Epithelial membrane antigen in differential diagnosis of malignant mesothelioma, metastatic adenocarcinoma, and reactive mesothelial hyperplasia

Sibel Arslan¹, Kemal Bakır², Levent Elbeyli³

¹Department of Pathology, Gaziantep Şehitkamil State Hospital, Gaziantep, Turkey
²Departments of Pathology, Medical Faculty of Gaziantep University, Gaziantep, Turkey
³Departments of Thoracic Surgery, Medical Faculty of Gaziantep University, Gaziantep, Turkey

DOI : 10.5606/tgkdc.dergisi.2016.11606

Abstract

Background: This study aims to investigate the usefulness of epithelial membrane antigen immunohistochemical marker in differential diagnosis of malignant mesothelioma, reactive mesothelial hyperplasia, and metastatic adenocarcinoma.

Methods: In this study, 67 malignant mesothelioma, 18 reactive mesothelial hyperplasia, and 33 metastatic adenocarcinoma specimens were examined and stained with epithelial membrane antigen.

Results: Staining with epithelial membrane antigen was observed in 45 of 67 malignant mesotheliomas cases (68.7%) and all the metastatic adenocarcinoma cases (100%). All metastatic adenocarcinomas showed prominently expressed epithelial membrane antigen, whereas weak positivity with epithelial membrane antigen was detected in only one case with reactive mesothelial hyperplasia. Metastatic adenocarcinoma showed a statistically significantly stronger expression compared to mesothelial hyperplasia (p<0.001). Sensitivity and specificity for malignant and benign cases were 78% and 94%, respectively.

Conclusion: Epithelial membrane antigen is a useful positive marker of malignant mesothelioma and metastatic adenocarcinoma. Also, it plays a significant role in distinguishing neoplastic mesothelium from non-neoplastic mesothelium.

Abstract

Malignant mesothelioma (MM) is a primary tumor of serous membranes, especially seen in the pleura which accounts for approximately 70% of all cases of MM.[1] Histologically, MM has been divided into epithelioid, sarcomatoid and biphasic types.[2] Median survival after diagnosis is 4.5 to 17 months, depending on histological type, tumor stage, performance status and treatment, and other factors such as sex and age.[3-5]

Histopathological differentiation of MM from metastatic adenocarcinoma (MA) and reactive mesothelial hyperplasia (RMH) is often difficult, particularly in small biopsy specimens.[6,7] Several morphological features have been identified,[8] but these are not specific criteria. Clinical, radiologic, histological, and histochemical studies can be helpful in differential diagnosis, but their diagnostic utility is limited.

Immunohistochemistry is valuable for the differential diagnosis.[9] There are several studies about immunohistochemical markers. International Mesothelioma Panel recommended using at least two mesothelial cell makers and two carcinoma-related markers in distinction of MM from MA.[10]

Epithelial membrane antigen (EMA) is expressed both in MM and MA,[11] and appears to be the most useful marker in distinguishing benign from malignant mesothelial proliferations since EMA appears to be preferentially expressed in neoplastic mesothelium.[12] Thus, in this study, we aimed to investigate the usefulness of EMA in differential diagnosis of MM, RMH, and MA.

Methods

Paraffin-embedded blocks from surgical material of 67 malign mesotheliomas, 18 reactive mesothelial hyperplasias, and 33 metastatic adenocarcinomas were retrieved between January 2005 and June 2011 from the archive of the Department of Pathology of the University of Gaziantep. Each case was reviewed, and all cases were diagnosed by currently accepted histological criteria combined with immunohistochemical features. Immunohistochemistry was performed on formalin-fixed, paraffin-embedded tissue sections using an avidin-biotin-peroxidase technique EMA (clone E29, Cellmarq, dilution 1: 200). Appropriate sections were cut 4 μm thick. Immunoreactivity was scored as negative (no immunostaining) or positive.

Differences of staining ratio between MM, RMH, and MA groups were tested by the chi-square. A p value o f < 0.05 was accepted as statistically significant.

Methods

Results

The immunohistochemical results are summarized in Table 1. Epithelial membrane antigen was stained in 45 of 67 MMs (68.7%) and in all the MAs (100%). All MA cases had strong positivity with EMA, whereas only one RMH case showed weak immunoreactivity with EMA. Immunohistochemical staining pattern did not make a differential diagnosis between MM, MA, and RMH cases. Both MM and MA cases showed membranous and cytoplasmic staining and cytoplasmic staining was prominent in most of cases. The positive staining of EMA in MA and negativity of EMA stain in RMH were statistically significant (p<0.000). Sensitivity and specificity for malignant and benign cases were 78% and 94%, respectively.

