| Abstract|| |
Background: De-differentiated liposarcomas (DDLSs) are uncommon tumors with a wide histopathological spectrum. Materials and Methods: Over an 8-year period (2003-2011), 25 DDLSs, after review, were critically analyzed for histopathological features. Results: Twenty-five tumors, in 14 men and 11 women (M: F = 1.2:1), with age range of 22-88 years (mean, 53.9), occurred in retroperitoneum (14) (56%), thigh (3) (12%), paratesticular region (3) (12%), chest wall (2) (8%), leg (1) (4%), shoulder (1) (4%), and groin (1) (4%). Tumor size (21 tumors) varied from 7.5 to 25 cm (mean, 17.5). Histopathologically, DD component was high grade in 19 (76%) and low grade in 6 (24%) tumors. Whereas the most common WD component was adipocytic type; the most common DD component was pleomorphic sarcomatous (13) (52%), followed by myxofibrosarcomatous (MFS)-type (6) (24%). Low-grade DD components included MFS (2), fibrosarcoma (2), myogenic/myofibroblastic type (1), and IMFT-type (1). Three tumors displayed meningothelial-like whorls and metaplastic bone formation. Heterologous elements, noted in 11 (44%) tumors, included bone (8) and rhabdomyoblastic differentiation (2). Two tumors displayed homologous differentiation, reinforced with MDM2 staining. S100-P was diffusely positive in WD components (5/7) and focally in DD components (2/9). All patients were treated with surgery, including 10, who underwent adjuvant radiotherapy. Outcomes (16 patients, 64%), over 1-48 months included 10 patients free of disease, 4 died of disease, and 2 patients alive with disease. Conclusions: This study forms the largest documentation of DDLSs, including its wide histopathological spectrum, from our country. Rare cases overlap with pleomorphic liposarcoma. S100-P and MDM2 are useful in substantiating adipocytic differentiation, especially in selected cases. Analysis of adequate tumor sections is vital for correct identification of a DDLS. Surgical excision with adjuvant RT forms optimal treatment.
Keywords: Dedifferentiated liposarcoma, liposarcoma, MDM2, myxofibrosarcoma, soft tissue tumors, well differentiated liposarcoma
|How to cite this article:|
Rekhi B, Navale P, Jambhekar N A. Critical histopathological analysis of 25 dedifferentiated liposarcomas, including uncommon variants, reviewed at a Tertiary Cancer Referral Center. Indian J Pathol Microbiol 2012;55:294-302
|How to cite this URL:|
Rekhi B, Navale P, Jambhekar N A. Critical histopathological analysis of 25 dedifferentiated liposarcomas, including uncommon variants, reviewed at a Tertiary Cancer Referral Center. Indian J Pathol Microbiol [serial online] 2012 [cited 2019 Dec 11];55:294-302. Available from: http://www.ijpmonline.org/text.asp?2012/55/3/294/101733
| Introduction|| |
Liposarcoma is a soft tissue sarcoma of adiopocytic origin, with various clinicopathological subtypes, some of which are characterized by distinct molecular/cytogenetic abnormalities, including a well-differentiated liposarcoma (WDLS)/atypical lipomatous tumor, dedifferentiated liposarcoma (DDLS), and myxoid/round cell liposarcoma. The other subtypes include pleomorphic liposarcoma and mixed liposarcoma. 
According to the western literature, liposarcoma is the most common malignant mesenchymal tumor and constitutes 20% of these tumors.  Among its various histological subtypes, myxoid type is the most common, followed by a WDLS with or without dedifferentiated (DD) component (25%): round cell/myxoid type (15%) and pleomorphic liposarcoma (10%).  DDLS occurs in 15% WDLS.  The concept of tumor dedifferentiation was introduced by Dahlin and Beabout  in 1975 in a low-grade chondrosarcoma. It is defined by the presence of an undifferentiated tumor component in proximity of a differentiated tumor, invariably of low-grade or of borderline malignant type. Subsequently, a DDLS was first described by Evans,  in 1979, as a combination of an atypical lipomatous tumor in juxtaposition to a "non-lipogenic" sarcomatous component, invariably of high grade with proclivity for tumor dissemination and therefore a much worse prognosis, in contrast to a WDLS that has minimal metastatic potential.  Following year, Shimoda et al.  documented liposarcoma with malignant fibrous histiocytoma (MFH)-like pattern. Thereafter, there have been studies, mostly from the west, describing clinicopathological features of a DDLS, unraveling its wide histopathological spectrum. Besides, a conventional high-grade DD component resembling a pleomorphic sarcoma/malignant fibrous histiocytoma (MFH) in a DDLS, other described DD components include low-grade dedifferentiation, resembling fibromatosis or well-differentiated fibrosarcoma; myxofibrosarcomatous dedifferentiation; rhabdomyofibrosarcomatous dedifferentiation; DDLS with a peculiar "neural-like" or "meningothelial-like" whorling associated with metaplastic bone formation; myogenic dedifferentiation, and lately, inflammatory myofibroblastic tumor (IMFT)-like dedifferentiation. ,,,,,,
Recent studies have shown that some DDLSs can show lipoblastic differentiation in the DD, high-grade component, resulting in areas indistinguishable from pleomorphic LPS, leading to consideration of revision of initial criteria for its diagnosis. , Besides, a diagnostic challenge, as a result of this varied spectrum, there is a therapeutic challenge in treatment of a DDLS with complete excision, especially in sites like retroperitoneum. Identification of various histological types within the DD component and marginal status of excised DDLSs has been observed to have prognostic relevance.  Till date, there have been occasional studies on clinicopathological features of DDLs from our continent. 
