, Roth CW, et al. Carboxypeptidase M: a
biomarker for the discrimination of well-differentiated liposarcoma from
lipoma. Mod Pathol 2009;22:1541-7.
[18] Oliner JD, Saiki AY, Caenepeel S. The role of MDM2 amplification
and overexpression in tumorigenesis. Cold Spring Harb Perspect Med
2016;6.
[19] Shimada S, Harada H, Ishizawa K, Hirose T. Retroperitoneal lipomatous
angiomyolipoma associated with amyloid deposition masquerading as
well-differentiated liposarcoma. Pathol Int 2006;56:638-41.
[20] Sirvent N, Coindre JM, Maire G, et al. Detection of MDM2-CDK4 am-
plification by fluorescence in situ hybridization in 200 paraffin-
embedded tumor samples: utility in diagnosing adipocytic lesions and
comparison with immunohistochemistry and real-time PCR. Am J Surg
Pathol 2007;31:1476-89.
[21] Clay MR, Martinez AP, Weiss SW, Edgar MA. MDM2 and CDK4 im-
munohistochemistry: should it be used in problematic differentiated lipo-
matous tumors?: A New Perspective. Am J Surg Pathol 2016;40:1647-52.
[22] Putri RI, Siregar NC, Siregar B. Overexpression and amplification of
Murine double minute 2 as a diagnostic tool in large lipomatous tumor
and its correlation with Ki67 proliferation index: an institutional experi-
ence. Indian J Pathol Microbiol 2014;57:558-63.
40
[23] Cappellesso R, d'Amore ES, Dall'Igna P, et al. Immunohistochemical
expression of p16 in lipoblastomas. HUM PATHOL 2016;47:64-9.
[24] Roma AA, Magi-Galluzzi C, Zhou M. Differential expression of melano-
cytic markers in myoid, lipomatous, and vascular components of renal
angiomyolipomas. Arch Pathol Lab Med 2007;131:122-5.
[25] Weaver J, Goldblum JR, Turner S, et al. Detection of MDM2 gene am-
plification or protein expression distinguishes sclerosing mesenteritis
and retroperitoneal fibrosis from inflammatory well-differentiated lipo-
sarcoma. Mod Pathol 2009;22:66-70.
[26] Creytens D. A subset of fat-predominant angiomyolipomas label for
MDM2: a potential diagnostic pitfall; comment on Asch-Kendrick
et al. 2016. HUM PATHOL 2017;64:232-3.
X. Lin et al.
[27] Ng W, Messiou C, Smith M, Thway K. p16 expression in fat necro-
sis: a potential diagnostic pitfall in the evaluation of differenti-
ated adipocytic neoplasms. Int J Surg Pathol 2015;23:544-8.
[28] Creytens D, van Gorp J, Ferdinande L, Speel EJ, Libbrecht L. Detection
of MDM2/CDK4 amplification in lipomatous soft tissue tumors from
formalin-fixed, paraffin-embedded tissue: comparison of multiplex ligation-
dependent probe amplification (MLPA) and fluorescence in situ hybridiza-
tion (FISH). Appl Immunohistochem Mol Morphol 2015;23:126-33.
[29] Sasaki T, Ogose A, Kawashima H, et al. Real-time polymerase chain re-
action analysis of MDM2 and CDK4 expression using total RNA from
core-needle biopsies is useful for diagnosing adipocytic tumors. BMC
Cancer 2014;14:468.