記住我
Abstract
Context: Ewing sarcoma (ES) are malignant small round cell tumors (MSRCT) characterized by rearrangements of EWSR1 gene. Although gold standard for diagnosis is detection of specific fusion genes by molecular testing, these ancillary tests are costly and only available in limited number of settings. There is a persuasive evidence for reliability of NKX2.2 immunohistochemistry (IHC) as a surrogate marker for EWSR1 gene rearrangement in ES. Aims: The aim of this study is to correlate the NKX2.2 immuno-expression with genetically confirmed ES cases and also to assess the reliability and accuracy of NKX2.2 along with combined positivity of NXX2.2 and CD99 in diagnosing ES and differentiating it from other relevant histological mimics. Settings and Design: The present study is a retrospective study conducted over a period of 6-year duration in a tertiary cancer care center. Methods and Material: We evaluated NKX2.2 immunoexpression in 35 genetically confirmed cases of ES and also in pertaining differential entities (n = 58) of ES including rhabdomyosarcoma (n = 20), lymphoblastic lymphoma (n = 14), Wilms tumor (n = 10), poorly differentiated synovial sarcoma (n = 4), small-cell osteosarcoma (n = 4), neuroblastoma (n = 5), and mesenchymal chondrosarcoma (n = 1). CD99 was performed in the category of MSRCTs showing NKX2.2 positivity to evaluate combined specificity for the diagnosis of ES. Results: Of the 35 genetically confirmed cases of ES, 29 cases (83%) showed NKX2.2-positive expression (83% sensitivity). Compared to ES, NKX2.2 was positive in only 05% cases (3/58 cases) of non-ES MSRCT. Only two of five cases of neuroblastomas and one case of mesenchymal chondrosarcoma showed NKX2.2 positivity. CD99 positivity was seen in 100% of ES and in the single case of mesenchymal chondrosarcoma. All five cases (100%) of neuroblastoma were negative for CD99. Conclusions: The presented study, which is the first from an Indian oncology center, showed NKX2.2 IHC is quite reliable in diagnosis of ES in the right clinicopathological context. With remarkable sensitivity and specificity of NKX2.2 IHC for diagnosis of ES, we propose that combined positivity of CD99 and NKX2.2 IHC can obviate or minimize the need of EWSR1 gene rearrangement molecular testing for diagnosis of ES.
Keywords: CD99, Ewing sarcoma, EWSR1 gene rearrangement, immunohistochemistry, NKX2.2
How to cite this article: Introduction 
Ewing sarcoma (ES) are highly aggressive malignant tumors affecting predominantly long bones and soft tissue of pediatric population and young adults.[1] Morphologically, ES belongs to a family of malignant small round blue cell tumors (MSRCTs) with a broad differential diagnosis including rhabdomyosarcoma (RMS), small-cell osteosarcoma, non-Hodgkin's lymphoma, neuroblastoma, poorly differentiated synovial sarcoma (PDSS), and Wilms tumor.[2] ES is characterized by reciprocal chromosomal translocation and gene rearrangement involving the ES breakpoint region 1 gene (EWSR1) at 22q12 and one of the members of the ETS family of transcription factors.[3],[4] The most frequent fusion partner of EWS is FLI1 (90–95%) followed by ERG (5–10%) and other rare targets such as ETV1, E1AF, and FEV (each <1%).[5],[6] Although molecular testing and demonstration of the gene fusion is the gold standard for the definite diagnosis of ES, however, such approaches are difficult to establish in routine pathology practice, especially in resource constraints nations.[5]
CD99/MIC2 has been used historically as a diagnostic immunohistochemistry (IHC) marker for ES but has a very poor specificity and is commonly expressed in the histological mimics of ES.[6],[7] Recently, there is a persuasive evidence that NKX2.2, a transcriptional target of EWSR1-FLI1, has good utility as an IHC marker for the diagnosis of ES. However, few studies have questioned its specificity for the diagnosis of ES.[1],[8]
ES tumors are highly sensitive to chemotherapy with a very specific chemotherapeutic regime and patients with localized disease have reported an overall survival rate of 70% with primary treatment.[2] Therefore, distinguishing ES from their broad range of differentials and providing an accurate diagnosis is crucial.[3],[6]
The present study was undertaken to correlate the NKX2.2 immuno-expression with genetically confirmed ES cases and also to assess the reliability and accuracy of NKX2.2 along with combined IHC positivity of NXX2.2 and CD99 in diagnosing ES and differentiating it from other relevant histological mimics.
Subjects and Methods A total of 35 genetically confirmed (with EWSR1 gene rearrangement) cases of ES were retrieved from the pathology records of our hospital during the last 6-year duration (2014–2020). All these cases were subjected to NKX2.2 and CD99 IHC.
Fluorescence in-situ hybridization (FISH) for EWSR1 rearrangement was performed using the Metasystem XL EWSR1 dual-color, break-apart probe (MetaSystems USA in Medford, MA) according to the manufacturer's recommendations. Analysis of EWSR1 signals was done using Leica DM 6000B, Germany. Only cell nuclei with one yellow (fusion), one green, and one red signal detected simultaneously were considered positive for EWSR1 rearrangement. Signals were considered to be colocalized when their distance was equal to or smaller than the size of the hybridization signal. Samples were evaluated for the presence of fused or split signals in tumor cells and an estimated percentage reported. EWSR1 break-apart was reported positive if more than 30% of the tumor cells have shown break-apart signals.
