Open Journal of Pathology, 2012, 2, 143-146
Published Online October 2012 (http://www.SciRP.org/journal/ojpathology)
http://dx.doi.org/10.4236/ojpathology.2012.24026
1
Molecular Particularity in Rare Tumour of Buttock: Case
Report and Literature Review*
El Fatemi Hinde1#, Florence Mishellany2, Pierre Gimbergues3, Frédérique Penault-Llorca2
1Department of Pathology, Hassan II Teaching Hospital, Fez, Morocco; 2Department of Pathology, Centre Jean-Perrin, Cler-
mont-Ferrand, France; 3Department of Surgery, Centre Jean-Perrin, Clermont-Ferrand, France.
Received June 26th, 2012; revised July 27th, 2012; accepted August 17th, 2012
ABSTRACT
Introduction: Hyalinizing spindle cell tumor with giant rosettes (HSCT) is a very uncommon mesenchymal tu mor that
has similar morphological and biological features to the low-grade fibromyxoid sarcoma (LGFMS). Case Report: Re-
ported herein is a case of primary tumour of buttock HSCT that had rare FUS-CREB3L1 fusion transcripts, a product of
characteristic chromosomal ab normality t (7, 16) (q33, p11) of HSCT and LG FMS. The patient was a 48-year-old man
who had a large solitary mass in the butto ck. Histologically, it was composed of b land spindle cells with variable cellu-
larity deposited in a densely hyalinized stroma alternating with myxoid areas. Characteristic collagen rosettes were
scattered in the cellular areas. Reverse transcription-polymerase chain reaction (RT-PCR) assay using formalin-fixed,
paraffin-embedded tissue detected FUS-CREB3L1 fusion transcripts. In our knowledge is the second case may display
a variant FUS/CREB3L1 fusion transcript in international literature. Conclusion: LGFMS and HSCT probably have a
wider spectrum of morphologic features than previou sly tho ught, the awareness of wh ich will help patho logists to avo id
diagnostic pitfalls. Demonstration of the t (7, 16) translocation will help to diagnose difficult cases with unusual his-
tologic features.
Keywords: Deep Soft Tissue; Fibromyxoid Sarcoma; Hyalinizing Spindle Cell Tumor; RT-PCR; Fusion Transcripts;
CREB3L1
1. Introduction
Hyalinizing spindle cell tumor with giant rosettes
(HSCTGR) is a recently described low-grade sarcoma
with a deceptively benign histological appearance [1].
Because it has similar clinicopathological features and
biological behavior to that of low-grade fibromyxoid
sarcoma (LGFMS), HSCT might be regarded as a variant
of LGFMS [2,3]. Low-grade fibromyxoid sarcoma
(LGFMS) is a rare soft-tissue tumor with a deceptively
benign histologic appearance affecting predominantly
young adults during the fourth decade of life [1]. Low-
grade fibromyxoid sarcoma has a predilection for in-
volving deep soft tissues of the thigh, inguinal region, or
chest wall, affecting less frequently th e shoulder or axilla
[1]. Local postsurgical recurrence and metastases to
lungs and bone are frequently seen [1]. Additionally,
both tumors have a characteristic FUS-CREB3L2 fusion
gene transcript. A product of reciprocal translocation, t (7,
16) (q33, p11), further supporting this concept [4,5].
HSCT arises in a variety of deep soft tissues, most com-
monly in the extremities. We describe here a new case of
primary tumour of buttock HSCT harboring a molecular
particularity with a rare FUS- CREB3L1 fusion tran-
scripts.
2. Case Report
A 48-year-old man presented with a 13-cm painful left
buttock mass that had been present for 1 year. The mass
was palpable by digital rectal examination but did not
involve the mucosa. A superficial FNAB of the subcuta-
neous mass was performed and diagnosed as “spindle
cell neoplasm, favor low grade sarcoma”. The lesion was
resected with negative surgical margins. Macroscopically,
the tumors were well circumscribed, firm, and yellow-tan.