The purpose of this study was to critically review series of DDLSs diagnosed at our center with regards to its highly evolved histopathological spectrum, since its inception. This forms the largest documented series of this tumor from our country and one of the few ones from our continent. 
| Materials and Methods|| |
Cases that were registered and or diagnosed as DDLS at our Hospital, over an 8-year period (2003-11), were included in the present study. The study material was in form of 18 (78.2%) surgical excision specimens (performed at our hospital, as well as referred from other Centers); 5 cases in the form of slides and paraffin blocks and a single case in the form of stained slides, that were representative of the surgical excisions performed at other centers. Clinical details were procured from the patient case files and across EMR (Electronic Medical Records). Radiological findings were obtained from Picture Archives and Communication System at Centricity (GE Healthcare, USA).
Surgical resection specimens were defined as R0 (wide excision with gross and microscopically free margins), R1 (marginal excision with grossly negative and microscopically close margin, less than 1 cm), and R2 (intracapsular excision with grossly and microscopically or only microscopically positive margins). Resection specimens with unavailable marginal status were labeled as Rx.
Hematoxylin and eosin (H & E) stained slides were available in all 25 cases. The number of tumor sections analyzed in every case was 24, 10, 30, 17, 11, 15, 8, 30, 14, 10, 19, 11, 21, 16, 12, 15, 5, 25, 21, 9, 10, 3, 22, 5, and 18, with an average nearing 15 slides per case. The various types of well-differentiated (WD) liposarcomatous components in every tumor were recorded. Dedifferentiation was histologically defined as a "non-lipogenic" tumor area occupying at least one low-power (x 10 objective) microscopic field in the proximity of a well-differentiated liposarcoma (WDLS). Further, DD tumor areas were defined as those with excess of tumor cellularity than that encountered in sclerosing WDLS, but varied in the degree of pleomorphism and the level of mitotic activity. These areas were defined as low grade when the cellularity approached that of a fibromatosis, low-grade fibrosarcoma or a low-grade myxofibrosarcoma and the nuclear atypia and mitotic activity were low, whereas high-grade DD areas were defined as those that revealed moderate to marked cellularity and pleomorphism and mitoses resembling high-grade fibrosarcoma or a high-grade pleomorphic sarcoma. ,,,,, Higher grade of the DD component was considered.
Out of the 34 cases initially reviewed by two authors (BR with PN), as per established histopathological criteria, 9 cases were excluded that comprised 3 pleomorphic liposarcomas, 2 spindle cell tumors (not otherwise classified) and a single pleomorphic sarcoma, myxofibrosarcoma, osteosarcoma, epithelioid leiomyosarcoma, respectively. ,,,,, Finally, 23 cases were included in the study.
Immunohistochemical (IHC) staining was performed by the immunoperoxidase method using the MACH 2 Universal HRP-Polymer Detection Kit (Biocare, USA), including 3′-3′-diaminobenzidine tetrahydrochloride (DAB) as the chromogen. Antibody markers performed in various cases are listed in [Table 1].
Treatment details were available in all 25 cases and outcomes were available in 16 (64%) cases.
| Results|| |
Twenty-five tumors occurred in 14 men and 11 women (M: F = 1.2:1) within age range of 28-88 years (mean 53.9 and median 54). Site-wise, 14 tumors (56%) occurred in the retroperitoneum and 11 (44%) at extra-abdominal sites, including thigh (3), chest wall (2), paratesticular region (3), leg (1), shoulder (1), and groin (1). Tumor (T) size, in 21 (84%) cases, varied from 7.5 to 25 cm (mean, 17.5, median 15). Most tumors (24) presented as de novo or primary DDLSs, whereas a single case of secondary DDLS developed in the setting of an existing WDLS after an interval of 21 months.
On gross examination, the tumors revealed a multinodular external surface. Cut surfaces were fleshy with variable grey-white and yellow areas.
