SEARCH

SEARCH BY CITATION

Keywords:

  • malignant pleural effusion;
  • cytology;
  • CRxA-01;
  • mammaglobin;
  • CK7/CK20

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

The most common causes of malignant pleural effusions in women are metastatic lung carcinomas and breast carcinomas. It is often very difficult to distinguish between breast carcinomas and other metastatic carcinomas when they share a similar morphology and a similar cytokeratin profile (CK7-positive/CK20-negative [CK7+/CK20−]). To better differentiate between metastatic mammary carcinomas and other metastatic carcinomas in pleural effusion cytology, the authors studied the potential use of a novel antibody, CRxA-01, which was identified by a cDNA subtraction library, together with a well characterized antibody against mammaglobin.

METHODS

A computer search for patients with malignant pleural effusion specimens between January 1992 and November 2002 generated 228 patients, 71 of whom had cell block material and a known clinical history. Primary malignancies among these patients included 20 breast carcinomas, 32 lung carcinomas, 4 endometrial carcinomas, 9 ovarian carcinomas, 4 gastrointestinal carcinomas, and 2 genitourinary carcinomas. All specimens were immunostained with anti-CK7, CK20, CRxA-01, and mammaglobin antibodies. Only CK7-positive/CK20-negative (CK7+/CK20−) specimens were included in the current study, and only definitive membranous staining for CRxA-01 and cytoplasmic staining for mammaglobin were considered to be positive.

RESULTS

For patients with metastatic breast carcinomas, mammaglobin was positive in 11 of 20 (55%) tissue specimens and CRxA-01 was positive in 12 of 20 (60%) tissue specimens. When CRxA-01 and mammaglobin were used together, 16 of 20 (80%) tissue specimens were positive for mammaglobin or/and CRxA-01 antibodies. This staining pattern was not seen for tissue specimens from patients with other metastatic carcinomas. Two of 4 (50%) uterine carcinoma specimens and 6 of 9 (67%) ovarian carcinoma specimens were positive for CRxA-01 only.

CONCLUSIONS

CRxA-01 and mammaglobin were expressed in most metastatic breast carcinoma specimens. Other CK7+/CK20− carcinoma specimens did not express mammaglobin and showed weak or negative staining for CRxA-01. When used together, CRxA-01 and mammaglobin greatly improved the sensitivity and specificity for the detection of metastatic breast carcinoma in pleural effusion specimens. Cancer (Cancer Cytopathol) 2004. © 2004 American Cancer Society.

To identify the origin of a metastatic malignant neoplasm in pleural effusions based on morphology alone is a diagnostic challenge even for the most experienced cytopathologist. Immunohistochemistry (IHC) can be helpful and is often used to determine the origin of a primary tumor. A particularly useful IHC panel consists of anti-cytokeratin 7 and anti-cytokeratin 20 (CK7/CK20) antibodies.1 CK7, a 54-kilodalton (kD) basic protein, is expressed in the epithelium of the lung, breast, endometrium, stomach, pancreatobiliary tract, urothelium, and skin adnexal glands.2, 3 CK20, a 46-kD acidic protein, is widely expressed in carcinomas of the gastrointestinal (GI) tract, pancreatobiliary tract, urothelium, and in mucinous ovarian tumors.3, 4 Different immunophenotypes are associated with different carcinomas. For example, the CK7-positive/CK20-positive (CK7+/CK20+) phenotype is found frequently in GI and genitourinary epithelial neoplasms, the CK7−/CK20+ phenotype is seen in most colorectal adenocarcinomas, and the CK7−/CK20− profile is found in adrenal cortical carcinomas, prostatic carcinomas, and hepatocellular carcinomas.1, 5

However, in tissue specimens of adenocarcinomas with an identical CK7+/CK20− immunophenotype, the dilemma is not resolved easily because numerous tumors exhibit the same immunohistochemistry profile. Nearly all breast carcinomas, ovarian adenocarcinomas (endometrioid, serous papillary, and clear cell types), lung, endometrial, thyroid (follicular, papillary, and medullary subtypes), and salivary gland carcinomas, and approximately 65% of mesotheliomas demonstrate a CK7+/CK20− phenotypic profile.1, 5, 6 For cytology specimens, the task can be even more challenging. The specimen is usually scarce, lacks good architectural features, and thus requires careful selection of IHC studies, when at all possible. Cytology specimens, however, are usually the first diagnostic modality to a clinical inquiry for a potential metastatic carcinoma. Determining the origin of a primary tumor will undoubtedly provide invaluable information for further management of patients.

