Detection of cancer cells in effusions from patients diagnosed with gynaecological malignancies. Evaluation of five epithelial markers

Our plain-English summary of Detection of cancer cells in effusions from patients diagnosed with gynaecological malignancies. Evaluation of five epithelial markers, by B. Davidson, B. Risberg, G. Kristensen, et al (Virchows Archiv, 1999).

The study

• Authors: B. Davidson, B. Risberg, G. Kristensen, G. Kvalheim, E. Emilsen, A. Bjåmer, A. Berner
• Journal: Virchows Archiv
• Published: 1999;435(1):43-9
• Cited by: 92 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

The detection of malignant cells in pleural, peritoneal, and pericardial fluids of cancer patients marks the presence of metastatic disease and is associated with a grave prognosis. We evaluated five epithelial markers for the detection of cancer cells in 94 fresh pleural, peritoneal and pericardial effusions. Eighty-four of the samples were regarded as adequate for analysis after evaluation of cytological smears, including 61 samples from patients known to have gynaecological neoplasms. The other 23 samples were from patients with various non-gynaecological malignancies or tumours of unknown origin. Our control cases were 10 fallopian tubes not affected by any malignancy and 12 malignant mesotheliomas. Cell blocks from all cases were stained for CA-125, BerEP4, carcinoembryonic antigen (CEA), BG8 (Lewis Y blood antigen), and B72.3 (TAG-72). Fifty-one of 84 cases were diagnosed as malignant or suggestive of malignancy in cytological smears and/or cell block sections. However, staining for epithelial markers highlighted the presence of malignant cells in 7 additional cases. When membrane staining was evaluated, the sensitivity of the markers studied in detecting malignant cells was as follows: CA-125: 88%, BerEP4: 78%, CEA: 26%, BG8: 86%, B72.3: 79%. Membrane positivity for CEA, B72.3 and BerEP4 was not detected in reactive mesothelial cells. However, membranous staining in mesothelial cells was evident in 13% and 31% of cases with the use of BG8 and CA-125, respectively. Weak cytoplasmic staining for CEA was observed in mesothelial cells in 2 cases. When Ber-EP4, B72.3, and BG8 staining results in cancer cells were combined, the following sensitivity levels were observed: BG8+B72.3: 91%; BG8+Ber-EP4: 90%; B72.3+Ber-EP4: 93%; BG8+Ber-EP4+B72.3: 95%. The detection of malignant cells in effusions is facilitated by the use of immunocytochemistry using a wide panel of antibodies. BerEP4 and B72.3 appear to be the best markers when both sensitivity and specificity are considered, followed by BG8, while CEA and CA-125 have a limited role in the detection of metastases from gynaecological tumours owing to the low sensitivity of the former and the low specificity of the latter. Analysis of all staining results should be based on a thorough morphological examination.