On microscopic examination, WD component revealed various subtypes of WDLS in varying proportions, including adipocytic (lipoma-like) (22), sclerosing (11), inflammatory (3) and spindle cell type (2), with more than one subtype in most tumors (12,48%). The adipocytic type of WDLS comprised sheets of mature fat cells and variable number of lipoblasts; sclerosing type revealed broad fibrous septa and areas of hyalinisation with stromal atypical cells and lipoblasts; inflammatory showed the presence of extensive chronic inflammatory infiltrate in the form of lymphoplasmacytic aggregates and spindle cell type showed bland spindle cells with a neural appearance in a fibro-myxoid background.
The interface between WD and DD areas was abrupt in most cases (18, 72%), while gradual in some (7, 28%), including occasional tumors with mosaic pattern. Perivascular whorls were observed in 7 (28%) tumors.
The DDLS component was of high-grade in 19 (76%) tumors and low grade in 6 (24%) tumors. High-grade DDLS included pleomorphic sarcoma, including malignant fibrous histiocytoma (13), myxofibrosarcoma (4), and fibrosarcoma (2). DD component in 6 low-grade DDLS comprised myxofibrosarcoma (2), low-grade fibrosarcoma (2), sarcoma with myogenic/myofibroblastic differentiation (1) and inflammatory myofibroblastic (IMFT)-like (1).
Three tumors exhibited peculiar meningothelial-like whorling and metaplastic bone formation with 2 of those exhibiting low-grade myxofibrosarcoma and the other displaying fibrosarcoma.
Heterologous bone formation was noted in 8 tumors, rhabdomyosarcomatous differentiation in 2, and myogenic dedifferentiation in a single tumor.
Two tumors, including a recently diagnosed one, displayed distinct lipoblasts (homologous dedifferentiation) within the DD component.
Among 25 DDLSs, 5 tumors were initially "misclassified" as myxoid liposarcomas (2), embryonal rhabdomyosarcoma (1), follicular dendritic sarcoma versus a gastrointestinal stromal tumor (GIST) (1), and as unremarkable fibrocollagenous tissue (1), the latter on review was diagnosed as WDLS. The reasons included misinterpretation, including in the latter tumor, wherein WDLS component was missed on a biopsy, but the final diagnosis was accomplished on the resected specimen that included DDLS component.
Immunohistochemistry (IHC) was performed in 13 (52%) tumors. S100-P expression was analyzed in 13 tumors. It was positive in WD components of 6/8 tumors, including a single tumor showing negative staining, but CD34 positive staining. It was negative in DD components of seven tumors, but displayed focal positivity two tumors that contained lipoblasts, reinforcing homolgous dedifferentiation. Desmin was positive in 4/11 tumors, including two tumors with rhabdomyoblastic dedifferentiation that additionally showed focal MyoD1positivity. SMA was positive in 2/8 tumors, CD68 in 4/4 tumors, and CD163 in 2/2 tumors. CD34 was positive in 3/11 tumors, including one where its expression was positive in WD component and negative in DD component (case 11) and the other case where it was positive in both components (case 24). MDM2 staining was performed in two tumors with homologous dedifferentiation, wherein it was diffusely positive in WD areas and focally in DD areas. CKIT/CD117 was negative in all six tumors and h-caldesmon in three tumors, wherever performed. CD21 and CD23 were negative in a single tumor (Case 23) [Figure 1],[Figure 2],[Figure 3],[Figure 4],[Figure 5],[Figure 6],[Figure 7],[Figure 8].
|Figure 1: (a) Classical dedifferentiated liposarcoma (DDLS) (Case 5). (a) Interface of well-differentiated liposarcomatous (WDLS) component (adipocytic type) with DD component. (hematoxylin and eosin, ×100). (b) WDLS component comprising lipoblasts (hematoxylin and eosin, ×400). Inset: Distinct lipoblast with atypical nucleus indented on its surfaces by fat vacuoles. (hematoxylin and eosin, ×1000). (c) Dedifferentiated (DD), "non-lipogenic" component comprising highgrade pleomorphic sarcoma with pleomorphic cells and numerous mitotic figures, including atypical forms. (hematoxylin and eosin, ×400). (d) S100 positivity in lipoblasts (arrows) (Case 6). (Diaminobenzidine, ×400)|
Click here to view
