Anti-CD10 (56C6) is expressed variably in adrenocortical tumors and cannot be used to discriminate clear cell renal cell carcinomas
Özgür Mete · Yersu Kapran · Mine G. Güllüoğlu · Işın Kılıçaslan · Yeşim Erbil · Yasemin Giles Şenyürek · Ferhunde Dizdaroğlu
Received: 26 December 2009 /Revised: 18 February 2010 / Accepted: 20 February 2010 /Published online: 14 April 2010 C Springer-Verlag 2010
Abstract In the evaluation of retroperitoneal masses, the practicing pathologist faces a dilemma when making a diagnosis based on histology given the often overlapping morphologic appearances of the adrenocortical carcinoma, renal cell carcinoma (RCC), and hepatocellular carcinoma (HCC). CD10 is expressed in a membranous fashion in the vast majority of clear cell RCCs; therefore, it is widely used for distinction from its mimics. However, its expression is not well-investigated in adrenal cortical tumors. We exam- ined CD10 expression in 47 surgically resected adrenocor- tical tumors (26 adenomas and 21 carcinomas) and compared with 20 clear cell RCCs and 25 HCCs. Twenty HCCs (80%), 18 RCCs (90%), 11 adrenocortical carcinomas (52%), and 18 adrenocortical adenomas (69%) were positive for CD10. HCCs were characterized by a canalicular staining, and clear cell RCCs exhibited membranous or mixed membranous- cytoplasmic staining. Adrenocortical tumors displayed mainly cytoplasmic staining. Four adrenocortical carcino- mas and one adenoma also displayed the membranous staining pattern. Despite the relatively small number of samples, our preliminary results revealed that adrenocortical tumors may express CD10 (Clone: 56C6). The most important point from this paper is the fact that anti-CD10
Ö. Mete · Y. Kapran · M. G. Güllüoğlu · I. Kılıçaslan · F. Dizdaroğlu Department of Pathology, Istanbul Faculty of Medicine, Istanbul University, Capa, 34800 Istanbul, Turkey
e-mail: ozgurmete77@gmail.com
Y. Erbil · Y. Giles Şenyürek Department of General Surgery, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey
expression has not been previously reported in adrenocorti- cal carcinomas. This suggests that CD10 does not seem to be a useful marker for discriminating clear cell RCCs from adrenocortical tumors since CD10 expression does not rule out the possibility of adrenocortical tumors. This feature should be kept in mind when constructing an antibody panel for an epithelial tumor that involves the adrenal gland and kidney, especially in small biopsy specimens.
Keywords CD10 · Adrenocortical carcinoma . Adrenocortical adenoma · Hepatocellular carcinoma · Renal cell carcinoma
Introduction
Although imaging studies provide important information for detection and localization of adrenocortical tumors, the definitive diagnosis must be established by pathological examination. Even at the time of the surgery, a clear determination of the exact origin of a particular tumor cannot always be confirmed, especially for large lesions. The differential diagnosis between metastatic renal cell carcinoma (RCC), hepatocellular carcinoma, and primary adrenocortical carcinoma, which may share some morpho- logical similarities, can be challenging, especially in the setting of limited tissue (e.g., core biopsies).
CD10, or the common acute lymphoblastic leukemia antigen, is a cell-surface neutral endopeptidase that inacti- vates bioactive peptides [1, 2]. After its initial description in neoplastic and nonneoplastic hematolymphoid tissue (lym- phoid precursor cells, some myeloid cells, and germinal center B cells), an increased number of human non- hematolymphoid neoplasms were shown to express CD10
| Parameters | Adenomas, n (%) | Carcinomas, n (%) |
|---|---|---|
| Total | 26 (100) | 21 (100) |
| Female | 20 (77) | 12 (57) |
| Male | 6 (23) | 9 (43) |
| Age (mean±stdª) (years) | 42.57±14.38 | 42.81±16.05 |
| Size (mean±stdª) (cm) | 3.48±1.05 | 10.55±4.60 |
| Weight (mean±stdª) (g) | 20.07±9.47 | 490.47±441.11 |
| Right/left side | 9/17 | 14/7 |
| Macroscopic necrosis | 0 (0) | 20 (95) |
| Cystic degeneration | 3 (12) | 1 (5) |
| Hemorrhage | 3 (12) | 15 (71) |
a
ª std Standard deviation
antigen [1-6]. Since CD10 is expressed in a membranous fashion in the vast majority of clear cell RCCs, it is widely used for the distinction of RCC [2-4, 7]. However, its expression has not been well-investigated in adrenocortical adenomas and carcinomas. For this reason, we examined the expression pattern of CD10 in adrenocortical tumors, and compared its expression profile with hepatocellular carcinomas and clear cell RCCs.