The most common malignant pleural effusions in adult women are caused by metastatic breast carcinomas or lung carcinomas. Differentiating these two carcinomas from one another and from other carcinomas that share the same CK7+/CK20− profile is often difficult. To better distinguish mammary carcinomas from other carcinomas in pleural fluid effusion cytology, more specific markers would be useful.

One such IHC marker is the novel CRxA-01 antibody, which has yielded excellent results as a breast tumor marker. The CRxA-01 antibody was identified by cDNA library subtraction and was expressed in > 70% of breast carcinoma specimens at the mRNA level by quantitative polymerase chain reaction. The antibody was generated by raising mouse monoclonal antibodies against Escherichia coli-derived CRxA-01 protein (unpublished data). We have combined this novel antibody with a well characterized antibody against mammaglobin, which is highly expressed in the breast epithelium,7–9 to increase the sensitivity and specificity in discriminating metastatic mammary carcinomas from other metastatic adenocarcinomas in pleural effusion specimens.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Specimen Selection

A computer search of the University of Massachusetts Memorial Medical Center database was performed to identify malignant pleural effusion specimens from January 1992 to November 2002. The search generated 228 specimens from patients, 71 of whom had cell block material and a known clinical history. The primary malignancies comprised 20 breast carcinomas, 32 lung carcinomas, 4 endometrial carcinomas, 9 ovarian carcinomas, 4 GI carcinomas, and 2 genitourinary carcinomas. All specimens were immunostained with anti-CK7, CK20, CRxA-01, and mammaglobin antibodies.

Immunohistochemistry

Immunoperoxidase staining was performed on 5-μm sections of formalin-fixed, paraffin-embedded tissue specimens. Slides were first deparaffinized and rehydrated. Antigen retrieval was carried out with 0.01 M citrate buffer at pH 6.0, in an 800-W microwave oven for 10 minutes. After cooling and rinsing, slides were stained on the Dako autostainer (Dako, Carpinteria, CA). Slides for CRxA-01 staining were digested with a proteinase-K solution (Dako) at a dilution of 1:40 for 4 minutes, then rinsed. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide. The slides were rinsed, and nonspecific sites were bound with blocking protein (Chemmate, Ventana, Tuscon, AZ). The sections were incubated with primary antibody (mouse anti–CRxA-01 and mouse anti-mammaglobin at 1 μg/mL) for 45 minutes. After brief buffer washes, the sections were treated with a cocktail of biotinylated goat anti-rabbit immunoglobulin (Ig)G and goat anti-mouse IgG/IgM (Chemmate, Ventana) for 30 minutes, then washed. Sections were then incubated in avidin/biotin/peroxidase complex (Chemmate, Ventana) for 30 minutes. After rinsing, the slides were treated with diaminobenzidine (Chemmate, Ventana) to visualize the end products. Finally, the sections were counterstained with hematoxylin.

Positivity of the primary antibodies was defined as strong brown membranous staining for CRxA-01, and cytoplasmic staining for mammaglobin. Negative staining was defined as no staining, or as weak diffuse background granules.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

For metastatic breast carcinoma specimens, mammaglobin was positive in 11 of 20 (55%) specimens, and CRxA-01 was positive in 12 of 20 (60%) specimens. When mammaglobin and CRxA-01 were used together, 16 of 20 (80%) specimens were positive for mammaglobin and/or CRxA-01 antibodies (Table 1; Fig. 1). This staining pattern was not seen in other metastatic carcinoma specimens. All 32 (100%) metastatic lung carcinoma specimens were negative for both mammaglobin and CRxA-01 antibodies (Fig. 1). Although 2 of the 4 (50%) uterine carcinoma specimens and 6 of the 9 (67%) ovarian carcinoma specimens exhibited CRxA-01 positivity (Table 1), they were negative for mammaglobin.

Table 1. CRxA-01 and Mammaglobin Expression in Metastatic Carcinoma
HistoryCRxA+/mam+ (%)CRxA−/ mam+ (%)CRxA+/ mam− (%)CRxA−/ mam− (%)
  1. Mam: mammaglobin; +: positive; −: negative; GI: gastrointestinal; GU: genitourinary.

