Differential Immunoprofiles of Hepatocellular Carcinoma, Renal Cell Carcinoma, and Adrenocortical Carcinoma A Systemic Immunohistochemical Survey Using Tissue Array Technique
Chin-Chen Pan, MD, Paul Chih-Hsueh Chen, MD, PhD, Shyh-Haw Tsay, MD, and Donald Ming-Tak Ho, MD
Abstract: The differential diagnoses of hepatocellular carcinoma (HCC), renal cell carcinoma (RCC), and adrenocortical carcinoma (ACC) are sometimes difficult due to their overlapping histologic features. Immunohistochemistry is a helpful adjunct in supporting the histologic diagnosis. In this study, the authors used the tissue array technique to systemically analyze the efficacy of different immuno- histochemical panels in discerning these neoplasms. Immunohisto- chemical stains were performed on a total of 895 tumors (including 170 HCCs, 176 RCCs, and 40 ACCs) using monoclonal antibodies against hepatocyte antigen (HPA), CD10, RCC marker, vimentin, a-inhibin, keratins (KL-1, CAM 5.2, 7, and 20), epithelial membrane antigen, and polyclonal antibodies against carcinoembryonic antigen (pCEA) and a-fetoprotein, and antibodies Melan-A (A103), MOC31, and BG8. HPA immunostain alone detected 85.9% of HCCs, and the addition of canalicular pattern of pCEA and CD10 immunostains raised the sensitivity to 94.7%. RCC marker was positive in 54.5% of RCCs but was negative in all non-RCC tumors. Using positive CD10 and negative HPA and pCEA together with RCC marker increased the sensitivity to 74.4%. Immunoreactivity for @-inhibin and A103 could be detected in 67.5% and 55% of ACCs, respectively. When the two antibodies were combined, 82.5% of ACCs were labeled. Proper selection of immunohistochemical stains aid in the differential diag- nosis of the three neoplasms. Using the tissue array technique, the authors also showed an effective model for comprehensive antibody testing.
Key Words: hepatocellular carcinoma, renal cell carcinoma, adrenocortical carcinoma, immunohistochemistry, tissue array (Appl Immunohistochem Mol Morphol 2005;13:347-352)
T he pathologic diagnoses of hepatocellular carcinoma (HCC), renal cell carcinoma (RCC), and adrenocortical carcinoma (ACC) may be difficult, especially on limited biopsy material from metastatic sites in patients with unknown
Received for publication March 24, 2004; accepted September 13, 2004. From the Department of Pathology, National Yang-Ming University, Taipei Veterans General Hospital, and Taipei City United Hospital, Taipei, Taiwan. Supported by a grant from Taipei Veterans General Hospital (V93-173).
Reprints: Chin-Chen Pan, MD, Department of Pathology, Taipei Veterans General Hospital, No. 201, Sec. 2, Shi-Pai Rd., Taipei, 11217 Taiwan (e-mail: ccpan@vghtpe.gov.tw).
Copyright @ 2005 by Lippincott Williams & Wilkins
primaries. The pathologist may not be confident enough to make a definitive diagnosis based on histology alone, given the often-overlapping morphologic appearances among the three entities. The constituent cells of the three tumors are usually polygonal and contain clear to eosinophilic cytoplasm. They may grow in acinar, trabecular, or diffuse patterns. In such cases, ancillary immunohistochemical (IHC) stains are helpful to confirm the impression based on histology. Recently, a few markers in addition to the conventional ones appear to be beneficial indicators, including hepatocyte antigen (HPA), also called Hep Par 1, for HCC1-5; renal cell carcinoma marker (RCC Ma) for RCC6-8; and «-inhibin and Melan-A (A103) for ACC.9-12 A systemic survey of these additional markers would provide valuable information regarding the sensitivity and specificity for practicing pathologists.
The recently developed tissue array technique is very suitable for in situ analysis of a large number of cases.13,14 This tissue array technique enables multiple cases to be constructed into single block, saving significant time, labor, and costs. We applied this method to test for 15 antibodies in HCC, RCC, and ACC, accompanied by a variety of other neoplasms, in a search for useful panels to identify the three tumors.
MATERIALS AND METHODS
Case Selection
One hundred seventy (150 primary and 20 metastatic) cases of HCC, 176 cases (154 primary and 22 metastatic) of RCC, and 40 cases (32 primary and 8 metastatic) of ACC were retrieved from the surgical pathology archive from 1991 to 2003 at Taipei Veterans General Hospital. The material con- sisted of excisional biopsy or radical surgical specimens. Small biopsies were not used. Cases with extensive necrosis rendering the viable tissue inadequate for analysis were also excluded.