Table 1: Epithelial membrane antigen reactivity in malignant mesothelioma, metastatic adenocarcinoma and reactive mesothelial hyperplasia groups

Results

Discussion

Malignant mesothelioma is a primary tumor of pleura, pericardium, and peritoneum. It is a very aggressive tumor and histopathological diagnosis is difficult. Differential diagnosis between MM, RMH, and MA is a problem for pathologists, especially on the basis of routine histology alone. Immunohistochemistry has an important role in the differential diagnosis of these lesions.[13,14]

Epithelial membrane antigen is a member of the glycosylated protein group (human milk fat globule).[1] Although EMA is often expressed by both epithelial mesotheliomas and adenocarcinomas, it has been reported to be a valuable marker in differential diagnosis between MM, MA and RMH.[5,12,15] Saad et al.[16] used Mc5 and E29 clones of EMA in their study. For the Mc5 clone, there was positive staining in 14/20 cases of MM (70%) and 12/20 (60%) cases of RMH; for the E29 clone, the corresponding results were 15/20 (75%) for MM and 0/20 RMH. They concluded that EMA clone E29 was a reliable marker in differential diagnosis of MM and RMH.

Attanoos et al.[12] have shown that EMA may be applied as a differential marker for reactive and neoplastic mesothelium, as a positive marker of neoplastic mesothelium.

In several studies, EMA expression has been shown to be between 58 to 100% in MM cases and 0 to 55% in RMH cases.[5,15,17,18] Walz and Koch.[19] detected EMA in 33 (76.7%) of 43 mesotheliomas, and Wick et al.[20] found that 43 (84.3%) of 51 epithelial mesothelioma expressed EMA. Hammar et al.[21] detected EMA in 50 (78.1%) of 64 mesotheliomas and in 37 (61.6%) of 60 adenocarcinomas. Gümürdülü et al.[22] found that EMA stained 50 of 55 MMs and all the adenocarcinoma cases. Similarly, in our series, EMA was expressed in 45 of 67 MMs (68.7%), one of 18 RMH (5.6%), and in all MAs (n=33) (Figure 1, 2 and 3).

Figure 1: Epithelial malignant mesothelioma. (a) Hematoxylin-eosin stain (H-E x 100); (b) Epithelial membrane antigen immunostain showing a strong cytoplasmic staining (x100).

Figure 2: Reactive mesothelial hyperplasia. (a) H-E x 100); (b) Epithelial membrane antigen immunostain is negative (x100).

Figure 3: Metastatic adenocarcinoma. (a) H-E x 100); (b) Membranous immunostaining with epithelial membrane antigen (x100).

Several studies showed that the EMA reactivity in mesotheliomas was often markedly concentrated on the cell membrane, whereas the staining pattern in adenocarcinomas was predominantly cytoplasmic.[12,22,23] Aydıner and Yerci[23] detected positive staining in all MM and MA cases (100%); all MM cases were stained in membranous pattern, 75% of MA cases were stained in cytoplasmic pattern, other 25% were stained in both cytoplasmic and membranous patterns. However, in our cases, there was no difference between staining patterns and most of the cases stained in cytoplasmic pattern.

Cury et al.[15] concluded that strong, diffuse, and linear staining for EMA is a good marker of malignancy, although weak focal staining may occur in reactive conditions. Ikeda et al.[24] showed that, staining intensity of EMA was weak in RMH cases, whereas it was strong in the malignant cells. In our study, EMA staining was weak in one of RMH cases.

King et al.[13] reviewed 15 papers to evaluate the ability of immunohistochemistry to distinguish between benign and malignant pleural disease, which revealed a sensitivity of 74% and specificity of 89% for EMA. In our study, sensitivity and specificity for malignant and benign cases were 78% and 94%, respectively.

In conclusion, our study suggests that epithelial membrane antigen is a useful positive marker for malignant mesothelioma and metastatic adenocarcinoma. Also, it is a useful marker for distinguishing neoplastic mesothelium from nonneoplastic mesothelium. Epithelial membrane antigen positivity supports malignant mesothelioma; however, epithelial membrane antigen negativity does not eliminate malignant mesothelioma.

Declaration of conflicting interests
The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Funding
The authors received no financial support for the research and/or authorship of this article.

Discussion