ORIGINAL ARTICLES
Expression of ProEx C in Primary and Metastatic Urothelial Carcinoma Lian Liu, M.D., PH.D., Cynthia Cohen, M.D., and Momin T. Siddiqui, M.D., FIAC*
Background: ProEx C is an antibody cocktail targeting the
expression of topoisomerase IIa and minichromosome maintenance protein-2. Both these proteins are over-expressed in the cell nucleus during aberrant S-phase induction of neoplastic cells, which leads to cell proliferation. The aim of this study was to determine whether ProEx C expression can detect primary and metastatic urothelial carcinoma (UC). Methods: Thirty one fine needle aspiration cell blocks (CB) with metastatic UC were identified. Immunohistochemical staining for ProEx C and thrombomodulin was performed. Additionally, staining for Pro Ex C was also performed in tissue microarrays (TMA) of 46 cases of primary UC and carcinomas from colon (80), stomach (31), pancreas (33), liver (92), ovary (24), endometrium (25), breast (60), lung (27), kidney (32), and prostate (44), as well as melanoma (22). Nuclear staining of ProEx C and membrane staining of thrombomodulin in at least 10% tumor cells was considered a positive result. Results: Both ProEx C and thrombomodulin have similar sensitivity for metastatic UC (84% vs. 77%, p50.75; whereas ProEx C yielded a higher sensitivity of 93% for primary UC than thrombomodulin (72%, p50.01). In addition to UC, ProEx C is also expressed in most of the malignant neoplasms tested in our TMA study, and has the highest sensitivity in colon and stomach carcinomas (94%). Conclusion: ProEx C has high sensitivity for UC. However, it is also expressed in carcinomas of colon, stomach, breast, and lung carcinomas and may not be a useful marker for workup of metastatic UC. Diagn. Cytopathol. 2015;43:181–187. VC 2014 Wiley Periodicals, Inc.
Key Words: ProEx C; urothelial carcinoma; urine; tissue microarray; thrombomodulin Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, Georgia *Correspondence to: Momin T. Siddiqui, M.D., FIAC, Department of Pathology, Room G179B, Emory University Hospital, 1364 Clifton Rd. NE, Atlanta, GA 30322, USA. E-mail: [email protected] Received 11 March 2014; revised 2 May 2014; Accepted 11 June 2014 DOI: 10.1002/dc.23193 Published online 27 June 2014 in Wiley Online Library (wileyonlinelibrary.com). C 2014 WILEY PERIODICALS, INC. V
ProEx C is a novel immunohistochemical cocktail consisting of monoclonal antibodies against topoisomerase IIa (TOP2A) and minichromosome maintenance protein-2 (MCM2). TOP2A is a nuclear enzyme that relaxes supercoiled DNA during replication through an adenosine triphosphate-dependent double strand break followed by strand passing.1 MCM2 is a protein which is required in an early stage of DNA replication to assemble replication origins onto DNA during the G1 phase of the cell cycle. In addition, it acts as a helicase to unwind DNA.2 MCM proteins are down-regulated in cells undergoing differentiation or quiescence, their expression present only in restricted proliferative compartments. In contrast, MCM proteins are expressed in the majority of cells in dysplastic and malignant tissues.3 Deregulation of the MCM2 function resulting in chromosomal defects has been suggested to contribute to tumorigenesis.4 Both TOP2A and MCM are over-expressed in the cell nucleus during aberrant S-phase induction of neoplastic cells, which leads to cell proliferation.3,5 TOP2A and MCM have been shown to be overexpressed in cervical neoplasia as a result of human papillomavirus (HPV) infection.5,6 It is widely recognized that HPV infection is the etiology for the initiation of cervical dysplasia. HPV oncoproteins E6 and E7 abrogate the critical cell cycle checkpoints and induce expression of Sphase genes through the constitutive action of the E2F transcription factor. This results in prolonged and active induction of S-phase genes and DNA replication. TOP2A and MCM are expressed during this aberrant gene transcriptional activation. The overexpression of ProEx C in cervical squamous dysplasia, in either histologic specimens or liquid-based cytology, are confirmed by multiple studies.7–10 Pinto et al. evaluated the performance of ProEx C in distinguishing high-grade squamous intraepithelial lesion (HSIL) from HSIL mimickers in histologic Diagnostic Cytopathology, Vol. 43, No 3
181
Diagnostic Cytopathology DOI 10.1002/dc
LIU ET AL.
sections compared to P16 and MIB 1.10 In their studies, the combination of p16/ProEx C showed similar sensitivity for HSIL compared to the p16/MIB 1 combination (92% vs. 94%, respectively); whereas the specificity appeared to be higher (61% vs. 43%, respectively), although this difference was not statistically significant. Our previous studies showed that in cervicovaginal liquid-based cytology samples that were diagnosed as atypical squamous cells-cannot exclude HSIL (ASC-H) or atypical squamous cells of undetermined significance (ASCUS), ProEx C had a higher sensitivity for detecting CIN21 lesions compared to high risk HPV, whereas the specificity was not statistically significant.7,8 Recently, it has been shown by Moatamed et al. that ProEx C is a useful marker for diagnosing urothelial carcinoma (UC).11 They reported that ProEx C expression has an overall 95% sensitivity and specificity for detection of UC in histologic sections. They suggested that ProEx C is effective in differentiating carcinoma in situ and benign urothelial lesions.11 The authors further demonstrated that ProEx C is also a useful marker for detecting UC in urine cytology specimens.12 In their study, ProEx C immunostain was performed on 60 urine specimens with histologic follow up. All 12 benign cytology samples showed negative staining with ProEx C. Twelve of 13 cases that had a malignant cytologic diagnosis showed a positive nuclear staining of the malignant cells. Of the 35 cases diagnosed as “atypical”, 17 of 24 cases, which were proven to be malignant histologically, stained positive with ProEx C. Whereas only one of the 11 cases that were confirmed as benign histologically, stained positive with ProEx C. The aim of our study was to determine whether ProEx C expression can detect primary and metastatic UC, and if so, how useful it would be for usage in work-up of metastatic disease.
