Metastatic cancers can be difficult to diagnose in a. CDH17 Is a More Sensitive Marker for Gastric Adenocarcinoma Than CK20 and CDX2

Transcription

1 CDH17 Is a More Sensitive Marker for Gastric Adenocarcinoma Than CK20 and CDX2 David Altree-Tacha, PhD; Jillian Tyrrell, PhD; Thomas Haas, DO Context. CDH17, which is expressed in the intestinal epithelium, is a novel oncogene involved in tumor invasion and metastasis. A panel consisting of cytokeratin (CK) 7, CD20, and CDX2 antibodies is typically used to diagnose gastrointestinal adenocarcinomas. However, studies have shown that CDH17 is a highly specific marker for gastrointestinal adenocarcinoma and may be important in clinical diagnosis. Objective. To evaluate the sensitivity and specificity of CDH17, CK20, and CDX2 antibodies in neoplastic tissues, with emphasis on colon, stomach, and esophageal gastrointestinal lineage. Design. Immunohistochemistry was performed with CDH17, CK20, and CDX2 antibodies on formalin-fixed, paraffin-embedded tissue microarrays from normal (n ¼ 26) and neoplastic (n ¼ 884) tissues. Results. CDH17 immunostaining was positive in 97.3% (145 of 149) of colon adenocarcinomas, whereas CK20 and CDX2 stained positively in 88.6% (132 of 149) and 93.3% (139 of 149), respectively. In metastatic colon cancers, CDH17, CK20, and CDX2 positive staining was observed in 90.6% (29 of 32), 59.4% (19 of 32), and 81.3% (26 of 32) of cases, respectively. In stomach adenocarcinomas, CDH17 positively stained 64.0% (112 of 175) of tissues, compared to CK20 and CDX2, where staining was observed in only 24.6% (43 of 175) and 46.9% (82 of 175), respectively. In esophageal adenocarcinomas, CDH17, CK20, and CDX2 stained 38.7% (12 of 31), 25.8% (8 of 31), and 29% (9 of 31) of specimens, respectively. Low or no expression was observed in other neoplastic tissues, except pancreatic cancers, where CDH17 displayed higher expression than CK20 and CDX2. Conclusions. CDH17 is a specific and more sensitive marker in the gastrointestinal tract than CK20 and CDX2. CDH17 may be especially valuable when gastrointestinal tumors are suspected in cancers of unknown primary. (Arch Pathol Lab Med. 2017;141: ; doi: / arpa oa) Metastatic cancers can be difficult to diagnose in a clinical setting with most being carcinomas. In the past, a panel consisting of cytokeratin (CK) 7, CK20, and CDX2 antibodies has been used to assess gastrointestinal (GI) tumors of unknown primary. 1 5 Expression of CK7 is observed in a wide variety of carcinomas, thus, it is an important marker in diagnosis. 5 CK20 expression is specific to certain types of cancer and is typically used in combination with CK7 to distinguish different types of tumors. CDX2 is a homeobox gene that encodes an intestine-specific transcription factor and is expressed in the nuclei of epithelial cells throughout the GI tract. Immunohistochemical expression of CDX2 protein in primary and metastatic colorectal carcinomas has been previously documented. 2,3 CK7 along with CK20 or CDX2 Accepted for publication May 4, Published as an Early Online Release October 20, From Chief Scientific Officer (Dr Altree-Tacha) and the Department of Research and Development (Dr Tyrrell), Biocare Medical, Concord, California; and the Department of Pathology, Mercy Health System, Janesville, Wisconsin (Dr Haas). Dr Altree-Tacha is a minor shareholder and chief scientific officer of Biocare Medical. Dr Tyrrell is an employee of Biocare Medical. Dr Haas is a consulting pathologist for Biocare Medical. The authors have no other relevant financial interest in the products or companies described in this article. Reprints: David Altree-Tacha, PhD, Biocare Medical, 4040 Pike Lane, Concord, CA ( dtacha@biocare.net). markers constitute an important panel for the diagnosis of GI tumors versus tumors of unknown origin. 