Int J Clin Exp Med 2018;11(9):9178-9188www.ijcem.com /ISSN:1940-5901/IJCEM0074439

Original ArticleApplication of previously reported lipo-related markers in atypical lipomatous tumor/well differentiated liposarcoma (ALT/WDLPS) and lipoma

Sung Hak Lee1, Hong Sik Park2, Soyoung Im2, Yinji Cui3, Su Young Kim3, Changyoung Yoo2, Jinyoung Yoo2

1Department of Hospital Pathology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea; 2Department of Hospital Pathology, St. Vincent’s Hospital, College of Medicine, The Catho-lic University of Korea, Seoul, Korea; 3Department of Pathology, College of Medicine, The Catholic University of Korea, Seoul, Korea

Received September 30, 2017; Accepted June 20, 2018; Epub September 15, 2018; Published September 30, 2018

Abstract: Background: In previous reports, several lipo-related markers have been introduced. Diagnosis of the atypical lipomatous tumor/well differentiated liposarcoma (ALT/WDLPS) is not easy, and we cannot exclude the in-fluence of subjectivity when a differential diagnosis from lipoma is concerned. The aim of this study was to know the expression patterns of these markers in ATL/WDLPS and lipoma, and to evaluate whether these markers are helpful for the differential diagnosis between them. Material and Methods: Total 11 cases of ALT/WDLPS and 5 cases of lipoma previously diagnosed at St. Vincent’s Hospital were reviewed, and immunohistochemistry for MDM2, CDK4, HMGI-C, FABP4, Calreticulin was performed on the specimen of these cases. Results: After slide review, four cases of previous ATL/WDLPS were proved as lipoma. With revision of the diagnosis, cases of ALT/WDLPS with charac-teristic histological findings showed positivity for both MDM2 and CDK4. In cases of benign lipoma, one of these markers were positive but not in both. HMGI-C, FABP4, and Calreticulin showed various and inconsistent positive reactions. Calreticulin was consistently expressed in ALT/WDLPS and lipoma. Conclusions: Histologic findings are most important in the diagnosis of ALT/WDLPS. In ambiguous cases, positivity for both MDM2 and CDK4 may be helpful for the diagnosis of ALT/WDLPS, but meticulous care is needed because positivity of MDM2 or CDK4 may also be observed in lipomas. HMGI-C, FABP4, and Calreticulin were inferior in terms of discrimination. Calreticulin is expected to be useful for identifying the lipogenic origin.

Keywords: Atypical lipomatous tumor, lipoma, immunohistochemistry, MDM2, CDX4, HMGI-C, FABP4, calreticulin

Introduction

Lipogenic tumors are the most common soft tissue tumors, and liposarcoma is the most common sarcoma [1]. Liposarcoma has been reported to account for 15% of all soft tissue sarcomas (STS), 24% of STS of the extremities and 45% of STS in the retroperitoneum [2]. There are some agonizing situations for the differential diagnosis, for example between benign lipoma and atypical lipomatous tumor/well differentiated liposarcoma (ALT/WDLPS), between dedifferentiated liposarcoma and other high grade sarcomas, or between myxoid liposarcoma and other myxoid tumors. Among these difficulties, the most common difficulty we guess is the discrimination between benign lipoma and ALT/WDLPS. Lipoma is a com- monly encountered tumor, and atypical cells

can be found even in this benign tumor. There- fore, there is always a need of an auxiliary tool to distinguish ALT/WDLPS from lipoma. In this study, immunohistochemical staining was per-formed with several markers that are known to give help to distinguish ALT/WDLPS from lipo-ma. The aim of this study was to know the expression patterns of these markers in ATL/WDLPS and lipoma, and to evaluate whether these markers are helpful for the differential diagnosis between ATL/WDLPS and lipoma.

Materials and methods

Study samples

We collected 11 cases of ALT/WDLPS and 5 cases of lipoma that were diagnosed at St. Vincent’s Hospital, The Catholic University of

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9179 Int J Clin Exp Med 2018;11(9):9178-9188

Korea, South Korea between 2001 and 2012. The ALT/WDLPS cases included 2 cases of atypical spindle cell lipomatous tumors. This study was approved by the institutional review board of St. Vincent hospital and the Catholic University of Korea (VC14SISI0263).

ing at each step, primary antibody reaction, secondary antibody reaction, diaminobenzidine (DAB) solution preparation, counter staining with haematoxylin, and sealing were perform- ed. We used the following primary antibodies for immunohistochemistry: MDM2 (monoclo-

