Fax: 0030 210 9480375
CK19 and CD10 expression in pancreatic neuroendocrine tumors diagnosed by endoscopic ultrasound-guided fine-needle aspiration cytology
Article first published online: 5 OCT 2009
Copyright © 2009 American Cancer Society
Volume 117, Issue 6, pages 516–521, 25 December 2009
How to Cite
Salla, C., Konstantinou, P. and Chatzipantelis, P. (2009), CK19 and CD10 expression in pancreatic neuroendocrine tumors diagnosed by endoscopic ultrasound-guided fine-needle aspiration cytology. Cancer Cytopathology, 117: 516–521. doi: 10.1002/cncy.20048
- Issue published online: 11 DEC 2009
- Article first published online: 5 OCT 2009
- Manuscript Accepted: 27 APR 2009
- Manuscript Revised: 11 MAR 2009
- Manuscript Received: 14 JAN 2009
- pancreatic neuroendocrine tumor;
- fine-needle aspiration cytology;
CK19 and CD10 are useful markers in the differential diagnosis of pancreatic tumors. The authors evaluated CK19 and CD10 expression in pancreatic neuroendocrine tumors (NETs) obtained by endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA).
Twenty-eight patients diagnosed with pancreatic NETs based on EUS-FNA cytology were studied retrospectively (2004-2007) for immunohistochemical expression of CK19 and CD10. Immunohistochemistry was performed on cell blocks for each case. The pattern of expression for CD10 (cytoplasmic or membranous) and its intensity (0-2) were noted. The staining of the stromal elements for CD10 was recorded as negative. Cytoplasmic staining in tumor cells and percentage distribution (1+ to 4+) for CK19 were regarded as positive.
Twenty-three of 28 (82.14%) NETs showed positive cytoplasmic and/or membranous staining for CD10, and 25 of 28 (89.29%) cases were positive for CK19.
The findings demonstrate the high expression of CD10 and CK19 in pancreatic NETs. This indicates that CD10 and CK19 cannot reliably differentiate NETs from other tumors with similar cytomorphologic features (solid pseudopapillary tumors, which frequently stain with CD10, and pancreatic adenocarcinoma, which stains with CK19). Cancer (Cancer Cytopathol) 2009. © 2009 American Cancer Society.
Neuroendocrine tumors (NETs) of the pancreas are relatively uncommon tumors that account for 1% to 2% of all pancreatic neoplasms.1 They originate from the multipotent ductular stem cells, and nonislet cell origin was also suggested in a recent study.2, 3 NETs are usually identified preoperatively because of their propensity to secrete various endocrine hormones and to exhibit hormone-related clinical signs and symptoms. These functional tumors are classified based on the hormones they produce and the associated endocrine syndrome. Nonfunctioning tumors are either an incidental finding or are associated with an expanding mass rather than a hormonal syndrome.
We have recently shown in a large series that endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA) cytology is a safe and accurate method, providing adequate samples for cytologic diagnosis of NETs, and in addition, immunohistochemistry (IHC) performed on cell blocks can confirm the diagnosis.4 IHC stains play an important role in confirming the neuroendocrine nature of tumor cells. Most NETs are positive for neuron-specific enolase (NSE), chromogranin A, synaptophysin, and CD56. Several studies have recently reported a variety of IHC markers, such as CK19,5, 6 COX2,7 p27,8 CD99,9 and CD10,10 in pancreatic NETs, with prognostic significance for some of them. CK19 is a low–molecular-weight (40 kDa) keratin belonging to a family of intermediate filaments of epithelial cells. It is a well-known marker for ductular epithelial cells in pancreas, and it is highly expressed in adenocarcinomas. CD10 is a type II cell surface metalloproteinase and a member of a family of peptides that inactivate bioactive peptides, such as inflammatory mediators and vasoactive peptides. It was formerly referred to as “common acute lymphoblastic leukemia antigen,” because it is commonly expressed on lymphoid precursor cells and acute lymphoblastic leukemia, as well as other forms of hematopoietic tumors.11 The antigen has also been detected in pancreatic ductal carcinomas and is frequently observed in solid pseudopapillary tumors.12
Our aim was to evaluate the IHC expression of CK19 and CD10 in pancreatic NETs obtained by EUS-FNA cytology, and moreover, to highlight the role of these markers in the differential diagnosis of NETs with other tumors with potentially similar cytomorphologic features (solid pseudopapillary, adenocarcinomas).