Materials and methods
Tissue samples
We collected 47 surgically resected adrenocortical tumors (26 adrenal cortical adenomas and 21 adrenal cortical carcino- mas), 25 hepatocellular carcinomas, and 20 clear cell RCCs, from the archives of the Department of Pathology, Istanbul Faculty of Medicine, Istanbul University. All cases were classified according to the standard diagnostic criteria including modified Weiss criteria for adrenocortical carcino- mas. Patient age, gender information, and the gross pathologic features (size, weight, necrosis, cystic degeneration, and hemorrhage) of the adrenocortical tumors were reviewed. Radiologically, all adrenocortical tumors were characterized
by unilateral mass. The gross and microscopic presence of non-tumorous adrenal tissue adjacent to the tumors helped us to ensure that these were primary adrenal tumors. Moreover, in the evaluation of adrenocortical carcinomas, the presence of positive immunostaining for synaptophysin, «-inhibin and Melan-A (Clone: A103) helped us to verify the adrenal cortical differentiation.
Immunohistochemistry
One paraffin block having the most representative area was selected for the immunohistochemical study for each tumor. Routine immunohistochemical staining was performed man- ually on formalin-fixed, paraffin-embedded tissues using the streptavidin-biotin-complex method (Lab Vision Corp., Fre- mont, CA, USA) after a pretreatment with heat-induced antigen unmasking by incubating the sections in 10 mmol/l of citrate buffer solution (pH: 6) using a domestic pressure cooker (25 min, 1 atm). The entire procedure was carried out in room temperature. Immunohistochemistry was performed using anti-CD10 antibody (Monoclonal, mouse anti-human, Clone: 56C6, 1:10, 120 min, Novocastra Laboratories, UK). Representative paraffin blocks that contain normal renal cortical parenchyma and non-tumorous liver except bile canaliculi were used as positive and negative controls, respectively. Moreover, the heat-induced antigen retrieval was also paralleled by serial sections run without primary antibody incubation. The results of immunohistochemical staining were recorded as positive (>10% of tumors cells staining) and the staining patterns (cytoplasmic, membranous, mixed cytoplasmic-membranous, and canalicular), and immu- nostaining area percentages (10-25, 25-50, 50-75, and 75- 100%) were noted.
Results
Clinical and gross pathological findings
The main clinical and gross findings of adrenocortical tumors are summarized in Table 1. A female predominance
| Patterns and rates | ACC n (%) 21 (100) | ACA n (%) 26 (100) | HCC n (%) 25 (100) | Clear cell RCC n (%) 20 (100) |
|---|---|---|---|---|
| Positivity rates | 11 (52) | 18 (69) | 20 (80) | 18 (90) |
| No staining | 10 (48) | 8 (31) | 5 (20) | 2 (10) |
| Membranous | 3 (14) | 0 (0) | 0 (0) | 15 (75) |
| Cytoplasmic | 7 (33) | 17 (65) | 0 (0) | 0 (0) |
| Canalicular | 0 (0) | 0 (0) | 20 (80) | 0 (0) |
| Membranous and cytoplasmic | 1 (5) | 1 (4) | 0 (0) | 3 (15) |
ACC adrenocortical carcinoma, ACA adrenocortical adenoma, HCC hepatocellular carcinoma, Clear cell RCC clear cell renal cell carcinoma
| Table 3 The distribution of CD10 immunostaining area percentages | CD10 immunostaining area percentages (%) | ACC n (%) 21 (100) | ACA n (%) 26 (100) | HCC n (%) 25 (100) | Clear cell RCC n (%) 20 (100) |
|---|---|---|---|---|---|
| ACC adrenocortical carcinoma, | 0-10 | 10 (48) | 7 (27) | 5 (20) | 2 (10) |
| 10-25 | 5 (24) | 17 (65) | 5 (20) | 3 (15) | |
| ACA adrenocortical adenoma, | 25-50 | 3 (14) | 2 (8) | 1 (4) | 3 (15) |
| HCC hepatocellular carcinoma, | 50-75 | 2 (9) | 0 (0) | 4 (16) | 8 (40) |
| Clear cell RCC clear cell renal cell carcinoma | 75-100 | 1 (5) | 0 (0) | 10 (40) | 4 (20) |
is observed in tumor groups. Female/male ratios in adenomas and carcinomas were 3.2 and 1.3, respectively. The mean weight of the cortical carcinomas (490.47± 441.11 g) is greater than that of adenomas (20.07±9.47 g). Carcinomas were located predominantly in the right adrenal gland (right/left: 2) and adenomas were located more in the left adrenal gland (right/left: 0.53). No necrosis was found in adenomas, and nearly all carcinomas (95%) exhibited varying proportions of necrosis.
