Novel marker D2-40, combined with calretinin, CEA, and TTF-1

An optimal set of immunodiagnostic markers for pleural mesothelioma




Malignant pleural mesothelioma is a challenging disease with regard to diagnosis and treatment; early and accurate diagnosis would lead to appropriate therapeutic strategies, including extrapleural pneumonectomy. Immunohistochemistry has proven valuable for the diagnosis of the most common epithelioid mesothelioma, although it is often difficult to differentiate it from pulmonary or metastatic adenocarcinoma with absolute certainty if a single antibody is employed. The current study was designed to identify an immunodiagnostic panel for pleural mesothelioma.


Large surgical specimens from 66 cases with pleural mesothelioma and 66 with lung adenocarcinoma were immunohistochemically reevaluated under uniform conditions. The antibodies examined were directed against the novel mesothelial marker D2-40, as well as calretinin, CEA, and TTF-1.


For mesothelioma the sensitivities of D2-40 and calretinin were 84.8% and 87.9%, respectively, and their specificities were both 95.5%. For adenocarcinoma, the sensitivities of CEA and TTF-1 were 95.5% and 92.4%, respectively, and their specificities were both 100%. Immunoreactivity to D2-40 and calretinin was observed in most areas of epithelioid differentiation in mesothelioma. Western blots also showed higher levels of D2-40 antigen in pleura invaded by epithelioid mesothelioma as compared with unaffected pleura.


These data strongly suggest the significant usefulness of D2-40 and calretinin as positive markers, and of CEA and TTF-1 as negative markers, for pleural mesothelioma. The 4-antibody immunohistochemical panel showed high sensitivity and specificity with regard to differentiation of epithelioid mesothelioma from lung adenocarcinoma. Cancer 2007. © 2007 American Cancer Society.

Malignant pleural mesothelioma (MPM) is an uncommon tumor of difficult diagnosis and wide prognostic implications whose occurrence has been increasing throughout the world. However, the controversy over diagnosis and treatment persists. Few malignancies have such a direct association with exposure to a definite carcinogen as mesothelioma has with asbestos exposure, and the great interest in asbestos has resulted in a level of awareness regarding mesothelioma that people do not have concerning most other cancers of equivalent occurrence, not to mention the social-legal-medical concerns about the disease. Based on the pattern of use of asbestos the occurrence of MPM can be expected to continue to increase in the following decades.

It is important, albeit not easy, to accurately diagnose MPM for an appropriate clinical management, including curative surgery. One of the most common diagnostic problems is the difficulty in discriminating epithelioid MPM, the most frequent subtype, from lung adenocarcinoma (LAD). Several authors have evaluated the use of various antibodies for the diagnosis of MPM, but the results are still controversial.

D2-40 is a newly monoclonal antibody that reacts with a 40-kD antigen in germ cells and germ cell tumors1 and recognizes the antigen selectively expressed in lymphatic endothelium. This antibody has proven useful in detecting tumors derived from lymphatic tissue as well as lymphatic involvement by a tumor.2 Recently, 2 studies suggested that D2-40 might be helpful in differentiating epithelioid mesothelioma from adenocarcinoma because it reacted with epithelioid MPM but not with other cancers.3, 4

In a clinical setting the diagnosis of MPM is so hard that benign diseases or other cancers could be contaminated with MPM. In this study the diagnosis was strictly based on the histopathologic features of large surgical specimens instead of tiny tissues obtained by needle biopsy or on the cytology of pleural effusion. The purpose of this study was to analyze the potential utility of D2-40 in the diagnosis of MPM, compared with the conventionally used immunohistochemical markers (calretinin, carcinoembryonic antigen [CEA], and thyroid transcription factor-1 [TTF-1]), as well as to define an immunohistochemical panel ideally suited for the definite diagnosis of MPM.


Paraffin-embedded blocks from pleural specimens obtained by major resection not by biopsy in 66 patients with MPM and 66 with LAD except bronchioloalveolar carcinoma were reviewed by more than 1 pathologist whose specialty is pulmonary oncology. Each case with MPM was definitely diagnosed based on currently accepted histopathologic criteria combined with immunohistochemical findings in addition to clinical and radiographic findings. LAD was diagnosed according to the World Health Organization histologic criteria. The characteristics of the 66 patients with MPM are shown in Table 1. The mean age was 60.6 years (range, 35–78 years) and 56 (84.8%) patients were male. The histology was epithelioid in 48 (72.7%) patients, biphasic in 14 (21.2%), and sarcomatoid in 4 (6.1%). The operative procedure performed was pleurectomy in 33 patients (50%), extrapleural pneumonectomy in 25 (37.9%), and partial resection of the tumor in 8 (12.1%). According to the International Mesothelioma Interest Group (IMIG) staging, there were 8 patients with p-stage I disease (12.1%), 15 with p-stage II (22.7%), 38 with p-stage III (57.6%), and 5 with p-stage IV (7.6%).