1 5 However, in difficult cases, expression patterns may vary and the presence or absence of these markers may not lead to a definitive diagnosis. 1,4 Cadherin 17 (CDH17), also called liver-intestine cadherin, is a cell-adhesion protein that is transcriptionally regulated by CDX2, and is involved in tumor invasion and metastasis. 6,7 CDH17 is a novel diagnostic marker for GI tract carcinoma with greater sensitivity than CDX2 in colon cancers. 7,8 Recently, a mouse monoclonal CDH17 (clone IH3) antibody has been developed to aid pathologists in interpreting GI adenocarcinomas, especially in cases of unknown primary origin. There have been conflicting results with CDH17 antibodies, even when using the same designated clone as presented in this study. 6 8 This study will present an immunohistochemical evaluation in a sideby-side comparison with antibodies CDH17, CK20, and CDX2 in various neoplastic tissues, with emphasis on GI adenocarcinomas. Our results confirm that CDH17 expression analysis may be of significant value in the diagnosis of GI tumors and those of unknown origin. DESIGN CDH17 mouse monoclonal (1:4000; clone IH3, Biocare Medical, Concord, California), CK20 (1:10 000; Ks20.8, Biocare Medical), and CDX2 (1:250, CDX2-88; BioGenex, Fremont, California) antibodies were evaluated with immu- 144 Arch Pathol Lab Med Vol 141, January 2017 CDH17 Is a Specific Marker in GI Adenocarcinoma Altree-Tacha et al

2 Table 1. Expression in Normal Tissue Types (n ¼ 26) a Tissue Type CDH17 CK20 CDX2 Adrenal gland Appendix þ þ þ Bladder Bone marrow Breast Colon þ þ þ Cervix b Esophagus Heart Kidney Liver Lung Lymph node Ovary Pancreas þ Pituitary Skeletal muscle Skin Small intestine þ þ þ Smooth muscle Spleen Stomach þ Testis Tonsil Thymus Uterus Abbreviation: CK, cytokeratin. a Three of each tissue type were tested. b Squamous components, specifically. nohistochemistry for specificity and sensitivity on formalinfixed, paraffin-embedded tissue microarrays (TMAs) consisting of normal tissues (n ¼ 26), colon adenocarcinoma (n ¼ 149), stomach adenocarcinoma (n ¼ 175), esophageal cancer (n ¼ 54), appendix cancer (n ¼ 5), pancreatic cancer (n ¼ 57), hepatocellular carcinoma (n ¼ 57), cholangiocarcinoma (n ¼ 12), ovarian cancer (n ¼ 60), endometrial adenocarcinoma (n ¼ 7), lung cancer (n ¼ 78), prostate adenocarcinoma (n ¼ 20), breast cancer (n ¼ 73), bladder cancer (n ¼ 63), renal cell carcinoma (n ¼ 10), thyroid cancer (n ¼ 12), seminoma (n ¼ 23), brain cancer (n ¼ 12), melanoma (n ¼ 6), and lymphoma (n ¼ 11). Antibody titers were optimized and applied for 30 minutes on TMA specimens, followed by a polymer-based horseradish peroxidase detection system (MACH 4 HRP-Polymer, Biocare Medical). Staining was visualized with 3,3 0 -diaminobenzidine chromogen (Betazoid DAB Chromogen Kit, Biocare Medical). To compare positive staining of CDH17 in tissues with CK20 and CDX2, statistical analyses were performed with GraphPad Prism version 6.0 (GraphPad Software, La Jolla, California). P values were determined by using the v 2 test, and values less than or equal to 0.05 were considered statistically significant. RESULTS