Table 1. Clinical information of the study cases and their original diagnoses

No Sex Age Sites DiagnosisStatistical summaries

ALT/WDLPS Lipoma Total 1 F 28 Buttock ALT/WDLPS Sex M 4 4 82 F 52 Thigh ALT/WDLPS F 7 1 83 F 64 Thigh ALT/WDLPS Total 11 5 164 F 49 Thigh ALT/WDLPS5 M 53 Buttock ALT/WDLPS Age 21-30 1 0 16 M 67 Thigh ALT/WDLPS 31-40 1 1 27 F 73 Thigh ALT/WDLPS 41-50 4 2 68 M 46 Buttock ALT/WDLPS 51-60 2 1 39 F 31 Axilla ALT/WDLPS > 61 3 1 410 F 50 Thigh ALT/WDLPS Total 11 5 1611 M 46 Thigh ALT/WDLPS

Sites Neck 0 1 112 M 49 Neck Lipoma Cheek 0 1 113 F 53 Back Lipoma Upper extremities 1 2 314 M 39 Forearm Lipoma Back 0 1 115 M 64 Cheek Lipoma Lower extremities 10 0 1016 M 50 Shoulder Lipoma Total 11 5 16ALT/WDLPS, Atypical Lipomatous Tumor/well differentiated liposarcoma.

Figure 1. Histology of atypical lipomatous tumor. In these tumors, the cell size variations of adipocytes (A) and the presence of atypical hyperchromatic stromal cells (B) are noted. In the case of atypical spindle cell lipomatous tumor cell size variation of adipocytes is not conspicuous (C) but there are atypical hyperchromatic stromal cells in the stroma (D).

Methods

Histological review: All cas- es were reviewed by autho-rized pathologists. We exam-ined the specimens with a focus on the size variation of adipocytes and the presence of atypical fibroblasts, which are necessary for establishing the diagnosis of ALT/WDLPS.

Immunohistochemistry: For immunohistochemical stain-ing of MDM2, CDK4, Calreti- culin, FABP4, and HMGI-C, tis-sue blocks were sectioned and mounted on glass slides (4-μm thickness). Deparaffi- nization, antigen retrieval us- ing a heated water bath, endogenous peroxidase bl- ockade with H2O2-methanol solution, phosphate-buffered saline (0.01 M, pH 7.4) wash-

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9180 Int J Clin Exp Med 2018;11(9):9178-9188

nal; 1:50; GeneTex, CA, USA), CDK4 (monoclo-nal; 1:100; GeneTex, CA, USA), Calreticulin (monoclonal; 1:100; GeneTex, CA, USA), FABP4 (monoclonal; 1:50; Abnova, Taipei, Taiwan), HMGI-C (monoclonal; 1:100; GeneTex, CA, USA).

Interpretation: Interpretation of IHC staining was divided in the following five semi-quantita-tive categories: No staining at all (negative), partial weak staining of cells (focally positive), weak staining of most cells (weakly positive), distinct staining of most cells (positive), and focal or diffuse intense staining (strongly posi-tive). After these quantification, weak positive, positive and strong positive were grouped as positive, and negative, focal positive were grouped as negative. Cytoplasmic or nuclear stainings were considered as positive results. The results of IHC staining were compared with the findings of haematoxylin and eosin (HE) slides, and a correlation was derived.

Statistical analysis

We used the chi-square test to analyze the sta-tistical significance of the results, using SAS software (version8; SAS Inc., Cary, NC, USA). A p-value of < 0.05 was considered significant.

Results

Clinical and histological summary

The age of patients with ALT/WDLPS ranged from 28 to 64 years. Among the 11 cases of

ALT/WDLPS, 4 cases were males and 7 cases were females. Tumor sites of ALT/WDLPS were lower extremities including buttock in 10 cas- es, and axillary area in 1 case. Lipomas were occurred in various sites (Table 1). Histologi- cally, definite cases of ALT/WDLPS showed vari-ation in the size of adipocytes and atypical fibroblasts. In cases of atypical spindle cell li- pomatous tumor, size variations in adipocytes were not obvious as in ALT/WDLPS, but atypi- cal fibroblasts were commonly found (Figure 1).