Table 1. Characteristics of Patients With Mesothelioma (N = 66)
FactorNo. (%)
  1. IMIG indicates International Mesothelioma Interest Group.

Age, y
Mean [range]60.6 [35–78]
 Men56 (84.8)
 Women10 (15.2)
Histologic subtype
 Epithelioid48 (72.7)
 Biphasic14 (21.2)
 Sarcomatoid4 (6.1)
 Pleurectomy33 (50.0)
 Extrapleural pnemonectomy25 (37.9)
 Incomplete resection8 (12.1)
IMIG staging
 I8 (12.1)
 II15 (22.7)
 III38 (57.6)
 IV5 (7.6)

Over a period of 15 months we prospectively performed pleural biopsy in 15 patients who were clinically suspected of MPM and obtained final diagnoses using D2-40, calretinin, CEA, and TTF-1 antibodies. Our policy for diagnosing MPM is to carry out pleural biopsy by video-assisted thoracoscopic surgery (VATS), which allowed us to collect large specimens under general anesthesia and to examine the tissues using this combination of immunohistochemical markers.


Tissue sections (4 μm thick) were air-dried overnight at 37°C, deparaffinized in xylene, and rehydrated in a descending ethanol series. Endogenous peroxidase activity was blocked by immersion for 10 minutes in 0.3% hydrogen peroxide in methanol followed by a single wash in phosphate-buffered saline (PBS), pH 7.4. Sections for D2-40 immunodetection were autoclaved for 7 minutes at 105°C in 10 mM citrate buffer solution (pH 6.0). Sections were incubated with D2-40 monoclonal antibody (DakoCytomation, Glostrup, Denmark; 1:50 dilution), calretinin monoclonal antibody (DakoCytomation, 1:50 dilution), CEA monoclonal antibody (DakoCytomation, 1:50 dilution), and TTF-1 monoclonal antibody (DakoCytomation, 1:100 dilution). The color was developed using 3,3′-diaminobenzidine (Sigma Chemical, St. Louis, MO) as the chromogen. The slides were then counterstained with Mayer hematoxylin, dehydrated, and mounted. Negative controls were prepared by substituting the specific primary antibody with nonimmune serum or control mouse IgG.

Western Blot Analysis

Frozen samples of parietal pleura and mesotheliomas were crushed into pieces and vigorously vortexed in a buffer solution containing 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% Triton X-100, and a protease inhibitor cocktail (Sigma Chemical). Impurities were removed by centrifugation. The resulting lysates were separated on 10% SDS-polyacrylamide gels and transferred to immobilon (Millipore, Bedford, MA). The blots were reacted with the D2-40 monoclonal antibody (1:1000 dilution) in a buffer containing 20 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.05% Tween 20, and 5% skim milk (Difco, Sparks, MD), and then with peroxidase-conjugated antimouse IgG antibody in the same buffer. The color was developed with Western Lighting reagents (PerkinElmer Life Sciences, Boston, MA) before exposure. After stripping, the blots were probed again with an anti-β-actin antibody (Sigma Chemical) according to similar procedures. The chemiluminescent intensity of specific signals was calculated with the FluorChem IS-8000 system (Alpha Innotech, San Leandro, CA). The relative D2-40 signal intensity was obtained by dividing the intensity of D2-40 signals by that of β-actin signals.


As illustrated in Figure 1, immunohistochemical evaluation of epithelioid MPM revealed strong expression of calretinin and D2-40 antigens and a distinct difference in the pattern of staining localization between calretinin (nuclear and cytoplasmic) and D2-40 (membrane). Conversely, both antigens were virtually undetectable in pulmonary adenocarcinoma. The reactivity of CEA and TTF-1 is observed in the cytoplasma and the nuclei of lung adenocarcinoma, respectively. To confirm the role of D2-40 as a potential new marker, we performed Western blot analysis and examined whether there were differences inD2-40 expression between the results of immunohistochemical and Western blot analyses. The latter showed high levels of D2-40 antigen in pleura from patients with MPM compared with low levels of D2-40 antigen in normal pleura (Fig. 2). Also, the level of D2-40 determined by immunohistochemical analysis correlated with that estimated by Western blot analysis regardless of the presence of MPM. These data strongly support a participatory role for D2-40 in MPM.