Results of staining for expression analysis of CDH17, CK20, and CDX2 are summarized in Tables 1 through 4. In normal tissues, CDH17 was expressed in colon mucosa, small-intestine mucosa, appendix, and pancreatic ducts, but not acinar cells or islets. CK20 was not found in normal pancreatic ducts, but was expressed in colon, small intestine, appendix, and stomach. CDX2 was expressed in normal colon, small intestine, and appendix (Table 1). In normal Figure 1. Expression patterns in normal colon. CDH17 (A) and CDX2 (B) stained full-length colonic crypts. C, CK20 stained the upper half of colonic crypts. Bottom half of crypts show incomplete or no staining with CK20 (arrow) (original magnification 320 [A through C]). colon, CDH17 and CDX2 staining patterns were observed in full-length colonic crypts, whereas CK20 only stained the upper half of colonic crypts (Figure 1, A through C). CDH17 was expressed in 97.3% (145 of 149) of colon adenocarcinomas, whereas CK20 and CDX2 were found in 88.6% (132 of 149) and 93.3% (139 of 149), respectively (Figure 2, A through F; Table 2). Specifically, in primary colon cancer, CDH17, CK20, and CDX2 immunostaining was positive in 99.1% (116 of 117), 95.7% (112 of 117), and 96.6% (113 of 117) of cases, respectively (Table 3). In certain cases, CDH17 Arch Pathol Lab Med Vol 141, January 2017 CDH17 Is a Specific Marker in GI Adenocarcinoma Altree-Tacha et al 145

3 Table 2. Neoplastic Tissues (n ¼ 884) Cancer Type CDH17, % (No.) CK20, % (No.) CDX2, % (No.) Colon adenocarcinoma 97.3 (145/149) 88.6 (132/149) a 93.3 (139/149) b Stomach adenocarcinoma 64.0 (112/175) 24.6 (43/175) c 46.9 (82/175) a Esophageal cancer (n ¼ 54) Esophageal adenocarcinoma 38.7 (12/31) 25.8 (8/31) b 29 (9/31) b Esophageal squamous cell carcinoma 0 (0/23) 0 (0/23) 0 (0/23) Appendiceal cancer (n ¼ 5) Adenocarcinoma 2/2 2/2 2/2 Undifferentiated carcinoma 0/2 0/2 0/2 Pancreatic cancer (n ¼ 57) Pancreatic ductal adenocarcinoma 39.3 (11/28) 10.7 (3/28) a 0 (0/28) c Pancreatic adenocarcinoma 24.1 (7/29) 13.8 (4/29) b 6.9 (2/29) b Hepatocellular carcinoma 1.8 (1/57) 7 (4/57) 0 (0/57) Cholangiocarcinoma 33.3 (4/12) 33.3 (4/12) 8.3 (1/12) Ovarian cancer (n ¼ 60) Serous papillary cystadenocarcinoma 6.4 (3/47) 8.5 (4/47) 4.4 (2/47) Endometrioid adenocarcinoma 28.6 (2/7) 28.6 (2/7) 14.3 (1/7) Mucinous adenocarcinoma 50 (3/6) 50 (3/6) 66.7 (4/6) Endometrial adenocarcinoma 28.6 (2/7) 57.1 (4/7) 0 (0/7) Lung cancer (n ¼ 78) Adenocarcinoma 11.1 (4/36) 5.6 (2/36) 2.8 (1/36) Squamous cell carcinoma 0 (0/29) 0 (0/29) 0 (0/29) Small cell carcinoma 0 (0/5) 0 (0/5) 0 (0/5) Large cell carcinoma 0 (0/5) 0 (0/5) 0 (0/5) Neuroendocrine carcinoma 0 (0/3) 0 (0/3) 0 (0/3) Prostate adenocarcinoma 0 (0/20) 0 (0/20) 0 (0/20) Breast cancer (infiltrating ductal) 0 (0/73) 2.7 (2/73) 0 (0/73) Bladder cancer (n ¼ 63) Urothelial carcinoma 0 (0/61) 52.5 (32/61) 4.9 (3/61) Bladder adenocarcinoma 100 (2/2) 100 (2/2) (0/2) Clear cell renal cell carcinoma 0 (0/10) 0 (0/10) 0 (0/10) Thyroid cancer (n ¼ 12) Papillary carcinoma 0 (0/10) 0 (0/10) 0 (0/10) Follicular carcinoma 0 (0/2) 0 (0/2) 0 (0/2) Seminoma 0 (0/23) 0 (0/23) 0 (0/23) Brain cancer (astrocytoma) 0 (0/12) 0 (0/12) 0 (0/12) Melanoma (classic) 0 (0/6) 0 (0/6) 0 (0/6) Lymphoma (n ¼ 11) B-cell lymphoma 0 (0/8) 0 (0/8) 0 (0/8) T-cell lymphoma 0 (0/3) 0 (0/3) 0 (0/3) Abbreviation: CK, cytokeratin. a P,.05; colon adenocarcinoma CK20: P ¼.003; stomach adenocarcinoma CDX2: P ¼.001; pancreatic ductal adenocarcinoma CK20: P ¼.01. b P..05; colon adenocarcinoma CDX2: P ¼.10; esophageal adenocarcinoma CK20: P ¼.28, CDX2: P ¼.42; pancreatic adenocarcinoma CK20: P ¼.32, CDX2: P ¼.07. c P,.001. Table 3. Primary Colon Cancer Versus Metastasis Colon Cancer CDH17, % (No.) CK20, % (No.) CDX2, % (No.) Primary 99.1 (116/117) 95.7 (112/117) a 96.6 (113/117) a Metastasis into lymph node b 90.6 (29/32) 59.4 (19/32) c 81.3 (26/32) a Abbreviation: CK, cytokeratin. a P..05; primary CK20: P ¼.10, CDX2: P ¼.18; metastasis into lymph node CDX2: P ¼.15. b The origin of metastatic carcinomas was determined by a board-certified pathologist before receiving the tissue for testing. c P,.05; metastasis into lymph node CK20: P ¼.004. Table 4. Primary Stomach Adenocarcinoma Versus Metastasis Stomach Cancer CDH17, % (No.) CK20, % (No.) CDX2, % (No.) Primary 63.3 (88/139) 23 (32/139) a 46 (64/139) b Metastasis c 66.7 (24/36) 30.5 (11/36) b 50 (18/36) d Abbreviation: CK, cytokeratin. a P,.001. b P,.05; primary CDX2: P ¼.004; metastasis CK20: P ¼.002. c The origin of metastatic carcinomas was determined by a board-certified pathologist before receiving the tissue for testing. d P..05; metastasis CDX2: P ¼ Arch Pathol Lab Med Vol 141, January 2017 CDH17 Is a Specific Marker in GI Adenocarcinoma Altree-Tacha et al

4 Figure 2. Expression in primary colon adenocarcinoma. A through C, Well-differentiated adenocarcinoma stained with CDH17 (A), CK20 (B), and CDX2 (C). D through F, Moderately differentiated adenocarcinoma stained with CDH17 (D), CK20 (E), and CDX2 (F). In some instances, CDH17 stained a higher percentage of tumor cells than CK20 and CDX2 (compare [D] with [E] and [F]) (original magnification 320 [A through F]). stained a higher percentage of tumor cells than CK20 and CDX2 (Figure 2, D through F). In metastatic colon cancers, CDH17 was a more sensitive marker than CK20 and CDX2, and positively stained 90.6% (29 of 32) of specimens, compared to 59.4% (19 of 32) and 81.3% (26 of 32), respectively (Figure 3, A through F; Table 3). In stomach adenocarcinoma, CDH17, CK20, and CDX2 positive staining was observed in 64% (112 of 175), 24.6% (43 of 175), and 46.9% (82 of 175) of specimens, respectively (Figure 4, A through F; Table 2). In primary stomach adenocarcinoma, CDH17, CK20, and CDX2 were expressed in 63.3% (88 of 139), 23% (32 of 139), and 46% (64 of 139) of tissues, respectively (Table 4). CDH17, CK20, and CDX2 also stained metastatic stomach adenocarcinoma (66.7% [24 of 36], 30.5% [11 of 36], and 50% [18 of 36], respectively). In esophageal adenocarcinoma, CDH17, CK20, and CDX2 immunostaining was positive in 38.7% (12 of 31), 25.8% (8 of 31), and 29% (9 of 31) of cases, respectively (Table 2). In esophageal squamous cell carcinoma (n ¼ 23), all 3 markers showed negativity. CDH17 stained 31.6% (18 of 57) of pancreatic cancers, compared to CK20 and CDX2, which showed positivity in 12.3%, (7 of 57) and 3.5% (2 of 57) of cases, respectively (Table 2). CDH17 and CK20 were expressed in pancreatic ductal adenocarcinoma (39.3% [11 of 28] and 10.7% [3 of 28]), while CDX2 was not detected (0 of 28). In pancreatic adenocarcinoma, CDH17, CK20, and CDX2 were expressed in 24.1% (7 of 29), 13.8% (4 of 29), and 6.9% (2 of 29) of tissues, respectively. In ovarian cancers, including serous papillary, endometrioid, and mucinous adenocarcinoma, CDH17 stained 13.3% (8 of 60) of cases, similar to CK20 and CDX2, which stained 15% (9 of 60) and 11.6% (7 of 60), respectively (Table 2). In lung