In plain English

This Oslo laboratory study tested whether five antibody stains (CA-125, BerEP4, CEA, BG8 and B72.3) help detect cancer cells in fluid around the lungs, abdomen and heart. It examined 94 consecutive fluid samples received from January to August 1998. Of 84 adequate samples, 61 came from patients with known gynaecological neoplasms. The case list included borderline ovarian tumours and benign ovarian cysts. Controls were 10 non-cancerous fallopian tubes and 12 malignant mesotheliomas. The article does not discuss asbestos. Microscope review found 51 samples malignant or suggestive of malignancy, and staining revealed cancer cells in 7 more. That made 58, of mixed tumour origins. Counting membrane staining, detection rates in these 58 were: CA-125 88%, BerEP4 78%, CEA 26%, BG8 86%, B72.3 79%. Adding cytoplasmic staining raised CEA to 34%. BerEP4, B72.3 and CEA showed no membrane staining of benign lining (mesothelial) cells. BG8 did in 13% of cases and CA-125 in 31%. CA-125 also stained 8 of 12 mesotheliomas. Combining BG8, BerEP4 and B72.3 detected cancer cells in 95% of the 58. The authors concluded this combination appears favourable for gynaecological cancer patients, while CA-125 and CEA cannot be recommended.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Laboratory diagnostic study comparing immunocytochemical staining with five antibodies against microscope (morphological) diagnosis in body-cavity fluid samples
• Who was studied: 94 consecutive fresh pleural, peritoneal and pericardial fluid samples were received at The Norwegian Radium Hospital, Oslo, from January to August 1998. They came from patients with a previous cancer diagnosis or suspected malignancy, aged 16 to 86. 84 samples were adequate: 61 from patients with known gynaecological neoplasms, 18 with non-gynaecological malignancies and 5 with unknown primary tumours. The case list (Table 1) includes 2 ovarian borderline tumours and 2 ovarian benign cysts. Controls were 10 nonmalignant fallopian tubes and 12 malignant mesotheliomas from tissue sections.
• Exposure: No asbestos or other environmental exposure was studied. The study tested cell-block staining with five antibodies: CA-125, BerEP4, carcinoembryonic antigen (CEA), BG8 (Lewis Y blood antigen) and B72.3 (TAG-72).
• Main results:
  • Microscope review of smears and cell blocks together identified 51 of 84 specimens (61%) as positive or suggestive for malignant cells.
  • Staining revealed malignant cells in 7 more specimens that had been judged benign on both smear and cell block, for a total of 58 malignant cases.
  • In the 58 malignant cases, sensitivity of membrane staining for detecting malignant cells was CA-125 88%, BerEP4 78%, CEA 26%, BG8 86% and B72.3 79%.
  • Counting cytoplasmic CEA staining as well raised CEA sensitivity to 34%.
  • In reactive (benign) mesothelial cells, no membrane staining was seen with CEA, B72.3 or BerEP4.
  • Benign mesothelial cells showed membrane staining in 13% of cases with BG8 (9/71) and in 31% with CA-125 (22/71).
  • In the 12 malignant mesothelioma controls, CA-125 stained 8/12 (67%). Table 3 shows CEA, Ber-EP4, BG8 and B72.3 each at 0/12.
  • Combined staining in the cancer cells of the 58 malignant cases gave these sensitivities: BG8+B72.3 91%, BG8+Ber-EP4 90%, B72.3+Ber-EP4 93%, and all three together 95%.
  • Staining profiles of gynaecological and breast carcinomas differed only for CA-125. CA-125 was positive in 32/33 (97%) ovarian carcinomas and 6/10 (60%) breast carcinomas.
  • No false-positive cases were seen.
• The authors' conclusion: The authors concluded that for patients diagnosed with gynaecological malignancies, the combined use of Ber-EP4, B72.3 and BG8 appears favourable, while CA-125 and CEA cannot be recommended for a diagnostic panel. They added that using all three gives the highest sensitivity.

Topics

Medical terms in this study, explained in our Lexicon: Pericardial, Prognosis, Pericardial Effusion, Neoplasm, Malignant Mesothelioma, Cell Block, CA-125, Carcinoembryonic Antigen (CEA).

Related studies

• Cytological differential diagnosis among adenocarcinoma, epithelial mesothelioma, and reactive mesothelial cells in serous effusions by immunocytochemistry — X.Y. Su, G.D. Li, et al. (2011)
• MMP-7 is a highly specific negative marker for benign and malignant mesothelial cells in serous effusions — B. Davidson, H.T. Stavnes, et al. (2016)
• Cytomorphological profile of neoplastic effusions: an audit of 10 years with emphasis on uncommonly encountered malignancies — S. Gupta, P. Sodhani, et al. (2012)
• Morphologic and immunocytochemical performances of effusion cell blocks prepared using 3 different methods — X. Jing, Q.K. Li, et al. (2013)
• IMP3/L523S, a novel immunocytochemical marker that distinguishes benign and malignant cells: the expression profiles of IMP3/L523S in effusion cytology — K. Ikeda, G. Tate, et al. (2010)
• The importance of EZH2 and MOC-31 expression in the differential diagnosis of benign and malignant effusions — C. Sadullahoglu, D. Nart, et al. (2017)

Browse more research in Virchows Archiv, or return to medical research.