Professional Documents
Culture Documents
Liposarcoma is the most common soft tissue sarcoma and accounts for approxi-
mately 20% of all mesenchymal malignancies. In the last decade, the results of
several studies have led to the delineation of new variants as well as to the
introduction of new concepts, mainly as a result of the fruitful interactions
between genetics and pathology. Spindle cell liposarcoma represents an uncommon
variant of well-differentiated liposarcoma. It tends to occur in adults and often
involves the subcutaneous soft tissue. However, from the observation of a larger
number of cases, the anatomic distribution of spindle cell liposarcoma seems to be
comparable to that of the other well-differentiated liposarcoma subtypes. Spindle
cell liposarcoma tends to recur locally and may dedifferentiate. Morphologically it
is composed of a fairly bland neural-like spindle cell proliferation set in a fibrous
and/or myxoid background and is associated with an atypical lipomatous compo-
nent. Great debate has been generated by the introduction of the term atypical
lipoma to emphasize the fact that well-differentiated liposarcoma shows risk of
local recurrence but no potential for metastasis. In our opinion well-differentiated
liposarcoma and atypical lipoma should be considered synonyms that describe
lesions identical both morphologically and kayotypically. Dedifferentiated liposar-
coma is a distinct type of liposarcoma in which transition from low-grade to
high-grade nonlipogenic morphology within a well-differentiated liposarcoma is
observed. The transition usually occurs in an abrupt fashion; however, in rare
cases it can be more gradual. Recently, it also has been proposed that dedifferenti-
ated liposarcoma should be further classified into low and high grade. Dedifferen-
tiated liposarcoma rarely exhibits heterologous (most often myoid) differentia-
tion. A peculiar ‘‘neural-like whorling pattern’’ of dedifferentiation also has been
described recently. Surprisingly, the clinical outcome of dedifferentiated liposar-
coma is less aggressive that in other high-grade pleomorphic sarcomas but genetic
as well as molecular data exist that may partially justify such a discrepancy. Myxoid
and round cell liposarcoma, even if still classified by the World Health Organiza-
tion as two distinct subtypes, share both clinical and morphologic features. Lesions
combining both patterns are very frequent and wide agreement exists in consider-
ing round cell liposarcoma as the high-grade counterpart of myxoid liposarcoma.
Furthermore, myxoid and round cell liposarcoma share the same characteristic
chromosome change. Albeit rare, it has been recently shown that liposarcoma
indeed can occur as a primary skin lesion. It often presents clinically as a
dome-shaped or polypoid lesion that, histologically, most frequently shows high-
grade morphologic features but carries a comparatively good prognosis. Consider-
ing currently available data, the most logical classification of liposarcoma is into
three main groups: (1) well-differentiated liposarcoma (including adipocytic, scle-
rosing, inflammatory, spindle-cell, and dedifferentiated variants), characterized
by ring or long markers chromosomes derived from the long arm of chromosome
12; (2) myxoid and round cell (poorly differentiated myxoid) liposarcoma, charac-
terized in most cases by a reciprocal translocation t(12;16)(q13;p11); and (3)
pleomorphic liposarcoma, characterized by complex karyotypes.
Ann Diagn Pathol 4: 252-266, 2000. Copyright r 2000 by W.B. Saunders Company
retroperitoneal lesions lacking lipoblasts and also sug- transition from low-grade to high-grade nonlipogenic
gested entirely abandoning the term well-differentiated morphology within a well-differentiated liposarcoma is
liposarcoma. More recently, Weiss and Rao16 suggested observed. First described in 1979 by Evans,22 who coopted
limiting use of the term atypical lipomatous tumors to the term from Dhalin et al’s23 description of tumor
subcutaneous lesions. The rationale for such a proposal progression in chondrosarcoma, dedifferentiation in well-
was the belief that liposarcoma, when occurring in the differentiated liposarcoma tends to occur more fre-
subcutis, does not dedifferentiate.16 However, bona fide quently in the primary tumor (90%) but can also be
examples of subcutaneous dedifferentiated liposarcoma observed in recurrences (10%). The transition usually
exist that make this approach less acceptable.8,17-19 The occurs in an abrupt fashion (Fig 8); however, in some
debate around the use of this alternative term to cases this can be more gradual and exceptionally low-
well-differentiated liposarcoma is still far from settle- grade and high-grade areas appear to be intermingled.
