Usefulness of GATA3 and p40 Immunostains in the Diagnosis of Metastatic Urothelial Carcinoma in Cytology Specimens

Transcription

1 Original Article Usefulness of GATA3 and p40 Immunostains in the Diagnosis of Metastatic Urothelial Carcinoma in Cytology Specimens Tamar C. Brandler, MD, MS 1 ; Mohamed S. Aziz, MD 1 ; Lisa M. Rosen, ScM 2 ; Tawfiqul A. Bhuiya, MD 1 ; and Oksana Yaskiv, MD 1 BACKGROUND: GATA3 (GATA-binding protein 3) expression in urothelial carcinoma (UC) and mammary carcinomas has been recently reported. However, to the authors knowledge, studies examining GATA3 staining of metastatic UC (MUC) in cytology specimens are lacking. Delta Np63 (p40) has been shown to be expressed highly selectively in squamous cell carcinomas (SCCs) but the literature concerning the expression of p40 in UC is limited and controversial. In the current study, the authors evaluated the usefulness of GATA3 and p40 in the diagnosis of MUC in cytology specimens. METHODS: Thirty-two MUC cytology cases and 44 controls (22 UC cases and 22 SCC cases) were stained for GATA3, p40, and p63 and nuclear staining intensity and the percentage of positive cells were recorded and compared. RESULTS: MUC cytology cases stained positive for GATA3, p40, and p63 in 78.13%, 80.65%, and 61.29% of cases, respectively, with moderate=strong staining intensity. MUC cases had a significantly higher percentage of GATA3 positivity compared with SCC controls (P <.001), but GATA3 positivity was not found to be significantly different from UC controls (90.91%) (P 5.28). For p40 positivity, there was no significant difference observed between MUC cases, UC controls (95.45%), and SCC controls (90.91%) (P 5.29). p63 positivity was found to be significantly lower in MUC cases compared with UC controls (95.45%) and SCC controls (95.45%) (P <.01). CONCLUSIONS: The results of the current study demonstrate that GATA3 is useful in confirming the diagnosis of MUC in cytology specimens and in distinguishing between MUC and SCC. p40 is a valuable adjunct to GATA3 in the diagnosis of MUC in cytology specimens, especially when SCC is not part of the clinical differential diagnosis. Cancer (Cancer Cytopathol) 2014;122: VC 2014 American Cancer Society. KEY WORDS: GATA3 (GATA-binding protein 3); p40; urothelial carcinoma; cytology; immunocytochemistry. INTRODUCTION The transcription factor GATA3 (GATA-binding protein 3) is a member of a zinc finger transcription factor family that plays a role in the development of hematopoietic tissue as well as the kidney, central nervous system, 1,2 skin, and mammary glands. 3 GATA3 expression in urothelial carcinoma (UC) has recently been reported in cdna array and monoclonal antibody staining. 4 To the best of our knowledge, only recently have studies begun to explore the expression of GATA3 immunostaining in low-grade and high-grade (and metastatic) UC in surgical pathology reporting differing rates of sensitivity. 2,4,5 However, to the best of our knowledge, studies examining the use of GATA3 staining in establishing metastatic UC (MUC) in cytology specimens are lacking. Corresponding author: Oksana Yaskiv, MD, Department of Pathology, Hofstra North Shore-LIJ School of Medicine, 6 Ohio Dr, Suite 202, Lake Success, NY 11042; Fax: (516) ; OYaskiv@nshs.edu 1 Department of Pathology, Hofstra North Shore-LIJ School of Medicine, Lake Success, New York; 2 Department of Biostatistics, Hofstra North Shore- LIJ School of Medicine, Lake Success, New York We thank Tanya Reid-Abreu for technical assistance. Received: February 11, 2014; Revised: March 11, 2014; Accepted: March 12, 2013 Published online April 7, 2014 in Wiley Online Library (wileyonlinelibrary.com) DOI: /cncy.21424, wileyonlinelibrary.com 468 Cancer Cytopathology June 2014