Materials and Methods Study Material The study protocol was approved by the Institutional Review Board, Emory University Hospital, Atlanta, GA. In this study 31 fine needle aspiration (FNA) specimens collected between 2005 and 2013 diagnosed with metastatic UC were identified. Immunohistochemical (IHC) staining for ProEx C (BD, Franklin Lakes, NJ, ready to use) was performed in formalin-fixed, paraffin-embedded cell blocks (CB). Staining of thrombomodulin (Cell Marque, Rocklin, CA, ready to use), an established urothelial marker, was performed as well for comparison. ProEx C staining was also performed in tissue microarrays (TMA) of 46 primary UC. Additionally, ProEx C staining was also performed in TMAs of 470 nonurothelial tumors, which include colon adenocarcinoma (80), 182
Diagnostic Cytopathology, Vol. 43, No 3
gastric adenocarcinoma (31), pancreatic ductal adenocarcinoma (33), hepatocellular carcinoma (92), ovarian carcinoma (24), endometrial carcinoma (25), breast carcinoma (60), lung adenocarcinoma (27), renal cell carcinoma (32), and prostatic adenocarcinoma (44), as well as melanoma (22).
CB Preparation All FNA specimens from metastatic UC were immediately Diff-Quik (Thermo-Fisher Scientific, Kalamazoo, MI) stained and evaluated for adequacy. The remainder of the FNA specimen was submitted for CB preparation by fixing in 10% neutral buffered formalin for a minimum of two hours and then poured undiluted into a 15mL glass centrifuge tube coated with collodion solution and centrifuged for 7 minutes. The supernatant was discarded and the resultant collodion bag (JT Baker, Phillipsburg, NJ) and centrifuged material underwent routine paraffin-embedding as a CB then staining with hematoxylin and eosin (H&E).
Immunohistochemistry Five micron sections of formalin-fixed, paraffin-embedded CB of metastatic UC and TMAs were deparaffinized and rehydrated to deionized water. Antigen retrieval was performed in citrate buffer (pH 6.0), using an electric pressure cooker for 3 minutes at 12–15 pounds/square inch (120 C), and cooled for 10 minutes prior to immunostaining. All slides were loaded on an automated system (Dako AutoStainer plus, Carpinteria, CA) and exposed to 3% H2O2 for 5 minutes, incubated with primary antibodies for 30 minutes, R 1 dual link) for 30 with labeled polymer (EnvisionV minutes, 3,30 -diaminobenzidine (DAB) as chromogen for 5 minutes, and hematoxylin as counterstain for 5 minutes. These incubations were performed at room temperature; between incubations, sections were washed with trisbuffered saline (TBS). Cover-slipping was performed using the Tissue Tek SCA (Sakura Finetek, USA) coverslipper. Antibody titers used were the same for both CB and TMA preparations. Positive and negative controls were also used for both CB and TMA samples. Nuclear staining of ProEx C and membrane staining of thrombomodulin in at least 10% tumor cells was considered a positive result.
Statistical Analysis Two 3 two contingency tables were used to calculate the sensitivity and specificity of ProEx C. Fisher’s exact test was used to compare the sensitivity of ProEx C and thrombomodulin in metastatic and primary UC.
Results In our studies, the expression of ProEx C was compared to that of thrombomodulin, a well accepted marker for
Diagnostic Cytopathology DOI 10.1002/dc
EXPRESSION OF PROEX C IN UROTHELIAL CARCINOMA
Fig. 1. Positive ProEx C (nuclear staining) and thrombomodulin (membrane staining) in metastatic and primary urothelial carcinoma (UC) (4003), respectively. A. ProEx C in a cell block of metastatic UC. B. ProEx C in a tissue microarray of primary UC. C. Thrombomodulin in a cell block of metastatic UC. D. Thrombomodulin in a tissue microarray of primary UC. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
UC. ProEx C shows nuclear staining and thrombomodulin membrane staining (Fig. 1). Both ProEx C and thrombomodulin have similar sensitivity for metastatic UC in FNA specimens; whereas ProEx C yielded a higher sensitivity of 93% for primary UC than thrombomodulin (72%, P 5 0.01) in TMA (Table I). In addition to UC, ProEx C is also expressed in most of the malignant neoplasms that we have tested in our TMA study (Fig. 2), and has the highest sensitivity in colon and stomach carcinomas (Table II). ProEx C
Table I. Sensitivity of ProEx C vs. Thrombomodulin in Primary and Metastatic Urothelial Carcinoma (UC)
ProEx C Expression in Primary UC TM Expression in Primary UC ProEx C Expression in Metastatic UC TM Expression in Metastatic UC
Positive/ total cases
Sensitivity
P (ProEx C vs. TM)
43/46
93%
5 0.01
33/46
72%
26/31
84%
24/31
77%
5 0.75
Diagnostic Cytopathology, Vol. 43, No 3
183
Diagnostic Cytopathology DOI 10.1002/dc
LIU ET AL.