HCCs were classified according to the cytologic features into conventional (146 cases), clear cell (20 cases), and fibro- lamellar (4 cases) type.15 RCCs were classified based on the 1997 UICC/AJCC classification into conventional (117 cases), papillary (26 cases), chromophobe (16 cases), collecting duct (11 cases), and unclassified type.16,17 The diagnosis of ACC was made based on the Weiss system. 18,19
Five hundred nine tumors of other kinds, preferentially epithelial tumors, were randomly selected from the same surgical pathologic archive.
Construction of Tissue Arrays
We constructed high-density tissue arrays as previously described.20 The hematoxylin and eosin-stained slides were examined under a light microscope. The areas of interest were circled with a color pen. One to three foci per case were selected according to the heterogeneity of the tumor. We used a 16-gauge bone marrow biopsy trephine apparatus to punch the blocks at the circled areas. A tissue cylinder 2.0 mm in diameter was extracted. The cylinder was carefully transferred with a forceps to a recipient metal paraffin block box. One recipient box could accommodate up to 88 (11 × 8) cylinders. After all the cylinders were aligned in the box, the box was covered with a plastic cassette and then liquid wax was gently poured into the box until it was full. Then the box was put on a hot plate for 1 minute to homogenize the wax, after which the box was removed from the hot plate and cooled to room temperature slowly. Four-micron sections were cut and mounted on silane-coated slides. The 895 cases were made into 18 blocks.
Immunohistochemistry
The slides were submitted for IHC staining using monoclonal antibodies against HPA, CD10, RCC Ma, vimentin, «-inhibin, cytokeratins (CKs) (KL-1, CAM 5.2, CK 7, CK 20), epithelial membrane antigen (EMA), mono- clonal antibodies A103, MOC31, and BG8, and polyclonal antibodies directed at «-fetoprotein (AFP) and carcinoem- bryonic antigen (pCEA). The clonal names, dilution, antigen retrieval methods, and commercial sources of the antibodies are listed in Table 1. The bound antibodies were detected by the DAKO Envision-plus system (Dako, Carpinteria, CA). The system is biotin-free. The immunopositivity was initially scored as 0, undetectable; 1+, equivocal individual single cell positivity in less than 1% tumor cells; 2+, heterogeneous
positivity in less than 50% of tumor cells; 3+, strong positivity in 50% to 90% of tumor cells; and 4+, diffuse positivity in more than 90% of tumor cells. The cases scored as 2 to 4 were regarded as being positive, while those scored as 0 or 1 were regarded as negative.
Sensitivity and Specificity
First, the positive rate of each antibody in different tumor categories was calculated. Then, the sensitivities and specificities of those antibodies showing high discriminatory power were determined individually or in combination. The specificities among the HCC, RCC, and ACC, which are the major differential diagnoses, and among all tumors were calculated separately.
RESULTS
The positive case numbers in different types of tumors are listed in Table 2. The typical IHC expression is illustrated as Figure 1.
Hepatocellular Carcinoma
One hundred forty-six of the 170 HCCs were positive for HPA. In 138 positive cases, the immunoreactivity was strong (3+ or 4+). A canalicular pattern for pCEA and CD10 was observed in 122 cases and 68 cases, respectively. Cytoplasmic or membranous positivity was also seen in 20 and 11 cases, respectively. Around 20% of the HCCs were reactive for AFP. The majority of HCCs (166/170) were positive for CAM5.2. The positive rates for pan-cytokeratin were lower. HCCs were infrequently positive for CK7, EMA, MOC31, and BG8, with positive rates ranging from 16% to 24%. HCCs rarely expressed CK20. The expression of these markers did not differ significantly among different subtypes of HCC.
| TABLE 1. Sources and Staining Conditions | ||||
|---|---|---|---|---|
| Antigen | Clone | Pretreatment | Dilution | Source |
| Hepatocyte antigen | OCH1E5.2.10 | MC | 1:50 | Dako, Carpinteria, CA |
| CEA | Polyclonal | Nil | 1:400 | Dako |
| CD10 | 270 | MC | 1:40 | Novocastra, Newcastle, UK |
| Alpha fetoprotein | Polyclonal | Nil | 1:100 | Dako |
| RCC marker | 66.4.C2 | TD | 1:50 | Novocastra |
| Vimentin | V9 | MC | 1:3000 | Dako |
| Alpha-inhibin | MCA951S | MC | 1:10 | Oxford Bio-Innovation Oxfordshire, UK |
| A103 | A103 | MC | 1:40 | Dako |
| Pan-cytokeratin | KL-1 | MC | Prediluted | Immunotech, Westbrook, ME |
| CAM 5.2 | CAM 5.2 | MC | 1:10 | Becton-Dickinson, Mountain View, CA |
| Cytokeratin 7 | OV-TL | MC | 1:50 | Dako |
| Cytokeratin 20 | Ks20.8 | MC | 1:30 | Dako |
| EMA | E29 | MC | 1:40 | Dako |
| MOC31 | MOC31 | MC2 | 1:50 | Dako |
| BG8 | F3 | Nil | 1:60 | Signet, Dedham, MA |
MC, microwave in pH 6.0, 10 mM citric acid buffer from 10 min; MC2, 98℃ hot bath in pH 6.0, 10 mM citric acid buffer for 40 min; TD, trypsin digestion (200 mg trypsin and 200 g calcium chloride in 200 mL Tris buffer saline), 37°℃, pH 7.8, for 30 min.