2 GATA3 and p40 in Urothelial Carcinoma/Brandler et al TABLE 1. Antibody Summary Antibody Predilute/Concentrate Titration Detection Kit Company Name GATA3 Concentrate 1:25 OptiView DAB IHC Detection Kit Ventana Medical Systems, Inc p40 Concentrate 1:2000 OptiView DAB IHC Detection Kit Ventana Medical Systems, Inc p63 Predilute Ready to use OptiView DAB IHC Detection Kit Ventana Medical Systems, Inc Abbreviations: DAB, diaminobenzidine; GATA3, GATA-binding protein 3; IHC, immunohistochemistry. Another antibody, Delta Np63 (p40), has recently been established as a sensitive and specific marker for squamous cell carcinoma (SCC) and has begun to replace p63. Although p63 has traditionally been used as a marker for UC, discrepant results have been reported with regard to the expression of its p40 isoform in UC. Koga et al have demonstrated p40 expression in low-grade UC with loss of expression in high-grade and invasive UCs. 6 Urist et al demonstrated p40 expression in patients with invasive UC. 7 Park et al stated that overexpression of p40 might contribute to the progression of bladder cancer, indicating that p40 would stain cases of higher-grade UC. 8 Brown et al reported a sensitivity of 88% for p40 in UC without addressing tumor grade. 9 p40 staining has been shown to be highly specific in SCC, in contrast with adenocarcinoma of the lung However, to the best of our knowledge, p40 staining in MUC in cytology specimens is not well documented. Therefore, we chose to study the usefulness of GATA3 antibody staining in conjunction with p40 staining in cytology specimens to help establish a diagnosis of MUC and to examine their usefulness in distinguishing MUC from SCC. MATERIALS AND METHODS The current study protocol was approved by the Hofstra North Shore-LIJ Institutional Review Board. A search was conducted in our information system for cytology cases of confirmed or suspected MUC in nongenitourinary organs between January 1, 2005 and June 16, 2013, which generated 48 potential cases. A total of 32 cases had cell blocks that were retrieved and unstained slides were cut and stained for GATA3, p63, and p40 (Table 1) at the immunohistochemistry laboratory of the study institution. A total of 22 UC bladder surgical specimens and 22 SCC surgical specimens were selected as controls. UC controls were bladder specimens obtained via transurethral resection. SCC controls included 7 SCCs of the lung, 3 SCCs of the cervix=ectocervix, 5 SCCs of the skin, 2 SCCs of the esophagus, 1 SCC of the uterine adnexa (SCC arising in ovarian teratoma), 1 SCC of the uterus, 1 SCC of the penis, 1 SCC of the lymph node (axillary), and 1 SCC of the parotid gland. MUC cases were obtained from the following: 14 from the lymph nodes (4 retroperitoneal, 3 peripancreatic, 2 abdominal, 2 paraaortic, 2 subcutaneous flank, 1 axillary, 1 mediastinal, 1 paracaval, 1 right external iliac, and 1 supraclavicular), 5 from the lung, 4 from the liver, and 1 from the neck. All MUC cases were high grade. Each cell block had a minimum of 50 cells. Three pathologists examined the slides to determine antibody positivity. Nuclear staining of GATA3, p63, and p40 were recorded as weak, moderate, or strong in intensity and were classified by percentages (< 1, 1-5, 6-20, 21-50, and > 50). Percentages < 1wereconsideredtobenegative and percentages 1wereconsideredtobepositive. Cell Block Preparation The needle from a fine-needle aspiration was rinsed in 10% buffered formalin. Standard cell block preparation was performed through refrigerated cytocentrifugation. The supernatant fluid was discarded and the pellet was hardened with 95% ethyl alcohol. The cell sediment was wrapped in filter paper and submitted in a tissue cassette for processing, embedding, and cutting in the histology laboratory. Statistical Analysis The sensitivity and specificity of GATA3 staining for MUC were calculated. Sensitivity and specificity were repeated for primary UC. The sensitivity of p63 and p40 staining for cases and controls were compared using the McNemar test. Logistic regression analysis was used to model specimen type (MUC vs SCC) as a function of each immunostain and pairs of immunostains (namely, GATA3 and p40). RESULTS MUC cytology cases (Figs. 1A-1C) were found to stain positively for GATA3, p40, and p63 in 78.13%, 80.65%, Cancer Cytopathology June