For microscopic examination, wet formalin-fixed tissue
was available in our case. Microscopically, a pseudo-
capsule surrounded the tumors. The tumors were com-
prised of bland, fibroblastic spindle cells arranged in a
whirling growth pattern within a hyalinized collagen
bundles and a myxoid matrix (Figure 1). The myxoid
areas were characterized by low cellularity with a
prominent capillary network. However, in some areas the
*The authors d eclare no conflict of interest.
#Corresponding author.

Molecular Particularity in Rare Tumour of Buttock: Case Report and Literature Review
144
Figure 1. formalin-fixed tissue with (hematoxylin eosin and
safran coloration). HES × 10. Histologic appearance show-
ing fusocellulaire proliferation of heterogeneous density.
spindle cells were concentrated around blood vessels.
Alternating with the myxoid regions were fibrous areas
with varying degrees of cellularity, including some hy-
pocellular areas comprised nearly entirely of dense col-
lagenous tissue. A large rosette-like structure with hya-
linized stroma was found (Figure 2), which is character-
istic of LGFMS. Immunohistochemistry in our case re-
vealed strong positivity with vimentin only; no staining
was observed with Smooth muscle actin, desmin, protein
S-100, or CD34 in the sections examined. PCR amplify-
cations were performed in a 50-lL reaction volume con-
taining 13 AccuPrime Pfx reaction mix, 1 unit Accu-
Prime Pfx DNA polymerase, 0.3 lM of each of the for-
ward and reverse primers and 200 ng template DNA. The
PCR was run on a PCT-200 DNA Engine. The cycling
included an initial denaturation at 958C. For 2 min, fol-
lowed by 30 cycles of 15 s at 958C, 30 s at 588C, and 2
min at 688C, and a final extension for 5 min at 728C. All
ligations were performed overnight at 168C in 10-lL re-
action volume containing 13 Ligase Reaction Buffer, 5
units T4 DNA ligase, and 1:3 vector to insert ratio. When
purification was required, the DNA fragments were puri-
fied using either the QIAquick gel extraction kit or the
QIAquick PCR purification kit. For sequence analysis,
the ABI Prism BigDye terminator v1.1 cycle sequencing
kit was used and the products were analyzed on an Ap-
plied Biosystems Model 3100-Avant DNA sequencing
system. RT-PCR demonstrated the FUS-CREB3L1 fu-
sion transcripts resulting from the reciprocal transloca-
tion, t (7, 16) (q33, p11). The patient received no addi-
tional therapy. Clinical fo llow-up, including radiographic
studies, had not detected recurrent tumor at 3 years of
follow-up.
3. Discussion
The HSCTGR was first reported in a series of 19 cases
Figure 2. Formalin-fixed tissue with (hematoxylin eosin and
safran coloration). HES × 40. A large rosette-like structure
with hyalinized stroma.
culled from a large soft tissue consultation service by
Lane et al. in 1997 [2]. Clinically, HSCTGR can develop
at almost any age (average age, 38 year) in the deep soft
tissues of the extremities, particularly the thigh [2,6].
Other sites of involvement are the chest wall, axilla,
rarely buttock, and the [2]. The most common symptom
is a painless, deep-seated, slowly enlarging mass [2,6].
Grossly, the tumor is an oval multilobulated mass, rang-
ing in size from 2 to 20 cm in diameter. Although most
of the lesions appear well circumscribed, they can exten-
sively infiltrate the surrounding soft tissue [6,7]. A prior
report on the imaging characteristics of LGFMS de-
scribed a heterogeneously hyperechogenic multi-nodular
sonographic appearance, a heterogeneous MR imaging
appearance with low to slightly high SI on T1-weighted
images, heterogeneously low to high SI on T2-weighted
images, and heterogeneous postcontrast enhancement [3].
The cut surface has a whorled white-tan appearance.