| TABLE 2. Immunoprofiles of 895 Cases | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total | HPA | PCEA Can | PCEA | CD10 | CD10 | AFP | RCC | VIM | Pan A103 CK CAM5.2 CK7 | CK20 | MOC31 | BG8 | ||||||
| Cy&M | Can | Cy&M | Ma | Inhibin | ||||||||||||||
| Hepatocellular carcinoma | 170 | 146 | 122 | 20 | 68 | 11 | 35 | 0 | 12 | 2 | 0 | 137 | 166 | 31 | 6 | 42 | 33 | 28 |
| Conventional | 146 | 123 | 101 | 19 | 55 | 10 | 33 | 0 | 11 | 2 | 0 | 122 | 143 | 25 | 6 | 39 | 30 | 23 |
| Clear cell | 20 | 19 | 18 | 0 | 11 | 1 | 2 | 0 | 0 | 0 | 0 | 11 | 19 | 4 | 0 | 3 | 1 | 3 |
| Fibrolamellar | 4 | 4 | 3 | 1 | 2 | 0 | 0 | 0 | 1 | 0 | 0 | 4 | 4 | 2 | 0 | 0 | 2 | 2 |
| Renal cell carcinoma | 176 | 1 | 0 | 16 | 0 | 123 | 0 | 94 | 117 | 1 | 3 | 153 | 156 | 38 | 2 | 135 | 28 | 11 |
| Conventional | 117 | 0 | 0 | 6 | 0 | 94 | 0 | 70 | 55 | 1 | 2 | 100 | 103 | 12 | 2 | 92 | 9 | 5 |
| Papillary | 26 | 0 | 0 | 4 | 0 | 16 | 0 | 17 | 34 | 0 | 0 | 24 | 24 | 13 | 0 | 17 | 4 | 3 |
| Chromophobe | 16 | 0 | 0 | 2 | 0 | 6 | 0 | 2 | 8 | 0 | 0 | 16 | 16 | 9 | 0 | 16 | 13 | 3 |
| Collecting duct | 11 | 0 | 0 | 2 | 0 | 5 | 0 | 3 | 9 | 0 | 0 | 9 | 9 | 2 | 0 | 7 | 1 | 0 |
| Unclassified | 6 | 1 | 0 | 2 | 0 | 2 | 0 | 2 | 8 | 0 | 1 | 4 | 4 | 2 | 0 | 3 | 1 | 0 |
| Adrenocortical carcinoma | 40 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 31 | 27 | 22 | 4 | 4 | 0 | 0 | 1 | 0 | 0 |
| Adrenocortical adenoma | 62 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 31 | 52 | 60 | 5 | 5 | 0 | 0 | 0 | 0 | 0 |
| Hepatoblastoma | 4 | 4 | 0 | 0 | 0 | 2 | 4 | 0 | 0 | 0 | 0 | 4 | 4 | 0 | 0 | 0 | 3 | 0 |
| Cholangiocarcinoma | 18 | 0 | 0 | 14 | 0 | 4 | 0 | 0 | 0 | 0 | 0 | 18 | 18 | 18 | 0 | 18 | 10 | 13 |
| Renal oncocytoma | 7 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 6 | 6 | 6 | 0 | 6 | 0 | 1 |
| Nephroblastoma | 7 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 0 | 0 | 4 | 4 | 4 | 0 | 2 | 1 | 1 |
| Urothelial carcinoma | 20 | 0 | 0 | 13 | 0 | 8 | 0 | 0 | 0 | 0 | 0 | 20 | 20 | 20 | 8 | 20 | 2 | 18 |
| Pheochromocytoma | 21 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 16 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| Lung SCC | 14 | 2 | 0 | 12 | 0 | 4 | 0 | 0 | 0 | 1 | 0 | 14 | 12 | 5 | 0 | 13 | 3 | 12 |
| Lunc adenoca | 12 | 0 | 0 | 12 | 0 | 0 | 0 | 0 | 2 | 0 | 0 | 12 | 12 | 12 | 0 | 12 | 11 | 7 |
| Esophagus SCC | 6 | 0 | 0 | 6 | 0 | 3 | 0 | 0 | 2 | 0 | 0 | 6 | 3 | 1 | 0 | 6 | 0 | 6 |
| Stomach adenoca | 16 | 4 | 0 | 15 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 16 | 16 | 8 | 3 | 16 | 15 | 13 |
| Colorectal adenoca | 17 | 2 | 0 | 17 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 17 | 17 | 1 | 17 | 17 | 17 | 13 |