adenocarcinoma, CDH17 positively stained 11.1% of cases and CDX2 and CK20 stained 5.6% and 2.8%, respectively (Table 2). All other lung cancer phenotypes, including squamous cell carcinoma, neuroendocrine carcinoma, large cell carcinoma, and small cell carcinoma, were negative for all 3 markers. CDH17 staining was positive in only 1 case of liver cancer, which was poorly differentiated, of 57 cases. To confirm the diagnosis in the 1 positive case, we stained with hepatocyte-specific antigen and arginase-1 and both antibodies showed negativity. Therefore, we could not confirm that this case was hepatocellular carcinoma. In cholangiocarcinoma, CDH17 and CD20 immunostaining was positive in 4 of 12 cases, and CDX2 was positive in 1 of 12 cases (Table 2). CDH17 showed negativity in all other cancer types tested (Table 2). Other staining results in neoplastic tissues are further summarized in Table 2. DISCUSSION In this study, we compared expression of CDH17, CK20, and CDX2 in colon, stomach, and esophageal adenocarcinomas, coupled with a large and comprehensive study of other various normal and neoplastic tissues. CDH17 was expressed in a higher percentage (97.3%) of colon adenocarcinomas than CK20 (88.6%, P ¼.003) and CDX2 (93.3%, P ¼.10). Similar results were observed in stomach adenocarcinoma, where CDH17 was expressed in 64.0% of cases, and CK20 and CDX2 were expressed in only 24.6% (P,.001) and 46.9% (P ¼.001), respectively. CDH17- Arch Pathol Lab Med Vol 141, January 2017 CDH17 Is a Specific Marker in GI Adenocarcinoma Altree-Tacha et al 147

5 Figure 3. Staining results in metastatic colon adenocarcinoma. A and D, Strong, positive staining was observed in a high percentage of specimens with CDH17. B, Focal staining was observed in CK20-positive tissue; and in specimens considered negative, CK20 was completely absent (E). Representative negative (C) and moderate positive (F) staining for CDX2 (original magnification 320 [A through F]). positive staining was also significantly higher in pancreatic cancers (31.6%, 18 of 57) than CK20 (12.3%, 7 of 57; P ¼.01) and CDX2 (3.5%, 2 of 57; P,.001). Expression of all 3 markers was similar in esophageal cancers. In most other neoplastic tissues, expression of all 3 markers was low or absent. In sum, CDH17 exhibited higher sensitivity in GI cancers (67%) than CK20 (44%) and CDX2 (53%). CDH17 and CDX2 showed similar specificity (96% and 97%, respectively) but were both more specific than CK20 (88%). Our results are consistent with those reported by Panarelli et al 7 in which 100% (161 of 161) of colon cancers were positive for expression of CDH17. Lin et al 8 showed CDH17 expression in 98% (123 of 125) of colon adenocarcinomas and 13% (26 of 198) of lung adenocarcinomas, which was almost identical to our findings of 97.3% and 12.1%, respectively. Their study also included other cancer types not included in our study, such as angiosarcoma, gastrointestinal stromal tumors, germ-cell tumors, pheochromocytoma, and mesothelioma. These cancers were all negative for CDH17, thus demonstrating high specificity. Additionally, our results confirm CDH17 expression in 50% of ovarian mucinous carcinomas, similarly to results obtained by Panarelli et al, 7 where expression was observed in 53% (8 of 15) of ovarian mucinous carcinomas. Su et al 6 also observed CDH17 (IH3) expression in 96% (46 of 48) of colon adenocarcinomas. However, their study showed that CDH17 was not expressed in lung adenocarcinoma (0 of 39), serous ovarian carcinoma (0 of 55), or in ovarian mucinous carcinoma (0 of 30). In contrast, our results indicate low expression in lung adenocarcinoma and in serous papillary cystadenocarcinoma cancers (Table 2). The discrepancy between findings could be due to low expression levels of CDH17 in these types of cancer or variation in tissue sampling. However, we cannot explain the discrepancy in ovarian mucinous carcinoma, as the same CDH17 designated clone was used in all 3 studies. 