Fig. 2. Positive ProEx C in representative malignant neoplasms in TMAs (2003). A. Colon. B. Stomach. C. Ovary. D. Endometrium. E. Breast. F. Lung. G. Pancreas. H. Melanoma. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
expression is strong and diffuse in more than half of the malignant cells in colon, stomach, bladder, ovary, and lung carcinomas. The positive cells tend to be fewer and scattered in breast, kidney, and prostate carcinoma. The poor ability to distinguish the type of malignancies by ProEx C staining is reflected by the low specificity (range from 34% to 49% in all of the tumors tested).
Discussion Bladder cancer is the fourth most common cancer in men in the United States. An estimated 74,690 new 184
Diagnostic Cytopathology, Vol. 43, No 3
cases of bladder cancer are expected to occur in 2014, and an estimated 15,580 deaths will occur in the same period (http://www.cancer.org). Bladder cancer is diagnosed by microscopic examination of urine samples or surgical biopsies. About ninety percent of bladder tumors are UC. High-grade UC typically shows a high nuclear to cytoplasmic ratio, irregular nuclear contour, nuclear hyperchromasia, and coarse chromatin. The cytologic features, however, sometimes overlap with reactive atypia such as those seen in urolithiasis. The diagnosis can also be difficult when the tumor cells are few and/ or degenerate.
Diagnostic Cytopathology DOI 10.1002/dc
EXPRESSION OF PROEX C IN UROTHELIAL CARCINOMA
Fig. 2. (Continued.)
Approximately 5% of patients with UC have metastasis at initial diagnosis.13 The accurate diagnosis is pivotal for appropriate management. Johnson and Kini reported that the most distinctive features for metastatic UC were the presence of “spindled, pyramidal, and/or racquet-shaped malignant cells with eccentric nuclei and cytoplasmic features of both squamous and glandular differentiation”.14 Powers and Elbadawi described a “cercariform cell”, which is characterized by a globular body and a unipolar cytoplasmic process with a blunt end, as a distinctive feature of metastatic UC.15 In addition, Melamed-Wolinska bodies, which are intracytoplasmic eosinophilic inclusions,
were shown to be a feature of primary/metastatic UC.16,17 Although the above cytological features give a clue to the origin of the metastatic carcinoma, they are neither sensitive nor specific. The confirmation of metastatic UC is mostly based on IHC stains when the primary tumor is not available for comparison. Several IHC markers have been used in the diagnosis of metastatic UC such as cytokeratin 7 and 20, p63, thrombomodulin, and uroplakin. However, those markers lack either sensitivity or specificity for the confirmation of metastatic UC. In our studies, we confirmed that ProEx C has high sensitivity for primary UC (93%), which is greater than Diagnostic Cytopathology, Vol. 43, No 3
185
Diagnostic Cytopathology DOI 10.1002/dc
LIU ET AL. Table II. Sensitivity and Specificity of ProEx C in Tissue Microarrays (TMAs) TMA Colon Stomach Bladder Ovary Endometrium Breast Lung Pancreas Kidney Prostate Hepatocellular Carcinoma Melanoma
Positive/ total cases
Sensitivity (%)
Specificity (%)
75/80 29/31 43/46 18/24 18/25 41/60 18/27 21/33 17/32 15/44 7/92 13/22
94% 94% 93% 75% 72% 68% 67% 64% 53% 34% 8% 59%
49% 39% 42% 37% 37% 42% 37% 37% 37% 36% 34% 36%
the sensitivity of thrombomodulin (72%, P 5 0.01). Although the two markers have similar sensitivity for metastatic UC. Our studies further demonstrated that ProEx C is a rather nonspecific biomarker that is positive in many other malignant neoplasms. In addition to UC, ProEx C has high sensitivity for gastrointestinal carcinomas including colonic and gastric adenocarcinomas. Tumors that show a moderate frequency of expression of ProEx C include ovarian, endometrial, breast, lung, pancreatic, and renal cell carcinomas, as well as melanomas. Only onethird of prostatic adenocarcinomas are positive for ProEx C and the positive cells tend to be few and focal. More than 90% of the hepatocellular carcinoma (HCC) are negative for ProEx C. Our studies demonstrated that although useful for diagnosing primary UC, ProEx C is not helpful for working up a metastatic carcinoma as it shows moderate to high expression in most of the common carcinomas. Even in urine cytology specimens, albeit rare, metastatic carcinomas such as colonic, prostatic, or renal cell carcinomas can be seen.18 When using ProEx C to aid in detecting UC in urine specimens, one should be aware that metastases, especially colonic carcinoma, can be positive for this marker as well. In addition to urine specimens, multiple studies have shown a potential important role that ProEx C may play in the diagnosis of squamous intraepithelial lesions in cervical liquid-based Papanicolaou (Pap) tests. ProEx C has been used to detect HSIL and squamous cell carcinoma, and to differentiate them from dysplasia mimickers such as immature squamous metaplasia and atrophy in cytology material. In Pap tests, evaluation of hyperchromatic crowded groups can be challenging, and the differential diagnosis ranges from benign cells, such as exfoliated endometrial cells and tubal metaplasia, to dysplastic lesions, such as HSIL, in situ and invasive endocervical adenocarcinoma, and endometrial adenocarcinoma. 186
Diagnostic Cytopathology, Vol. 43, No 3
Ge et al. suggested that ProEx C is helpful in interpreting hyperchromatic crowded groups in menstrual Pap specimens.19 In their studies, a strong and diffuse staining pattern for ProEx C was observed in cases of HSIL; whereas benign endometrial cells had either negative or focal patchy staining, which is often associated with tubal metaplasia. However, our studies suggest that caution should be taken when using ProEx C in evaluating hyperchromatic crowded groups in Pap tests as endometrial carcinoma also expresses ProEx C. Additionally, expression of ProEx C has been reported in situ and invasive endocervical adenocarcinoma.20 In conclusion, ProEx C has high sensitivity for UC, and can be useful for detecting UC in metastatic disease identified in CB preparations as well as different preparations such as TMA samples prepared from primary UC specimens. However, ProEx C is a rather nonspecific marker, and is expressed in many tumors such as carcinomas such as colon, stomach, breast, and lung carcinomas. Therefore, ProEx C may not a useful marker for workup of metastatic disease.
