Chung-Chin Yao

True cytokeratin 8/18 immunohistochemistry is of no use in distinguishing between primary endocervical and endometrial adenocarcinomas in a tissue microarray study.

The choice of appropriate therapeutic plans for primary endocervical adenocarcinomas and endometrial adenocarcinomas depends on the site of origin of the tumor. The purpose of this study was to make clear whether the immunohistochemistry of the true cytokeratin 8/18 monoclonal antibody (Leica Microsystems, Newcastle, United Kingdom), instead of CAM 5.2 (Becton Dickinson Biosciences, San Jose, CA), has potential use in distinguishing between endocervical adenocarcinomas and endometrial adenocarcinomas. A tissue microarray was constructed using paraffin-embedded, formalin-fixed tissues from 34 hysterectomy specimens, including 14 endocervical adenocarcinomas and 20 endometrial adenocarcinomas. Using the Bond-Max autostainer (Leica Microsystems) and the associated Bond Refine Polymer Detection Kit, tissue array sections were immunostained with cytokeratin 8, 18, and 8/18 commercially available antibodies. The immunohistochemical expressions of all 3 markers, cytokeratin 8, 18, and 8/18 showed nonsignificant (P>0.05) frequency differences between the immunostaining results (positive vs. negative) in tumors of both gynecologic adenocarcinomas. Although CAM 5.2 has been reported to be helpful in distinguishing between primary endocervical adenocarcinomas and endometrial adenocarcinomas, we could not verify this point of view using the true cytokeratin 8/18 monoclonal antibody (Leica Microsystems). It has often been mistakenly cited that CAM 5.2 reacts with cytokeratin 8 and 18, and the results herein confer that there is a wrong impression that cytokeratin 8/18 is differentially expressed in these 2 gynecologic malignancies. In conclusion, the true cytokeratin 8/18 monoclonal antibody is of no use in distinguishing between primary endocervical adenocarcinomas and endometrial adenocarcinomas.

HER2 gene amplification in primary mucinous ovarian cancer: a potential therapeutic target

Sir: New cancer therapies can be directed at specific signalling pathways of tumour cells. Trastuzumab (Herceptin), a humanized monoclonal antibody that binds selectively to the HER2 protein, is currently being used in breast and some other types of tumours. The purpose of this study was to investigate whether HER2 could be a potential therapeutic target for the treatment of primary mucinous and non-mucinous epithelial ovarian cancer (EOC). Twenty-seven epithelial ovarian cancers, four mucinous and 23 non-mucinous, were analysed in a tissue microarray (TMA) study. In addition to its usefulness in surveys of specific biomarkers, TMA has become a popular tool for tissue-based research, because it allows for the massive acceleration of studies correlating molecular in-situ findings with clinicopathological information. However, the limitation of TMA seems to be its insufficiency in demonstrating tumour heterogeneity, because of the small sample size used. In 2007, the American Society of Clinical Oncology ⁄ College of American Pathologists (ASCO ⁄ CAP) established guidelines for the HER2 scoring algorithm for breast cancers. Although there are no valid data so far that correlate this HER2 scoring mechanism with clinical outcomes in ovarian cancers, we tried to apply the current ASCO ⁄ CAP guideline recommendations to interpret the results of HER2 HercepTest immunohistochemical (IHC) assays and PathVysion fluorescence in-situ hybridization (FISH) tests in EOC. The IHC results were evaluated by three pathologists (C.P.H., J.D.H. and H.C.) who did not know the patients’ clinical and FISH data. All of the FISH ⁄ IHC paired results confirmed that 19 of the 27 specimens gave concordant negative results and three gave concordant positive results. Of the remaining five specimens, all gave equivocal IHC results, whereas one gave FISH-positive and four gave FISHnegative results (Table 1). Surprisingly, we found that the overall rate of HER2 gene amplification was high in four of four (100%) mucinous EOC, but low in 0 out of 23 (0%) non-mucinous EOC. Figure 1A,B shows typical IHC and FISH results for one representative sample. The genetic basis for the oncogenic process of each histological subtype of EOC is very complex. For example, amplification and ⁄ or altered expression of oncogenes and tumour suppressor genes can play an important role. In 2003, Bookman et al. reported both a low frequency of HER2 overexpression (11.3%) and low response rates (7.3%) for single-agent trastuzumab therapy in patients with persistent or recurrent EOC in a phase II clinical trial. However, they applied oldfashioned semiquantitative IHC scoring scales, with a cut-off of more than 10% of tumour cells with total membranous rimmed staining. Additionally, they did not provide FISH data to confirm or validate the eligible cases, and there was no histological breakdown of their tested cases. Accordingly, we think that definitive HER2 data still remain inadequate for tumours originating from the ovarian surface epithelium. Until now, HER2 gene amplification has been the only established mechanism for the pathological HER-2 protein overexpression seen in human breast cancers. So far, no other common ⁄ fatal cancers have been studied as thoroughly or as comprehensively as breast cancer. Thus, adopting the more stringent ASCO ⁄ CAP guideline recommendations for breast cancer in this study, we investigated and evaluated the true state of HER2 overexpression in both mucinous and non-mucinous EOC. With careful exclusion of borderline and metastatic lesions, primary mucinous EOCs might be substantially less common than previously reported, and appear to constitute <3% of EOC. In this study, all four cases of mucinous EOC were strictly selected as primary ovarian tumours after careful review of the patients’ clinical histories of tumour staging, size and laterality, as well as vigilant performance of the appropriate immunohistochemical panel. Our results conclusively reveal that Table 1. Result categories of immunohistochemical (IHC) and fluorescence in-situ hybridization (FISH) tests obtained from 27 epithelial ovarian carcinomas