6,7 The overall staining for CDH17 in stomach adenocarcinoma cases was significantly higher than for CK20 (P,.001) and CDX2 (P ¼.004) (Table 2). Of the 175 cases of stomach adenocarcinoma, 97 cases were poorly differentiated and 36 cases were metastatic cancers (data not shown). In cases of metastatic stomach adenocarcinoma, CDH17 was expressed in a higher percentage of tissues than CK20 (P ¼.002) and CDX2 (P ¼.15) (Table 4). Su et al 6 demonstrated CDH17 expression in 56% (25 of 45) of gastric cancers. Fan et al 9 demonstrated nuclear CDX2 expression in 36.7% (40 of 107) of gastric cancers, and Kim et al 10 showed CK20 expression in 30% of gastric cancers (n ¼ 329). 10 In other neoplastic tissues such as lung and ovarian cancer, we observed low staining percentages of CDH17, CK20, and CDX2 expression (Table 2). However, it is well documented that CK20 and CDX2 are expressed in ovarian mucinous tumors. 11,12 In this study, CDH17, CK20, and CDX2 were expressed in 3 of 6 cases, 3 of 6 cases, and 4 of 6 cases, respectively, consistent with previous reports. In endometrial adenocarcinoma, CDH17 stain showed positivity in 2 of 7 cases, while CDX2 showed negativity. In a larger study, Panarelli et al 7 evaluated 58 endometrial carcinomas of which 7% (4 of 58) were positive for 148 Arch Pathol Lab Med Vol 141, January 2017 CDH17 Is a Specific Marker in GI Adenocarcinoma Altree-Tacha et al

6 Figure 4. Representative staining results in stomach adenocarcinoma. A, CDH17. B, CK20. C, Negative staining for CDX2. D, CDH17-positive staining at higher magnification. E, Negative staining for CK20. F, CDX2 (original magnifications 310 [A through C] and 320 [D through F]). CDH17 and 5% (3 of 58) were positive for CDX2; however, staining for CDX2 was mainly limited to squamous morules. In lung cancers, CDH17 was expressed in 11.1% (4 of 36) of lung adenocarcinomas. Only well-differentiated or moderately differentiated lung adenocarcinomas stained positively with CDH17; and in those cases, lung adenocarcinoma was confirmed by staining with thyroid transcription factor 1 (TTF-1) and napsin A (data not shown). CK20 was expressed in 2 cases of poorly differentiated lung adenocarcinomas and CDX2 was expressed in 1 well-differentiated lung adenocarcinoma. In bladder cancers, all cases were previously diagnosed as urothelial carcinoma or bladder adenocarcinoma. In urothelial carcinomas, CDH17 expression was absent, while CK20 and CDX2 stained 52.5% (32 of 61) and 4.9% (3 of 61) of cases, respectively. CDH17 and CK20 were expressed in 2 of 2 bladder adenocarcinomas and CDX2 showed negativity (0 of 2). Together, our results are consistent with the findings from a study by Rao et al 13 in which CDH17 was expressed in 24 of 26 bladder adenocarcinomas and 0 of 11 urothelial carcinomas. The authors concluded that CDH17 is a specific marker for primary bladder adenocarcinoma, distinguishing it from urothelial carcinoma with glandular differentiation. While this study did not correlate CDH17 