ment, with the result that the term atypical lipoma is Dedifferentiated areas exhibit a variable histologic pic-
heterogeneously applied by different institutions. In our ture but most frequently it overlaps with storiform and
opinion well-differentiated liposarcoma and atypical pleomorphic malignant fibrous histiocytoma (Fig 9) as
lipoma should be considered synonyms. As a matter of well as myxofibrosarcoma. Recently, it also has been
fact, both terms describe lesions identical both morpho- proposed that dedifferentiated liposarcoma should be
logically and karyotypically. Their use therefore should further classified into low-grade and high-grade sub-
be determined by the degree of reciprocal comprehen- types.18,24 Low-grade dedifferentiation was defined by
sion between the surgeon and the pathologist to prevent the presence of bland spindle cells organized in a
either inadequate or excessive treatment.20 Impor- fascicular pattern and exhibiting a cellularity intermedi-
tantly, pleomorphic lipoma, which represents an en- ate between well-differentiated sclerosing liposarcoma
tirely benign lesion closely related to spindle cell lipoma, and usual high-grade areas. The very existence of
should be kept separated from the atypical lipoma low-grade dedifferentiation raises the question of the
category as it rarely recurs and has no potential to differential diagnosis with spindle cell liposarcoma.
dedifferentiate. The presence in spindle cell and pleomor- However, it should be noted that spindle cell liposar-
phic lipomas of chromosome 16q rearrangements repre- coma represents a lipogenic lesion (eg, it contains
sents further support of this view.21 atypical adipocytes or lipoblasts) whereas dedifferenti-
ated areas, both low- and high-grade, are generally
Dedifferentiated Liposarcoma
nonlipogenic.
Dedifferentiated liposarcoma represents both a mor- Dedifferentiated liposarcoma may exhibit heterolo-
phologically and biologically fascinating lesion in which gous differentiation in approximately 5% to 10% of
cases, which apparently does not affect the clinical Surprisingly, the biological behavior of dedifferenti-
outcome. Most often the line of heterologous differentia- ated liposarcoma tends to be less aggressive than that of
tion is myogenic (Figs 10-12) or osteo/chondrosarcoma- other types of high-grade pleomorphic sarcomas.17,18 At
tous, but angiosarcomatous elements also have been variance with well-differentiated liposarcoma, dediffer-
reported. Recently, a peculiar ‘‘neural-like’’ or ‘‘meningo- entiation is associated with a 15% to 20% metastatic
thelial-like whorling pattern of dedifferentiation has rate. However, long-term survival rates for retroperito-
been described but both immunohistochemistry and neal dedifferentiated liposarcoma is not significantly
electron microscopy failed to elucidate the line of differ- worse than that observed in ordinary well-differentiated
entiation in these lesions25,26 (Fig 13). This pattern is liposarcoma occurring at the same site, mortality being
often associated with ossification (Fig 14). more related to repeated destructive local recurrences
than to metastatic spread. The fact that dedifferenti- tains the same basic cytogenetic anomalies as well-
ated liposarcoma can recur as an entirely well-differenti- differentiated liposarcoma represented by the presence
ated liposarcoma also represents a morphologic indica- of ring and/or giant marker chromosomes,27 although
tor of the unique biologic behavior of this liposarcoma there may be superimposed additional aberrations.28 A
subtype. significant increase in the level of both MDM2 overex-
Genetics as well as molecular pathology have recently pression and amplification in the high-grade areas has
provided data that, at least in part, may account for the been observed that may account for the tumor progres-
discrepancy observed between morphology and biologic sion in this subset of sarcomas.29,30 However, at variance
aggressiveness. Interestingly, in stark contrast with the with high-grade pleomorphic sarcomas, in which con-
complex karyotypic aberrations observed in pleomor- comitant MDM2 amplification and TP53 mutation are
phic sarcomas, dedifferentiated liposarcoma usually re- detected31 and that are significantly related to poor
clinical outcome, retention of TP53 integrity is almost classified by WHO as two distinct subtypes. How-
always observed in dedifferentiated liposarcoma.29 There- ever, clinical, morphologic, and cytogenetic evidence
fore, dedifferentiated liposarcoma appears to be just one clearly supports the existence of a single spectrum of
more step along the spectrum of well-differentiated lesions in which purely myxoid and round cell lipo-
adipocytic neoplasms, showing only an increased dosage sarcomas represent the well and poorly differentiated
of the same molecular targets involved in the molecular ends, respectively. Clinically, myxoid and round cell
pathogenesis of well-differentiated liposarcoma. liposarcoma tends to occur in the limbs (especially the
thigh) of patients ranging in age between 35 and 55
Myxoid and Round Cell Liposarcoma
years, while in contrast with the well-differentiated
Myxoid and round cell liposarcoma represents approxi- group, the retroperitoneal location seems to be excep-
mately 30% to 35% of liposarcomas. They are still tional.