p16INK4 and CEA can be mutually exchanged with confidence between both relevant three-marker panels (ER/Vim/CEA and ER/Vim/p16INK4) in distinguishing primary endometrial adenocarcinomas from endocervical adenocarcinomas in a tissue microarray study

The accurate distinction between primary endocervical adenocarcinomas (ECA) and endometrial adenocarcinomas (EMA) may require the use of multiple ancillary monoclonal antibodies in panels of immunohistochemistry stains. In addition to reappraising the expressions of four commonly used individual monoclonal antibodies [estrogen receptor (ER), Vimentin (Vim), carcinoembryonic antigen (CEA), and p16INK4], this study was designed to investigate whether CEA and p16INK4 can be effectively exchanged between two relevant three-marker panels (ER/Vim/CEA vs. ER/Vim/p16INK4) in distinguishing ECA from EMA. A tissue microarray was constructed using paraffin-embedded, formalin-fixed tissues from 35 hysterectomy specimens, including 14 ECA and 21 EMA. Utilizing the avidin–biotin technique, tissue array sections were immunostained with the four aforementioned individual markers (ER, Vim, CEA, and p16INK4). In addition to the four individual monoclonal antibodies, both their respective three-marker panels, proposed here, showed statistically significant (p < 0.05) frequency differences between these two gynecologic tumors (ECA vs. EMA). The panel performance and test effectiveness revealed that both three-marker panels are promising and very helpful. According to our data, when histomorphological and clinical doubt exists as to the primary site of origin, we recommend using either of these two conventional three-marker panels, which consist of ER/Vim/CEA and ER/Vim/p16INK4. CEA and p16INK4 can be interchanged with confidence without significantly influencing the panel presentations and efficiencies in distinguishing between adenocarcinomas of endocervical and endometrial origin.

Distinguishing between primary endocervical and endometrial adenocarcinomas: is a 2-marker (Vim/CEA) panel enough?

Gynecological pathologists are used to operating many panels of various markers in combination for the diagnostic distinction between primary endocervical and endometrial adenocarcinomas. The conventional 3-marker (ER/Vim/CEA) panel is the most promising tool. In this study, our aim is to investigate whether a 2-marker panel is enough to distinguish between these two gynecologic malignancies. Additionally, we wish to determine which one is the most favorable among eight panels tested, including six 2-marker (ER/CEA, PR/CEA, Vim/CEA, ER/p16INK4a, PR/p16INK4a, Vim/p16INK4a) and two 3-marker (ER/Vim/CEA, ER/Vim/p16INK) panels. A tissue microarray was constructed using paraffin-embedded, formalin-fixed tissues from 35 hysterectomy specimens, including 14 primary endocervical adenocarcinomas and 21 primary endometrial adenocarcinomas. Utilizing the avidin-biotin complex (ABC) method, tissue array sections were immunostained with five commercially available antibodies (ER, Vim, CEA, PR, and p16INK4a) to evaluate their individual frequencies of expression. We found that all eight aforementioned panels showed an encouraging range of overall accuracy (69.2% to 78.3%). However, one panel of 2-markers (Vim, CEA) exhibited the most efficiency (78.3%) in the diagnostic distinction between primary endocervical and endometrial adenocarcinomas. Based on the analyzed data, we conclude that the 2-marker (Vim/CEA) panel seems adequate to be an appropriate, convenient, and efficient means to distinguish between primary endocervical and endometrial adenocarcinomas. Even though there were a limited number of cases, this study still provides valuable references to help avoid wasting resources and unnecessary marker testing.