Results of immunohistochemistry

Staining status in immunohistochemistry was variable according to the cases, and generally, not only the nucleus and cytoplasm of adipo-cytes but also fibroblasts in the stroma were stained (Figure 2). Immunohistochemically, st- aining for MDM2 was positive in the 10 cases of ALT/WDLPS, and among them, 4 cases showed strongly positive staining and 6 cases showed weakly positive staining. A case of ALT/WDLPS showed negative staining. In the 5 cases of lipoma, 1 case showed positive stain-ing, 1 case showed weakly positive staining, and 3 cases showed negative staining. Stain- ing for CDK4 was positive in 7 cases of ALT/WDLPS, but it was negative in 4 cases of ALT/WDLPS. In cases of lipoma, 1 case showed positive staining, 1 case showed weakly posi-tive staining, and 3 cases showed negative staining. Staining for HMGI-C was strongly posi-tive in 4 cases of ALT/WDLPS, positive in 3 cases of ALT/WDLPS, weakly positive in 2

Figure 2. Immunohistochemistry of atypical lipomatous tumors. Stain-ing interpretations are as follow: No staining at all (negative) (A), partial weak staining of cells (focally posi-tive) (B), weak staining of most cells (weakly positive) (C), distinct stain-ing of most cells (positive) (D), and focal or diffuse intense staining (strongly positive) (E).

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9181 Int J Clin Exp Med 2018;11(9):9178-9188

Table 2. Results of immunohistochemistry

No DiagnosisMDM2 CDK4 HMGI-C FABP4 Calreticulin

ICH results Interpretationa ICH

results Interpretation ICH results Interpretation ICH

results Interpretation ICH results Interpretation

1 ALT/WDLPS N - N - WP + N - P +2 ALT/WDLPS P + P + P + WP + P +3 ALT/WDLPS P + N - SP + N - P +4 ALT/WDLPS WP + P + N - N - P +5 ALT/WDLPS P + P + SP + N - P +6 ALT/WDLPS WP + N - P + N - WP +7 ALT/WDLPS P + P + P + WP + P +8 ALT/WDLPS WP + WP + SP + P + SP +9 ALT/WDLPS WP + P + WP + FP - P +10 ALT/WDLPS WP + WP + N - N - WP +11 ALT/WDLPS WP + N - SP + N - SP +12 Lipoma WP + N - SP + P + P +13 Lipoma N - P + N - N - P +14 Lipoma N - N _ N - P + WP +15 Lipoma P + N _ WP + WP + WP +16 Lipoma N - WP + SP + N - N -ALT/WDLPS, Atypical Lipomatous Tumor/well differentiated liposarcoma; N, Negative; FP, Focal positive; WP, Weak positive; P, Positive; SP, Strong positive. a. After semi-quantification, weak positive, positive and strong positive were grouped as positive (+), and negative and focal positive were grouped as negative (-).

Table 3. Revised diagnosis according to the histological features and immunohistochemical results of MDM2 and CDK4

No Original diagnosisHistological features Immunohistochemistry

Revised diagnosisVariation of cell size Atypical cells MDM2 CDK4

1 ALT/WDLPS Suspicious Absent - - Intramuscular lipoma2 ALT/WDLPS Present Present + + ASCLT3 ALT/WDLPS Present Present + - ALT/WDLPS4 ALT/WDLPS Present Present + + ALT/WDLPS5 ALT/WDLPS Present Present + + ALT/WDLPS6 ALT/WDLPS Suspicious Absent + - Lipoma 7 ALT/WDLPS Suspicious Present + + ASCLT8 ALT/WDLPS Present Absent + + Lipoma with fat necrosis9 ALT/WDLPS Present Present + + ALT/WDLPS10 ALT/WDLPS Present Present + + ALT/WDLPS11 ALT/WDLPS Suspicious Absent + - Suspicious lipoma12 Lipoma Suspicious Absent + - Lipoma 13 Lipoma Absent Absent - + Lipoma14 Lipoma Absent Absent - - Lipoma15 Lipoma Suspicious Absent + - Lipoma16 Lipoma Absent Absent - + LipomaN, Negative; WP, Weak positive; P, Positive; SP, Strong positive; ALT/WDLPS, Atypical Lipomatous Tumor/well differentiated liposarcoma; ASCLT, Atypical spindle cell lipomatous tumor.

cases of ALT/WDLPS, and negative in 2 cases of ALT/WDLPS. In cases of lipoma, 2 cases showed strongly positive staining, 1 case showed positive staining, and 2 cases showed negative staining. Staining for FABP4 was posi-

tive in 1 case of ALT/WDLPS, weakly positive in 4 cases of ALT/WDLPS, and negative in 6 cases of ALT/WDLPS. In cases of lipoma, 2 cases showed positive staining, 1 case showed weakly positive staining, and 2 cases showed

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9182 Int J Clin Exp Med 2018;11(9):9178-9188

negative staining. Stainings for Calreticulin in ALT/WDLPS were strongly positive in 2 cases, positive in 7 cases, and weakly positive in 2 cases. In cases of lipoma, positive staining was observed in 2 cases and weakly positive staining was observed in 3 cases (Table 2).