| Pancreas adenoca | 11 | 1 | 0 | 11 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 11 | 11 | 10 | 0 | 11 | 11 | 8 |
| Salivary gland tumors | 40 | 0 | 0 | 31 | 0 | 5 | 0 | 0 | 18 | 0 | 0 | 38 | 39 | 37 | 1 | 36 | 38 | 24 |
| Female genital tract | ||||||||||||||||||
| Cervix SCC | 10 | 0 | 0 | 10 | 0 | 3 | 0 | 0 | 5 | 0 | 0 | 10 | 10 | 7 | 0 | 10 | 0 | 8 |
| Cervix adenoca | 10 | 0 | 0 | 9 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 10 | 10 | 9 | 0 | 10 | 10 | 7 |
| Mucinous adenoca | 7 | 0 | 0 | 5 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 7 | 7 | 6 | 3 | 7 | 6 | 5 |
| Serous adenoca | 6 | 0 | 0 | 3 | 0 | 0 | 0 | 0 | 2 | 0 | 0 | 6 | 6 | 6 | 0 | 6 | 3 | 4 |
| Endometrial adenoca | 14 | 0 | 0 | 9 | 0 | 1 | 0 | 0 | 6 | 0 | 0 | 14 | 14 | 13 | 0 | 14 | 7 | 12 |
| Clear cell adenoca | 3 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 2 | 3 | 0 | 3 | 2 | 2 |
| Breast adenoca | ||||||||||||||||||
| Ductal | 21 | 0 | 0 | 17 | 0 | 1 | 0 | 1 | 1 | 0 | 0 | 21 | 18 | 20 | 0 | 21 | 6 | 15 |
| Lobular | 12 | 0 | 0 | 6 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 12 | 12 | 11 | 0 | 12 | 0 | 11 |
| Thymic carcinoma | 23 | 0 | 0 | 18 | 0 | 6 | 0 | 0 | 1 | 0 | 0 | 23 | 18 | 8 | 0 | 23 | 23 | 17 |
| Thymoma | 25 | 0 | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 25 | 19 | 9 | 0 | 10 | 25 | 6 |
| Prostate adenoca | 25 | 0 | 0 | 19 | 0 | 11 | 0 | 0 | 0 | 0 | 0 | 25 | 25 | 2 | 0 | 12 | 16 | 21 |
| Neuroendocrine tumor | 44 | 1 | 0 | 25 | 0 | 8 | 0 | 0 | 5 | 1 | 0 | 43 | 42 | 17 | 2 | 35 | 20 | 28 |
| Thyroid | ||||||||||||||||||
| Papillary carcinoma | 11 | 0 | 0 | 4 | 0 | 0 | 0 | 0 | 5 | 0 | 0 | 11 | 11 | 11 | 0 | 10 | 7 | 1 |
| Follicular carcinoma | 5 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 0 | 0 | 2 | 5 | 3 | 0 | 0 | 1 | 0 |
| Mesothelioma | 11 | 0 | 0 | 2 | 0 | 2 | 0 | 0 | 7 | 0 | 0 | 11 | 8 | 8 | 0 | 0 | 0 | 0 |
| Germ cell tumor | ||||||||||||||||||
| Seminoma | 11 | 0 | 0 | 3 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 4 |
| Yolk sac tumor | 8 | 0 | 0 | 3 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 8 | 0 | 0 | 0 | 0 | 7 | 4 |
| Choriocarcinoma | 4 | 0 | 0 | 3 | 0 | 4 | 0 | 0 | 0 | 1 | 0 | 4 | 4 | 4 | 0 | 2 | 2 | 0 |
| Embryonal carcinoma | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 3 | 0 | 0 | 0 | 1 | 1 |
HPA, hepatocyte antigen; pCEA, polyclonal carcinoembryonic antigen; Can, canalicular; Cy&M, cytoplasmic and membranous; AFP, «-fetoprotein; RCC Ma, RCC marker; VIM, vimentin; EMA, epithelial membrane antigen; SCC, squamous cell carcinoma; Adenoca, Adenocarcinoma.