Anti-cytokeratin CAM 5.2 does not act as a surrogate of the cytokeratin 8/18 monoclonal antibody. Comment on: "Utility of p63 in the differential diagnosis of atypical fibroxanthoma and spindle cell squamous cell carcinoma", in J Cutan Pathol 2009; 36: 543

To the Editor, This letter focuses on the important issue of how we interpret immunohistochemistry staining results using the anti-cytokeratin CAM 5.2 reagent (Becton, Dickinson and Company. In the recent article, ‘‘Utility of p63 in the differential diagnosis of atypical fibroxanthoma and spindle cell squamous cell carcinoma’’ in J Cutan Pathol 2009; 36: 543, the authors may have inadvertently cited that CK8/18 is derived from CAM 5.2 clone (Becton Dickinson, San Jose, CA, USA) in Table 1, as well as anticytokeratin CAM 5.2 stands for cytokeratin 8/18 antibody in Table 2 of their article.1 However, Becton Dickinson Biosciences revised the data sheet for anti-cytokeratin CAM 5.2 (Catalog No. 349205) in 1997 and lists the primary reactivity as being to cytokeratin 8 and to a lesser extent to cytokeratin 7. Furthermore, it shows no reactivity with cytokeratin 18 or 19.2,3 We conclude that Becton Dickinson, San Jose, CA, USA, does not manufactured cytokeratin 8/18 monoclonal antibody. In addition, this letter is written to clarify that anti-cytokeratin CAM 5.2 reagent does not act as a surrogate of cytokeratin 8/18 monoclonal antibody. Jih-Tseng Chen1,†, Jeng-Dong Hsu2,3,†, Chung-Chin Yao4,5,†, Lih-Wen Han6, Chih-Ping Han7,8,9 1Department of Rehabilitation, Dajia Lee’s General Hospital, Lee’s Medical Corporation, Taichung, Taiwan, 2Department of Pathology, Chung Shan Medical University Hospital, Taichung, Taiwan, 3Department of Pathology, School of Medicine, Chung Shan Medical University, Taichung, Taiwan, 4Department of Surgery, Chung-Shan Medical University Hospital, Taichung, Taiwan, 5Department of Surgery, School of Medicine, Chung Shan Medical University, Taichung, Taiwan, 6Clinical Medicine Program, Peking University Health Science Center, Beijing, People’s Republic of China, 7Department of Obstetrics and Gynecology, Chung-Shan Medical University Hospital, Taichung, Taiwan, 8Institute of Medicine, Chung-Shan Medical University, Taichung, Taiwan, and 9Clinical Trial Center, Chung-Shan Medical University Hospital, Taichung, Taiwan e-mail: hanhaly@gmail.com

Monoclonal antibody CAM5.2 can detect the cytokeratin 8, not cytokeratin 18. Comment on: “Occult disseminated tumor cells in lymph nodes of patients with gastric carcinoma. A critical appraisal of assessment and relevance. Langenbecks Arch Surg. 2009 Jan; 394(1): 105-113.”

To the Editor: We read with great interest the contribution by Scheunemann, P. and colleagues regarding their article entitled, “Occult disseminated tumor cells in lymph nodes of patients with gastric carcinoma, a critical appraisal of assessment and relevance” [1]. However, the authors may have mistakenly interpreted that the monoclonal antibody CAM5.2 can be used to detect both the cytokeratin (CK) 8 and 18 in reference to a few misguided reports. We would like to comment and clarify that Becton Dickinson (BD) Biosciences has developed and produced the anti-cytokeratin CAM5.2 (clone CAM5.2) reagent, which reacted with human cytokeratin intermediate filament proteins of 48 and 52 kD, as well as identified as cytokeratins 7 and 8, respectively. In addition, Becton Dickinson Biosciences has also revised the data sheet for CAM5.2 (BD) to have a primary reactivity with CK8 as

CAM5.2 (Becton Dickinson Biosciences) is not synonymous with CK8/18 (Leica Microsystems)

