Indian Journal of Cancer Home 

[Download PDF]
Year : 2015  |  Volume : 52  |  Issue : 3  |  Page : 331--335

Histological diversity and clinical characteristics of Ewing sarcoma family of tumors in children: A series from a tertiary care center in South India

D Priya1, Rekha V Kumar1, L Appaji2, BS Aruna Kumari2, M Padma2, Prasanna Kumari1,  
1 Department of Pathology, Kidwai Memorial Institute of Oncology, Bengaluru, Karnataka, India
2 Department of Paediatric Oncology, Kidwai Memorial Institute of Oncology, Bengaluru, Karnataka, India

Correspondence Address:
Rekha V Kumar
Department of Pathology, Kidwai Memorial Institute of Oncology, Bengaluru, Karnataka


Background: The Ewing sarcoma family of tumors (ESFT) are aggressive malignant tumors with small round cell morphology affecting mainly children and adolescents. The aim of this study is to classify the histological diversity and clinical characteristics of ESFT in children from a Tertiary Care Center in South India. Materials And Methods: This retrospective descriptive study includes 51 cases of ES in children aged below 15 years. Clinical details were collected from case files. Histomorphological features were reviewed and tumors were subtyped into classic, primitive neuroectodermal tumor (PNET) and atypical variants along with immunohistochemical markers, cytogenetics, and fluorescence in situ hybridization (FISH). Results: Fifty-three percent were female and 47% were male with mean age of 10 years. The most common site of involvement was skeletal involvement in 71%, followed by soft tissue involvement in 23%, and visceral involvement in 6%. Localized disease at presentation was seen in 44%, locally advanced disease in 28%, and metastatic disease in 28%. Recurrence was documented during follow-up in 18% of the cases. Histomorphologically, classic type was the most common (72%) followed by PNET (20%) category and atypical variant (8%). All cases were immunoreactive for CD99. Cytogenetic study in 12 cases showed translocation t(11;22) (q24;12) in 80% and variant translocations such as t(3;16), t(3;11) with nonspecific numerical abnormalities in 20%. FISH was carried out for documentation of four cases with atypical histomorphology. Conclusion: ESFT had wide histological variation which required confirmation by ancillary studies.

How to cite this article:
Priya D, Kumar RV, Appaji L, Aruna Kumari B S, Padma M, Kumari P. Histological diversity and clinical characteristics of Ewing sarcoma family of tumors in children: A series from a tertiary care center in South India.Indian J Cancer 2015;52:331-335

How to cite this URL:
Priya D, Kumar RV, Appaji L, Aruna Kumari B S, Padma M, Kumari P. Histological diversity and clinical characteristics of Ewing sarcoma family of tumors in children: A series from a tertiary care center in South India. Indian J Cancer [serial online] 2015 [cited 2021 Jun 15 ];52:331-335
Available from:

Full Text


Ewing sarcoma (ES)/primitive neuroectodermal tumor (PNET) is one of the aggressive malignant small round cell tumors that occur in bone, soft tissue, and parenchyma, and the second most common tumor of bone in childhood and adolescence. This tumor was first described by Ewing in 1921 as “diffuse endothelioma of bone.”[1] The soft tissue counterpart was first reported by Angervall and Enzinger in 1975.[2] In 1979, Askin et al. reported identical tumors in the thoracopulmonary region which came to be known as Askin tumor.[3] Further work on the molecular characteristics revealed that both ES and PNET shared identical features and these were designated as “ES family of tumors” (ESFT). They represented the primitive mesenchymal neoplasm with limited capacity for multidirectional differentiation.[4] Seventy–80% of cases exhibit classic Ewing morphology and up to 20% display atypical features including large cell, adamantinoma-like, spindle cell sarcoma-like, sclerosing, clear cell, or vascular-like patterns.[5] A panel of immunomarkers such as CD99, leukocyte common antigen (LCA), cytokeratin, desmin, and neural markers such as neuron specific enolase (NSE), S100, and synaptophysin are usually used to distinguish the different tumors with small round cell morphology. Molecular testing is the gold standard and is recommended in some of the cases with atypical features. Approximately, three quarters of patients have initially localized disease and about two-thirds survive disease-free.[6] The advent of multimodality treatment, which includes local control by surgery and radiotherapy and systemic control by chemotherapy, has improved the overall survival in these cases. This study was performed to analyze the histological heterogeneity with ancillary techniques including immunohistochemistry, cytogenetics, and fluorescence in situ hybridization (FISH) and the clinical characteristics of pediatric cases from a Tertiary Care Center in South India.

