Main

From a clinical perspective, testicular neoplasms of germ cell origin are generally categorized as either seminomas or nonseminomatous germ cell tumors, based on established differences between these groups with respect to biologic behavior and treatment. In addition to clinical stage, the appropriate therapeutic management of germ cell tumors of the testis is thus largely predicated upon accurate histologic classification.1

The histologic diagnosis of seminoma is relatively straightforward in most circumstances due to its characteristic morphologic features. However, some seminomas may exhibit increased nuclear pleomorphism, cell crowding, and lack a lymphocytic infiltrate, resulting in confusion with embryonal carcinoma.2, 3 Seminoma may also be difficult to distinguish from embryonal carcinoma in the context of a limited biopsy specimen or poor tissue fixation resulting in artifactual morphologic alterations.3

In these settings, immunohistochemistry may be utilized to facilitate distinction between seminoma and embryonal carcinoma. Evidence suggests that CD30 and KIT (CD117) are useful markers in this regard. CD30 is expressed by the majority of embryonal carcinomas and is generally negative in seminomas,4, 5, 6, 7 while KIT is regularly expressed in seminomas and only rarely positive in embryonal carcinomas.7, 8, 9, 10, 11 However, CD30 immunoreactivity has been observed in a number of pure seminomas and seminomatous components of mixed germ cell tumors,2, 7, 12, 13, 14 and KIT positivity in seminomas has been noted to be variable and weak,15 somewhat limiting the diagnostic utility of these particular markers.

D2-40 is a monoclonal antibody that reacts with an oncofetal membrane antigen known as the M2A antigen.16 The M2A antigen is present in fetal germ cells of the testis, as well as lymphatic endothelial cells and mesothelial cells.17, 18, 19 In the context of germ cell neoplasia, the distribution of the M2A antigen has been shown to be largely restricted to intratubular germ cell neoplasia and seminoma, with limited to absent expression in non seminomatous germ cell tumors.16, 17, 18, 19, 20, 21 The selective expression of the M2A antigen in seminoma suggests potential use of the D2-40 antibody for distinguishing this particular tumor from embryonal carcinoma. In the present study, the immunohistochemical expression of D2-40 was evaluated in testicular germ cell tumors and compared with CD30 and KIT to determine the relative usefulness of this particular marker in discriminating between seminoma and embryonal carcinoma.

Materials and methods

Forty cases of testicular germ cell tumor were identified from the files of the Department of Pathology, City of Hope National Medical Center, Duarte, CA, USA. Patients ranged in age from 18 to 41 years, with a mean age of 30 years. Hematoxylin and eosin-stained sections from all cases were reviewed to confirm the diagnosis. The testicular germ cell tumors were classified according to the World Health Organization criteria.22 The cases included 19 seminomas, three embryonal carcinomas, three teratomas, one yolk sac tumor, and 14 mixed germ cell tumors. The 14 cases of mixed germ cell tumors consisted of the following components: seven seminomas, 11 embryonal carcinomas, 10 teratomas, two yolk sac tumors, and one choriocarcinoma.

A representative tissue block containing tumor was selected from each case for immunohistochemical study. Immunohistochemical staining was performed using the following antibodies: D2-40 (clone D2-40, dilution 1:2, Signet Laboratories, Inc., Dedham, MA, USA), KIT (polyclonal, dilution 1:50, DakoCytomation, Carpinteria, CA, USA), and CD30 (clone Ber-H2, dilution 1:120, DakoCytomation). Sections were deparaffinized in xylene and rehydrated in a graded ethanol series. Antigen retrieval was performed by heating slides in EDTA buffer (pH 8.0) in a pressure cooker (Biocare Medical, Concord, CA, USA). Staining was performed using an automated immunostainer (DAKO), followed by antibody detection using the DAKO EnVision+ System and 3,3′-diaminobenzidine as a chromogen. The slides were counterstained with hematoxylin and coverslipped. Appropriate positive and negative tissue controls were used throughout.