Immunohistochemical findings
The immunoreactivity patterns, rates, and percentages of immunostained areas are summarized in Tables 2 and 3. Nonneoplastic renal cortical parenchyma revealed positive immunostaining in the proximal tubular epithelium and glomerular cells. Normal liver parenchyma exhibited CD10 expression in bile canaliculi. Non-tumorous adrenal cortex revealed focal and weak cytoplasmic staining in two cases.
a
C
b
d
Adrenocortical carcinomas
Eleven cases (52%) were positive for CD10 (Fig. 1). Membranous staining was present in three cases (14%). One case (5%) showed mixed membranous-cytoplasmic staining, while in seven cases (33%) cytoplasmic staining was present with accentuation in the paranuclear region. Cytoplasmic immunoreactivity was generally weak, and the staining percentage ranged from 20 to 50%. However, the staining percentage of the case that exhibited mixed membranous-cytoplasmic pattern was categorized as 75- 100%, and other three cases, which exhibited membranous staining pattern, revealed a staining percentage of 25-50, 50-75, and 50-75%, respectively.
Adrenocortical adenomas
Eighteen cases (69%) were positive for CD10 (Fig. 2). Membranous and cytoplasmic staining was present in 1
case (4%) and 17 cases (65%) showed cytoplasmic staining. The majority of cytoplasmic staining was weak, and the staining percentage ranged from 10 to 30%.
Hepatocellular carcinomas
Twenty cases (80%) were positive for anti-CD10 with a canalicular pattern, which is characterized by a branching or short linear positivity between two cells rather than within them (Fig. 3a). The staining percentage ranged from 20 to 100%.
Clear cell RCCs
Eighteen cases (90%) were positive for CD10 (Fig. 3b). Pure membranous staining was present in 15 cases (75%), and 3 cases (15%) showed mixed membranous-cytoplasmic stain- ing. The majority of cytoplasmic staining was moderate to strong, and the staining percentage ranged from 20 to 100%.
a
C
b
Discussion
In the evaluation of retroperitoneal masses, adrenocortical carcinomas, RCCs, hepatocellular carcinomas, angiomyoli-
pomas, gastrointestinal stromal tumors, soft tissue tumors, lymphoid malignancies, and metastatic tumors should be considered in the differential diagnosis. Because of the close anatomical relation of the adrenal gland with kidney and surrounding retroperitoneal fat, adrenal neoplasm may pose clinical and pathological diagnostic challenges. In this context, the distinction of adrenocortical carcinomas from renal cell carcinomas and hepatocellular carcinomas has been recognized as particularly difficult especially in core biopsies [5-8]. The practicing pathologist may not be confident enough to make a definitive diagnosis based only on histology, given the often overlapping morphologic appearances of these tumors.
CD10 has been considered a useful marker of RCC, because of its high level of expression in clear cell and papillary RCCs [2-4, 7]. On the other hand, a number of recent studies have shown that cytoplasmic CD10 is variably expressed in renal oncocytomas and in other eosinophilic variants of RCC [6]. Moreover, the membra- nous CD10 expression pattern, which is reported to range from 82 to 100% in clear cell renal cell carcinomas, is commonly accepted to distinguish renal carcinomas from adrenocortical carcinomas [2-5, 7]. Chu and Arber [2] studied CD10 (clone: 56C6) expression in 505 non- hematopoietic tumors, including ten adrenocortical carci- nomas; in that analysis, all ten adrenocortical carcinomas were negative for CD10, whereas 41 of 46 cases (89%) of RCC were positive. According to their results, CD10 immunohistochemistry is extremely helpful in the differen- tial diagnosis of RCC and adrenocortical carcinoma [2]. Subsequently, Pan et al. [5] investigated 15 markers in a tissue microarray study of 895 different tumors including 40 adrenocortical carcinomas and 62 adrenocortical adeno- mas. Among the antibodies studied, anti-CD10 (clone: 270) revealed cytoplasmic and membranous staining in only one adrenocortical adenoma, and no CD10 positivity was demonstrated in adrenocortical carcinomas [5]. The present study revealed that adrenocortical tumors may display
| Study | Pan et al. [5] | Chu and Arber [2] | Present study (Mete et al.) |
|---|---|---|---|
| Antigen | CD10 | CD10 | CD10 |
| Clone | 270 | 56C6 | 56C6 |
| Source | Novocastra, Newcastle, UK | Novocastra, Burlingame, CA | Novocastra Laboratories, UK |
| Pretreatment | Microwave, citrate (pH: 6) | Food steamer, EDTA (pH:8) | Pressure cooker, citrate (pH:6) |
| Dilution | 1/40 | 1/20 | 1/10 |
| Incubation time | Not available | 25 min | 120 min |
| Application method | High density tissue array slides | Automated immunostainer, whole section slides | Non-automated staining, whole section slides |
| Detection system | Biotin-free DAKO Envision-plus system | Streptavidin-biotin-complex technique | Streptavidin-biotin-complex technique |
variable immunoreactions with anti-CD10 antibody (clone: 56C6). Furthermore, four adrenocortical carcinomas and one adrenocortical adenoma revealed a membranous stain- ing pattern, which is regarded typical of clear cell RCCs.