Breast (n = 20)7 (35)4 (20)5 (25)4 (20)
Lung (n = 32)00032 (100)
Uterus (n = 4)002 (50)2 (50)
Ovary (n = 9)006 (67)3 (33)
GI (n = 4)0004 (100)
GU (n = 2)0002 (100)
thumbnail image

Figure 1. (A) Hematoxylin and eosin (H & E)–positive pleural effusion cytology in a patient with metastatic breast carcinoma. (B) Strong (3+) cytoplasmic staining for mammaglobin. (C) Strong membranous staining for CRxA-01. (D) H & E staining–positive pleural effusion cytology in a patient with metastatic lung carcinoma. (E) Neoplastic cells in Figure 1D are negative for mammaglobin. (F) Neoplastic cells in Figure 1D are negative for CRxA-01.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Mammaglobin is a member of the epithelial secreting protein family,7 which also includes rat prostatic protein subunit C3, human Clara cell protein, and rabbit uteroglobin. Mammaglobin has been studied widely since it was first discovered by Watson et al.7, 8 who characterized the structure of the protein and subsequently located the gene on chromosome 11q3. Mammaglobin has been shown to be an excellent marker for circulating breast carcinoma cells in peripheral blood and for lymph node metastasis.9–13 Overexpression of mammaglobin has also been correlated with the clinicobiologic behavior of breast carcinomas.14–16 Previously, our group showed that mammaglobin was expressed in > 87% and in < 22% of breast carcinoma and lung carcinoma surgical resection specimens, respectively.17

CRxA-01 belongs to the B7 family of ligands and has a role in inhibiting early-stage T-cell response by a variety of pathways including cell cycle arrest and nonselective CK inhibition.18 CRxA-01 has been shown to be overexpressed in breast, ovarian, and endometrial carcinomas. In our previous study, overexpression of CRxA-01 was observed in 69% of breast carcinoma specimens, 48% of ovarian serous papillary carcinoma specimens, and in 28% of endometrial carcinoma specimens.17

In the current study, we extended our analysis of mammaglobin and CRxA-01 antibodies to pleural effusion cytology specimens and showed that 35% of metastatic breast carcinoma specimens were stained with both markers, 20% was stained with mammaglobin only, and 25% was stained with CRxA-1 only. When these two antibodies were used together, 80% of metastatic breast carcinoma specimens were stained with ≥ 1 marker and only 20% of specimens were negative for both markers. None of the metastatic carcinoma specimens from the lung, GI tract, or gentitourinary tract showed positive staining with either antibody. Although some of the metastatic carcinoma specimens from the ovary or endometrium showed positive CRxA-01 staining, none of them stained positive with the mammaglobin antibody.

Several IHC markers have been described to aid in the distinction between breast carcinomas and other epithelial carcinomas. The most commonly used markers include CK7, CK20,1, 5 and thyroglobulin transcription factor (TTF-1).3 TTF-1 is a recently described marker that stains both thyroid tissue and a high percentage of pulmonary adenocarcinoma specimens and small cell carcinoma specimens.3, 19 In the current study, inclusion of TTF-1 immunostain, together with mammaglobin and CRxA-01, as a panel in the workup of malignant pleural effusion specimens further increased the specificity in distinguishing between metastatic breast carcinomas and metastatic lung carcinomas (data not shown).