Purely myxoid liposarcoma is defined by the presence chicken wire) pattern represents one of the most useful
of a hypocellular spindle cell proliferation set in a diagnostic clues when dealing with myxoid sarcomas
myxoid background (Fig 15), often with mucin pooling (Fig 15).
that sometimes is so extensive to produce the so-called The presence of hypercellular areas sometimes featur-
‘‘pulmonary edema’’ pattern (Fig 16). Lipoblasts tend ing an undifferentiated round cell morphology and
to be small and often monovacuolated and to cluster ranging in extent between 5% and 80% permits the
around vessels or at the periphery of the lesion (Fig 17). recognition of the myxoid/round cell liposarcoma sub-
Nuclear pleomorphism is minimal and mitoses are type (Fig 18). Pure round cell liposarcoma is a rare
rare. The presence of distinctive capillary-sized blood neoplasm in which hypercellularity or round cell differ-
vessels organized in a plexiform (eg, crow’s feet, entiation account for more than 80% of tumor tissue
(Fig 19). Most frequently, early foci of hypercellularity cell liposarcoma represents a morphologic continuum of
begins to form in a perivascular distribution. Adipocytic myxoid adipocytic neoplasia. Evans22 was the first to
differentiation in pure round cell liposarcoma is some- suggest abandoning the distinction between these sub-
times hard to appreciate, but it is worth noting that types. Recently, these morphologic observations have
approximately 80% of round liposarcomas exhibit con- gained the conclusive support of genetic analysis. Myx-
vincing S-100 immunopositivity, which represents a oid and round cell liposarcoma share exactly the same
useful diagnostic clue in this context.32 The fact that characteristic chromosome translocation that, at the
transition to hypercellular/round cell areas are com- molecular level, fuses the CHOP gene on 12q13 with the
monly observed in myxoid liposarcoma provides strong FUS (or TLS) gene on 16p11 or with EWS on 22q12.33,34
evidence in favor of the concept that myxoid and round When dealing with poorly differentiated lesions as well
Conclusion
Figure 21. Pleomorphic liposarcoma can be entirely repre-
During the last decade, the integration of morphology sented by a proliferation of pleomorphic monovacuolated as
and genetics has greatly contributed to a more accurate well as multivacuolated lipoblasts.
Liposarcoma: New Entities & Evolving Concepts 265
diagnosis of liposarcoma. Certainly, 135 years after in situ hybridization investigation of ring chromosomes characterizing
Virchow’s42 first detailed description of liposarcoma, a specific pathologic subgroup of adipose tissue tumors. Cancer Genet
Cytogenet 1993;68:85-90
there remain many unsolved problems. The need for 12. Dei Tos AP, Doglioni C, Piccinin S, et al: Coordinated expres-
identification of less-subjective prognostic parameters sion and amplification of the MDM2, CDK4 and HMGI-C genes in
in myxoid liposarcoma represents only one example of atypical lipomatous tumors. J Pathol 2000:190:531-536
such problematic areas. Nonetheless, while waiting for 13. Evans HL, Soule EH, Winkelmann RK: Atypical lipoma, atypi-
further diagnostic advances, it is important to remem- cal intramuscular lipoma, and well differentiated retroperitoneal
liposarcoma. A reappraisal of 30 cases formerly classified as well-
ber that a better knowledge of the mechanisms involved
differentiated liposarcoma. Cancer 1979;43:574-584
in the pathogenesis of adipocytic tumors also may have a 14. Kindblom LG, Angervall L, Fassina AS: Atypical lipoma. AP-
significant impact over the delineation of innovative MIS 1982;90:27-36
therapeutic approaches. This concept is exemplified by 15. Evans HL: Liposarcoma and atypical lipomatous tumors: A