It is of interest that previous studies used different methods in terms of antibody clone, incubation time, dilution ratio, pretreatment, detection system, and applica- tion (Table 4). Our study highlighted that CD10 expression (clone: 56C6) was heterogeneous in adrenocortical tumors (Table 3). Therefore, the use of it in limited tissue samples such as tissue microarrays may not reflect this heterogene- ity well enough. Furthermore, we observed that the heat- induced antigen retrieval process with EDTA (pH: 8) revealed weaker staining in some of the positive control tissues, when compared with the staining in tissues processed with citrate (pH: 6). Therefore, the use of different antibody clones, antigen retrieval methods, incu- bation time, and concentration may explain possible discrepancies between the results of different studies.
On the other hand, CD10 is developmentally regulated in precursor epithelial cells, and CD10 loss from methyl- ation leads to increased cell migration, cell growth, and cell survival in epithelial cell, contributing to neoplastic development and progression [9]. It is clear that CD10 possesses multiple biological effects; therefore CD10 may be more than just a marker for discrimination of some tumors [9]. However, the latter requires further investiga- tions in larger series in order to illuminate whether the loss or presence of CD10 is of clinical significance in adrenocortical tumors.
Pan et al. highlighted that polyclonal-CEA and CD10 were valuable because of their high specificity for hepato- cellular carcinomas, which share some overlapping mor- phological features with adrenocortical tumors [5]. However, only the canalicular pattern, the thick, waxy, branching, or short linear, dot-like positivity between cells rather than within them, was regarded as specific for hepatocellular carcinomas [5]. This pattern is regarded to have a diagnostic importance, as the membranous and/or cytoplasmic CD10 staining pattern is not observed in hepatocellular carcinomas. In comparison with HCCs, we did not observe any canalicular staining in adrenocortical tumors.
Melan-A (A103) and «-inhibin immunoreactivities have been advocated as being useful in distinguishing adreno- cortical tumors [5, 10]. However, certain types of non- epithelial tumors including sex cord stromal tumors and some soft tissue tumors have been reported to express x- inhibin and Melan-A (A103) [11-14]. Soft tissue tumors with epithelioid differentiation, may also mimic the morphology of primary adrenocortical carcinomas, and are usually both negative for epithelial markers such as pankeratin and epithelial membrane antigen in the majority
of cases. Furthermore, it is of interest that RCCs, particularly those associated with t(6;11) and t(X;1p11)/ TFE3 translocations may be negative for keratins and may exhibit Melan-A [13]. Another diagnostic challenge in the adrenal region is the distinction of adrenocortical carcino- mas from metastatic or primary melanoma and pure epithelioid angiomyolipoma (EAML), which may show Melan-A (A103) positivity [12, 15]. However, this distinc- tion can be usually made by the demonstration of the absence of pan-melanocytic antigens (S100 protein, HMB- 45, Melan-A (A103 and other clones), CD63, tyrosinase and microphthalmia-associated transcription factor), and muscle markers. In this context, it is worth stressing that CAM5.2 and S-100 protein may occasionally be positive, albeit focal, in EAMLs [12]. In special situations, the distinction of an adrenocortical carcinoma may require the use of other markers such as SF-1 (steroidogenic factor-1) and DAX-1 (dosage-sensitive sex reversal adrenal hypo- plasia congenital critical region on the X chromosome gene 1) [16-18]. Recently, Kaneko et al. [18] showed that DAX- 1 and SF-1 can be used to diagnose the site of origin of metastatic adrenocortical carcinoma in patients clinically presenting with widespread metastasis and bulky retroper- itoneal masses.