MATERIALS AND METHODS
Twenty-eight patients with pancreatic NETs diagnosed by EUS-FNA cytology were retrospectively selected between 2004 to 2007, through a computer search of the database in the Department of Cytology, General Hospital of Athens, Greece (during this period we diagnosed 48 NETs). Our primary aim was to include all cases (n = 48) in our study. However, a combination of technical difficulties and most importantly the lack of adequate material in our cell blocks compelled us to limit our series. We managed to include all cases in a previous study, regarding biologic behavior of NETs.13 All cytology specimens and microbiopsies were performed in the endoscopy suite. EUS was performed in a sequential manner by which the order of structures examined was the same for each patient. No informed consent was obtained from patients, because the studies were carried out during routine clinical examinations. A linear-array echoendoscope (EG3630 UR; Pentax, Tokyo, Japan) was advanced under indirect visualization through the cricopharyngeus into the esophagus. The echoendoscope was connected to a monitoring device (6000 Victor; Hitachi, Tokyo, Japan), and the EUS-guided FNA procedure was performed using 22-gauge needles (Medi-Globe, GmbH, Achenmuhle, Germany). The aspirated samples were assessed immediately by an on-site attending cytopathologist in all cases (C.S.). The aspirated material was smeared onto slides, and smear preparation was followed by either air drying or immediate fixation in 95% alcohol for subsequent Papanicolaou staining methods. Additional aspirated material was fixed in formalin, embedded in paraffin, and processed for routine histologic examination using standard techniques.
IHC studies were performed in cells blocks. For this purpose, 5-μm sections were cut, deparaffinized, and mounted on precoated slides. The following antibodies were used for this study: CK19 (monoclonal, 1:500 dilution, Novocastra, Newcastle, UK) and CD10 (monoclonal 56C6, Novocastra). CK19 staining was assessed, as proposed by Ali et al,9 as follows: 1+, 5% to 10% tumor-cell immunolabeling; 2+, 11% to 25%; 3+, 26% to 50%; and 4+, >50%. Cytoplasmic and/or membranous staining was evaluated as CK19 positive. The intensity of staining for CD10 was assessed as follows: negative scored as 0, weak or moderate scored as 1, and strong scored as 2. The cytoplasmic and/or membranous staining was evaluated as CD10 positive. The immunolabeling of the stromal elements (lymphocytes, fibro/myofibroblasts, endothelial cells, dispersed ductal epithelial cells) for CD10 was recorded as negative. All these stains were evaluated by 3 cytopathologists (P.C., C.S, P.K.). Residual pancreatic ducts and scattered lymphocytes served as positive controls for CK19 and CD10, respectively. Also, 4 tumors diagnosed as solid pseudopapillary tumors of the pancreas were stained for CD10 for comparison.
Twenty-five of 28 (89.29%) cases were CK19 positive. Eight cases showed 4+ positivity (>50% of tumor cells), 8 were 3+ (26%-50%), 5 were 2+ (11%-25%), and 4 were 1+ (5%-10%). The staining was cytoplasmic and membranous (Fig. 1). In addition, in 4 cases we observed a distinct dot-like staining pattern. Twenty-three tumors showed CD10 positivity (82.14%) in cytoplasm and/or in their membranes (Fig. 2). Sixteen cases showed strong positivity (2+), and 7 showed weak or moderate positivity (1+). In nontumoral pancreatic tissue, CD10 was detected focally in ductal cells, perineurally, in some endothelial cells, and within lymphocytes. Scattered cells forming clusters of 3 to 4 cells resembling endocrine cells were not evaluated. All solid pseudopapillary tumors of the pancreas were consistently CD10 positive.
In daily practice, most pancreatic NETs do not induce diagnostic problems for an experienced cytopathologist. In the presence of the typical cytomorphologic features, as previously described,4 the diagnosis of NETs is easy to make. However, many times, cytomorphologic samples obtained from pancreatic NETs by the EUS-FNA procedure tend to be more hemorrhagic, with gastrointestinal contaminants and with scant material. For these reasons, the reported accuracy for pancreatic NETs is lower compared with that reported for pancreatic exocrine adenocarcinoma.14 The diagnostic accuracy for NETs is enhanced when combined with IHC.