 Materials and Methods

This retrospective descriptive study includes 51 pediatric patients, of age range 2–14 years, treated for ES in our hospital from August 2009 to April 2014. The clinical details were collected from case files. The clinical features such as age, sex, site of involvement, radiological findings, serum lactate dehydrogenase (LDH) value, soft tissue extension, metastasis, and recurrence were evaluated. The paraffin blocks of all cases were retrieved. Histopathological features were reviewed and the diagnosis of ES was confirmed. The tumors were further categorized into classic, PNET, and atypical subtypes. Immunohistochemical (IHC) marker panels which included CD99, cytokeratin (CK), synaptophysin, chromogranin, NSE, S100, desmin, and LCA were also reviewed [Table 1]. A diagnosis of PNET was considered when Homer Wright rosettes were seen or when any two different neural markers were positive. The atypical category included cases with large and pleomorphic cells. Presence of necrosis and mitotic activity were noted in all cases. Immunohistochemistry analysis for all the above markers was done on 4 µm thick sections collected on silane-coated slides by immunoperoxidase methods with respective antibodies as per manufacturers' instructions. Antigen retrieval was achieved by the heat-induced epitope retrieval method in Tris-ethylenediaminetetraacetic acid buffer pH 6.0 after Tris buffer wash. The endogenous avidin-binding activity was blocked by immersing in skimmed milk powder. The sections were incubated in primary antibody for 1 h 30 min. Enhancer with horse radish peroxidase polymer (Biogenex) was used as secondary antibody and 3'3-diaminobenzidine as the chromogenic substrate. Appropriate positive and negative controls were included. IHC markers were scored as follows: No staining or very exceptional positive cells were considered negative. Staining in <25% of the tumor cells was considered a mildly positive immunoreaction (1+), staining in 25–50% as moderately positive (2+), and in >50% as strongly positive (3+).[7]{Table 1}

For cytogenetic analysis, the fine needle aspiration biopsy material was cultured in RPMI-1640 medium supplemented with fetal bovine serum and harvested, and slides were prepared and Giemsa-Trypsin-Giemsa banding procedure was followed. FISH was carried out on formalin-fixed paraffin embedded tissue. Break apart probes were used for three cases (Vysis EWSR1 Break Apart FISH probe kit) and dual fusion probe was used for one case (Cytocell Aquarius FLI1/EWSR1 Translocation Dual fusion probe kit). The dual fusion probe kit contained FLI1/EWSR1 probe mix which consisted of green probes flanking the breakpoint region at the EWSR1 gene locus and red probes flanking the breakpoint region at FLI1 locus. The Vysis EWSR1 (22q12) Break Apart Probe Kit consists of a mixture of two FISH DNA probes. The first probe, a ~ 500 kb probe labeled in Spectrum Orange, flanks the 5' side of the EWSR1 gene and extends inward into intron 4. The second probe, a ~ 1100 kb probe labeled in Spectrum Green, flanks the 3' side of the EWSR1 gene. The test was interpreted positive when more than 10% of the nuclei showed the split or fused signals.

Data analysis

The data were analyzed using the statistical software, SPSS (version 15.0; SPSS, Chicago, Illinois, USA) for Windows. The correlation between the discrete variables was performed using Chi-square test and log-rank test for pair wise comparisons. P <0.05 was considered to be statistically significant.