3 Original Article Figure 1. Cell block of metastatic urothelial carcinoma (A: H & E, 3 40) with (B) strong and diffuse nuclear GATA3 (GATA-binding protein 3) (3 10) and (C) p40 staining (3 10) is shown. (D) Squamous cell carcinoma surgical control is shown (H & E, 3 10). (E) GATA3-negative staining (3 10) and (F) p40 with strong diffuse nuclear staining (3 10) are shown. and 61.29% of cases, respectively, with moderate=strong staining intensity noted in 96%, 100%, and 84.21% of positive cases, respectively. MUC cases had a significantly higher percentage of GATA3 positivity compared with SCC controls (Figs. 1D-1F) (9.09%) in rare cells with weak staining (P <.001), but GATA3 positivity was not found to be significantly different from that of UC controls (90.91%) (P 5.28). For p40 positivity, there was no significant difference observed between MUC cases (80.65%), UC controls (95.45%), and SCC controls (90.91%) (P 5.29). p63 positivity was found to be significantly lower in MUC cases compared with UC controls 470 Cancer Cytopathology June 2014

4 GATA3 and p40 in Urothelial Carcinoma/Brandler et al (95.45%) (P <.01) and SCC controls (95.45%) (P <.01). Staining Intensity Among those MUC cases with a positive GATA3 stain, 4%, 32%, and 64%, respectively, had weak, moderate, and strong GATA3 staining intensity. GATA3 staining intensity was 95% strong and 5% moderate for primary UC controls and 100% weak for SCC controls that stained positive. Among those MUC cases with a positive p63 stain, 15.79%, 15.79%, and 68.42% of cases, respectively, had weak, moderate, and strong p63 staining intensity. p63 staining intensity was weak in 14.29%, moderate in 33.33%, and strong in 52.38% of primary UC controls and 9.52% weak, 52.38% moderate, and 38.10% strong for SCC controls. Among those with a positive p40 stain, 24.00% and 76.00%, respectively, of MUC cases had moderate and strong p40 staining intensity. p40 staining intensity was 4.76% moderate and 95.24% strong for primary UC controls and 5.00% moderate and 95.00% strong for SCC controls. Sensitivity and Specificity of GATA3 GATA3 sensitivity and specificity for MUC were 78.13% (95% confidence interval [95% CI], 60.03%-90.72%) and 90.91% (95% CI, 70.84%-98.88%), respectively. The sensitivity and specificity for GATA3 in primary UC were 90.91% (95% CI, 70.84%-98.88%) and 90.91% (95% CI, 70.84%-98.88%), respectively. Sensitivity of p40 For MUC, p40 had a significantly higher sensitivity (80.65%) compared with p63 (61.29%) (P <.01, McNemar test). For primary UC, the sensitivities of p63 and p40 were identical. In addition, the p63 and p40 stains were in complete agreement. Therefore, the McNemar test could not be calculated. For SCC, there was no significant difference noted with regard to the sensitivity of p40 (90.91%) and p63 (95.45%) (P 5.31, McNemar test). Logistic Regression Models Model 1 Specimens with positive GATA3 staining had an increased odds of being MUC (cytology cases) compared with SCC (odds ratio, 34.28; 95% CI, [P <.001]). Likewise, specimens with positive GATA3 staining had a decreased odds of being SCC (Table 2). TABLE 2. Logistic Regression Models Used to Predict MUC Model OR (95% CI) P Model 1: GATA ( ) <.001 Model 2: p ( ).31 Abbreviations: 95% CI, 95% confidence interval; GATA3, GATA-binding protein 3; MUC, metastatic urothelial carcinoma; OR, odds ratio. Model 2 There was no significant association noted between p40 staining and specimen type (MUC vs SCC) (P 5.31) because p40 stained diffusely and strongly in both MUC and SCC (Table 2). DISCUSSION MUC can be difficult to distinguish from other entities such as SCC based on morphology alone. MUC may demonstrate squamous features such as keratinization and intercellular bridges. It is therefore imperative that MUC be distinguished from other entities including SCC due to differences in prognosis and treatment. The results of the current study demonstrated that GATA3 is a highly sensitive and specific solution to this diagnostic difficulty. The Delta Np63 (p40) isoform has proven to be a very useful immunostain for MUC, and p40 appears to be an even more sensitive marker than the traditional p63. The results of the current study demonstrated that MUC cytology cases stained positively for GATA3 in 78.13% of specimens, with the majority of tumor cells (> 95%) staining with moderate to strong intensity. Conversely, it stained 2 of 22 SCC controls (9.09%), expressing weak intensity in rare cells. This confirms the specificity of GATA3 of 90.91% in differentiating MUC from SCC. p40 demonstrated strong staining in 80.65% and 90.91% of MUC cases and SCC controls, respectively. Therefore, it appears to be ineffective in differentiating between these 2 entities. However, in a metastatic setting, p40 as a strong surrogate antibody along with GATA3 can contribute to confirming a urothelial origin in cases in which the clinical differential diagnosis does not include entities with known GATA3 positivity. GATA3 expression has been reported in UC as well as other malignancies, including paragangliomas, Brenner tumors and Walthard cell nests, breast carcinomas, choriocarcinomas, and endodermal sinus tumors as well as skin adnexal tumors. 