Matching between histological findings and immunohistochemical results

We reviewed the slides of the study cases that were previously diagnosed as ALT/WDLPS and lipoma according to the diagnostic criteria for

the diagnosis of ALT/WDLPS such as variation in the size of adipocytes and presence of at- ypical fibroblasts in the stroma. Among the 11 cases that were previously diagnosed as ALT/WDLPS, only 7 cases showed these 2 histolo- gic findings. In the remaining 4 cases, there were variations in the size of adipocytes but the presence of atypical fibroblasts was debat-able. In a case of previously diagnosed ALT/WDLPS, We found that the variations in the size of adipocytes were resulted from inflammation and fibrosis. The cases of lipoma showed typi-cal benign histologic findings (Table 3).

Figure 3. Immunohistochemistry of MDM2 and CDK4 in ALT/WDLPS. MDM2 positive and CDK4 positive ALT/WDLPS (A), MDM2 positive and CDK4 positive atypical spindle cell lipomatous tumor (B), MDM2 positive and CDK4 nega-tive ALT/WDLPS (C), and MDM2 weak positive and CDK4 negative originally ALT/WDLPS but diagnosis revised as lipoma (D).

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9183 Int J Clin Exp Med 2018;11(9):9178-9188

Diagnosis revised

When comparisons were made between the results of histologic review and those of immu-nohistochemistry, among the 7 cases of ALT/WDLPS which satisfied the diagnostic criteria, 6 cases showed positive staining or at least weakly positive staining for both MDM2 and CDK4. The remaining 1 case showed positive staining for MDM2 but negative staining for CDK4. Among the 4 cases that were previously diagnosed as ALT/WDLPS but did not match with the diagnostic criteria in the slide review, negative staining for both MDM2 and CDK4 was observed in 1 case, weakly positive stain-ing for MDM2 but negative staining for CDK4 was observed in 2 cases, and weakly positive staining for both MDM2 and CDK4 was ob- served in 1 case (Table 3; Figure 3). In cases of lipoma, 2 cases showed positive staining for MDM2 and negative staining for CDK4, 2 cases showed negative staining for MDM2 and positive staining for CDK4, and 1 case

showed negative staining for both MDM2 and CDK4 (Table 3; Figure 4). Staining for HMGI-C and FABP4 did not show consistency with the histologic findings. Lipogenic tumors stained relatively well for Calreticulin, irrespective of whether they were ALT/WDLPS or lipomas (Figure 5).

Statistical significances of lipo-related markers

Statistically, Individual markers didn’t show dis-criminative power for the differential diagnosis. Positive results of both MDM2 and CDK4 pro-vided statistically significant help for the diag-nosis of ALP/WDLPS as was already well kno- wn. But these significances were found not only in the original diagnosis but also in the revised diagnosis. Pathologist should pay attention when he interprets these positive results. Positive result of just one of MDM2 or CDK4 did not show significant effect. Positive results at least one of HMGIC, FABP4 and calreticulin if combined with positive results of

Figure 4. Immunohistochemistry of MDM2 and CDK4 in lipomas. MDM2 negative and CDK4 negative lipoma (A), MDM2 negative and CDK4 positive lipoma (B), and MDM2 positive and CDK4 negative lipoma (C).

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9184 Int J Clin Exp Med 2018;11(9):9178-9188

both of MDM2 and CDK4 showed statistical significance for the differential diagnosis, but it is not certain whether these three markers have direct effect for this diagnosis (Table 4).

Discussion

ALT/WDLPS including atypical spindle cell lipo-matous tumor [3] is the most frequently found soft tissue malignancy [4]. One report empha-sized that patients who have a lipomatous tumor with clinical findings such as age more than 55 years, size more than 10 cm, history of previous surgery, and tumors arising in the extremities have a higher possibility of ALT/WDLPS [5]. Diagnosis of ALT/WDLPS is made on the basis of histologic findings such as sig-nificant variation in cell size and presence of atypical hyperchromatic stromal cells [4]. But these histologic findings can be estimated subjectively and there is a possibility of fai- lure to notice [6]. Clinically, because MRI find-ings are overlap, differentiation between ATL/WDLPS and lipoma is not easy task radiologi-cally [7]. There also is an attempt of scoring system based on tumor size, findings of MRI etc [8]. For making an accurate diagnosis of ALT/