Clinical data

This retrospective study comprised 51 children, of whom 53% (27/51) were female and 47% (24/51) were male. The age ranged between 2 and 14 years (mean: 10 years). The most common site of involvement was skeletal involvement in 71% (36/51), followed by soft tissue in 23% (12/51), and viscera in 6% (3/51). Parenchymal involvement included two cases in the kidney and one case in the adrenal gland. The most common bones involved were the femur and humerus, each in 11.8% (6/51) of the cases. The patients presented with local swelling as the most common complaint followed by pain, fever, limb weakness, and fracture. The size of the lesion varied from 1 cm to a maximum of 17 cm. Eighty-three percent (15/18) of the lesions in the long bones were located in the metadiaphyseal region.

Detailed clinical data were available only in 39 out of 51 patients. Localized disease at presentation was seen in 44% (17/39), locally advanced disease in 28% (11/39), and metastatic disease in 28% (11/39). Soft tissue extension was seen in 63% (15/24) of the osseous cases, of which five had metastasis and three had recurrence. The serum LDH value at diagnosis ranged from 142 U/L to 1965 U/L with a median value of 363 U/L. Out of 14 cases with LDH levels above the median value of 363 U/L, two had recurrence and six had metastasis. The increased levels of LDH correlated significantly with the incidence of metastasis (P = 0.03). Bone marrow involvement was seen in 10% (4/39) of the cases. Metastasis to other sites such as lung, liver, bone, and lymphnode was noted in 28% (11/39). Metastasis was more common in extraosseous tumors (36%). Recurrence was documented in 18% of the cases. The patients received multimodality treatment which included surgery, chemotherapy, and radiotherapy in accordance with their disease status.

Histological characteristics

Seventy-two percent (37/51) of cases exhibited classic morphology. Microscopically, the tumor had diffuse or lobular pattern of arrangement of cells. The tumor cells were small, with uniform round to oval or indented nuclei, and fine chromatin [Figure 1]a. The cytoplasm was pale, clear, and sometimes vacuolated. Some of the cases showed darker crushed cells. The mitotic activity was low (1–2/10 hpfs). The cases with necrosis exhibited increased mitotic activity ranging from 3 to 7/10 hpfs. All of them showed strong membranous CD99 expression [Figure 1]b.{Figure 1}

Twenty percent cases (10/51) were classified as PNET according to the criteria proposed by Schmidt et al.[8] A diagnosis of PNET was considered when Homer Wright rosettes were seen or when any two different neural markers (NSE/synaptophysin/chromogranin/S100) were positive. The number of rosettes seen in these cases varied from a few to many. The rosettes had a central core of neurofibrillary material, which was surrounded by cells with wreath-like nuclear arrangement [Figure 1]c.

There were four atypical cases in this cohort, two of which were in bone and two in soft tissue. Three of them had large cells arranged in diffuse sheets and in nests. The tumor cells had moderate to marked nuclear pleomorphism, hyperchromatic nuclei, and one tumor demonstrated multilobulated nuclei and prominent nucleoli [Figure 1]d. These cells had a moderate amount of eosinophilic cytoplasm. Mitotic activity was quite high (>10/10 hpf) in contrast to the classic cases. Large areas of necrosis were seen in two of three cases.

Among the atypical cases, one case was categorized as sclerosing variant of ES since the tumor had abundant hyalinized matrix. The tumor cells were arranged in trabeculae and cords with the small round cell morphology resembling a desmoplastic small round cell tumor. CD99 positivity and WT1 negativity supported the diagnosis of ES, which was further confirmed by FISH.