Results

The immunohistochemical results are summarized in Table 1. All cases of pure seminoma and the seminomatous components of mixed germ cell tumors reacted with the D2-40 antibody. D2-40 immunoreactivity in seminomas was characterized by strong and diffuse membrane staining of the neoplastic cells (Figure 1). Twenty-four of the 26 (92%) seminoma samples expressed KIT, exhibiting strong membrane, or membrane and cytoplasmic staining patterns. All seminomas were negative for CD30.

Table 1 D2-40, KIT (CD117), and CD30 immunoreactivity in testicular germ cell tumors
Full size table
Figure 1
figure 1

D2-40 immunoreactivity in seminoma. (a) Seminoma, hematoxylin and eosin stain. (b) The seminoma cells exhibit membrane positivity for the D2-40 antibody.

Full size image

Of the 14 samples of embryonal carcinoma studied, 13 were positive for CD30. Three cases of pure embryonal carcinoma were negative for D2-40, while the embryonal carcinoma component of mixed germ cell tumors were positive for D2-40 in four of 11 cases. In contrast to the strong membrane staining observed in seminomas, D2-40 immunoreactivity in the embryonal carcinomas was typically weak, focal, and distributed along the apical or luminal surfaces of the neoplastic cells (Figure 2). All embryonal carcinomas in their pure form and as a component of mixed germ cell tumors were negative for KIT.

Figure 2
figure 2

D2-40 immunoreactivity in embryonal carcinoma. (a, c) Embryonal carcinoma, hematoxylin and eosin stain, exhibiting solid (a) and glandular (c) growth patterns. (b, d) Positive staining with the D2-40 antibody is focal in embryonal carcinomas, with immunoreactivity present along the apical/luminal surface of the neoplastic cells.

Full size image

Other germ cell tumor components including yolk sac tumor, teratoma, and choriocarcinoma were immunohistochemically negative for D2-40, KIT, and CD30.

Discussion

D2-40 is a monoclonal antibody which recognizes a 40 kDa O-glycosylated sialoglycoprotein with a simple mucin type carbohydrate structure known as the M2A antigen.16 Among normal tissues, this antigen has been shown to be present in fetal germ cells, lymphatic endothelium, and mesothelial cells.17, 18, 19 Initial immunohistochemical investigations performed on frozen tissue specimens using a monoclonal antibody produced against a human ovarian epithelial adenocarcinoma cell line demonstrated M2A antigen expression in all seminomas and seminomatous components of mixed germ cell tumors, but in none of the nonseminomatous germ cell tumors studied.17, 18

In the present investigation utilizing the D2-40 antibody on paraffin-embedded tissue samples, immunoreactivity was observed in all cases of seminoma, in agreement with the distribution of the M2A antigen established by previous studies using immunohistochemical analysis on frozen sections.17, 18 However, in contrast to the results of these aforementioned reports indicating an absence of M2A in nonseminomatous germ cell tumors, D2-40 immunoreactivity was observed in four of 14 (29%) samples of embryonal carcinoma in the present study. Unlike the diffuse membrane staining present in seminomas, D2-40 immunoreactivity in embryonal carcinomas was always focal and confined to the apical or luminal surfaces of the tumor cells. These findings are comparable to those obtained by other investigators using the D2-40 antibody.15, 16 Marks et al16 reported D2-40 immunoreactivity in 98% of seminomas, and were also able to demonstrate positivity for this antibody in 69% of embryonal carcinomas. As in the present study, D2-40 positivity in embryonal carcinomas was noted by these investigators to be focal and concentrated at the luminal surface of the neoplastic cells, whereas a uniform membrane pattern of staining was observed in seminomas.16

Distinguishing seminoma from embryonal carcinoma is important, as differences exist between these two conditions in regards to therapeutic modalities and prognosis. In the majority of cases, distinction between these two entities can be made on a morphologic basis using conventional histologic methods. However, in some situations appropriate histologic classification of testicular germ cell tumors may be problematic. Studies addressing the impact of central histopathologic review of previously diagnosed testicular tumors have demonstrated major discrepancy rates of 4–11%, with the majority of discrepancies attributed to errors in distinguishing seminoma from embryonal carcinoma.23, 24, 25 The use of immunohistochemistry may therefore be of value in differentiation of these tumor types.