WDLPS, we should perform careful examina-tion and occasionally many histologic slides may be needed. In these cases, we should bear the economic loss. ALT/WDLPS has a local recurrence rate of 10% and in a recurrent tumor, this rate becomes higher. The possibility of progression to dedifferentiated liposarcoma is estimated to be about 4% [9]. ALT/WDLPS and dedifferentiated liposarcoma are closely related. Dedifferentiated liposarcoma is a non-lipogenic sarcoma arising in ALT/WDLPS [4]. Dedifferentiated liposarcoma should be differ-entiated not only from a benign tumor such as pleomorphic lipoma but also from other high grade soft tissue sarcomas [4].

At present, a few biologic markers are known to be related to lipogenic tumors. Murine double-minute 2 (MDM2), is a well-known marker ex- pressed in ALT/WDLPS and dedifferentiated liposarcoma [4, 6]. Expression of MDM2 was best identified using the fluorescent in situ hybridisation (FISH) method with the highest sensitivity and specificity [10, 11]. But this marker was not always identified in ALT/WDLPS, and in such a case, the molecular bio-logical method can be useful [12].

Figure 5. Immunohistochemistry of HMGI-C, FABP4 and calreticulin. Results of HMGI-C and FABP4 were not very informative: HMGI-C positive (A) and negative (B) in ALT/WDLPS, positive (C) and negative (D) in lipomas; FABP4 positive (E) and negative (F) in ALT/WDLPS, positive (G) and negative (H) in lipomas. Calreticulin was stained well not only in ALT/WDLPS but also in lipomas: strong positive (I) and positive (J) in ALT/WDLPS, positive (K) and weakly positive (L) in lipomas.

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9185 Int J Clin Exp Med 2018;11(9):9178-9188

Table 4. Chi-square analysis of the association of the markers or combination of the markers with the differential diagnosis of ATL/WDLPS and lipoma

MarkersOriginal diagnoses Revised diagnoses

ATL/WDLPS (%) Lipoma (%) Total p-valuea ATL/WDLPS (%) Lipoma (%) Total P-valuea

MDM2 Posb 10 (90.9)c 2 (40.0) 12 7 (100.0) 5 (55.6) 12Neg 1 (9.1) 3 (60.0) 4 0 (0.0) 4 (44.4) 4Total 11 5 16 0.063 7 9 16 0.088

CDK4 Pos 7 (63.6) 2 (40.0) 9 6 (85.7) 3 (33.3) 9Neg 4 (36.4) 3 (60.0) 7 1 (14.3) 6 (66.7) 7Total 11 5 16 0.596 7 9 16 0.060

HMGIC Pos 9 (81.8) 3 (60.0) 12 5 (71.4) 7 (71.8) 12Neg 2 (18.2) 2 (40.0) 4 2 (18.6) 2 (22.2) 4Total 11 5 16 0.547 7 9 16 1.000

FABP4 Pos 3 (27.3) 3 (60.0) 6 2 (28.6) 4 (44.4) 6Neg 8 (72.7) 2 (40.0) 10 5 (71.4) 5 (55.6) 10Total 11 5 16 0.299 7 9 16 0.633

Calreticulin Pos 11 (100.0) 4 (80.0) 15 7 (100.0) 8 (88.9) 15Neg 0 (0.0) 1 (20.0) 1 0 (0.0) 1 (11.1) 1Total 11 5 16 0.313 7 9 16 1.000

M+/C+d Pos 7 (63.6) 0 (0.0) 7 6 (85.7) 1 (11.1) 7Neg 4 (36.4) 5 (100.0) 9 1 (14.3) 8 (88.9) 9Total 11 5 16 0.034 7 9 16 0.009

M+/C- or M-/C+e Pos 10 (90.9) 4 (80.0) 14 7 (100.0) 7 (77.8) 14Neg 1 (9.1) 1 (20.0) 2 0 (0.0) 2 (22.2) 2Total 11 5 16 1.000 7 9 16 0.475

M+/C+ (+) H+F+Cal+f Pos 3 (27.3) 0 (0.0) 3 2 (28.6) 1 (11.1) 3Neg 8 (72.7) 5 (100.0) 13 5 (71.4) 8 (88.9) 13Total 11 5 16 0.506 7 9 16 0.550

M+/C+ (+) at least two are + among HFCalg Pos 5 (45.5) 0 (0.0) 5 4 (57.1) 1 (11.1) 5Neg 6 (54.5) 5 (100.0) 11 3 (42.9) 8 (88.9) 11Total 11 5 16 0.119 7 9 16 0.106