|Figure 2: Other variants of WDLS in different cases. (a) (Case 2). WDLS component showing admixture of adipocytic and sclerosing types with peculiar perivascular collagenous whorls. hematoxylin and eosin, ×200). (b) (Case 7). Spindle cell type of WDLS areas with interspersed lipoblasts (arrow). (hematoxylin and eosin, ×200). Inset: lipoblast at high magnification. (hematoxylin and eosin, ×1000). (c) (Case 22). Infl ammatory type WDLS comprising lymphoplasmactyic aggregates. (hematoxylin and eosin, ×200). (d) Interspersed lipoblasts. (hematoxylin and eosin, ×400)|
Click here to view
|Figure 3: DDLS with low-grade DDLS of myxofibrosarcomatous type (Case 11). (a) Multiple nodules of non-lipogenic myxofibrosarcomatos appearance. (hematoxylin and eosin, ×100). (b) Higher magnification showing curvilinear vessels within the myxofibrosarcomatous appearance. (hematoxylin and eosin, ×400). (c) Distinct areas displaying lipoblasts (arrows) reminiscent of WDLS component within the same tumor. (hematoxylin and eosin, ×400). (d) CD34 positivity in WDLS component. (hematoxylin and eosin, ×400). Inset: CD34 positive lipoblast. (Diaminobenzidine, ×1000)|
Click here to view
|Figure 4: Low-grade DDLS with meningothelial like whorls and metaplastic bone formation. (Case 23). (a) Gross appearance displaying multinodular tumor. (b) Cut surface showing multilobulated tumor with grey-white and yellowish/"fatiy" areas|
Click here to view
|Figure 5: Microscopic features (Case 23). Distinct WDLS (Left) and DDLS (Right) components. (hematoxylin and eosin, ×100). (b) WDLS showing lipoblasts (arrow). (hematoxylin and eosin, ×400). (c) Low-grade myxofi brosarcomatous areas in the DDLS component. (hematoxylin and eosin, ×200). (d) Prominent areas showing meningothelial-like whorls. (hematoxylin and eosin, ×200). (e) Metaplastic bone formation in proximity to meningothelial-like whorls. (hematoxylin and eosin, ×200). (f) Diff use CD 68 positivity. (Diaminobenzidine, ×200)|
Click here to view
|Figure 6: DDLS with homologous differentiation (Case 22). (a) Paratesticular tumor with grey-white and yellow areas. (b) WD component (sclerosing type) with atypical cells, including lipoblast (thick arrow) and perivascular whorling (thin arrow). (hematoxylin and eosin, ×400). (c) MDM2 positive atypical cells. Inset: MDM2 positive lipoblast. (Diaminobenzidine, ×400). (d) DDLS with discrete lipoblasts (hematoxylin and eosin, ×200). (e) Pleomorphic sarcomatous DDLS with "mulberry"-shaped lipoblasts (arrow) and mitotic figures (arrow head). (hematoxylin and eosin, ×400). Inset: S100-P positive lipoblast. (Diaminobezidine, ×400). (f) MDM2 positive lipoblasts within DD component. (Diaminobezidine, ×400)|
Click here to view
|Figure 7: DDLS with IMFT-like areas. (Case 12). (a) WDLS. (hematoxylin and eosin, ×200). Inset: lipoblast. (hematoxylin and eosin, ×400). (b) DDLS with IMFT-like areas including lymphoplasmacytic aggregates and myofibroblastic cells. (hematoxylin and eosin, ×200). Inset: myofi broblasts. (hematoxylin and eosin, ×400). (c) Lymphoplasmacytic cells and myofi broblastic cells. (hematoxylin and eosin, ×400). (d) Desmin positivity within myofi broblastic cells. (Diaminobenzidine ×400). Inset: Desmin positive myofibroblastic cells at higher magnification. (Diaminobenzidine ×400)|
Click here to view
|Figure 8: DDLS with heterologous and homologous differentiation. (a) WDLS component. (hematoxylin and eosin, ×200). (b) Rhabdomyosarcomatous differentiation within DD component. (hematoxylin and eosin, ×400). Inset: rhabdomyoblasts at higher magnification. (hematoxylin and eosin, ×1000). (c) Discrete desmin positivity reinforcing muscle differentiation. (Diaminobenzidine ×400). (d) Homologous differentiation within DDLS component. (hematoxylin and eosin, ×200). Inset showing lipoblasts within DDLS. (hematoxylin and eosin, ×1000)|
Click here to view
Therapeutically, all patients underwent surgical excisions. Thirteen patients underwent surgery and 10 underwent surgery and adjuvant radiotherapy (RT); a single patient underwent surgery and adjuvant chemotherapy (CT) and another patient underwent surgery with adjuvant RT and CT.
Tumor excisions from metastatic specimens, in those cases, displayed DD areas.
Out of 25 surgical excisions, 6 patients underwent R0, 4 underwent R2, 2 underwent R1, and 13 patients underwent Rx excisions, the latter mostly operated at centers, elsewhere. Out of 6 R0 excisions, 5 were extra-abdominal and a single excision was that of a retroperitoneal tumor.