In conclusion, our preliminary results revealed that adrenocortical tumors may also express CD10. Despite the relatively small number of samples, the most important point from this paper is the fact that anti-CD10 expression has not been previously reported in adrenocortical carcino- mas. This suggests that anti-CD10 (56C6) does not seem to be a useful marker for the distinction of renal cell carcinomas from adrenocortical tumors; since its expression does not rule out the possibility of adrenocortical tumors. Therefore, it is important to remember this feature when constructing an antibody panel for a retroperitoneal epithe- lial neoplasm that involves the adrenal gland and kidney; especially in the setting of limited tissue sample (e.g. core biopsies or FNA specimens).
Acknowledgements This study was supported by the research fund of Istanbul University (Project number: UDP-2768/23072008). The preliminary results of this study were partially presented during the poster session of the XXVII. Congress of the International Academy of Pathology in Athens (October, 12-17, 2008).
Conflict of interest statement We declare that we have no conflict of interest.
References
1. McIntosh GG, Lodge AJ, Watson P et al (1999) NCL-CD10-270: a new monoclonal antibody recognizing CD10 in paraffin- embedded tissue. Am J Pathol 154:77-82
2. Chu P, Arber DA (2000) Paraffin-section detection of CD10 in 505 non-hematopoietic neoplasms: frequent expression in renal cell carcinoma and endometrial stromal sarcoma. Am J Clin Pathol 113:373-382
3. Langner C, Ratschek M, Rehak P et al (2004) CD10 is a diagnostic and prognostic marker in renal malignancies. Histo- pathol 45:460-467
4. Avery AK, Beckstead J, Renshaw AA et al (2000) Use of antibodies to RCC and CD10 in the differential diagnosis of renal neoplasms. Am J Surg Pathol 24:203-210
5. Pan CC, Chen PC, Tsay SH et al (2005) Differential immunopro- files of hepatocellular carcinoma, renal cell carcinoma, and adrenocortical carcinoma: a systemic immunohistochemical sur- vey using tissue array technique. Appl Immunohistochem Mol Morphol 13:347-352
6. Mete O, Kilicaslan I, Uysal V (2009) Does CD10 immunoex- pression have a diagnostic utility in the differential diagnosis of renal oncocytomas and eosinophilic variants of chromophobe renal cell carcinomas? Pathol 41:191-193
7. Pan CC, Chen PCH, Ho DMT (2004) The diagnostic utility of MOC31, BerEP4, RCC marker and CD10 in the classification of renal cell carcinoma and renal oncocytoma: an immunohisto- chemical analysis of 328 cases. Histopathol 45:452-459
8. Kato I, Inayama Y, Yamanaka S et al (2009) Epithelioid angiomyolipoma of the kidney. Pathol Int 59:38-43
9. Papandreou CN, Nanus DM (2010) Is methylation the key to CD10 loss? J Pediatr Hematol Oncol 32:2-3
10. DeLellis RA, Lloyd RV, Heitz PU, Eng C (2004) World Health Organization classification of tumours. Pathology and genetics of tumours of endocrine organs. IARC, Lyon
11. Stewart CJ, Nandini CL, Richmond JA (2000) Value of A103 (Melan-A) immunostaining in the differential diagnosis of ovarian sex cord stromal tumours. J Clin Pathol 53:206-211
12. Aydin H, Magi-Galluzzi C, Lane BR et al (2009) Renal angiomyolipoma: clinicopathologic study of 194 cases with emphasis on the epithelioid histology and tuberous sclerosis association. Am J Surg Pathol 33:289-297
13. Zamecnik M, Michal M, Mukensnabl P (2003) Reactivity of granular cell tumors for inhibin and other markers of sex cord and steroid cell differentiation. Am J Surg Pathol 27:413-414
14. Schraith DF, Hahm GK, Niemann TH et al (2003) Alpha-inhibin immunoreactivity in soft-tissue neoplasia. Mod Pathol 16:1205- 1209
15. Banerjee SS, Eyden B (2008) Divergent differentiation in malignant melanomas: a review. Histopathol 52:119-129
16. Doghman M, Cazareth J, Douguet D et al (2009) Inhibition of adrenocortical carcinoma cells proliferation by SF-1 inverse agonists. J Clin Endocrinol Metab 94:2178-2183
17. Bakke M, Zhao L, Hanley NA et al (2001) SF-1: a critical mediator of steroidogenesis. Mol Cell Endocrinol 171:5-7
18. Kaneko T, Kojima Y, Umemoto Y et al (2008) Usefulness of transcription factors Ad4BP/SF-1 and DAX-1 as immunohisto- logic markers for diagnosis of advanced adrenocortical carcinoma. Horm Res 70:294-299