Cystic degeneration is an uncommon finding [7]. His-
tologically, the tumor cells have a deceptively bland ap-
pearance, consisting of short, ill-defined fascicles of
fusiform to spindled cells with minimal atypia and barely
perceptible levels of mitotic activity [6 ]. In a small num-
ber of cases there may be foci that are more cellular and
atypical, which would mimic an intermediate grade fi-
brosarcoma [3,6]. The mitotic figures were difficult to
identify, usually with less than one mitosis per 50
high-power fields [6]. The most characteristic feature
was the presence of a variable number of large rosettes
like structures. These structures either occupy only a
small portion of the lesion or are so prominent as to ob-
scure the other features of the neoplasm [2,6]. The ro-
settes, which tended to cluster, were made up of a central
collagenous core flanked by an irregular rim of rounded
neoplastic cells. The immunohistochemical profile, espe-

Molecular Particularity in Rare Tumour of Buttock: Case Report and Literature Review 145
cially in HSCTGR, has suggested a neural phenotype,
with the cells forming the rosettes being Leu-7, S-100,
and pgp 9.5 positive and the spindle cells usually being
negative [2,6,7]. The significance of the rosettes is un-
clear. However, cells forming rosettes have the ultra-
structural features of fibroblasts [6]. Moreover, they ap-
pear to be part of the neoplasm rather than a necrobiotic
granuloma-like response within the tumor [6]. Similar
appearing giant collagen-containing rosettes may be seen
in other mesenchymal tumors, including a neuroblas-
toma-like neurilemmoma, in which the rosettes are com-
prised of a core of collagen, flanked by small, round,
differentiated Schwann cells [6]. The microscopic dif-
ferential diagnoses of LGFMS include low-grade myxo-
fibrosarcoma, myxoid n eurofibroma, and myxo id solitary
fibrous tumor. Low-grade myxofibrosarcoma has more
cellular atypia and less swirling of tumor cells in a uni-
formly myxoid stroma. Myxoid neurofibroma shows
more wavy nuclei of the spindle cells and strong S-100
positivity. The osteoid, in rare cases of osteosarcoma,
may be confused for the collagen rosettes. Calcification
in the material and the surrounding marked nuclear
pleomorphism will help in settling the issue [2,8]. In the
original report by Lane et al., follow-up information was
available for 12 of the 19 patien ts, with a mean fo llow-up
period of 39 months [2,6]. Of the 12 patients, 7 were
treated with a simple local excision and 5 by a wide ex-
cision. Although one of the patients treated by simple
excision developed a local recurrence 20 months after the
initial surgery, none of the remaining patients had a re-
current or metastatic disease within the follow-up period.
Therefore, this lesion was considered to be a low-grade
sarcoma, and a wide surgical excision is the treatment of
choice [6]. Though the lesion appears bland in morphol-
ogy, long-term follow-up has revealed that it has a poten-
tial for metastasis, especially to the lungs [3,7,9]. Im-
proved recognition and treatment have improved the
prognosis of the lesion but, nevertheless, prolonged fol-
low-up is necessary. The discovery that LGFMS has a
specific translocation t(7, 16) (q33, p11), or in rare cases
t(11, 16) (p11, p11), has greatly facilitated the diagnosis
[10]. Through these translocations, the chimeric genes
FUS-CREB3L2 or FUS-CREB3L1, respectively, are
created. At the molecular level, the 5’-part of FUS, en-
coding a transactivation domain, is fused to the 3’-part of
CREB3L2 or CREB3L1, encoding a basic leucine zipper
(bZIP) DNA-binding domain [10]. A subset of LGFMS
cases expresses the FUS-CREB3L2 fusion transcript but
lacks the typical t (7, 16) and instead harbors a supernu-
merary ring chromosome, which may contain the fusion
gene, as the sole aberration [1,6]. Karyotypic information
on LGFMS reveals few other recurrent aberrations, sug-
gesting that the chromosomal translocations are tumori-
genic events.
In conclusion, we report a case of primary HSCT aris-
ing in the buttock harboring a molecular particularity
with a rare FUS-CREB3L1 fusion transcripts. The rec-
ognition of typical histological features and behavior of
HSCT of the soft tissue would be helpful for the man-
agement of patients, and detection of the characteristic
FUS-CREB3L2 fusion transcript or rarely FUS-CREB-
3L1 is necessary for the differential diagnosis of HSCT
involving unusual site. Therefore, an accurate tissue di-
agnosis with wide surgical excision and a close fol-
low-up are essential.