M+/C+ (+) at least one is + among HFCalh Pos 7 (63.6) 0 (0.0) 7 6 (85.7) 1 (11.1) 7Neg 4 (36.4) 5 (100.0) 9 1 (14.3) 8 (88.9) 9Total 11 5 16 0.034 7 9 16 0.009

a. Fisher’s Exact Test. b. Pos, positive; Neg, negative. c. % within diagnosis. d. M+/C+, MDM2+/CDK4+. e. M+/C- or M-/C+, MDM2+/CDK4- or MDM2-/CDK4+. f. M+/C+ (+) H+F+Cal+, MDM2+/CDK4+ plus all positive of HMCIC, FABP4 and calreticulin. g. M+/C+ (+) at least two are + among HFCal, MDM2+/CDK4+ plus at least two are positive within HMCIC, FABP4 and calreticulin. h. M+/C+ (+) at least one is + among HFCal, MDM2+/CDK4+ plus at least one is positive within HMCIC, FABP4 and calreticulin.

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9186 Int J Clin Exp Med 2018;11(9):9178-9188

In ALT/WDLPS and dedifferentiated liposarco-ma, the expressions of HMGI-C, CDK4, and MDM2 are increased [13, 14]. Adipocyte P2/ fatty acid-binding protein 4 (aP2/FABP4) is known to be expressed in benign lipomas, hi- bernomas, spindle cell/pleomorphic lipomas, ALT/WDLPS, myxoid liposarcomas, and imma-ture fat cells or lipoblasts from pleomorphic liposarcoma. Therefore, this marker can be expected to be a useful marker for differentia-tion between lipogenic tumors and other soft tissue tumors [15]. Calreticulin is a Ca (2+)-buff-ering protein and is known to be an inhibitor of adipocyte differentiation. Calreticulin is not expressed in the normal adipose tissue and benign lipoma, but it is well expressed in atypical stromal cells of ALT/WDLPS and dedif-ferentiated areas of the dedifferentiated sar-coma [16]. P16 is expressed in the ALT/WDLPS and dedifferentiated liposarcoma as CDK4 and MDM2, and it can be used as a marker for differentiation between ALT/WDLPS and be- nign lipomatous tumors [17]. It was known that it’s diagnostic specificity is greatly enhance when combined with MDM2 and CDK4 [18]. But regrettably p16 was not included in this study. From the molecular biological aspect, ALT/WDLPS is characterized by amplification of the 12q13-15 chromosome, and this amplifica-tion is known to be associated with increased expression of MDM2, CDK4, and HMGI-C [2, 3].

In this study, we reviewed the cases that were previously diagnosed as ALT/WDLPS and lipo-ma according to the diagnostic criteria for ALT/WDLPS including the variation in size of adipocytes and the presence of atypical stro-mal cells, and we compared with the results of immunohistochemistry. Among the 11 cases of ALT/WDLPS, we found slight lack of data for the diagnosis of ALT/WDLPS in 4 cases, and these cases showed negative staining for both MDM2 and CDK4 or positive staining for only one of the markers. In these 4 cases, we could note the variation in size of adipocytes but the presence of atypical hyperchromatic stromal cells was debatable. We interpreted that the variation in the cell size in a case was due to inflammation and fibrosis. Therefore, we decid-ed that for these 4 cases, the diagnosis of lipoma is more appropriate. On the contrary, 6 cases which satisfied the diagnostic criteria for ALT/WDLPS showed positive staining for both MDM2 and CDK4. Among them 2 cases

were diagnosed as atypical spindle cell lipoma-tous tumor. Some studies suggested that this disease entity should be considered as a kind of ATL/WDLPS [3, 19]. In this study, we includ-ed this entity in ATL/WDLPS. A case of ALT/WDLPS showed positive staining for only CDK4 and not for MDM2. The cases of lipoma showed typical benign histologic findings and immuno-histochemically, they showed negative staining for both MDM2 and CDK4, or positive staining for one of the markers.