13,14 Some studies may therefore Cancer Cytopathology June

5 Original Article consider GATA3 to be nonspecific 13 ; however, these aforementioned entities are morphologically distinct from MUC and therefore do not pose a diagnostic dilemma. Entities that are morphologically similar to MUC such as poorly differentiated SCC of the lung and adenocarcinoma of the prostate stain poorly for GATA3, and therefore a diffuse and strong GATA3 stain can be used as a specific marker in a diagnostic panel. 13,14 To our knowledge, the current study is the first study in the English literature to systematically examine GATA3 and p40 immunostaining in MUC cytology specimens. Other studies have examined GATA3 immunostaining in low-grade, high-grade, and metastatic UC in surgical pathology, resulting in differing rates of sensitivity (67%-93%) and specificity (92%-97%). 2,4,5,13 15 Uroplakin III has been reported to be a more specific marker for UC, but its sensitivity ranges from 22% to 75% in invasive and metastatic UC. 2,4,5,13 15 In addition, GATA3 has been shown to be more sensitive than thrombomodulin (sensitivity of 63%-78%). 2,4 S100P has been shown to have a sensitivity of 71% to 78%, but low specificity. 4 Cytokeratin 7 (CK7) has been shown to have comparable sensitivity to GATA3 but, in combination with GATA3, is 100% sensitive for UC. 15 According to Liu et al, GATA3 is the most sensitive and specific marker for UCs. 3 p40 has been shown to be 96% sensitive for UC. 16 To the best of our knowledge, studies examining GATA3 staining of MUC in cytology specimens are lacking. p40 immunostaining used in MUC is even less well elucidated than GATA3. There are conflicting reports of p40 expression, with studies reporting either decreasing or increasing expression with progression of UC grade. 6 9,12,16 p40 staining has been shown to be highly specific in SCC. 9 12,16,17 However, p40 staining in MUC, particularly in cytology specimens, has not been well studied to the best of our knowledge. Therefore, we studied the usefulness of GATA3 antibody staining in conjunction with p40 staining in cytology specimens to help establish a diagnosis of MUC and to examine their usefulness in distinguishing MUC from SCC. Conversely, cases in which GATA3 is positive but p40 is negative might negate a diagnosis of UC and suggest the need to consider other GATA3-positive entities. In addition, we examined the expression of p63 because it is a more established immunostain used for staining UC. Thus, we were able to compare the standard p63 staining of MUC with the p40 isoform. The results of the current study demonstrated that for MUC, p40 was found to have significantly higher sensitivity (80.65%) compared with p63 (61.29%) (P <.01, McNemar test), and therefore p40 may be a more useful marker than p63 in panels for MUC in the future. p40 has already begun to replace p63 in immunohistochemical investigations of SCC of the lung and may be superior to p63 in UC as well. 16 The most challenging clinical paradigm is to differentiate metastatic UC to the lung from primary lung SCC and to establish the primary tumor site in lymph node metastasis in cases of known primary SCC of the lung and UC. To the best of our knowledge, current immunohistochemistry panels including CK7, CK20, and p63 are inadequate to achieve this goal. Other available antibodies include uroplakin III and thrombomodulin, which failed to gain wide acceptance due to their limited sensitivity and specificity. Cytomorphology alone, and even in combination with the standard panel of CK7, CK20, and p63, is usually not sufficient for a