References 1. Champoux JJ. DNA topoisomerases: structure, function, and mechanism. Annu Rev Biochem 2001;70:369–413. 2. Lei M, Tye BK. Initiating DNA synthesis: from recruiting to activating the MCM complex. J Cell Sci 2001;114(Pt 8): 1447–1454. 3. Freeman A, Morris LS, Mills AD, et al. Minichromosome maintenance proteins as biological markers of dysplasia and malignancy. Clin Cancer Res 1999;5:2121–2132. 4. Lei M. The MCM complex: its role in DNA replication and implications for cancer therapy. Curr Cancer Drug Target 2005;5: 365–380. 5. Malinowski DP. Molecular diagnostic assays for cervical neoplasia: emerging markers for the detection of high-grade cervical disease. Biotechniques 2005;Suppl:17–23. 6. Malinowski DP. Multiple biomarkers in molecular oncology. I. Molecular diagnostics applications in cervical cancer detection. Expert Rev Mol Diagn 2007;7:117–131. 7. Siddiqui MT, Cohen C, Nassar A. Detecting high-grade cervical disease on ASC-H cytology: role of BD ProEx C and Digene Hybrid Capture II HPV DNA testing. Am J Clin Pathol 2008;130: 765–770. 8. Siddiqui MT, Hornaman K, Cohen C, Nassar A. ProEx C immunocytochemistry and high-risk human papillomavirus DNA testing in papanicolaou tests with atypical squamous cell (ASC-US) cytology: correlation study with histologic biopsy. Arch Pathol Lab Med 2008;132:1648–1652. 9. Kelly D, Kincaid E, Fansler Z, Rosenthal DL, Clark DP. Detection of cervical high-grade squamous intraepithelial lesions from cytologic samples using a novel immunocytochemical assay (ProEx C). Cancer 2006;108:494–500. 10. Pinto AP, Schlecht NF, Woo TY, Crum CP, Cibas ES. Biomarker (ProEx C, p16(INK4A), and MiB-1) distinction of high-grade squamous intraepithelial lesion from its mimics. Mod Pathol 2008;21: 1067–1074. 11. Moatamed NA, Vergara-Lluri ME, Lu D, Apple SK, Kerkoutian S, Rao JY. Utility of ProEx C in the histologic evaluation of the neoplastic and nonneoplastic urothelial lesions. Hum Pathol 2013;44: 2509–2517.
Diagnostic Cytopathology DOI 10.1002/dc
EXPRESSION OF PROEX C IN UROTHELIAL CARCINOMA 12. Moatamed NA, Rao JY, Alexanian S, et al. ProEx C as an adjunct marker to improve cytological detection of urothelial carcinoma in urinary specimens. Cancer Cytopathol 2013;121:320–328. 13. Al-Sukhun S, Hussain M. Current understanding of the biology of advanced bladder cancer. Cancer 2003;15;97(Suppl 8):2064–2075. 14. Johnson TL, Kini SR. Cytologic features of metastatic transitional cell carcinoma. Diagn Cytopathol 1993;9:270–278. 15. Powers CN, Elbadwi A. Cercariform cells: a clue to the cytodiagnosis of urothelial origin of metastatic neoplasms? Diagn Cytopathol 1995; 13:15–21. 16. Rashidi B, Tongson-Ignacio JE. Melamed-Wolinska bodies in urine cytology an interesting aggregate in a degenerated urothelial cell. Diagn Cytopathol 2011;39:117. 17. Renshaw AA, Madge R, Granter SR. Intracytoplasmic eosinophilic inclusions (Melamed-Wolinska bodies). Association with metastatic
transitional cell carcinoma in pleural fluid. Acta Cytol 1997;41: 995–998. 18. Bardales RH, Pitman MB, Stanley MW, Korourian S, Suhrland MJ. Urine cytology of primary and secondary urinary bladder adenocarcinoma. Cancer 1998;84:335–343. 19. Ge Y, Mody DR, Smith D, Anton R. p16(INK4a) and ProEx C immunostains facilitate differential diagnosis of hyperchromatic crowded groups in liquid-based Papanicolaou tests with menstrual contamination. Acta Cytol 2012;56:55–61. 20. Aximu D, Azad A, Ni R, Colgan T, Nanji S. A pilot evaluation of a novel immunohistochemical assay for topoisomerase II-alpha and minichromosome maintenance protein 2 expression (ProEx C) in cervical adenocarcinoma in situ, adenocarcinoma, and benign glandular mimics. Int J Gynecol Pathol 2009;28: 114–119.