In this study, we tried to identify reliable ancil-lary markers for the diagnosis of ALT/WDLPS and we applied several markers which are known to be related to lipogenic tumors. But besides MDM2 and CDK4, immunohistoche- mical results for HMGI-C and FABP4 did not show a consistent relation with histologic find-ings. Statistical analysis of immunohistochemi-cal findings showed that positive reactions of both MDM2 and CDK4 had meaningful associ-ation with diagnosis of ATL/WDLPS, but when compared with the revised diagnoses we can see that their diagnostic power is not decisive. As was well known, MDM2 and CDK4 were most valuable markers as the ancillary tool for the diagnosis of ATL/WDLPS [20], although a study suggested that immunohistochemistry of MDM2 and CDK4 were insensitive method because of their low sensitivity [21]. As men-tioned above, FABP4 was known as a useful marker for the identification of lipogenic tumor [15]. But in this study, positive and negative staining for FABP4 was observed in not only ALT/WDLPS but also in lipomas. Calreticulin showed relatively good staining results not only in ALT/WDLPS but also in lipoma. Therefore, this marker could be useful for the discrimina-tion of lipogenic tumors from other soft tissue tumors.

Conclusions

Collectively, on the basis of these results, we conclude that to make the diagnosis of ALT/WDLPS, histologic findings are the gold stan-dard, and in ambiguous cases which are diffi-cult to discriminate from lipoma, positive im- munohistochemical staining for both MDM2 and CDK4 could be helpful for establishing the diagnosis of ALT/WDLPS. but meticulous care is needed because positivity of MDM2 or CDK4 may also be observed in lipomas. HMGI-C,

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9187 Int J Clin Exp Med 2018;11(9):9178-9188

FABP4, and Calreticulin were inferior in terms of discrimination. Calreticulin is expected to be useful for identifying the lipogenic origin among undifferentiated tumors.

Disclosure of conflict of interest

None.

Address correspondence to: Dr. Changyoung Yoo, Department of Hospital Pathology, St. Vincent Hospital, The Catholic University of Korea, 93-6, Jungbudaero, Paldal-gu, Suwon, Gyeonggi-do 442-723, Korea. Tel: 031-249-7839; Fax: 031-244-6786; E-mail: yoochangyoung@gmail.com

References

[1] Hogg ME, Wayne JD. Atypical lipomatous tu-mor/well-differentiated liposarcoma: what is it? Surg Oncol Clin N Am 2012; 21: 333-340.

[2] De Vita A, Mercatali L, Recine F, Pieri F, Riva N, Bongiovanni A, Liverani C, Spadazzi C, Mise-rocchi G, Amadori D, Ibrahim T. Current cla- ssification, treatment options, and new per-spectives in the management of adipocytic sarcomas. Onco Targets Ther 2016; 9: 6233-6246.

[3] Mariño-Enriquez A, Nascimento AF, Ligon AH, Liang C, Fletcher CD. Atypical spindle cell lipo-matous tumor: clinicopathologic characteriza-tion of 232 cases demonstrating a morpho-logic spectrum. Am J Surg Pathol 2017; 41: 234-244.

[4] Fletcher CD, Bridge JA, Hogendoorn PC, Mertens F. WHO classification of tumours of soft tissue and bone. 4th edition. Lyon: IARC Press; 2013.

[5] Fisher SB, Baxter KJ, Staley CA 3rd, Fisher KE, Monson DK, Murray DR, Oskouei SV, Weiss SW, Kooby DA, Maithel SK, Delman KA. The General Surgeon’s quandary: atypical lipoma-tous tumor vs lipoma, who needs a surgical oncologist? J Am Coll Surg 2013; 217: 881-888.

[6] Zhang H, Erickson-Johnson M, Wang X, Oliveira JL, Nascimento AG, Sim FH, Wenger DE, Zamolyi RQ, Pannain VL, Oliveira AM. Molecu-lar testing for lipomatous tumors: critical anal-ysis and test recommendations based on the analysis of 405 extremity-based tumors. Am J Surg Pathol 2010; 34: 1304-1311.

[7] Brisson M, Kashima T, Delaney D, Tirabosco R, Clarke A, Cro S, Flanagan AM, O’Donnell P. MRI characteristics of lipoma and atypical lipo- matous tumor/well-differentiated liposarco-ma: retrospective comparison with histology and MDM2 gene amplification. Skeletal Radiol 2013; 42: 635-647.

[8] Nagano S, Yokouchi M, Setoguchi T, Ishidou Y, Sasaki H, Shimada H, Komiya S. Differentia-tion of lipoma and atypical lipomatous tumor by a scoring system: implication of increased vascularity on pathogenesis of liposarcoma. BMC Musculoskelet Disord 2015; 16: 36.

[9] Mavrogenis AF, Lesensky J, Romagnoli C, Alberghini M, Letson GD, Ruggieri P. Atypical lipomatous tumors/well-differentiated liposar-comas: clinical outcome of 67 patients. Ortho-pedics 2011; 34: e893-898.