IHC stains play an important role in confirming the neuroendocrine nature of tumor cells. Most tumors are positive for chromogranin A, NSE, synaptophysin, and CD56. In addition, functional tumors can express other markers: insulin, glucagon, gastrin, vasoactive intestinal polypeptide, and somatostatin. However, not every pancreatic NET secretes a hormone that results in a clinicopathologic syndrome. Recently, new markers were included in the immunoprofile of NETs: CK19,5, 6 COX2,7 p27,8 CD99,9 and CD10.10 CK19 and CD99 have emerged as predictors of survival and biologic behavior.5, 6 Moreover, according to a study by Schmitt et al,6 CK19 should be included in routine diagnostics, because it improves the prognostic power of the World Health Organization 2004 classification. In these studies, CK19 positivity was well correlated with histopathologic factors like mitoses, necrosis, solid histological pattern, vascular invasion, and Ki-67 labeling index. At present, CK19 has not been included in the largest published series regarding cytomorphology and IHC profile in pancreatic NETs in samples obtained by EUS-FNA cytology.4, 15 In our study, CK19 positivity (cytoplasmic and membranous) was very high (89.29%), and most of the cases showed intense staining (3+, 4+). These findings indicate that a ductular epithelial type cytokeratin is also frequently detected in neuroendocrine cells.
According to some authors, NETs are likely to originate from the hypothetical multipotent ductular stem cells, and recently nonislet cell origin of these tumors was demonstrated in patients with multiple endocrine neoplasia type 1.2, 3 The high expression percentage of CK19 in our study reinforces this hypothesis. Also, it seems that CK19 cannot be reliably used as a specific diagnostic marker in ductular adenocarcinomas—especially in those tumors that have cytomorphologic features similar to those of NETs—because NETs can express it also. A wider immunohistochemical panel for cytokeratins (CK7, CAM5.2) should be used for differentiating these tumors.
CD10 expression has been demonstrated in several normal tissues and in numerous tumors, particularly neoplasms of the gastrointestinal and genitourinary tracts, such as renal cell, transitional cell, and prostate carcinomas.16 It is a marker of trophoblastic tumors and endometrial stromal neoplasms, and it is also expressed by serous ovarian carcinomas.11 In the gastrointestinal system, CD10 is expressed in colon and hepatocellular carcinomas.16 The antigen has been detected in pancreatic ductal carcinomas and is frequently observed in solid pseudopapillary tumors of the pancreas.12 Notohara et al12 have showed positivity of CD10 in some pancreatic endocrine tumors in a small series (5 of 20 NETs). Chu and Arber16 reported CD10 expression in only 1 of 21 NETs. Moreover, in the most recently published study by Deschamps et al,10 CD10 was expressed in 33% of NETs, and interestingly, staining was increased significantly with tumor stage. In this study, authors also showed that in normal pancreas the islets did not express CD10, which suggests de novo expression in NETs.
In our study, CD10 was expressed in 82.14% of cases (23 of 28), with similar distribution in cytoplasm and in membrane of the tumor cells. In well-differentiated carcinomas of the colon, pancreas, and prostate, its expression is mostly apical, whereas it is cytoplasmic in most poorly differentiated tumors.16 In normal tissues, the distribution of CD10 is primarily apical, and this is in accordance with its involvement in normal cellular secretory processes. CD10 is an endopeptidase that regulates signal transduction by hydrolyzing various peptide hormones, and it is implicated in the regulation of cell growth.17 The significance of frequent identification of CD10 in our cases and its functional role in NETs remain undetermined.
Another aspect of our findings regarding CD10 is that we cannot use it with great certainty in differential diagnosis of solid pseudopapillary tumors of the pancreas with NETs. Solid pseudopapillary tumors of the pancreas are benign or low-grade malignant tumors that may show the same cytomorphologic features as NETs. Immunostaining with CD10 may not clarify the diagnosis because of possible positivity in both of them.
In summary, we have shown that another 2 markers, CK19 and CD10, may be expressed in pancreatic NETs obtained by EUS-FNA cytology. This might represent a sign of stem cell capabilities of some cells in NETs. Particularly, the CK19 expression could be indicative of ductal differentiation. These findings should be addressed in further studies. Finally, in the daily practice of cytopathology, we should not base our final diagnosis exclusively on immunopositivity of these 2 markers. A further IHC panel of markers needs to be implemented.
Conflict of Interest Disclosures
The authors made no disclosures.
- 2Pancreatic neuroendocrine neoplasms. A current summary of diagnostic, prognostic, and differential diagnostic information. Am J Clin Pathol. 2001; 115: S28-S45., .
- 13The role of cytomorphology and proliferative activity in predicting biologic behavior of pancreatic neuroendocrine tumors: a study by endoscopic ultrasound-guided fine-needle aspiration cytology. Cancer Cytopathol. 2009; 117: 211-216., , , , .