Outcome details were available for 16 patients (64%) over a period ranging from 1 to 48 months (Mean 20.9 and Median 22.5 months). Tumor recurrences occurred in eight patients and metastasis in five, most commonly to the lungs. Finally, four patients died of disease (DOD) (1-24 months), including a single case who presented with metastasis and died within a month. Ten patients were free of disease (FOD) (11-48 months) and two patients were alive with disease (AWD) (12 and 21 months). Out of four patients with low-grade DD component, three developed tumor recurrences (AWD), a single patient DOD and two patients have been FOD for 11 and 25 months, respectively [Table 2].
|Table 2: Clinicopathological features of 25 dediff erentiated liposarcomas|
Click here to view
| Discussion|| |
The present study testifies earlier documented clinicopathological variables of a DDLS, such as male predisposition, older age, retroperitoneum as its most common site; uncommon sites like paratesticular region and an invariably larger tumor size, in different studies. ,,, Most tumors presented as de novo, except a single case that secondarily developed over a previously existing WDLS. ,
The focus of this case series is to share a wide histopathological spectrum within a DDLS, in concert with various studies till date. ,,,,,,,,,,,,,,,, On histopathology, the transition between the WD and DD components was mostly abrupt, uncommonly gradual, especially with sclerosing type of WD component and rarely mosaic, as noted earlier. ,, The most commonly observed WD component was adipocytic, followed by sclerosing and inflammatory types.  The DD components included 5 low-grade and 18 high-grade DD tumor types, including pleomorphic high-grade sarcoma/MFH that was most common, followed by myxofibrosarcoma. ,,,,, Earlier, Henricks et al. documented that neither low-grade dedifferentiation, nor low-percentage of dedifferentiation were associated with an improved outcome. In the present study, behavior of a low-grade was similar as that of high-grade DDLSs. Singer et al. documented a six-fold increased risk of death in cases with DDLS, in contrast to WDLS. Dei Tos  and McCormick et al. pointed that clinical outcome of DDLS is relatively better among other high-grade pleomorphic sarcomas, whereas Henricks et al.  observed its similar clinical behavior as that of other pleomorphic sarcomas. In a way, these results underpin the value of identification of DD components, including low-grade types that can otherwise be "missed," resulting in diagnosis of "mere" WDLS or myxoid liposarcoma. At the same time, the value of adequate tumor sampling in these tumors cannot be overemphasized, considering these tumors are heterogeneous. Noteworthy, in cases of retroperitoneal tumors with a biopsy diagnosis of pleomorphic sarcoma/MFH, adequate sampling should be done, especially from the adjacent adipocytic component of the excised tumor mass that could unravel a WDLS, leading to a timely, correct identification of a DDLS. At the same time, a WDLS should be adequately sampled and diligently analyzed with knowledge of wide spectrum of DD components so that the DD component is not missed. Most of our study tumors were excision specimens, with an average of 15 tumor sections studied per case.
Six low-grade DDLSs in the present study comprised areas of low-grade myxofibrosarcoma (2), low-grade fibrosarcoma (2), sarcoma with myofibroblastic differentiation (1), and IMFT (1), as described in individual reports on these discrete morphological variants. ,, Three tumors displayed peculiar neural-like/meningothelial-like whorling and metaplastic bone, as described by Nascimento et al. and Fanburg-Smith et al., in different studies. Similar to these cases, one of our study cases was initially suspected for a follicular dendritic cell sarcoma that forms the closest mimic of this tumor pattern. However, negativity of CD21, CD35, and CD23 ruled out this diagnosis. Negative EMA staining ruled out meningothelial or perineurial differentiation. CD68 positivity in one of these two tumors was indicative of fibrohistiocytic differentiation. It has been postulated that meningothelial like perivascular whorls represent mesenchymal dedifferentiation that may be myofibroblastic, pericytic, or osteoblastic, but not dendritic, meningothelial, or perineurial. These whorls might represent an early sign of dedifferentiation that should raise suspicion for a DDLS. Ten of our study tumors displayed perivascular whorls, especially with sclerosing WDLS.
The other uncommon components in two tumors included myofibroblastic/myosarcomatous and IMFT-like. Similar to a recent documentation on the latter subtype of DDLSs, we did not identify ALK positivity, but focal SMA reactivity was noted .  Identification of this subtype can lead to timely identification of DDLS, rather than diagnosis of "only" IMFT. Unlike some authors, we did not objectively identify smooth muscle dedifferentiation within DD components. , H-caldesmon and desmin were negative in two tumors composed of myofibroblasts that were diffusely SMA positive in certain DD areas. Among various IHC markers S100-P was useful in substantiating adipocytic differentiation and was negative in areas of dedifferentiation, except in two tumors. ,
Heterologous elements were noted in 11 tumors (44%), including heterotopic bone formation as the most common form, followed by rhabdomyosarcomatous differentiation. Earlier, Henricks et al. observed heterologous differentiation in 3.87% cases, whereas Hasegawa et al. noted the same in 28% of their study cases. A higher percentage in the present series is attributable to the fact that most cases present at a locally advanced stage, considering ours is a tertiary cancer referral center, catering to most states within the country. At the same time, extensive tissue sampling could have been reasons for identification of more number of tumors with heterologous components. In an earlier study, Evans et al. did not identify any different tumor behavior in DDLSs with or without heterologous differentiation. In the present study, heterologous differentiation was observed, irrespective of DD grade, but was associated with unfavorable events like recurrences, metastasis, and death, in cases wherever follow-up was available.