Necrosis was seen in 27.8% of cases, commonly in the atypical variant (2/4 cases), followed by PNET (3/10) and classical subtypes (9/37). All tumors had strong, diffuse, and membranous expression of CD99. The expression of various markers varied among the tumors, while synaptophysin was expressed in 16.7% tumors, CK in 8% of cases, S100 expression was seen in 8%, and NSE expression in 10% of the cases [Figure 2]a,[Figure 2]b,[Figure 2]c,[Figure 2]d. Desmin, LCA, and chromogranin were not expressed in any of the tumors.{Figure 2}

Cytogenetic analysis was done in 12 cases. The translocation classic for ES t(11;22) (q24;12) was observed in 80% and variant translocations such as t(3;16), t(3;11) along with nonspecific numerical abnormalities were seen in 20% [Figure 3]a. FISH was carried out for documentation of three cases with atypical histomorphology. Break apart probes were applied for three cases [Figure 3]b and dual fusion probe for one case with signal positivity of 20%, 30%, 20%, and 80% of interphase nuclei, respectively.{Figure 3}


This series comprised pediatric patients (age below 15 years) who were enrolled and treated in a Tertiary Care Center in South India. There was female preponderance and the mean age was 10 years. The most common sites of involvement were bones of both upper and lower extremities (appendicular skeleton). The patients presented most often with swelling followed by pain. A few cases had fever and weakness of limbs which is in accordance with other studies.[9]

The evaluated prognostic factors included soft tissue extension, metastasis, recurrence, and serum LDH value. We noted that among 15 patients with soft tissue extension, five had metastasis and three patients presented with recurrence, which is higher compared to the rest of the cohort, though not reaching statistical significance. According to a study by Mendenhall et al., cases with extraosseous extension often had metastasis with prognostic significance.[10]

The serum LDH level had clinical value in predicting the course of the disease. In our study, the values ranged from 142 U/L to 1965 U/L with a median of 363 U/L and

75% of cases with LDH value above this median had metastasis with statistical significance (P < 0.05), in concurrence with a previous report by Bacci et al. in 1999.[11]

The incidence of metastasis at presentation was 10%, while it was 28.2% (11/39) throughout the disease course. Lung 8% (3/39) and bone marrow 10% (4/39) were the most common sites of metastasis. Local recurrence was seen in 18% (7/39) of the cases.

Recurrence and metastatic rates were higher in extraosseous sites than osseous sites (29% vs. 12% and 36% vs. 24%), respectively, in this study while Orr et al. had reported an incidence of metastasis in 13% of extraosseous tumors.[12]

The prognosis for patients with ES/PNET had steadily improved. About 75% of patients presented with localized disease, and the combination of surgery and/or radiotherapy and systemic chemotherapy leads to a cure rate near 75% in this group.[13],[14],[15] In our hospital, the patients were given multimodality treatment according to their disease status.

Histomorphologically, the classic type was the most common followed by PNET and atypical variants, the incidence of which is similar to a study done by Folpe et al.[16] Llombart-Bosch et al. reported 19.2% of cases to be atypical.[17] All of our atypical cases showed diffuse and strong membranous CD99 expression. One of them had strong CK positivity. None of them were positive for neuroectodermal immunomarkers. The differential diagnosis that could be considered for atypical variant on histomorphology included rhabdomyosarcoma, lymphoma, and small cell osteosarcoma. Folpe et al. in 2005 analyzed the importance of confirmation of atypical variants of ES by molecular cytogenetics. Accordingly, all the atypical variants in this study were confirmed by FISH.[16] In ESFT, 90–95% of the tumors have a balanced translocation involving chromosomes 11 and 22, which fuse portions of the EWS gene on 22q12 with the FLI1 gene on 11q24, thus creating a novel fusion gene with oncogenic properties.[18]