Diagnostic Cytopathology, Vol. 43, No 3
187
Expression of ProEx C in Primary and Metastatic Urothelial Carcinoma Lian Liu, M.D., PH.D., Cynthia Cohen, M.D., and Momin T. Siddiqui, M.D., FIAC*
Background: ProEx C is an antibody cocktail targeting the
expression of topoisomerase IIa and minichromosome maintenance protein-2. Both these proteins are over-expressed in the cell nucleus during aberrant S-phase induction of neoplastic cells, which leads to cell proliferation. The aim of this study was to determine whether ProEx C expression can detect primary and metastatic urothelial carcinoma (UC). Methods: Thirty one fine needle aspiration cell blocks (CB) with metastatic UC were identified. Immunohistochemical staining for ProEx C and thrombomodulin was performed. Additionally, staining for Pro Ex C was also performed in tissue microarrays (TMA) of 46 cases of primary UC and carcinomas from colon (80), stomach (31), pancreas (33), liver (92), ovary (24), endometrium (25), breast (60), lung (27), kidney (32), and prostate (44), as well as melanoma (22). Nuclear staining of ProEx C and membrane staining of thrombomodulin in at least 10% tumor cells was considered a positive result. Results: Both ProEx C and thrombomodulin have similar sensitivity for metastatic UC (84% vs. 77%, p50.75; whereas ProEx C yielded a higher sensitivity of 93% for primary UC than thrombomodulin (72%, p50.01). In addition to UC, ProEx C is also expressed in most of the malignant neoplasms tested in our TMA study, and has the highest sensitivity in colon and stomach carcinomas (94%). Conclusion: ProEx C has high sensitivity for UC. However, it is also expressed in carcinomas of colon, stomach, breast, and lung carcinomas and may not be a useful marker for workup of metastatic UC. Diagn. Cytopathol. 2015;43:181–187. VC 2014 Wiley Periodicals, Inc.
Key Words: ProEx C; urothelial carcinoma; urine; tissue microarray; thrombomodulin Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, Georgia *Correspondence to: Momin T. Siddiqui, M.D., FIAC, Department of Pathology, Room G179B, Emory University Hospital, 1364 Clifton Rd. NE, Atlanta, GA 30322, USA. E-mail: [email protected] Received 11 March 2014; revised 2 May 2014; Accepted 11 June 2014 DOI: 10.1002/dc.23193 Published online 27 June 2014 in Wiley Online Library (wileyonlinelibrary.com). C 2014 WILEY PERIODICALS, INC. V
ProEx C is a novel immunohistochemical cocktail consisting of monoclonal antibodies against topoisomerase IIa (TOP2A) and minichromosome maintenance protein-2 (MCM2). TOP2A is a nuclear enzyme that relaxes supercoiled DNA during replication through an adenosine triphosphate-dependent double strand break followed by strand passing.1 MCM2 is a protein which is required in an early stage of DNA replication to assemble replication origins onto DNA during the G1 phase of the cell cycle. In addition, it acts as a helicase to unwind DNA.2 MCM proteins are down-regulated in cells undergoing differentiation or quiescence, their expression present only in restricted proliferative compartments. In contrast, MCM proteins are expressed in the majority of cells in dysplastic and malignant tissues.3 Deregulation of the MCM2 function resulting in chromosomal defects has been suggested to contribute to tumorigenesis.4 Both TOP2A and MCM are over-expressed in the cell nucleus during aberrant S-phase induction of neoplastic cells, which leads to cell proliferation.3,5 TOP2A and MCM have been shown to be overexpressed in cervical neoplasia as a result of human papillomavirus (HPV) infection.5,6 It is widely recognized that HPV infection is the etiology for the initiation of cervical dysplasia. HPV oncoproteins E6 and E7 abrogate the critical cell cycle checkpoints and induce expression of Sphase genes through the constitutive action of the E2F transcription factor. This results in prolonged and active induction of S-phase genes and DNA replication. TOP2A and MCM are expressed during this aberrant gene transcriptional activation. The overexpression of ProEx C in cervical squamous dysplasia, in either histologic specimens or liquid-based cytology, are confirmed by multiple studies.7–10 Pinto et al. evaluated the performance of ProEx C in distinguishing high-grade squamous intraepithelial lesion (HSIL) from HSIL mimickers in histologic Diagnostic Cytopathology, Vol. 43, No 3
181
Diagnostic Cytopathology DOI 10.1002/dc
LIU ET AL.
sections compared to P16 and MIB 1.10 In their studies, the combination of p16/ProEx C showed similar sensitivity for HSIL compared to the p16/MIB 1 combination (92% vs. 94%, respectively); whereas the specificity appeared to be higher (61% vs. 43%, respectively), although this difference was not statistically significant. Our previous studies showed that in cervicovaginal liquid-based cytology samples that were diagnosed as atypical squamous cells-cannot exclude HSIL (ASC-H) or atypical squamous cells of undetermined significance (ASCUS), ProEx C had a higher sensitivity for detecting CIN21 lesions compared to high risk HPV, whereas the specificity was not statistically significant.7,8 Recently, it has been shown by Moatamed et al. that ProEx C is a useful marker for diagnosing urothelial carcinoma (UC).11 They reported that ProEx C expression has an overall 95% sensitivity and specificity for detection of UC in histologic sections. They suggested that ProEx C is effective in differentiating carcinoma in situ and benign urothelial lesions.11 The authors further demonstrated that ProEx C is also a useful marker for detecting UC in urine cytology specimens.12 In their study, ProEx C immunostain was performed on 60 urine specimens with histologic follow up. All 12 benign cytology samples showed negative staining with ProEx C. Twelve of 13 cases that had a malignant cytologic diagnosis showed a positive nuclear staining of the malignant cells. Of the 35 cases diagnosed as “atypical”, 17 of 24 cases, which were proven to be malignant histologically, stained positive with ProEx C. Whereas only one of the 11 cases that were confirmed as benign histologically, stained positive with ProEx C. The aim of our study was to determine whether ProEx C expression can detect primary and metastatic UC, and if so, how useful it would be for usage in work-up of metastatic disease.