Fifty-eight cases of non-ES MSRCTs that mimic ES on conventional morphology were also retrieved. Both H and E and complete IHC panel slides were reviewed. All these cases were subjected to NKX2.2 IHC. These included RMS (n = 20), lymphoblastic lymphoma (n = 14), Wilms tumor (n = 10), neuroblastoma (n = 5), PDSS (n = 4), small-cell osteosarcoma (n = 4), and mesenchymal chondrosarcoma (n = 1). Subsequently, CD99 was performed in the category of non-ES MSRCTs showing NKX2.2 positivity to evaluate combined specificity for the diagnosis of ES.
Immunohistochemistry
IHC was performed on 4-μm sections using anti-NKX2.2 monoclonal antibody (clone: NX2/294) and CD99 monoclonal antibody (clone: 12E7) on Ventana Benchmark XT autostainer. IHC results were recorded as NKX2.2 positive if distinct nuclear immuno-expression of at least moderate intensity was seen in >50% of tumor cells. CD99-positive immunoexpression was recorded if there is distinct membranous accentuation evident. The sensitivity, specificity, and positive predictive value (PPV) were further evaluated.
Results Of the total 35 genetically confirmed cases of ES, there was a male preponderance (n = 27; 77%). The age of the patients ranged from 7 to 53 years with a median age group of 28 years. The cases involved both extraskeletal and skeletal sites. In 54.3% of cases (n = 19), the tumor was located in the extraskeletal location while bone was involved in rest 46% cases (n = 16). The anatomic distribution of the cases included lower limb (43%), vertebral column (17%), soft tissue (11%), upper limb (9%), and head and neck (6%). The other rare sites such as kidney, stomach, and retroperitoneum comprise 14% of cases. All tumors on histology showed diffuse sheets of uniform round blue cells in solid or lobular architecture [Figure 1]. Frequent mitosis and focal-to-extensive areas of necrosis were evident in nearly all cases. Few of the cases showed prominent rosette formation.
Figure 1: (a) Microscopic image showing malignant small cell tumor with closely packed uniform round blue cells with inconspicuous nucleoli and scant cytoplasm (hematoxylin and eosin 200×); (b) strong diffuse membranous expression for CD99 (400×); (c) diffuse nuclear immunoexpression for NKX2.2 (200×); and (d) FISH image showing positive analysis for EWSR1 gene rearrangement with one fused signal (yellow), one red and one green signal (arrow)The IHC results have been documented in [Table 1]. Among the 35 genetically confirmed cases of ES, 83% (29 cases) showed diffuse NKX2.2-positive immuno-expression (moderate-to-strong intensity). The sensitivity and specificity of NKX2.2 IHC for diagnosis of ES were 83% and 95%, respectively, with PPV of 90.6%.
Compared to ES, NKX2.2 was positive in only 05% cases (3/58 cases) of non-ES MSRCT. Only two of five cases of neuroblastomas and one case of mesenchymal chondrosarcoma showed NKX2.2 positivity, while all cases (100%) of RMS, lymphoblastic lymphoma, PDSS, Wilms tumor, and small-cell osteosarcoma were negative.
We also evaluated the CD99 immunoexpression in categories of tumors showing NKX2.2 positivity [Figure 2]. CD99 positivity was seen in 100% of ES (35 cases) and in the single case of mesenchymal chondrosarcoma [Figure 3]. All five cases (100%) of neuroblastoma were negative for CD99 [Table 2].
