2020 NPB-12

This case was originally published in 2020. The information provided in this case was accurate and correct at the time of initial program release. Any changes in terminology since the time of initial publication may not be reflected in this case.

The patient is a 25-year-old woman presenting with back pain and an intradural cauda equina region tumor. Intraoperatively, the tumor was noted to be highly vascular and adherent to nerve roots.

Tissue Site
Spinal cord, cauda equina

The whole slide image provided is an H&E-stained section from the cauda equina tumor resection.

Ewing sarcoma (ES) is the second most common solid tumor of bone after osteosarcoma, but roughly 12% are extraosseous neoplasms presenting in soft tissue, including within the craniospinal axis as a meningeal (intracranial or spinal), paraspinal, or peripheral nerve associated mass. ES is most commonly encountered in children and young adults, the latter age group more often represented in extraosseous cases such as this one. The histopathology in this cauda equina region resection showed sheets of primitive appearing and mitotically-active small round blue cells admixed with mature ganglion cells (Image A, Image B, and Image C). As such, one could consider a diagnosis of neuroblastoma in the differential, though the patient’s age of 25 years would be an exceptionally-late presentation for that tumor type. A drop metastasis from a medulloblastoma could also be considered based on H&E appearance alone. In ES, neuronal features most commonly manifest as Homer Wright rosettes (ie, primitive tumor cells surrounding central neuropil) and such cases were previously considered to represent “peripheral primitive neuroectodermal tumor” (pPNET), although this term is no longer utilized (discussed further below). Although rare, ganglion cell maturation has also been reported in ES. IHC workup can be extremely helpful in this diagnostic setting, although it does not always provide a definitive final diagnosis, and molecular confirmation is often necessary.

The ganglion cell differentiation in the current case was confirmed by immunoreactivity for neurofilament protein (Image D) and NeuN (Image E). There was no evidence of glial differentiation as one might see in tumors of CNS origin, highlighted by the lack of GFAP staining (Image F). However, there was limited evidence of a peripheral nerve sheath component in the form of focal SOX10-positive Schwann cells (Image G). There was extensive membrane positivity for CD99 in the primitive cells, but this was mostly lost in areas of ganglion cell maturation (Image H and Image I). Although this marker is considerably less specific for ES than originally touted, it is consistently negative in neuroblastoma, thus helping to exclude this possibility from further consideration. Strong nuclear staining for FLI1 (Image J) and cyclin D1 (Image K) provide further support for the diagnosis of ES. Rough estimates of IHC antibody sensitivities and specificities for ES in modern series are listed in Table 1, although some markers have not been studied in the full range of small round blue cell tumors.

Table 1: Useful IHC stains for the diagnostic workup of Ewing sarcoma versus other small round blue cell tumors.

An additional test performed in this case was EWSR1 breakapart FISH, which was positive for gene rearrangement. Although this assay has the disadvantage of not specifying the fusion partner, it is nonetheless diagnostic of Ewing sarcoma in this particular case where other small round blue cell tumors with EWSR1 gene rearrangement have essentially been excluded due to the presence of ganglion cells; these include desmoplastic small round cell tumor (WT1 fusion partner), extraskeletal myxoid chondrosarcoma (NR4A3 fusion partner), angiomatoid fibrous histiocytoma (CREB1 and ATF1 fusion partners most common), and myoepithelial neoplasms (POU5F1 and PBX1 fusion partners most common). Furthermore, given the diffuse immunoreactivity for both CD99 and FLI1 in the current case, the fusion partner is most likely FLI1 (see molecular discussion). Nevertheless, in other cases where the differential diagnosis includes these other entities, further studies to identify the specific fusion partner are recommended. Such studies could include reverse transcription PCR, fusion specific FISH, multiplex PCR panels, and “RNA seq” or other next-generation sequencing strategies that allow fusion detection. Moreover, because the fusions of ES lead to a highly-specific epigenomic signature, methylation profiling can also reliably detect these sarcomas.

Ewing sarcoma and pPNET were first described nearly a hundred years ago by James Ewing and others, with the former considered an undifferentiated small round cell tumor and the latter showing “neuroectodermal features,” mostly in the form of Homer Wright rosettes. ES most commonly affected bone, and in comparison, pPNET more commonly affected older patients, was localized to peripheral nerve or soft tissue, and was thought to have a worse prognosis. Another related tumor – “Askin tumor of the thoracopulmonary region” – was not described until 1979; reportedly, they more often recurred, but did not metastasize and lacked the glycogen accumulation of ES. However, these distinctions did not stand the test of time, and we now know that these three tumors represent a single genetically and biologically-distinct entity, with the diagnosis of ES now subsuming the pPNET and Askin tumor concepts. Of interest, the term “PNET” has been completely discarded in the latest versions of both the soft tissue/bone and CNS World Health Organization classification schemes and, therefore, is not a valid diagnosis in either the central or peripheral nervous systems.