Though there has been much emphasis on the application of cytogenetics for the confirmation of ES, many centers in the developing countries still do not have this facility in-house. Thus, in these centers, immunohistochemistry plays an important role in differentiating between the various small blue round cell tumors. In the panel of immunomarkers, CD99 is an important and “traditionally used” one. In our study, all the cases showed strong and diffuse membranous expression of CD99. Among the other markers, the most commonly expressed were the markers for neuroectodermal differentiation. This included synaptophysin which was expressed in 16.7%, NSE in 9.3%, and S100 in 7.4% of the cases. The expression of neural markers was less compared to the data in a previous study by Kavalar et al.,[7] where NSE and S100 were expressed in 66.6% and 25.4%, respectively. However, the proportion of marker positivity in ES in the present study fell within the range of other studies.[19],[20] In our cases, the expression of neural immunomarkers did not correlate with the presence of rosettes. Some of the cases that lacked light microscopic evidence of neural differentiation could be identified with these immunomarkers. Pancytokeratin expression that was seen in 9.3% of the cases highlights the epithelial differentiation in the Ewing family. There is a variability in the reported positive rates by different authors which range from 20% to 40%.[21],[22],[23] Other markers such as desmin and LCA were not expressed in our cases.

We had two cases involving the renal parenchyma. Primary renal involvement is extremely rare in the pediatric age group. Both of them exhibited classic morphology. They need to be differentiated from blastemal predominant Wilms tumor and other primitive renal tumors due to different therapeutic and prognostic implications. Strong CD99 positivity and WT-1 negativity with cytogenetics confirmed the diagnosis of ES. One of these cases presented with metastasis to liver, spleen, and bone which highlights the aggressive behavior of renal ES described in a previous study.[24]

One interesting and rare case of primary adrenal ES/PNET was encountered in a 2-year-old child who had a mass in adrenal gland with hepatomegaly. Microscopy of adrenal tumor showed small round cells in sheets with focal calcification and liver showed infiltration by tumor in multiple small nests. Absence of neuropil, rosette formation, ganglionic differentiation, and presence of strong membranous CD99 expression favored a diagnosis of ES which was further confirmed by the presence of t(11;22). The tumor at this site is prone to recur and metastasize as described by Komatsu et al. in 2006.[25] This case presented with metastasis which reflected the aggressive nature of the lesion at this site.