Materials and Methods Study Material The study protocol was approved by the Institutional Review Board, Emory University Hospital, Atlanta, GA. In this study 31 fine needle aspiration (FNA) specimens collected between 2005 and 2013 diagnosed with metastatic UC were identified. Immunohistochemical (IHC) staining for ProEx C (BD, Franklin Lakes, NJ, ready to use) was performed in formalin-fixed, paraffin-embedded cell blocks (CB). Staining of thrombomodulin (Cell Marque, Rocklin, CA, ready to use), an established urothelial marker, was performed as well for comparison. ProEx C staining was also performed in tissue microarrays (TMA) of 46 primary UC. Additionally, ProEx C staining was also performed in TMAs of 470 nonurothelial tumors, which include colon adenocarcinoma (80), 182
Diagnostic Cytopathology, Vol. 43, No 3
gastric adenocarcinoma (31), pancreatic ductal adenocarcinoma (33), hepatocellular carcinoma (92), ovarian carcinoma (24), endometrial carcinoma (25), breast carcinoma (60), lung adenocarcinoma (27), renal cell carcinoma (32), and prostatic adenocarcinoma (44), as well as melanoma (22).
CB Preparation All FNA specimens from metastatic UC were immediately Diff-Quik (Thermo-Fisher Scientific, Kalamazoo, MI) stained and evaluated for adequacy. The remainder of the FNA specimen was submitted for CB preparation by fixing in 10% neutral buffered formalin for a minimum of two hours and then poured undiluted into a 15mL glass centrifuge tube coated with collodion solution and centrifuged for 7 minutes. The supernatant was discarded and the resultant collodion bag (JT Baker, Phillipsburg, NJ) and centrifuged material underwent routine paraffin-embedding as a CB then staining with hematoxylin and eosin (H&E).
Immunohistochemistry Five micron sections of formalin-fixed, paraffin-embedded CB of metastatic UC and TMAs were deparaffinized and rehydrated to deionized water. Antigen retrieval was performed in citrate buffer (pH 6.0), using an electric pressure cooker for 3 minutes at 12–15 pounds/square inch (120 C), and cooled for 10 minutes prior to immunostaining. All slides were loaded on an automated system (Dako AutoStainer plus, Carpinteria, CA) and exposed to 3% H2O2 for 5 minutes, incubated with primary antibodies for 30 minutes, R 1 dual link) for 30 with labeled polymer (EnvisionV minutes, 3,30 -diaminobenzidine (DAB) as chromogen for 5 minutes, and hematoxylin as counterstain for 5 minutes. These incubations were performed at room temperature; between incubations, sections were washed with trisbuffered saline (TBS). Cover-slipping was performed using the Tissue Tek SCA (Sakura Finetek, USA) coverslipper. Antibody titers used were the same for both CB and TMA preparations. Positive and negative controls were also used for both CB and TMA samples. Nuclear staining of ProEx C and membrane staining of thrombomodulin in at least 10% tumor cells was considered a positive result.
Statistical Analysis Two 3 two contingency tables were used to calculate the sensitivity and specificity of ProEx C. Fisher’s exact test was used to compare the sensitivity of ProEx C and thrombomodulin in metastatic and primary UC.
Results In our studies, the expression of ProEx C was compared to that of thrombomodulin, a well accepted marker for
Diagnostic Cytopathology DOI 10.1002/dc
EXPRESSION OF PROEX C IN UROTHELIAL CARCINOMA
Fig. 1. Positive ProEx C (nuclear staining) and thrombomodulin (membrane staining) in metastatic and primary urothelial carcinoma (UC) (4003), respectively. A. ProEx C in a cell block of metastatic UC. B. ProEx C in a tissue microarray of primary UC. C. Thrombomodulin in a cell block of metastatic UC. D. Thrombomodulin in a tissue microarray of primary UC. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
UC. ProEx C shows nuclear staining and thrombomodulin membrane staining (Fig. 1). Both ProEx C and thrombomodulin have similar sensitivity for metastatic UC in FNA specimens; whereas ProEx C yielded a higher sensitivity of 93% for primary UC than thrombomodulin (72%, P 5 0.01) in TMA (Table I). In addition to UC, ProEx C is also expressed in most of the malignant neoplasms that we have tested in our TMA study (Fig. 2), and has the highest sensitivity in colon and stomach carcinomas (Table II). ProEx C
Table I. Sensitivity of ProEx C vs. Thrombomodulin in Primary and Metastatic Urothelial Carcinoma (UC)
ProEx C Expression in Primary UC TM Expression in Primary UC ProEx C Expression in Metastatic UC TM Expression in Metastatic UC
Positive/ total cases
Sensitivity
P (ProEx C vs. TM)
43/46
93%
5 0.01
33/46
72%
26/31
84%
24/31
77%
5 0.75
Diagnostic Cytopathology, Vol. 43, No 3
183
Diagnostic Cytopathology DOI 10.1002/dc
LIU ET AL.