In the current molecular era, ES is now defined as a small round cell sarcoma containing a gene fusion that combines a FUS/EWS/TAF15 (FET) RNA-binding protein family member (mostly EWSR1 and rarely FUS) with an E26 transformation specific (ETS) transcription factor family member (FLI1 > ERG >> ETV1, EIAF / ETV4, and FEV). The t(11;22)(q24;q12) that results in the EWSR1-FLI1 fusion transcript and protein product accounts for roughly 85% of all cases. Additional STAG2 (15% to 22%), CDKN2A (12%), and/or TP53 (7%) mutations are occasionally found and may be associated with a worse prognosis. Previously referred to as part of the “Ewing sarcoma family,” other round cell tumors with overlapping features, such as CIC-fused and BCOR-rearranged sarcomas are now considered unique diagnostic entities, despite still being treated on ES clinical protocols in most cases due to their rarity. Potential diagnostic clues in these mimics include atypical findings for ES, such as spindling, plasmacytoid and rhabdoid features, greater nuclear pleomorphism, and less extensive CD99 immunoreactivity. In contrast, the diffuse CD99 positivity encountered in mesenchymal chondrosarcoma (MCS) presents a potential diagnostic pitfall. However, the presence of hyaline cartilage, along with the HEY1-NCOA2 gene fusion now known to be specific for MCS, allows one to render a diagnosis of MCS.

In general, patient outcome has greatly improved using induction chemotherapy followed by tumor resection and radiation therapy. Nonetheless, about one-fourth of all cases present with metastatic disease, and this is the single most powerful prognostic variable, reducing the estimated five-year overall survival rates from ~70% to 80% in patients with localized disease to ~30% in those with disseminated disease. Also, in localized disease, complete response to induction chemotherapy (<10% viable tumor in the post-therapy resection specimen) is associated with a more favorable prognosis.

• ES is a biologically-distinct round cell sarcoma of children and young adults, which requires aggressive multimodality therapy.
• Rare cases of ES present with intracranial or spinal disease.
• Neuronal differentiation in ES is most commonly seen in the form of Homer Wright rosettes but may include ganglion cell maturation in rare examples.
• ES is molecularly defined by a characteristic FET-ETS gene fusion product, most commonly EWSR1-FLI1.

1. Chen J, Li M, Zheng Y, Zheng L, Fan F, Wang Y. Treatment outcomes and prognostic factors of patients with primary spinal ewing sarcoma/peripheral primitive neuroectodermal tumors. Front Oncol. 2019;9:555.
2. Grunewald TGP, Cidre-Aranaz F, Surdez D, et al. Ewing sarcoma. Nat Rev Dis Primers. 2018;4(1):5.
3. Hung YP, Fletcher CD, Hornick JL. Evaluation of NKX2-2 expression in round cell sarcomas and other tumors with EWSR1 rearrangement: imperfect specificity for Ewing sarcoma. Mod Pathol. 2016;29(4):370-80.
4. Kilpatrick SE, Reith JD, Rubin B. Ewing sarcoma and the history of similar and possibly related small round cell tumors: From whence have we come and where are we going? Adv Anat Pathol. 2018;25(5):314-26.
5. Koelsche C, Hartmann W, Schrimpf D, et al. Array-based DNA-methylation profiling in sarcomas with small blue round cell histology provides valuable diagnostic information. Mod Pathol. 2018;31(8):1246-56.
6. Machado I, Yoshida A, Lopez-Guerrero JA, et al. Immunohistochemical analysis of NKX2.2, ETV4, and BCOR in a large series of genetically confirmed Ewing sarcoma family of tumors. Pathol Res Pract. 2017;213(9):1048-53.
7. Magro G, Salvatorelli L, Alaggio R, et al. Diagnostic utility of cyclin D1 in the diagnosis of small round blue cell tumors in children and adolescents. Hum Pathol. 2017;60:58-65.
8. VandenHeuvel KA, Al-Rohil RN, Stevenson ME, et al. Primary intracranial Ewing's sarcoma with unusual features. Int J Clin Exp Pathol. 2015;8(1):260-74.
9. Weissferdt A, Kalhor N, Moran CA. Ewing sarcoma with extensive neural differentiation: a clinicopathologic, immunohistochemical, and molecular analysis of three cases. Am J Clin Pathol. 2015;143(5):659-64.