Several immunohistochemical markers with potential utility in differentiating seminoma from embryonal carcinoma have been evaluated previously, among which include antibodies towards various keratins, KIT, and CD30. Early immunohistochemical studies suggested antikeratin antibodies could be useful for the distinction between seminoma and embryonal carcinoma based on observations that, unlike embryonal carcinomas, seminomas lacked keratin intermediate filaments.26, 27, 28 However, more recent studies utilizing contemporary immunohistochemical techniques on formalin-fixed tissues have indicated keratin expression in seminomas is not unusual, with keratin positivity ranging from 4 to 45% depending on the particular antibody employed.2, 5, 13, 14 Thus, keratin expression alone cannot be used to definitively distinguish seminoma from embryonal carcinoma. More effective immunohistochemical markers in this regard appear to be KIT and CD30.

In the context of testicular germ cell neoplasia, KIT immunoreactivity has been shown to be largely limited to seminomas, with positivity in non seminomatous germ cell tumors considered uncommon, and characterized by focal cytoplasmic, rather than the typical membrane pattern of staining.7, 8, 9, 10, 11 Previous immunohistochemical studies have demonstrated KIT expression in up to 100% of cases of seminoma.9, 10, 11 In the present study, 92% of seminomas were positive for KIT, while none of the embryonal carcinomas or other histologic types of non seminomatous germ cell tumors were noted to express this marker, in keeping with the findings of previous investigators. Although one study has observed heterogeneous and weak staining of seminomas using the KIT antibody,15 in our experience, KIT reactivity in seminomas was typically strong and diffuse.

CD30 is regarded as a sensitive as well as a specific maker for embryonal carcinoma, with other types of germ cell tumors, including seminoma, generally lacking CD30 expression.4, 5, 6, 7 Similar to previous reports, CD30 immunoreactivity was restricted to only pure embryonal carcinomas and the embryonal carcinoma components of mixed germ cell tumors in the current study.

Although not observed in the present investigation, it should be recognized that focal CD30 expression has been described in a subset of seminomas,2, 7, 12, 13, 14 thus somewhat limiting the specificity of CD30 as an individual marker for embryonal carcinoma. As such, an antibody panel consisting of CD30 in combination with KIT has been suggested by Leroy et al,7 to account for any shortcomings in the use of these particular markers individually in the distinction of seminoma from embryonal carcinoma. In that study, a KIT positive, CD30 negative immunophenotype was characteristic of seminoma, while a KIT negative, CD30 positive immunophenotype was indicative of embryonal carcinoma. In addition, none of the seminomas expressed CD30 in the absence of KIT, and none of the embryonal carcinomas expressed KIT in the absence of CD30. The results of the present study confirm these previous observations regarding the differential expression of these particular markers in seminoma and embryonal carcinoma. In all, 92% of the seminoma samples were positive for KIT and negative for CD30, while conversely 93% of the embryonal carcinoma samples were positive for CD30 and negative for KIT.

The data obtained in the current investigation suggest D2-40 may serve as a useful immunohistochemical marker for the identification of seminoma. D2-40 was commonly expressed in seminomas, with an observed sensitivity higher than that of KIT in the present study. However, D2-40 immunoreactivity was also observed in a subset of embryonal carcinomas, making this a less specific marker for seminoma than KIT. It is nonetheless important to note that the pattern of D2-40 positivity in seminomas was distinct from that observed in embryonal carcinomas. Reactivity for D2-40 in seminomas was typically diffuse and membranous, while in embryonal carcinomas, positivity was always focal and limited to the apical or luminal surface of the cells. However, despite these differences, the fact that nearly one-third of the embryonal carcinomas were D2-40 positive in the present investigation limits the diagnostic utility of this marker in differentiating embryonal carcinoma and seminoma.

In summary, antibodies to KIT and CD30 constitute a useful panel of markers that allows for seminoma to be distinguished from embryonal carcinoma. KIT expression in the absence of CD30 immunoreactivity would substantiate a diagnosis of seminoma, while conversely, CD30 positivity with lack of KIT expression would be consistent with embryonal carcinoma. Although a highly sensitive marker of seminomas, focal D2-40 immunoreactivity can be seen in a subset of embryonal carcinomas, thus limiting the practical value of this marker for discriminating between these particular malignancies.