[10] Kimura H, Dobashi Y, Nojima T, Nakamura H, Yamamoto N, Tsuchiya H, Ikeda H, Sawada-Kitamura S, Oyama T, Ooi A. Utility of fluores-cence in situ hybridization to detect MDM2 amplification in liposarcomas and their mor-phological mimics. Int J Clin Exp Pathol 2013; 6: 1306-1316.

[11] Kashima T, Halai D, Ye H, Hing SN, Delaney D, Pollock R, O’Donnell P, Tirabosco R, lanagan AM. Sensitivity of MDM2 amplification and unexpected multiple faint alphoid 12 (alpha 12 satellite sequences) signals in atypical lipo-matous tumor. Mod Pathol 2012; 25: 1384-1396.

[12] Neuville A, Ranchère-Vince D, Dei Tos AP, Mon-tesco MC, Hostein I, Toffolatti L, Chibon F, Pis-saloux D, Alberti L, Decouvelaere AV, Albert S, Rossi CR, Blay JY, Coindre JM. Impact of mo-lecular analysis on the final sarcoma diagno-sis: a study on 763 cases collected during a European epidemiological study. Am J Surg Pathol 2013; 37: 1259-1268.

[13] Alshenawy H. Can HMGI-C be used as an aid with MDM2 and CDK4 to differentiate liposar-coma subtypes from their mimics? J Cancer Res Clin Oncol 2013; 139: 1073-1081.

[14] Saâda-Bouzid E, Burel-Vandenbos F, Ranchère-Vince D, Birtwisle-Peyrottes I, Chetaille B, Bou-vier C, Château MC, Peoc’h M, Battistella M, Bazin A, Gal J, Michiels JF, Coindre JM, Pedeu-tour F, Bianchini L. Prognostic value of HMGA2, CDK4, and JUN amplification in well-differenti-ated and dedifferentiated liposarcomas. Mod Pathol 2015; 28: 1404-1414.

[15] Kashima TG, Turley H, Dongre A, Pezzella F, Athanasou NA. Diagnostic utility of aP2/FABP4 expression in soft tissue tumours. Virchows Arch 2013; 462: 465-472.

[16] Hisaoka M, Matsuyama A, Nakamoto M. Aber-rant calreticulin expression is involved in the dedifferentiation of dedifferentiated liposarco-ma. Am J Pathol 2012; 180: 2076-2083.

[17] Thway K, Flora R, Shah C, Olmos D, Fisher C. Diagnostic utility of p16, CDK4, and MDM2 as an immunohistochemical panel in distin-guishing well-differentiated and dedifferen- tiated liposarcomas from other adipocytic tu-mors. Am J Surg Pathol 2012; 36: 462-469.

Lipo-related markers for the differential diagnosis of ALT/WDLPS and lipoma

9188 Int J Clin Exp Med 2018;11(9):9178-9188

[18] Kammerer-Jacquet SF, Thierry S, Cabillic F, Lannes M, Burtin F, Henno S, Dugay F, Bouzillé G, Rioux-Leclercq N, Belaud-Rotureau MA, Stock N. Differential diagnosis of atypical li- pomatous tumor/well-differentiated liposarco-ma and dedifferentiated liposarcoma: utility of p16 in combination with MDM2 and CDK4 immunohistochemistry. Hum Pathol 2017; 59: 34-40.

[19] Creytens D, Mentzel T, Ferdinande L, Lecoutere E, van Gorp J, Atanesyan L, de Groot K, Savola S, Van Roy N, Van Dorpe J, Flucke U. “Atypical” pleomorphic lipomatous tumor: a clinicopatho-logic, immunohistochemical and molecular study of 21 cases, emphasizing its relationship to atypical spindle cell lipomatous tumor and suggesting a morphologic spectrum (atypical spindle cell/pleomorphic lipomatous tumor). Am J Surg Pathol 2017; 41: 1443-1455.

[20] Creytens D, van Gorp J, Ferdinande L, Speel EJ, Libbrecht L. Detection of MDM2/CDK4 amplifi-cation in lipomatous soft tissue tumors from formalin-fixed, paraffin-embedded tissue: com-parison of multiplex ligation-dependent probe amplification (MLPA) and fluorescence in situ hybridization (FISH). Appl Immunohistochem Mol Morphol 2015; 23: 126-133.

[21] Clay MR, Martinez AP, Weiss SW, Edgar MA. MDM2 and CDK4 immunohistochemistry: should it be used in problematic differentiat- ed lipomatous tumors?: A new perspective. Am J Surg Pathol 2016; 40: 1647-1652.