recent studies that have shown the induction of terminal study of 66 cases followed for a minimum of 10 years. Surg Pathol
1988;1:41-54
differentiation on human liposarcoma cells, both in
16. Weiss SW, Rao VK: Well-differentiated liposarcoma (atypical
vitro43 and in vivo,44 by ligands for peroxisome prolifera- lipoma) of deep soft tissue of the extremities, retroperitoneum and
tor-activated receptor and the retinoid X receptor. miscellaneous sites. A follow-up study of 92 cases with analysis of the
However, while looking with enthusiasm to the valu- incidence of dedifferentiation. Am J Surg Pathol 1992;16:1051-1058
able information provided by these new methodologic 17. McCormick D, Mentzel T, Beham A, et al: Dedifferentiated
approaches, it has to be underlined that a consistent as liposarcoma. Clinicopathologic analysis of 32 cases suggesting a better
prognostic subgroup among pleomorphic sarcomas. Am J Surg Pathol
well as accurate morphologic classification still repre- 1994;18:1213-1223
sents the key step in the clinical management of 18. Henricks WH, Chu YC, Goldblum JR, et al: Dedifferentiated
liposarcoma patients. Any effort to confer reproducibil- liposarcoma: A clinicopathologic analysis of 155 cases with proposal for
ity to liposarcoma diagnosis, as well as to reduce at an expanded definition of dedifferentiation. Am J Surg Pathol 1997;21:
minimum the existence of terminologic disagreements, 271-281
19. Yoshikawa H, Ueda T, Mori S: Dedifferentiated liposarcoma of
will prove beneficial to achieve the most effective thera-
the subcutis. Am J Surg Pathol 1996;20:1525-1530
peutic approach. 20. Mentzel T, Fletcher CDM: Lipomatous tumours of soft tissue:
An update. Virchows Arch 1995;427:353-363
21. Mandahl N, Mertens F, Willem H, et al: A new cytogenetic
References subgroup in lipomas: Loss of chromosome 16 material in spindle cell
1. Enzinger RM, Weiss SW (eds): Soft Tissue Tumors (ed 3). St and pleomorphic lipomas. J Cancer Res Clin Oncol 1994;120:707-711
Louis, MO, Mosby, 1995 22. Evans HL: Liposarcoma: A study of 55 cases with a reassess-
2. Fletcher CDM: Soft tissue tumors, in Fletcher CDM (ed). ment of its classification. Am J Surg Pathol 1979;3:507-523
Diagnostic Histopathology of Tumors. Edinburgh, UK, Churchill 23. Dhalin DD, Unni KK, Matsuno T: Malignant (fibrous) histiocy-
Livingstone, 1995, pp 1043-1096 toma of bone: Fact or fancy? Cancer 1977;39:1509-1516
3. Weiss SW (ed): Histologic Typing of Soft Tissue Tumors. World 24. Elgar F, Goldblum JR: Well-differentiated liposarcoma of the
Health Organization Histological Classification of Tumors. Berlin, retroperitoneum: A clinicopathologic analysis of 20 cases, with particu-
Germany, Springer, 1994 lar attention to the extent of low-grade dedifferentiation. Mod Pathol
4. Enzinger FM, Winslow DJ: Liposarcoma. A study of 103 cases. 1997;10:113-120
Virchows Arch Pathol Anat 1962;335:367-388 25. Nascimento AG, Kurtin PJ, Guillou L, et al: Dedifferentiated
5. Stout AP: Liposarcoma: The malignant tumor of lipoblasts. Ann liposarcoma. A report of nine cases with a peculiar neurallike whorling
Surg 1944;119:86-107 pattern associated with metaplastic bone formation. Am J Surg Pathol
6. Kraus MD, Guillou L, Fletcher CDM: Well-differentiated inflam- 1998;22:945-955
matory liposarcoma: An uncommon and easily overlooked variant of a 26. Fanburg-Smith JC, Mietinen M: Liposarcoma with meningothe-
common sarcoma. Am J Surg Pathol 1997;21:518-527 lial-like whorls: A study of 17 cases of a distinctive histological pattern
7. Argani P, Facchetti F, Inghirami G, et al: Lymphocyte-rich associated with dedifferentiated liposarcoma. Histopathology 1998;33:
well-differentiated liposarcoma: Report of nine cases. Am J Surg 414-424
Pathol 1997;21:884-895 27. Fletcher CDM, Akerman M, Dal Cin P, et al: Correlation