Figure 1.

Representative images of hematoxylin and eosin staining (H&E) and immunostaining for D2-40, calretinin, CEA, and TTF-1 in adjacent sections of epithelioid mesothelioma (left) and pulmonary adenocarcinoma (right). In mesothelioma, strong D2-40 reactivity was observed along the apical surface of the cell, whereas the nuclei and cytoplasma show strong positive staining for calretinin. In adenocarcinoma the cytoplasma demonstrates positivity for CEA and the nuclei reveals reactivity for TTF-1. In contrast, there is no staining for D2-40 and calretinin in adenocarcinoma, or no reactivity for CEA and TTF-1 in mesothelioma. Original magnification ×200.

Figure 2.

D2-40 expression in parietal pleura with and without epithelioid mesothelioma. Representative Western blots are shown. Quantitative densitometric data (expressed as D2-40/β-actin antigen levels, n = 6 for each group) from multiple samples are shown as the mean value ± standard deviation.

The results of immunohistochemical analyses using D2-40, calretinin, CEA, and TTF-1 are summarized in Table 2. Fifty-six (84.8%) and 58 (87.9%) of the 66 MPM tissue specimens expressed D2-40 and calretinin, respectively, whereas none expressed CEA and TTF-1. Both D2-40 and calretinin were similarly observed in most areas of epithelioid differentiation, although in areas of sarcomatoid differentiation (either biphasic mesothelioma or pure sarcomatoid subtype) they were detected less frequently and the immunoreactivity was less strong. In the cases with LAD, neither D2-40 nor calretinin was detected, except in only 3 cases (4.5%) that showed weak focal staining for both the antibodies. In contrast, CEA and TTF-1 staining was present in 95.5% and 92.4% of LADs, respectively. For mesothelioma, the specificities of D2-40 and calretinin were both 95.5% and the specificities of CEA and TTF-1 for adenocarcinoma were both 100%.

Table 2. Immunohistochemical Positivity in Mesothelioma and Lung Adenocarcinoma
No. (%)No. (%)No. (%)No. (%)
Mesothelioma6656 (84.8)58 (87.9)0 (0)0 (0)
 Epithelioid4843 (89.6)43 (89.6)0 (0)0 (0)
 Biphasic1411 (78.6)14 (100)0 (0)0 (0)
 Sarcomatoid42 (50.0)1 (25.0)0 (0)0 (0)
Adenocarcinoma663 (4.5)3 (4.5)63 (95.5)61 (92.4)

More recently, we prospectively used these 4 antibodies for distinguishing MPM from LAD in 15 patients subjected to VATS biopsy of the pleura for differential diagnosis; the results are shown in Table 3. The final diagnosis was MPM in 8 (53.3%) patients containing 5 epithelioid subtypes and 3 biphasic subtypes, and adenocarcinoma in 7 patients (46.7%) including 5 LADs. In all the patients with MPM the immunoreactivity for D2-40 and calretinin was seen, whereas that for CEA and TTF-1 was never observed. Conversely, there were no adenocarcinomas with positive immunoreactivity for D2-40, positive for calretinin, or negative for CEA. TTF-1 expression is considered to be present only in LAD but not in metastatic adenocarcinoma from other organs, which was supported by our data. Only 1 case with LAD was judged negative for TTF-1, which was very weak focal expression. These data powerfully confirm significant function of D2-40 and calretinin as positive markers and CEA and TTF-1 as negative markers in MPM.

Table 3. Results of VATS Pleural Biopsy Performed for Suspicious Pleural Mesothelioma During the Last 15-Month Period (N = 15)
Case No.Age/SexFinal PathologyImmunohistochemistry
  • VATS indicates video-assisted thoracoscopic surgery; M, male; F, female; MPM, malignant pleural mesothelioma; Epi, Epithelioid; Bi, biphasic; LAd, lung adenocarcinoma; Adenoca, adenocarcinoma; P, positive; N, negative.

  • #1: Pleural metastasis originated from gastric carcinoma.

  • #2: Metastatic adenocarcinoma with unknown primary lesion.

  • *

    Very weak focal reactivity suggesting negative result.