Table 2: Evaluation of CD99 immunoexpression in the category of NKX2.2-positive MSRCTs
Figure 2: (a) Case of Ewing's sarcoma with EWSR1 gene rearrangement (hematoxylin and eosin 200×); (b) strong nuclear expression of NKX2.2; (c) strong membranous expression of CD99; (d) case of olfactory neuroblastoma (hematoxylin and eosin 200×); (e) diffuse nuclear expression of NKX2.2; and (f) negative expression of CD99 in tumor cells
Figure 3: (a) Case of mesenchymal chondrosarcoma showing undifferentiated mesenchymal cells with uniform round blue cells admixed with areas of chondroid differentiation (hematoxylin and eosin 200×); (b) diffuse positive expression for CD99; (c) positive expression for NKX2.2 (moderate intensity); and (d) FISH (break-apart) analysis negative for EWSR1 gene rearrangement Discussion ES is the second most common aggressive sarcoma in children and young adults characterized by balanced translocation involving EWSR1 gene.[9] The fusion partner of EWSR1 is FLI1 in approximately 90% of the cases, followed by most commonly ERG.[2],[8],[10],[11] CD99/MIC2 has been utilized as a diagnostic IHC marker and is still prevailing in many diagnostic centers.[9] CD99 has been found to be sensitive but has poor specificity as variety of non-ES MSRCTs show CD99 immunoexpression including lymphoblastic lymphoma, Wilms tumor, small-cell osteosarcoma, RMS, and PDSS.[2],[8],[12],[13] FLI-1 also gained popularity in the past for diagnosis of ES as a sensitive IHC marker; however, it lacks specificity which is even lower than CD99. Therefore, relying on CD99 or FLI1 as an independent diagnostic marker may prompt an erroneous diagnosis and further management.[12],[14],[15]
During last few years, many studies have shown the diagnostic utility of NKX2.2 IHC in ES with much better specificity. NKX2.2, an oncogenic transcriptional target of EWSR1/FLI1, is a member of NK2 family of transcription factors and plays an important role in development and differentiation of the gastrointestinal/pancreatic endocrine cells and central nervous system.[5],[6],[12],[16],[17]
In the present study, we found that NKX2.2 is a useful IHC marker for the diagnosis of ES with sensitivity of 83% and a high specificity of 95%. The specificity and diagnostic accuracy were found to be further enhanced with combined positivity of NKX2.2 and CD99, as two cases of neuroblastoma which were NKX2.2 positive were CD99 negative [Table 2]. CD99 and NKX2.2 dual positivity was seen in all the NKX2.2-positive cases (83%) of ES. Out of total 58 cases of non-ES MSRCT, only a single case (mesenchymal chondrosarcoma) showed combined positivity for CD99 and NKX2.2, hence reinforcing the high specificity of dual positivity.
The remaining six cases of ES (17%) which were negative for NKX2.2 on IHC showed classical morphology similar to the cases immunoreactive for NKX2.2. Though the negative immuno-expression is still unclear, it can be explained on the basis of variant fusion partners other than EWS-FLI1.
Shibuya et al.[5] evaluated NKX2.2 and CD99 expression in 46 genetically confirmed cases of ES and found the NKX2.2 diagnostic sensitivity and specificity of 80% and 84%. They also found CD99 expression in all the cases of ES similar to our study. With combined NKX2.2 and CD99 positivity, the sensitivity was 80% but the specificity substantially increased to 98%. The majority of cases of non-ES MSRCTs which showed NKX2.2 positivity were olfactory neuroblastoma and small-cell carcinoma; with combined assessment of CD99 and NKX2.2, only 1 out of 15 cases of small-cell carcinoma and 1 of 3 cases of mesenchymal chondrosarcoma showed combined positivity.
Yoshida et al.[18] evaluated NKX2.2 expression in 30 genetically confirmed cases of ES and 130 non-ES MSRCTs. Their study showed NKX2.2 diagnostic sensitivity and specificity of 93% and 89%, respectively.
Russell-Goldman et al.[2] and Hung et al.[8] reported NKX2.2 diagnostic sensitivity of 100% and 93% while modest specificity of 85% and 88%, respectively. However, combined CD99 and NKX2.2 dual positivity was not evaluated in both studies.[2],[8] Significant cases of the non-ES MSRCTs showing NKX2.2 positivity comprised mostly of neuroblastoma, which are characteristically CD99 negative, hence would be easily distinguished by dual CD99 and NKX2.2 immunoexpression evaluation.[15] The other major non-ES MSRCTs expressing NKX2.2 were small-cell carcinoma, well-differentiated neuroendocrine tumor, and Merkel cell carcinoma, which are rarely considered as differentials in germane to ES of bone and soft tissue tumors in clinicopathological practice.
Mesenchymal chondrosarcoma can show both CD99 and NKX2.2 positivity, thus posing a diagnostic challenge on Tru-cut biopsy if chondroid area is not included.[2],[6],[8]
In the present study, one and only single case of mesenchymal chondrosarcoma showed diffuse NKX2.2 positivity. Similar results in mesenchymal chondrosarcoma were shown by Hung et al.[8] in 75% cases and Yoshida et al.[18] in 33% cases. Features favoring the diagnosis of mesenchymal chondrosarcoma over ES are the presence of biphasic population of small round blue cells admixed with islands of cartilage, hemangiopericytoma-like growth pattern, and SOX-9 positivity.[6],[19] These tumors lack EWSR1 gene rearrangement and express defining HEY1–NCOA2 gene fusion.[1],[6],[8],[20]
Conclusion We conclude from our study that NKX2.2 IHC is a valuable and reliable IHC marker for diagnosing ES and differentiating it from other MSRCTs in a relevant clinicoradiological setting. It can be utilized as a convenient surrogate to molecular testing in the diagnosis of ES. Combined NKX2.2 and CD99 positivity has very high specificity than NKX2.2 alone. Testing for EWSR1 gene rearrangement with FISH or fusion RT-PCR can be reserved for few selected cases which are not resolved on IHC, as these molecular techniques are laborious, expensive, require expertise, and not accessible in resource-limited settings. With remarkable sensitivity and specificity of NKX2.2 IHC for diagnosis of ES, we propose that combined positivity of CD99 and NKX2.2 IHC can obviate or minimize the need of EWSR1 gene rearrangement molecular testing for diagnosis of ES.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References 
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/ijpm.ijpm_535_21
留言 (0)