Two tumors displayed "homologous" adipocytic differentiation within DD component. The lipoblasts within the DD components were further reinforced with MDM2 and S100-P positive staining. In a recent study,  it has been concluded that a subset of pleomorphic liposarcomas might arise through tumor differentiation of WDLS or might represent a transitory differentiated step towards a DDLS. In a similar study, Mariño-EnrÍquez et al. attributed the presence of lipoblasts in the DD component as a "homologous" dedifferentiation that may be included in the spectrum of DDLS and such cases need not to be designated as mixed liposarcomas. In the present study, MDM2 staining was performed for only these two tumors, in order to reinforce lipoblastic differentiation within DDLS components. In view of unavailability, it could not be tested in other tumors that were unequivocal DDLSs. Aleixo et al. documented 90% and 65%, and 70% and 96.3%, sensitivity and specificity of MDM2 and CDK4, respectively, in distinguishing DDLSs from other mesenchymal tumors. In cases with limited biopsies, when the WD component cannot be identified, these IHC markers, as well as application of ancillary techniques like fluorescent in-situ hybridization (FISH) ad real time-PCR, for identifying MDM2-CDK4 amplification can be utilized to substantiate a diagnosis of DDLS from among its other the differentials. , These include gastrointestinal stromal tumor (GIST), lipoleiomyosarcoma, β mesenteric fibromatosis, lipoleiomyosarcoma, and other pleomorphic sarcomas. It is noteworthy that focal CKIT positivity has been documented in 30% DDLS.  In our study, CKIT was negative in six tumors, wherever performed. B-catenin positivity helps in diagnosis of fibromatosis, while lack of definite bands of smooth muscle differentiation traversing across the WDLS ruled out a lipoleiomyosarcoma in a case that showed diffuse SMA positive, but was desmin and h-caldesmon negative.  Occasional studies have shown correlation of increased expression of certain IHC markers like p53 and Ki 67 with tumor progression and relatively aggressive clinical course, respectively. 
Dedifferentiation in liposarcomas is attributable to tumor progression, at least in certain cases, considering the occurrence of areas resembling low-grade MFS imperceptibly blending with high-grade ones; gradual transition in some cases from WDLS to DDLS and homologous dedifferentiation in some of our cases, as well as noted in earlier studies. ,, The stromal cells and lipoblasts within WD components are comparable with spindly sarcomatous elements of DD areas. CD34 (interstitial dendritic fibroblastic cell antigen) positivity in atypical spindle or floret-type lipoblasts in WD components and also within DD components, as noted in two of our study cases, as well as in an earlier study, is suggestive that atypical spindle cell might be precursors for progression into DD component.  Earlier, Rosai et al. demonstrated the presence of supernumerary ring or giant chromosome in DDLS, as noted in WDLS, thereby supporting the proposed "pathogenetic link" between WDLS and DDLS. Recently, Hisaoka et al. have demonstrated overexpression of calreticulin, a major Ca(2+)-buffering protein that can inhibit adipocyte differentiation, in DDLSs and WDLS, in contrast to normal adipocytes, subsequent to a gene expression study by Singer et al. They have suggested that aberrantly expressed calreticulin in a DDLS is involved in its dedifferentiation and/or tumor progression. Further, they have postulated that down-regulation of calreticulin by small-interfering RNA could induce adipogenesis in dedifferentiated liposarcoma cells and reduce cell proliferation. 
Therapeutically, most of our study cases underwent surgery. Adjuvant treatment, mostly in the form of RT, was offered in 11 cases. The primary treatment for DDLSs is surgical excision. Adjuvant RT is offered in cases where complete tumor resection is not possible. The importance of identification of DD component in an otherwise WDLS, especially in locations like retroperitoneum, where complete resection is difficult to achieve, is that adjuvant RT can be added in those cases for better locoregional clearance. Chemotherapy offers no significant benefit except in cases where it is offered as multimodal therapy in few large sized tumors.  In their study on retroperitoneal liposarcomas, Singer et al. concluded that DD histological subtype and need for contiguous organ resection (excluding nephrectomy) was associated with increased risk of local recurrence and metastasis. Further, they postulated that nephrectomy in such cases may be added, but has no measurable influence on disease-specific survival.
In the present study, outcomes were available in 16 patients (64%) (Median, 22.5 months). Tumor recurred in 8 (50%) patients and metastasized in 5 (31.2%), most commonly to the lungs.  Four (25%) patients died of disease (DOD), including a single patient, who presented with an advanced stage and died within a period of 1 month. In different studies, McCormick et al. observed a local recurrence rate of 43.7% and metastatic rate of 12.5%, and Henricks, et al. observed a recurrence rate of 41% and a metastatic rate of 17%, in their series. In the former study, there were 21.2% tumor-related deaths, while in the latter the same was 28%. Both these studies had a longer follow-up, unlike the present study where follow-up was available in limited patients, considering poor patient compliance. Ten patients free of disease (11-48 months) included 7 who underwent surgery (irrespective of complete resection) and adjuvant RT.