In our study, the analysis of histological heterogeneity of ESFT emphasizes the vital role of immunohistochemistry, cytogenetics, and FISH in supporting its diagnosis.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Ewing J. Diffuse endothelioma of bone. Proc N Y Path Soc 1921;7:17-24.
2Angervall L, Enzinger FM. Extraskeletal neoplasm resembling Ewing's sarcoma. Cancer 1975;36:240-51.
3Askin FB, Rosai J, Sibley RK, Dehner LP, McAlister WH. Malignant small cell tumor of the thoracopulmonary region in childhood: A distinctive clinicopathologic entity of uncertain histogenesis. Cancer 1979;43:2438-51.
4Moll R, Lee I, Gould VE, Berndt R, Roessner A, Franke WW. Immunocytochemical analysis of Ewing's tumors. Patterns of expression of intermediate filaments and desmosomal proteins indicate cell type heterogeneity and pluripotential differentiation. Am J Pathol 1987;127:288-304.
5Wu HT, Govender D. Ewing sarcoma family of tumours: Unusual histological variants and immunophenotypic characteristics. Diagn Histopathol 2012;18:348-55.
6Bernstein M, Kovar H, Paulussen M, Randall RL, Schuck A, Teot LA, et al. Ewing's sarcoma family of tumors: Current management. Oncologist 2006;11:503-19.
7Kavalar R, Pohar Marinsek Z, Jereb B, Cagran B, Golouh R. Prognostic value of immunohistochemistry in the Ewing's sarcoma family of tumors. Med Sci Monit 2009;15:CR442-52.
8Schmidt D, Herrmann C, Jürgens H, Harms D. Malignant peripheral neuroectodermal tumor and its necessary distinction from Ewing's sarcoma. A report from the kiel pediatric tumor registry. Cancer 1991;68:2251-9.
9Hoffmann C, Ahrens S, Dunst J, Hillmann A, Winkelmann W, Craft A, et al. Pelvic Ewing sarcoma: A retrospective analysis of 241 cases. Cancer 1999;85:869-77.
10Mendenhall CM, Marcus RB Jr., Enneking WF, Springfield DS, Thar TL, Million RR. The prognostic significance of soft tissue extension in Ewing's sarcoma. Cancer 1983;51:913-7.
11Bacci G, Ferrari S, Longhi A, Rimondini S, Versari M, Zanone A, et al. Prognostic significance of serum LDH in Ewing's sarcoma of bone. Oncol Rep 1999;6:807-11.
12Orr WS, Denbo JW, Billups CA, Wu J, Navid F, Rao BN, et al. Analysis of prognostic factors in extraosseous Ewing sarcoma family of tumors: Review of St. Jude Children's Research Hospital experience. Ann Surg Oncol 2012;19:3816-22.
13Grier HE, Krailo MD, Tarbell NJ, Link MP, Fryer CJ, Pritchard DJ, et al. Addition of ifosfamide and etoposide to standard chemotherapy for Ewing's sarcoma and primitive neuroectodermal tumor of bone. N Engl J Med 2003;348:694-701.
14Rodriguez-Galindo C, Spunt SL, Pappo AS. Treatment of Ewing sarcoma family of tumors: Current status and outlook for the future. Med Pediatr Oncol 2003;40:276-87.
15Rodriguez-Galindo C. Pharmacological management of Ewing sarcoma family of tumours. Expert Opin Pharmacother 2004;5:1257-70.
16Folpe AL, Goldblum JR, Rubin BP, Shehata BM, Liu W, Dei Tos AP, et al. Morphologic and immunophenotypic diversity in Ewing family tumors: A study of 66 genetically confirmed cases. Am J Surg Pathol 2005;29:1025-33.
17Llombart-Bosch A, Machado I, Navarro S, Bertoni F, Bacchini P, Alberghini M, et al. Histological heterogeneity of Ewing's sarcoma/PNET: An immunohistochemical analysis of 415 genetically confirmed cases with clinical support. Virchows Arch 2009;455:397-411.
18Turc-Carel C, Aurias A, Mugneret F, Lizard S, Sidaner I, Volk C, et al. Chromosomes in Ewing's sarcoma. I. An evaluation of 85 cases of remarkable consistency of t(11;22) (q24;q12). Cancer Genet Cytogenet 1988;32:229-38.
19Carter RL, al-Sams SZ, Corbett RP, Clinton S. A comparative study of immunohistochemical staining for neuron-specific enolase, protein gene product 9.5 and S-100 protein in neuroblastoma, Ewing's sarcoma and other round cell tumours in children. Histopathology 1990;16:461-7.
20Amann G, Zoubek A, Salzer-Kuntschik M, Windhager R, Kovar H. Relation of neurological marker expression and EWS gene fusion types in MIC2/CD99-positive tumors of the Ewing family. Hum Pathol 1999;30:1058-64.
21Gu M, Antonescu CR, Guiter G, Huvos AG, Ladanyi M, Zakowski MF. Cytokeratin immunoreactivity in Ewing's sarcoma: Prevalence in 50 cases confirmed by molecular diagnostic studies. Am J Surg Pathol 2000;24:410-6.
22Collini P, Sampietro G, Bertulli R, Casali PG, Luksch R, Mezzelani A, et al. Cytokeratin immunoreactivity in 41 cases of ES/PNET confirmed by molecular diagnostic studies. Am J Surg Pathol 2001;25:273-4.
23Elbashier SH, Nazarina AR, Looi LM. Cytokeratin immunoreactivity in Ewing sarcoma/primitive neuroectodermal tumour. Malays J Pathol 2013;35:139-45.
24Oliva E, Amin MB, Jimenez R, Young RH. Clear cell carcinoma of the urinary bladder: A report and comparison of four tumors of mullerian origin and nine of probable urothelial origin with discussion of histogenesis and diagnostic problems. Am J Surg Pathol 2002;26:190-7.
25Komatsu S, Watanabe R, Naito M, Mizusawa T, Obara K, Nishiyama T, et al. Primitive neuroectodermal tumor of the adrenal gland. Int J Urol 2006;13:606-7.