263/MMPM (Epi)PPNN
468/MMPM (Epi)PPNN
567/MMPM (Epi)PPNN
646/FMPM (Epi)PPNN
762/MMPM (Epi)PPNN
969/FAdenoca #1NNPN
1059/FAdenoca #2NNPN


At present, the diagnosis of MPM depends basically on the histopathology of the lesion and the assessment of clinical and radiological findings. Pleural fluid cytology or pleural needle biopsy seldom provides specimens large enough to perform the immunohistochemical analyses that are crucial for an accurate diagnosis. To confidently obtain an adequate amount of tissue from macroscopically diseased sites we routinely practice VATS biopsy. In the present study, large specimens obtained at the time of major resection of the tumor were reevaluated under uniform conditions by more than 1 pathologist expert in pulmonary oncology.

Immunohistochemistry has proven the most useful method for conclusively diagnosing MPM but it is in general agreed that no single antibody is sufficiently sensitive or specific. In a clinical setting the histologic distinction between epithelioid MPM and adenocarcinoma is often difficult and requires ancillary studies. Therefore, several antibodies must be used for the differential diagnosis of a pleural tumor, and nowadays the use of a panel of antibodies is a clinically accepted practice. However, the choice of antibodies varies. In this study, a tumor that shows immunoreactivity for D2-40 and calretinin antibodies, but not for CEA and TTF-1 antibodies, is unequivocally MPM, and a tumor negative for D2-40 and calretinin, but positive for CEA and TTF-1, is unquestionably not MPM, but LAD. Of note, D2-40 and calretinin act as positive markers, whereas CEA and TTF-1 function as negative markers in the differential diagnosis of MPM, although great attention should be given to the fact that each antibody has individual deficiencies.

D2-40, a recently developed, commercially available monoclonal antibody could be helpful both in the diagnosis of lymphatic derived tumors and in determining lymphatic invasion by tumors.1, 5 Chu et al3 demonstrated strong membranous reactivity in 33 of 33 epithelioid mesotheliomas (100%) and in the epithelioid component of 15 of 16 biphasic mesotheliomas (94%). Ordonez4 also reported positivity for D2-40 in 25 of 29 epithelioid mesotheliomas. The membranous reaction of D2-40 was diffuse and intense in the majority of epithelioid MPM, whereas weak focal membranous positivity was observed in some serous carcinomas and no membranous staining was seen in any of the other carcinomas.6 In the present study we demonstrated D2-40 protein expression in the pleura invaded by epithelioid MPM through immunohistochemical examinations and Western blotting analyses, and a positive correlation in the level of the protein evaluated by different methods. In addition, our large series showed D2-40 positivity for MPM (56/66, 84.8%), especially for epithelioid MPM (43/48, 89.6%). These findings suggest that D2-40 is one of the most informative markers for the diagnosis of MPM, particularly of the epithelioid subtype.

Among the so-called ‘positive’ mesothelioma markers, calretinin is one of the most used in the diagnosis of MPM because of its high sensitivity and specificity,6–8 and is frequently expressed in all histologic types of MPM, in contrast with other highly sensitive mesothelioma markers, which are commonly expressed in the epithelioid component but not in the sarcomatoid component, such as D2-40.7, 8 As well, calretinin is one of the few antibodies that are much more frequently reactive in MPM than in LAD, and thus we included it in the panel of antibodies employed in our series.

CEA is one of the negative mesothelioma markers available for the differential diagnosis of epithelioid MPM and LAD. Approximately 80% of LADs are reported to express CEA, whereas MPMs are typically negative for this marker.6, 8 However, it should be noted that the diagnostic value of CEA immunoreactivity when differentiating between epithelioid MPM and metastatic, nonpulmonary adenocarcinomas depends mainly on the site of origin of the adenocarcinoma. In contrast, TTF-1, which is expressed in adenocarcinomas originating only in the lungs or thyroid, is an important negative marker.9–12 Previous studies have shown that TTF-1 expression is retained in thyroid carcinomas and in up to 75% of LADs.9, 10 We have never detected CEA nor TTF-1 expression in any of the MPMs investigated so far; therefore, positive staining for either marker should be considered an indication against such a diagnosis.

Therefore, our results show that the use of D2-40, calretinin, CEA, and TTF-1 should allow us to achieve a definite diagnosis in the large majority of suspected cases of MPM. This approach decreases the number of antibodies required in the majority of cases and contributes to the establishment of a more accurate diagnosis of MPM or LAD.