8. Dei Tos AP, Mentzel T, Newman PL, et al: Spindle cell between clinicopathological features and karyotype in lipomatous
liposarcoma: A hitherto unrecognized variant of well-differentiated tumors. Am J Pathol 1996;148:623-630
liposarcoma: Analysis of six cases. Am J Surg Pathol 1994;18:913-921 28. Mertens F, Fletcher CDM, Dal Cin P, et al: Cytogenetic
9. Nascimento AG: Spindle cell liposarcoma. Adv Anat Pathol analysis of 46 pleomorphic soft tissue sarcomas and correlation with
1995;2:320-325 morphologic and clinical features: A report of the CHAMP study
10. Rosai J, Akerman M, Dal Cin P, et al: Combined morphologic group. Chromosomes and morphology. Genes Chromosomes Cancer
and karyotypic study of 59 atypical lipomatous tumors: Evaluation of 1998;22:16-25
their relationship and differential diagnosis with other adipose tissue 29. Dei Tos AP, Doglioni C, Piccinin S, et al: Molecular abnormali-
tumors. Am J Surg Pathol 1996;20:1182-1189 ties of the p53 pathway in dedifferentiated liposarcoma. J Pathol
11. Dal Cin P, Kools P, Sciot R, et al: Cytogenetic and fluorescence 1997;181:8-13
266 Angelo P. Dei Tos
30. Nakayama T, Toguchida, Wadayama B, et al: mdm2 gene logic spectrum of myxoid and round cell liposarcoma. A study of 95
amplification in bone and soft tissue tumours: Association with cases. Cancer 1996;77:1450-1458
tumour progression in differentiated adipose tissue tumours. Int J 38. Smith TA, Easley KA, Goldblum JR: Myxoid/round cell liposar-
Cancer 1995;64:342-346 coma of the extremities. A clinicopathologic study of 29 cases with
31. Cordon Cardo C, Latres E, Drobnjac M, et al: Molecular particular attention to extent of round cell liposarcoma. Am J Surg
abnormalities of MDM2 and p53 genes in adult soft tissue sarcomas. Pathol 1996;20:171-180
Cancer Res 1994;54:794-799 39. Fletcher CDM: Will we ever reliably predict prognosis in a
32. Dei Tos AP, Wadden C, Fletcher CDM: S-100 protein staining patient with myxoid and round cell liposarcoma? Adv Anat Pathol
in liposarcoma. Its diagnostic utility in the high grade myxoid (round 1997;2:108-113
40. Mentzel T, Bosemberg M, Fletcher CDM: Pleomorphic liposar-
cell) variant. Appl Immunohistochem 1996;4:95-101
coma: Clinicopathologic and prognostic analysis of 31 cases. Mod
33. Knight JC, Renwick PJ, Dal Cin P, et al: Translocation t(12;
Pathol 1999;12:13A (abstr)
16)(q13;p11) in myxoid liposarcoma and round cell liposarcoma:
41. Dei Tos AP, Mentzel T, Fletcher CDM: Primary liposarcoma of
Molecular and cytogenetic analysis. Cancer Res 1995;55:24-27
the skin: A rare neoplasm with unusual high grade features. Am J
34. Dal Cin P, Sciot R, Panagopoulos I, et al: Additional evidence of
Dermatopathol 1998;20:332-338
a variant translocation t(12;22) with EWS/CHOP fusion in myxoid 42. Virchow R: Myxoma lipomadotes malignum. Virchow Arch
liposarcoma: Clinicopathologic features. J Pathol 1997;182:437-441 Pathol Anat 1865;32:545-546
35. Tallini G, Akerman M, Dal Cin P, et al: Combined morphologic 43. Tontonoz P, Singer S, Froman BM, et al: Terminal differentia-
and karyotypic study of 28 myxoid liposarcomas. Implications for a tion of human liposarcoma cells induced by ligands for peroxisome
revised morphologic typing. A report from the CHAMP group. Am J proliferator-activated receptor gamma and the retinoid ⫻ receptor.
Surg Pathol 1996;20:1047-1055 Proc Natl Acad Sci U S A 1997;94:237-241
36. Dei Tos AP, Doglioni C, Piccinin S, et al: Molecular aberrations 44. Demetri GD, Fletcher CDM, Mueller E, et al: Induction of solid
of the G1-S cell cycle checkpoint in myxoid and round cell liposarcoma. tumor differentiation by the peroxisome proliferator-activated ligand
Am J Pathol 1997;151:1531-1539 troglitazone in patients with liposarcoma. Proc Natl Acad Sci U S A
37. Kilpatrick SE, Doyon J, Choong PFM, et al: The clinicopatho- 1999;96:3951-3956