with prognosis or patient survival, it should be noted that other studies have shown prognostic value for CDH17 in colon and in gastric adenocarcinoma. CDH17 was expressed in 97.3% of colon adenocarcinomas (Table 2); however, we did observe a wide range of high and low expression of CDH17 (data not shown). CONCLUSIONS In conclusion, CDH17 is a highly sensitive and specific marker for colon and stomach adenocarcinoma in routine immunohistochemistry. In comparison to CK20 and CDX2, CDH17 was a more sensitive marker. The use of CDH17 þ tumors in combination with CK7 tumors should improve overall specificity, especially when colon or stomach adenocarcinomas are suspected in cancers of unknown primary. CDH17 may also be helpful in cases that are negative for CK7, CDX2, and CK20. References 1. Park SY, Kim HS, Hong EK, et al. Expression of cytokeratins 7 and 20 in primary carcinomas of the stomach and colorectum and their value in the differential diagnosis of metastatic carcinomas to the ovary. Hum Pathol. 2002; 33(11): Werling RW, Yaziji H, Bacchi CE, et al. CDX2, a highly sensitive and specific marker of adenocarcinomas of intestinal origin: an immunohistochemical survey of 476 primary and metastatic carcinomas. Am J Surg Pathol. 2003; 27(3): Bayrak R, Haltas H, Yenidunya S, et al. The value of CDX2 and cytokeratins 7 and 20 expression in differentiating colorectal adenocarcinomas from extraintestinal gastrointestinal adenocarcinomas: cytokeratin 7-/20þ phenotype is more specific than CDX2 antibody. Diagn Pathol. 2012;7:9. 4. Bayrak R, Yenidünya S, Haltas H. Cytokeratin 7 and cytokeratin 20 expression in colorectal adenocarcinomas. Pathol Res Pract. 2011;207(3): Chu P, Wu E, Weiss LM. Cytokeratin 7 and cytokeratin 20 expression in epithelial neoplasms: a survey of 435 cases. Mod Pathol. 2000;13(9): Su MC, Yuan RH, Lin CY, et al. Cadherin-17 is a useful diagnostic marker for adenocarcinomas of the digestive system. Mod Pathol. 2008;21(11): Arch Pathol Lab Med Vol 141, January 2017 CDH17 Is a Specific Marker in GI Adenocarcinoma Altree-Tacha et al 149

7 7. Panarelli NC, Yantiss RK, Yeh MM, et al. Tissue-specific cadherin CDH17 is a useful marker of gastrointestinal adenocarcinomas with higher sensitivity than CDX2. Am J Clin Pathol. 2012;138(2): Lin F, Shi J, Zhu S, Chen Z, et al. Cadherin-17 and SATB2 are sensitive and specific immunomarkers for medullary carcinoma of the large intestine. Arch Pathol Lab Med. 2014;138(8): Fan Z, Li J, Dong B, et al. Expression of Cdx2 and hepatocyte antigen in gastric carcinoma: correlation with histologic type and implications for prognosis. Clin Cancer Res. 2005;11(17): Kim MA, Lee HS, Yang HK, et al. Cytokeratin expression profile in gastric carcinomas. Hum Pathol. 2004;35(5): Vang R, Gown AM, Wu LS, et al. Immunohistochemical expression of CDX2 in primary ovarian mucinous tumors and metastatic mucinous carcinomas involving the ovary: comparison with CK20 and correlation with coordinate expression of CK7. Mod Pathol. 2006;19(11): Shin JH, Bae JH, Lee A, et al. CK7, CK20, CDX2 and MUC2 immunohistochemical staining used to distinguish metastatic colorectal carcinoma involving ovary from primary ovarian mucinous adenocarcinoma. Jpn J Clin Oncol. 2010;40(3): Rao Q, Williamson SR, Lopez-Beltran A, et al. Distinguishing primary adenocarcinoma of the urinary bladder from secondary involvement by colorectal adenocarcinoma: extended immunohistochemical profiles emphasizing novel markers.mod Pathol. 