Fig. 2. Positive ProEx C in representative malignant neoplasms in TMAs (2003). A. Colon. B. Stomach. C. Ovary. D. Endometrium. E. Breast. F. Lung. G. Pancreas. H. Melanoma. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
expression is strong and diffuse in more than half of the malignant cells in colon, stomach, bladder, ovary, and lung carcinomas. The positive cells tend to be fewer and scattered in breast, kidney, and prostate carcinoma. The poor ability to distinguish the type of malignancies by ProEx C staining is reflected by the low specificity (range from 34% to 49% in all of the tumors tested).
Discussion Bladder cancer is the fourth most common cancer in men in the United States. An estimated 74,690 new 184
Diagnostic Cytopathology, Vol. 43, No 3
cases of bladder cancer are expected to occur in 2014, and an estimated 15,580 deaths will occur in the same period (http://www.cancer.org). Bladder cancer is diagnosed by microscopic examination of urine samples or surgical biopsies. About ninety percent of bladder tumors are UC. High-grade UC typically shows a high nuclear to cytoplasmic ratio, irregular nuclear contour, nuclear hyperchromasia, and coarse chromatin. The cytologic features, however, sometimes overlap with reactive atypia such as those seen in urolithiasis. The diagnosis can also be difficult when the tumor cells are few and/ or degenerate.
Diagnostic Cytopathology DOI 10.1002/dc
EXPRESSION OF PROEX C IN UROTHELIAL CARCINOMA
Fig. 2. (Continued.)
Approximately 5% of patients with UC have metastasis at initial diagnosis.13 The accurate diagnosis is pivotal for appropriate management. Johnson and Kini reported that the most distinctive features for metastatic UC were the presence of “spindled, pyramidal, and/or racquet-shaped malignant cells with eccentric nuclei and cytoplasmic features of both squamous and glandular differentiation”.14 Powers and Elbadawi described a “cercariform cell”, which is characterized by a globular body and a unipolar cytoplasmic process with a blunt end, as a distinctive feature of metastatic UC.15 In addition, Melamed-Wolinska bodies, which are intracytoplasmic eosinophilic inclusions,
were shown to be a feature of primary/metastatic UC.16,17 Although the above cytological features give a clue to the origin of the metastatic carcinoma, they are neither sensitive nor specific. The confirmation of metastatic UC is mostly based on IHC stains when the primary tumor is not available for comparison. Several IHC markers have been used in the diagnosis of metastatic UC such as cytokeratin 7 and 20, p63, thrombomodulin, and uroplakin. However, those markers lack either sensitivity or specificity for the confirmation of metastatic UC. In our studies, we confirmed that ProEx C has high sensitivity for primary UC (93%), which is greater than Diagnostic Cytopathology, Vol. 43, No 3
185
Diagnostic Cytopathology DOI 10.1002/dc
LIU ET AL. Table II. Sensitivity and Specificity of ProEx C in Tissue Microarrays (TMAs) TMA Colon Stomach Bladder Ovary Endometrium Breast Lung Pancreas Kidney Prostate Hepatocellular Carcinoma Melanoma
Positive/ total cases
Sensitivity (%)
Specificity (%)
75/80 29/31 43/46 18/24 18/25 41/60 18/27 21/33 17/32 15/44 7/92 13/22
94% 94% 93% 75% 72% 68% 67% 64% 53% 34% 8% 59%
49% 39% 42% 37% 37% 42% 37% 37% 37% 36% 34% 36%
the sensitivity of thrombomodulin (72%, P 5 0.01). Although the two markers have similar sensitivity for metastatic UC. Our studies further demonstrated that ProEx C is a rather nonspecific biomarker that is positive in many other malignant neoplasms. In addition to UC, ProEx C has high sensitivity for gastrointestinal carcinomas including colonic and gastric adenocarcinomas. Tumors that show a moderate frequency of expression of ProEx C include ovarian, endometrial, breast, lung, pancreatic, and renal cell carcinomas, as well as melanomas. Only onethird of prostatic adenocarcinomas are positive for ProEx C and the positive cells tend to be few and focal. More than 90% of the hepatocellular carcinoma (HCC) are negative for ProEx C. Our studies demonstrated that although useful for diagnosing primary UC, ProEx C is not helpful for working up a metastatic carcinoma as it shows moderate to high expression in most of the common carcinomas. Even in urine cytology specimens, albeit rare, metastatic carcinomas such as colonic, prostatic, or renal cell carcinomas can be seen.18 When using ProEx C to aid in detecting UC in urine specimens, one should be aware that metastases, especially colonic carcinoma, can be positive for this marker as well. In addition to urine specimens, multiple studies have shown a potential important role that ProEx C may play in the diagnosis of squamous intraepithelial lesions in cervical liquid-based Papanicolaou (Pap) tests. ProEx C has been used to detect HSIL and squamous cell carcinoma, and to differentiate them from dysplasia mimickers such as immature squamous metaplasia and atrophy in cytology material. In Pap tests, evaluation of hyperchromatic crowded groups can be challenging, and the differential diagnosis ranges from benign cells, such as exfoliated endometrial cells and tubal metaplasia, to dysplastic lesions, such as HSIL, in situ and invasive endocervical adenocarcinoma, and endometrial adenocarcinoma. 186
Diagnostic Cytopathology, Vol. 43, No 3
Ge et al. suggested that ProEx C is helpful in interpreting hyperchromatic crowded groups in menstrual Pap specimens.19 In their studies, a strong and diffuse staining pattern for ProEx C was observed in cases of HSIL; whereas benign endometrial cells had either negative or focal patchy staining, which is often associated with tubal metaplasia. However, our studies suggest that caution should be taken when using ProEx C in evaluating hyperchromatic crowded groups in Pap tests as endometrial carcinoma also expresses ProEx C. Additionally, expression of ProEx C has been reported in situ and invasive endocervical adenocarcinoma.20 In conclusion, ProEx C has high sensitivity for UC, and can be useful for detecting UC in metastatic disease identified in CB preparations as well as different preparations such as TMA samples prepared from primary UC specimens. However, ProEx C is a rather nonspecific marker, and is expressed in many tumors such as carcinomas such as colon, stomach, breast, and lung carcinomas. Therefore, ProEx C may not a useful marker for workup of metastatic disease.