In conclusion, our study demonstrates that a high percentage of metastatic breast carcinomas detected in pleural effusion specimens express mammaglobin and CRxA-01 proteins. Using mammaglobin and CRxA-01 as an immunostaining panel will greatly facilitate the distinction of metastatic breast carcinomas from other CK7+/CK20− carcinomas.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Chu P, Wu E, Weiss ML. Cytokeratin 7 and cytokeratin 20 expression in epithelial neoplasms: a survey of 435 cases. Mod Pathol. 2000; 13: 962972.
  • 2
    Moll R, Lowe A, Laufer J, Franke WW. Cytokeratin 20 in human carcinomas. A new histodiagnostic marker detected by monoclonal antibodies. Am J Pathol. 1992; 140: 427447.
  • 3
    Chhieng CD, Cangiarella FJ, Zakowski FM, Goswami S, Cohen JM, Yee HT. Use of thyroid transcription factor 1, PE-10, and cytokeratins 7 and 20 in discriminating between primary lung carcinomas and metastatic lesions in fine-needle aspiration biopsy specimens. Cancer (Cancer Cytopathol). 2001; 93: 330336.
  • 4
    van Niekerk CC, Jap PH, Ramaekers FC, van de Malengraft F, Poels LG. Immunohistochemical demonstration of keratin 7 in routinely fixed paraffin-embedded human tissues. J Pathol. 1991; 165: 145152.
  • 5
    Wang NP, Zee S, Zarbo RJ, Bacchi CE, Gown AM. Coordinate expression of cytokeratins 7 and 20 defines unique subsets of carcinomas. Appl Immunohistochem. 1995; 3: 99107.
  • 6
    Kaufmann O, Deidesheimer T, Muehlenberg M, Deick P, Dietel M. Immunohistochemical differentiation of metastatic breast carcinomas from metastatic adenocarcinomas of other common primary sites. Histopathology. 1996; 29: 233240.
  • 7
    Watson MA, Fleming TP. Mammaglobin, a mammary-specific member of the uteroglobin gene family, is overexpressed in human breast cancer. Cancer Res. 1996; 56: 860865.
  • 8
    Watson MA, Darrow C, Zimonjic DB, Popescu NC, Fleming TP. Structure and transcriptional regulation of the human mammaglobin gene, a breast cancer associated member of the uteroglobin family localized to chromosome 11q3. Oncogene. 1998; 16: 817824.
  • 9
    Watson MA, Dintzis S, Darrow CM, et al. Mammaglobin expression in primary, metastatic, and occult breast cancer. Cancer Res. 1999; 59: 30283031.
  • 10
    Leygue E, Snell L, Dotwzlaw H, et al. Mammaglobin, a potential marker of breast cancer metastasis. J Pathol. 1999: 189: 2833.
  • 11
    Aihara T, Fujiwara Y, Ooka M, Sakita I, Tamaki Y, Monden M. Mammaglobin B as a novel marker for detection of breast cancer micrometastases in axillary nodes by reverse transcription-polymerase chain reaction. Breast Cancer Res Treat. 1999; 58: 137140.
  • 12
    Zach O, Kasparu H, Krieger O, Hehenwarter W, Girschikofsky M, Lutz D. Detection of circulating mammary carcinoma cells in peripheral blood of breast cancer patients via a nested reverse transcriptase polymerase chain reaction assay for mammaglobin mRNA. J Clin Oncol. 1999; 17: 20152019.
  • 13
    Lin YC, Wu Chou YH, Liao IC, Cheng AJ. The expression of mammaglobin mRNA in peripheral blood of metastatic breast cancer patients as an adjunct to serum tumor markers. Cancer Lett. 2003; 191: 9399.
  • 14
    Silva JM, Dominguez G, Silva J, et al. Detection of epithelial messenger RNA in the plasma of breast cancer patients is associated with poor prognosis tumor characteristics. Clin Cancer Res. 2001; 7: 28212825.
  • 15
    Nunez-Villar MJ, Martinez-Arribas F, Pollan M, et al. Elevated mammaglobin (h-MAM) expression in breast cancer is associated with clinical and biological features defining a less aggressive tumor phenotype. Breast Cancer Res. 2003; 5: R65R70.
  • 16
    Lin YC, Chen SC, Hsueh S, et al. Lack of correlation between expression of human mammaglobin mRNA in peripheral blood and known prognostic factors for breast cancer patients. Cancer Sci. 2003; 94: 99102.
  • 17
    Stachurski D, Fanger G, Xu B, et al. CRxA-01 and mammaglobin expression in breast cancer: potential role in differentiating breast cancer from other CK7+/CK20− tumors (abstract). Mod Pathol. 2003; 16: 48A.
  • 18
    Sica GL, Choi IH, Zhu G, et al. B7-H4, a molecule of the B7 family, negatively regulates T cell immunity. Immunity. 2003; 18: 849861.
  • 19
    Fujita J, Ohtsuki Y, Bandoh S, et al. Expression of thyroid transcription factor-1 in 16 human lung cancer cell lines. Lung Cancer. 2003; 39: 3136.