histology to nodular fasciitis. Even though FNAC is regarded as a reliable diagnostic tool in the assessment of salivary gland neoplasms, diagnostic pitfalls exist, one of which lies between nodular fasciitis and pleomorphic adenoma. The histological heterogeneity of pleomorphic adenoma is reflected in its cytology. Problems in interpretation occur for two reasons: the predominance of one element over other components, or the presence of atypical cytomorphological features. If myoepithelial cells are numerous and mesenchymal material not readily apparent, the tumour may be classified as myoepithelioma. Clinically, this is of little significance. If the myxoid component is abundant, it may overwhelm the few epithelial cells present and the lesion may be mistaken for a retention cyst, schwannoma, intravenous pyogenic granuloma or nodular fasciitis. This is more likely to happen if only May–Grünwald–Giemsa-stained slides are examined, as strong metachromasia may mask other cellular components. Factors favouring pleomorphic adenoma over nodular fasciitis include the presence of a chondroid matrix, lack of spindle cells, or inflammatory cells and positive immunohistochemical staining with S-100 and cytokeratin. History of sudden appearance and rapid growth, and SMA, vimentin and CD68 positivity favour the latter. In the context of a parotid gland, awareness of the possible mimics of a pleomorphic adenoma is crucial to avoid missing rarer entities for which conservative management may suffice. A. M. Silvanto*, L. Melly*, S. A. Hannan and G. Kocjan *Department of Cellular Pathology, Royal Free Hospital, Department of ENT, Royal National Throat, Nose & Ear Hospital and Department of Cytopathology, University College London Hospital, London, UK

CAM5.2 Is Not Identical to Cytokeratins 8 and 18

In an article entitled “Prognostic Significance of CXCL12 Expression in Patients With Colorectal Carcinoma,”1 we found that the authors may have inadvertently annotated CAM5.2 as cytokeratins (CKs) 8 and 18 by using a pair of square …

Anti-cytokeratin CAM 5.2 (B.D. Biosciences) has a primary reactivity with CK 7 and CK 8 (48 and 52 kD) rather than with the smaller molecular weight keratin breakdown fractions (50, 43, and 38 kD)

To the Editor We read with great interest the recently published article by Dr. Amemiya Y and colleagues, in the journal Breast Cancer Research and Treatment, entitled ‘‘Insulin like growth factor binding protein-7 reduces growth of human breast cancer cells and xenografted tumors’’ [1]. The authors implemented an excellent preclinical study, and presented a good report. However, this article contains an incorrect message in the 2nd paragraph (IGFBP7 is downregulated in xenografted primary human breast tumors) of the ‘‘Results’’ section which, I believe, may misguide readers. The authors seem to have possibly overlooked the manufacturer’s data sheet of the monoclonal antibody, Becton–Dickinson Biosciences (B.D. Biosciences) anti-cytokeratin CAM5.2. Additionally, they may have inadvertently annotated that CAM 5.2 monoclonal antibody identifies lower molecular weight cytokeratin proteins (50, 43, and 38 kD) [1]. Since previous results by Makin et al. were attributed to breakdown products of cytokeratin 8 (CK 8), giving smaller molecular weight fractions (50, 43, and 38 kD) on immunoblots, Dr. Amemiya Y et al. seem to introduce the B.D. Biosciences CAM5.2 monoclonal antibody on the basis of a few limited, inappropriate historical viewpoints and references [1, 2]. We would like to clarify that anticytokeratin CAM 5.2, derived from the clone CAM5.2 was developed, manufactured and sold by B.D. Biosciences (Franklin Lakes, NJ). This clone recognizes human cytokeratin intermediate filament proteins at 48 and 52 kDa, which are identified as cytokeratins 7 and 8 (CK 7 and CK 8), respectively. Moreover, B.D. Biosciences revised the data sheet for the anti-cytokeratin CAM5.2 in 1997. It is now well-known that CAM 5.2 shows specific reactivity for CK 8, but less strict reactivity for CK 7 and no reactivity for CK 18 or CK 19 [3–5]. Chung-Chin Yao, Wea-Lung Lin, and Ming-Yung Lee contributed equally to this work.

Anti-Cytokeratin CAM5.2 (BD) does not act as a substitute of the CK8/18 monoclonal antibody. Comment on: 'Expression of c-MET, low-molecular-weight cytokeratin, matrix metalloproteinases-1 and -2 in spinal chordoma'. Histopathology 2009; 54; 607-613.

Sir: I wish to address the important issue of how we interpret the immunohistochemical results obtained with the monoclonal antibody CAM5.2 (Becton Dickinson, BD, USA). In the discussion in their recently published paper on ‘Expression of c-MET, low-molecular-weight cytokeratin, matrix metalloproteinases-1 and -2 in spinal chordoma’, the authors may have inadvertently cited CAM 5.2 (BD) as immunoreacting with CK8 and 18. In 1997, Becton Dickinson Biosciences revised the data sheet for CAM5.2 (BD) to indicate that the antibody has primary reactivity for CK8 as well as a weaker but distinct reactivity for CK7. It also shows no reactivity with 18 or 19. We would like to clarify that anti-cytokeratin CAM5.2 can recognize the CK8, not CK18. Anti-cytokeratin CAM 5.2 (BD) neither stands for CK 8/18 monoclonal antibody, nor acts as a substitute for the CK8 ⁄ 18 monoclonal antibody.