201;26(5): Kwak JM, Min BW, Lee JH, et al. The prognostic significance of E-cadherin and liver intestine-cadherin expression in colorectal cancer. Dis Colon Rectum. 2007;50(11): Qiu HB, Zhang LY, Ren C, et al. Targeting CDH17 suppresses tumor progression in gastric cancer by downregulating Wnt/b-catenin signaling. PLoS One. 2013;8(3):e Huang LP, Yu YH, Sheng C, et al. Up-regulation of cadherin 17 and downregulation of homeodomain protein CDX2 correlate with tumor progression and unfavorable prognosis in epithelial ovarian cancer. Int J Gynecol Cancer. 2012; 22(7): Arch Pathol Lab Med Vol 141, January 2017 CDH17 Is a Specific Marker in GI Adenocarcinoma Altree-Tacha et al

8 Template for Reporting Results of Biomarker Testing of Specimens From Patients With Thyroid Carcinoma Simon Chiosea, MD; Sylvia L. Asa, MD; Michael A. Berman, MD; Sally E. Carty, MD; Louanne Currence, RHIT, CTR; Steven Hodak, MD; Yuri E. Nikiforov, MD, PhD; Mary S. Richardson, MD, DDS; Raja R. Seethala, MD; Lynette M. Sholl, MD; Lester D. R. Thompson, MD; Bruce M. Wenig, MD; Frank Worden, MD; for the Members of the Cancer Biomarker Reporting Committee, College of American Pathologists ORIGINAL ARTICLES SP174, NRAS Q61R Mutant-Specific Antibody, Cross-Reacts With KRAS Q61R Mutant Protein in Colorectal Carcinoma Jerzy Lasota, MD; Artur Kowalik, PhD; Anna Felisiak-Golabek, PhD; Shingo Inaguma, MD; Zeng-Feng Wang, PhD; Liliana Pięciak, MS; Sebastian Zięba, MS; Rafał Pęksa, MD; Janusz Kopczynski, MD; Krzysztof Okoń, MD; Piotr Waloszczyk, MD; Stanislaw Gozdz, MD; Wojciech Biernat, MD; Markku Miettinen, MD New-Generation Thromboelastography: Comprehensive Evaluation of Citrated and Heparinized Blood Sample Storage Effect on Clot-Forming Variables Joao D. Dias, PhD; Elaine I. Haney, BSc, MT; Blesy A. Mathew, MSc; Carlos G. Lopez-Espina, MSc; Adrian W. Orr, BSc, MBA; Mark A. Popovsky, MD Characteristics of a Breast Pathology Consultation Practice Ellen G. East, MD; Lili Zhao, PhD; Judy C. Pang, MD; Julie M. Jorns, MD REVIEW ARTICLE The Effects of Computerized Clinical Decision Support Systems on Laboratory Test Ordering: A Systematic Review Nicolas Delvaux, MD; Katrien Van Thienen, MD; Annemie Heselmans, PhD; Stijn Van de Velde, PhD; Dirk Ramaekers, MD, PhD; Bert Aertgeerts, MD, PhD RESIDENT SHORT REVIEWS Follicular Dendritic Cell Sarcoma Tiffany Chen, MD; Purva Gopal, MD, MS Acquired Cystic Disease Associated Renal Cell Carcinoma: Review of Pathogenesis, Morphology, Ancillary Tests, and Clinical Features Michelle Foshat, MD; Eduardo Eyzaguirre, MD ERRATUM An article in the January 2017 issue of the Archives (Altree-Tacha D, Tyrrell J, Haas T. CDH17 Is a More Sensitive Marker for Gastric Adenocarcinoma Than CK20 and CDX2. Arch Pathol Lab Med. 2017;141[1]: ; doi: /arpa OA) incorrectly refers to the use of a CD20 antibody to diagnose gastrointestinal adenocarcinomas. In the second sentence of the Context section of the abstract (page 144) and in the last paragraph of the Results section (page 147), both instances of CD20 should have been shown as CK20. The sentence in the Context section of the abstract should have read, A panel consisting of cytokeratin (CK) 7, CK20, and CDX2 antibodies is typically used to diagnose gastrointestinal adenocarcinomas. Similarly, the sentence in the Results section should have read, In cholangiocarcinoma, CDH17 and CK20 immunostaining was positive in 4 of 12 cases, and CDX2 was positive in 1 of 12 cases (Table 2). 484 Arch Pathol Lab Med Vol 141, April 2017