definitive diagnosis. Therefore, we often must rely on the clinical picture and the patient history when considering our diagnosis. The addition of GATA3 and=or p40 can help take us one step closer toward a conclusive diagnosis. Due to the high specificity of GATA3 in distinguishing MUC from SCC, GATA3 is a valuable adjunct in differentiating between the 2 when clinical history and morphology are ambiguous. Strengths of the current study include that to our knowledge this is the first study of GATA3 and p40 in MUC in cytopathology. In addition, the results of the current study are helpful in that they demonstrated that GATA3 and p40 are both stains of diagnostic value in UC in cytology. Study limitations include a small sample size and that the nature of the study was retrospective. The results of the current study demonstrated that GATA3 can be very valuable in confirming the diagnosis of MUC in cytology specimens and in distinguishing between MUC and SCC. We conclude that p40 markers can be very useful in future panels with GATA3 to distinguish MUC from other types of carcinoma. FUNDING SUPPORT Supported by departmental funds. CONFLICT OF INTEREST DISCLOSURES The authors made no disclosures. 472 Cancer Cytopathology June 2014

6 GATA3 and p40 in Urothelial Carcinoma/Brandler et al REFERENCES 1. So JS, Epstein JI. GATA3 expression in paragangliomas: a pitfall potentially leading to misdiagnosis of urothelial carcinoma. Mod Pathol. 2013;26: Chang A, Amin A, Gabrielson E, et al. Utility of GATA3 immunohistochemistry in differentiating urothelial carcinoma from prostate adenocarcinoma and squamous cell carcinomas of the uterine cervix, anus, and lung. Am J Surg Pathol. 2012;36: Liu H, Shi J, Wilkerson ML, Lin F. Immunohistochemical evaluation of GATA3 expression in tumors and normal tissues: a useful immunomarker for breast and urothelial carcinomas. Am J Clin Pathol. 2012;138: Higgins JP, Kaygusuz G, Wang L, et al. Placental S100 (S100P) and GATA3: markers for transitional epithelium and urothelial carcinoma discovered by complementary DNA microarray. Am J Surg Pathol. 2007;31: Gulmann C, Paner GP, Parakh RS, et al. Immunohistochemical profile to distinguish urothelial from squamous differentiation in carcinomas of urothelial tract. Hum Pathol. 2013;44: Koga F, Kawakami S, Kumagai J, et al. Impaired Delta Np63 expression associates with reduced beta-catenin and aggressive phenotypes of urothelial neoplasms. Br J Cancer. 2003;88: Urist MJ, Di Como CJ, Lu ML, et al. Loss of p63 expression is associated with tumor progression in bladder cancer. Am J Pathol. 2002;161: Park BJ, Lee SJ, Kim JI, et al. Frequent alteration of p63 expression in human primary bladder carcinomas. Cancer Res. 2000;60: Brown AF, Sirohi D, Fukuoka J, et al. Tissue-preserving antibody cocktails to differentiate primary squamous cell carcinoma, adenocarcinoma, and small cell carcinoma of lung. Arch Pathol Lab Med. 2013;137: Bishop JA, Teruya-Feldstein J, Westra WH, Pelosi G, Travis WD, Rekhtman N. p40 (dnp63) is superior to p63 for the diagnosis of pulmonary squamous cell carcinoma. Mod Pathol. 2012;25: Nonaka D. A study of dnp63 expression in lung non-small cell carcinomas. Am J Surg Pathol. 2012;36: Nobre AR, Albergaria A, Schmitt F. p40: a p63 isoform useful for lung cancer diagnosis-a review of the physiological and pathological role of p63. Acta Cytol. 2013;57: Miettinen M, McCue PA, Sarlomo-Rikala M, et al. GATA3: a multispecific but potentially useful marker in surgical pathology: a systematic analysis of 2500 epithelial and nonepithelial tumors. Am J Surg Pathol. 2014;38: Ordonez NG. Value of GATA3 immunostaining in tumor diagnosis: a review. Adv Anat Pathol. 2013;20: Zhao L, Antic T, Witten D, et al. Is GATA3 expression maintained in regional metastases?: a study of paired primary and metastatic urothelial carcinomas. Am J Surg Pathol. 2013;37: Gailey MP, Bellizzi AM. Immunohistochemistry for the novel markers glypican 3, PAX8, and p40 (dnp63) in squamous cell and urothelial carcinoma. Am J Clin Pathol. 2013;140: Pelosi G, Rossi G, Cavazza A, et al. dnp63 (p40) distribution inside lung cancer: a driver biomarker approach to tumor characterization. Int J Surg Pathol. 2013;21: Cancer Cytopathology June