To conclude, DDLS is an uncommon sarcoma with a wide histopathological spectrum and an unfavorable clinical course. In every WDLS, especially occurring in retroperitoneum, one should carefully examine for DD component that has a wide spectrum, including of low and high grade types, considering this tumor is aggressive, irrespective of its grade of dedifferentiation. Identification of perivascular whorls, especially in sclerosing and inflammatory WDLS should raise an index of suspicion for identification of DD components. At the same time, tumors diagnosed as pleomorphic or myxoid sarcomas, on biopsy specimens in retroperitoneum should be examined for possibility of an adjacent WDLS, for correct identification of a DDLS. Rare DDLS can show homologous differentiation. S100-P and MDM2 are useful IHC markers for reinforcing adipocytic differentiation within atypical cells, the latter especially in DDLS with homologous differentiation. Dedifferentiation in liposarcomas is a progressive event, in many cases. Surgery with free resection margins is the treatment mainstay. Adjuvant RT seems to offer benefit in cases with incomplete resection. Identification of "newer" gene targets such as calreticulin and their immunohistochemical expression in DDLSs and in WDLSs seems future area of research within DDLSs, both diagnostically and therapeutically.
| References|| |
|1.||Nielsen GP, Mandahl L, Rosenberg AE, Sciot R, Meis-Kindblom JM, Kinblom LG, et al. Chapter 1 Adipocytic tumors. In: Fletcher CDM, Unni K, Mertens F, editors. Tumors of soft tissue and bone. Pathology and genetics. World Health Organization classification of tumors. Lyon: IARC Press; 2002. p. 17-44. |
|2.||Dei Tos AP. Liposarcoma. New entities and evolving concepts. Ann Diagn Pathol 2000;4:252-66. |
|3.||Brooks JSJ. Disorders of soft tissue. In: Sternberg's. editor. Diagnostic surgical pathology. 3 rd edition. Philadelphia PA: Lippincott Williams and Wilkins; 1999. p. 131-221. |
|4.||Weiss SW, Rao VK. Well-differentiated liposarcoma (atypical lipoma) of deep soft tissue of the extermities, retroperitoneum and miscellaneous sites. A follow-up of 92 cases with analysis of the incidence of "dedifferentiation". Am J Surg Pathol 1992;16:1051-8. |
|5.||Dahlin DC, Beabout JW. Dedifferentiation of low-grade chondrosarcomas. Cancer 1971;28:461-6. |
|6.||Evans HL. Liposarcoma: A study of 55 cases with a reassessment of its classification. Am J Surg Pathol 1979;3:507-23. |
|7.||Shimoda T, Yamashita H, Furusato M, Kirino Y, Ishikawa E, Miyagawa A, et al. Light and Electron Microscopic Study with Comments on Their Relation to Malignant Fibrous Histiocytoma and Angiosarcoma. Pathol Int 1980;30:779-97. |
|8.||McCormick D, Metzel T, Beham A, Fletcher CDM. Dedifferentiated liposarcoma: Clinicopathologic analysis of 32 cases suggesting a better prognostic group among pleomorphic sarcomas. Am J Surg Pathol 1994;18:1213-23. |
|9.||Henricks WH, Chu YC, Goldblum JR, Weiss SW. Dedifferentiated liposarcoma: A clinicopathological analysis of 155 cases with a proposal for an expanded definition of dedifferentiation. Am J Surg Pathol 1997;121:271-81. |
|10.||Nascimento AG, Kurtin PJ, Guillou L, Fletcher CD. Dedifferentiated liposarcoma: A report of nine cases with a peculiar neural like whorling pattern associated with metaplastic bone formation. Am J Surg Pathol 1998;122:945-55. |
|11.||Fanburg -Smith JC, Mietdnen M. Liposarcoma with meningothelial -like whorls: A study of 17 cases of a distinctive histological pattern associated with dedifferentiated liposarcoma. Histopathology 1998;33:414-24. |
|12.||Russell MJ, Flynt FL, Harroff AL, Fadare O. Dedifferentiated Liposarcoma of the Retroperitoneum with Extensive Leiomyosarcomatous Differentiation and β-Human Chorionic Gonadotropin Production. Sarcoma 2008;2008:658090. |
|13.||Lucas DR, Shukla A, Thomas DG, Patel RM, Kubat AJ, McHugh JB. Dedifferentiated liposarcoma with inflammatory myofibroblastic tumor-like features. Am J Surg Pathol 2010;34:844-51. |