References 1. Champoux JJ. DNA topoisomerases: structure, function, and mechanism. Annu Rev Biochem 2001;70:369–413. 2. Lei M, Tye BK. Initiating DNA synthesis: from recruiting to activating the MCM complex. J Cell Sci 2001;114(Pt 8): 1447–1454. 3. Freeman A, Morris LS, Mills AD, et al. Minichromosome maintenance proteins as biological markers of dysplasia and malignancy. Clin Cancer Res 1999;5:2121–2132. 4. Lei M. The MCM complex: its role in DNA replication and implications for cancer therapy. Curr Cancer Drug Target 2005;5: 365–380. 5. Malinowski DP. Molecular diagnostic assays for cervical neoplasia: emerging markers for the detection of high-grade cervical disease. Biotechniques 2005;Suppl:17–23. 6. Malinowski DP. Multiple biomarkers in molecular oncology. I. Molecular diagnostics applications in cervical cancer detection. Expert Rev Mol Diagn 2007;7:117–131. 7. Siddiqui MT, Cohen C, Nassar A. Detecting high-grade cervical disease on ASC-H cytology: role of BD ProEx C and Digene Hybrid Capture II HPV DNA testing. Am J Clin Pathol 2008;130: 765–770. 8. Siddiqui MT, Hornaman K, Cohen C, Nassar A. ProEx C immunocytochemistry and high-risk human papillomavirus DNA testing in papanicolaou tests with atypical squamous cell (ASC-US) cytology: correlation study with histologic biopsy. Arch Pathol Lab Med 2008;132:1648–1652. 9. Kelly D, Kincaid E, Fansler Z, Rosenthal DL, Clark DP. Detection of cervical high-grade squamous intraepithelial lesions from cytologic samples using a novel immunocytochemical assay (ProEx C). Cancer 2006;108:494–500. 10. Pinto AP, Schlecht NF, Woo TY, Crum CP, Cibas ES. Biomarker (ProEx C, p16(INK4A), and MiB-1) distinction of high-grade squamous intraepithelial lesion from its mimics. Mod Pathol 2008;21: 1067–1074. 11. Moatamed NA, Vergara-Lluri ME, Lu D, Apple SK, Kerkoutian S, Rao JY. Utility of ProEx C in the histologic evaluation of the neoplastic and nonneoplastic urothelial lesions. Hum Pathol 2013;44: 2509–2517.
Diagnostic Cytopathology DOI 10.1002/dc
EXPRESSION OF PROEX C IN UROTHELIAL CARCINOMA 12. Moatamed NA, Rao JY, Alexanian S, et al. ProEx C as an adjunct marker to improve cytological detection of urothelial carcinoma in urinary specimens. Cancer Cytopathol 2013;121:320–328. 13. Al-Sukhun S, Hussain M. Current understanding of the biology of advanced bladder cancer. Cancer 2003;15;97(Suppl 8):2064–2075. 14. Johnson TL, Kini SR. Cytologic features of metastatic transitional cell carcinoma. Diagn Cytopathol 1993;9:270–278. 15. Powers CN, Elbadwi A. Cercariform cells: a clue to the cytodiagnosis of urothelial origin of metastatic neoplasms? Diagn Cytopathol 1995; 13:15–21. 16. Rashidi B, Tongson-Ignacio JE. Melamed-Wolinska bodies in urine cytology an interesting aggregate in a degenerated urothelial cell. Diagn Cytopathol 2011;39:117. 17. Renshaw AA, Madge R, Granter SR. Intracytoplasmic eosinophilic inclusions (Melamed-Wolinska bodies). Association with metastatic
transitional cell carcinoma in pleural fluid. Acta Cytol 1997;41: 995–998. 18. Bardales RH, Pitman MB, Stanley MW, Korourian S, Suhrland MJ. Urine cytology of primary and secondary urinary bladder adenocarcinoma. Cancer 1998;84:335–343. 19. Ge Y, Mody DR, Smith D, Anton R. p16(INK4a) and ProEx C immunostains facilitate differential diagnosis of hyperchromatic crowded groups in liquid-based Papanicolaou tests with menstrual contamination. Acta Cytol 2012;56:55–61. 20. Aximu D, Azad A, Ni R, Colgan T, Nanji S. A pilot evaluation of a novel immunohistochemical assay for topoisomerase II-alpha and minichromosome maintenance protein 2 expression (ProEx C) in cervical adenocarcinoma in situ, adenocarcinoma, and benign glandular mimics. Int J Gynecol Pathol 2009;28: 114–119.
Diagnostic Cytopathology, Vol. 43, No 3
187