4. Acknowledgements
The authors thank Prof . J. M. Coindre for its opinion and
diagnostic follow-up study in molecular biology.
REFERENCES
[1] G. P. Nielsen, M. K. Selig, J. X. O’Connell, S. B. Keel, G.
R. Dickersin and A. E. Rosenberg, “Hyalinizing Spindle
Cell Tumor with Giant Rosettes: A Report of Three Cases
with Ultrastructural Analysis,” American Journal of Sur-
gical Pathology, Vol. 23, No. 10, 1999, pp. 1227-1232.
[2] K. L. Lane, R. J. Shannon and S. W. Weiss, “Hyalinizing
Spindle Cell Tumor with Giant Rosettes. A Distinctive
Tumor Closely Resembling Low-Grade Fibromyxoid Sar-
coma,” American Journal of Surgical Pathology, Vol. 21,
No. 12, 1997, pp. 1481-1488.
[3] A. L. Folpe, K. L. Lane, G. Paull and S. W. Weiss,
“Low-Grade Fibromyxoid Sarcoma and Hyalinizing Spin-
dle Cell Tumor with Giant Rosettes. A Clinicopathologic
Study of 73 Cases Supporting Their Identity and Assess-
ing the Impact of High-Grade Areas,” American Journal
of Surgical Pathology, Vol. 24, No. 10, 2000, pp. 1353-
1360.
[4] R. Reid, C. Silva, L. Paterson, E. Ryan and C. Fisher,
“Low-Grade Fibromyxoid Sarcoma and Hyalinizing
Spindle Cell Tumor with Giant Rosettes Share a Common
t(7, 16)(q34, p11) Translocation,” American Journal of
Surgical Pathology, Vol. 27, No. 9, 2003, pp. 1229-1236.
[5] A. Matsuyama, M. Hisaoka, S. Shimajiri, et al., “Mo-
lecular Detection of FUS-CREB3L2 Fusion Transcripts
in Low-Grade Fibromyxoid Sarcoma Using Formalin-
Fixed, Paraffin-Embedded Tissue Specimens,” American
Journal of Surgical Pathology, Vol. 30, No. 9, 2006, pp.
1077-1084.
[6] S. W. Weiss and J. R. Goldblum, “Enzinger and Weiss’s
Soft Tissue Tumors,” 4th Edition, Mosby, St. Louis, 2001,
pp. 431-435.
[7] S. E. Vernon and P. A. Bejarano, “Low Grade Fibro-
myxoid Sarcoma: A Brief Review,” Archives of Pathol-
ogy & Laboratory Medicine, Vol. 130, No. 9, 2006, pp.
1358-1360.
[8] T. Hansen, K. Katenkamp, M. Brodhun and D. Katenk-
amp, “Low-Grade Fibrosarcoma—Report on 39 Not Oth-
erwise Specified Cases and Comparison with Defined
Low-Grade Fibrosarcoma Types,” Histopathology, Vol.

Molecular Particularity in Rare Tumour of Buttock: Case Report and Literature Review
146
39, No. 4, 2006, pp. 152-160.
[9] J. M. Woodruff, C. R. Antonescu, R. A. Erlandson and P.
J. Boland, “Low Grade Fibrosarcoma with Palisaded
Granulomalike Bodies (Giant Rosettes): Re port of a Case
That Metastasized,” American Journal of Surgical Pa-
thology, Vol. 23, 1999, pp. 1423-1428.
doi:10.1097/00000478-199911000-00015
[10] F. Mertens, C. D. Fletcher, C. R. Antonescu, J. M. Coin-
dre, M. Colecchia, et al., “Clinicopathologic and Mo-
lecular Genetic Characterization of Low-Grade Fibro-
myxoid Sarcoma, and Cloning of a Novel FUS/CREB3L1
Fusion Gene,” Laboratory Investigation, Vol. 85, No. 3,
2005, pp. 408-415. doi:10.1038/labinvest.3700230