|14.||Fabre-Guillevin E, Coindre JM, Somerhausen Nde S, Bonichon F, Stoeckle E, Bui NB. Retroperitoneal liposarcomas: Follow up analysis of dedifferentiation after clinicopathological reexamination of 86 liposarcomas and malignant fibrous histiocytoma .Cancer 2006;106:2725-33. |
|15.||Boland J, Weiss S, Oliveira AM, Erickson Johnson ML, Folpe A. Liposarcomas with mixed well-differentiated and pleomorphic features: A clinicopathologic study of 12 cases. Am J Surg Pathol 2010;34:837-43. |
|16.||Mariño-Enríquez A, Fletcher CD, Dal Cin P, Hornick JL. Dedifferentiated liposarcoma With "homologous" lipoblastic (pleomorphic liposarcoma-like) differentiation: Clinicopathologic and molecular analysis of a series suggesting revised diagnostic criteria. Am J Surg Pathol 2010;34:1122-31. |
|17.||Singer S, Antonescu CR, Riedel E, Brennan MF. Histologic subtype and margin of Resection predict pattern of recurrence and survival for retroperitoneal liposarcoma. Ann Surg 2003;238:358-71. |
|18.||Hasegawa T, Seki K, Hasegawa F, Matsuno Y, Shimoda T, Hirose T, et al. Dedifferentiated liposarcoma of retroperitoneum and mesentery: Varied growth patterns and histological grades -A clinicopathologic study of 32 cases. Hum Pathol 2000;31:717-27. |
|19.||Chan YF, Yuen MY, Ma LT, Li MK. Recurrent dedifferentiated liposarcoma of the spermatic cord simulating malignant fibrous histiocytoma: An immunohistochemical and ultrastructural study. Pathology 1987;19:99-102. |
|20.||Fukuaga M. Histologically low-grade dedifferentiated liposarcoma of the retroperitoneum. Pathol Int 2001;51:392-5. |
|21.||Hashimoto H, Daimaru Y, Enjoji M. S-100 protein distribution in liposarcoma. An immunoperoxidase study with special reference to the distinction of liposarcoma from myxoid malignant fibrous histiocytoma. Virchows Arch A Pathol Anat Histopathol 1984;405:1-10. |
|22.||Evans HL, Khurana KK, Kemp BL, Ayala AG. Heterologous elements in the dedifferentiated component of dedifferentiated liposarcoma. Am J Surg Pathol 1994;18:1150-7. |
|23.||Aleixo PB, Hartmann AA, Menezes IC, Meurer RT, Oliveira AM. Can MDM2 and CDK4 make the diagnosis of well differentiated/dedifferentiated liposarcoma? An immunohistochemical study on 129 soft tissue tumours. J Clin Pathol 2009:62:1127-35. |
|24.||Sirvent N, Coindre JM, Maire G, Hostein I, Keslair F, Guillou L, et al. Detection of MDM2-CDK4 amplification 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. |
|25.||Tayal S, Classen E, Bemis L, Robinson WA. c-kit Expression in Dedifferentiated and Well-differentiated Liposarcomas; Immunohistochemistry and Genetic Analysis. Ant Cancer Res 2005;25:2215-20. |
|26.||Folpe AL, Weiss SW. Lipoleiomyosarcoma (well-differentiated liposarcoma with leiomyosarcomatous differentiation): A clinicopathologic study of nine cases including one with dedifferentiation. Am J Surg Pathol 2002;26:742-9. |
|27.||Rosai J, Akerman M, Dal Cin P, DeWever I, Fletcher CD, Mandahl N, et al. Combined morphologic and karyotypic study of 59 atypical lipomatous tumors. Evaluation of their relationship and differential diagnosis with other adipose tissue tumors (a report of the CHAMP Study Group). Am J Surg Pathol 1996;20:1182-9. |
|28.||Hisaoka M, Matsuyama A, Nakamoto M. Aberrant Calreticulin Expression Is Involved In the Dedifferentiation of Dedifferentiated Liposarcoma. Am J Pathol 2012;180:2076-83. Epub 2012 Mar 17 |
|29.||Singer S, Socci ND, Ambrosini G, Sambol E, Decarolis P, Wu Y, et al. Gene expression profiling of liposarcoma identifies distinct biological types/subtypes and potential therapeutic targets in well-differentiated and de-differentiated liposarcoma. Cancer Res 2007;67:6626-36. |
|30.||Nascimento AG. Dedifferentiated Liposarcoma. Semin Diagn Pathol 2001;18:263-6. |
|31.||Dalal KM, Antonescu CR, Singer S. Diagnosis and management of lipomatous tumors. J Surg Oncol 2008:97:298-313. |
Department of Pathology, Tata Memorial Hospital, Dr. E.B. Road, Parel, Mumbai, 400012
Source of Support: None, Conflict of Interest: None
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8]
[Table 1], [Table 2]