HCC
HCC
RCC
RCC
ACC
ACC
H & E
HPA
CD10
pCEA
RCC Ma
a-Inhibin
A103
Renal Cell Carcinoma
Around 55% of the RCCs revealed immunoreactivity for RCC Ma. The positive rates in conventional and papillary RCC were higher than those in other variants. Cytoplasmic or membranous positivity for CD10 and vimentin was present in 123 (70%) and 117 (66%) cases, respectively. Most of the RCCs were diffusely positive for pan-cytokeratin, CAM5.2, and EMA. They uncommonly expressed CK7 and BG8 and rarely expressed CK20.
Adrenocortical Carcinoma
Twenty-seven (67.5%) and 22 cases (55%) showed heterogeneous (2+ or 3+) or diffuse (4+) cytoplasmic reactivity for «-inhibin and A103, respectively. The positive
rates for the two markers were higher in adrenocortical adenoma (84% for a-inhibin, 97% for A103). Immunoreac- tivity for epithelial markers was hardly detected. Around 78% of cases expressed vimentin.
Other Neoplasms
The detail of IHC expression are given in Table 2. Briefly, all hepatoblastomas were positive for HPA. HPA was detected occasionally in nonhepatic tumors, chiefly adeno- carcinomas from the digestive tract. Single exceptional cases could be labeled by RCC Ma, a-inhibin, or A103. The expression of epithelial markers was generally concordant with the nature of tumors.
Sensitivity and Specificity
Table 3 shows the selected panels we considered to be efficacious. The use of HPA by itself resulted in a sensitivity of 85.9% and a specificity of 99.5% for HCC, and the addition of canalicular pCEA and CD10 raised the final sensitivity to 94.7%. The sensitivity and specificity of RCC Ma was 54.5% and virtually 100% for RCC, respectively. Using positive CD10 and negative HPA and pCEA together increased the sensitivity to 74.4% almost without impact on the specificity. With ACC, positive a-inhibin and/or positive A103 resulted in a sensitivity of 82.5% and specificity of 98.1%.
DISCUSSION
The aim of this study was to address integrally the sensitivity and specificity of different panels of antibodies to facilitate a precise diagnosis of HCC, RCC, and ACC. Toward that end, we did not attempt to scrutinize the staining pattern of individual antibodies, but rather to access them as a whole. We believe such information can help the practicing pathologist to tailor an appropriate panel to approach the differential diagnosis.
Among the most widely used markers for HCC, HPA is considered effective. Past reports regarding this antigen revealed sensitivities for HCC ranging from 66% to 99.7%.1-5 Aside from HCC, hepatoblastoma also has a high positive rate (100% in this series). Other neoplasms that might reveal positivity include adenocarcinomas that chiefly arise from the digestive system.5 The latter tumors differ from HCC by histology and their common immunoreactivity for CK7 and/or CK20.21 Therefore, combined use of CK7 and CK20 would increase the specificity if the occasional immunoreactivity in extrahepatic tumors is a concern.
The IHC stains for pCEA and CD10 are valuable in their high specificity for HCC1,3; however, only the canalicular pattern, the thick, waxy, branching or short linear, dot-like positivity between two cells rather than within them, should be
| Diagnosis | Sensitivity (%) | Specificity (%) | |
|---|---|---|---|
| Among HCC, RCC & ACC | Among All Tumors | ||
| Hepatocellular carcinoma | |||
| A. HPA alone | 85.9 | 99.5 | 97.9 |
| B. Canalicular pCEA | 71.8 | 100 | 100 |
| C. Canalicular CD10 | 40.0 | 100 | 100 |
| A and/or B | 93.5 | 99.5 | 97.4 |
| A and/or C | 90.0 | 99.5 | 97.4 |
| A and/or B and/or C | 94.7 | 99.5 | 97.4 |
| Renal cell carcinoma | |||
| D. RCC Marker alone | 54.5 | 100 | 99.6 |
| E. CD10+ & HPA- | 61.9 | 99.0 | 97.4 |
| & pCEA- | |||
| D and/or E | 74.4 | 99.0 | 97.0 |
| Adrenocortical carcinoma | |||
| F. «-inhibin alone | 67.5 | 98.0 | 98.6 |
| G. A103 alone | 55.0 | 98.2 | 99.6 |
| F and/or G | 82.5 | 97.4 | 98.1 |
regarded as being specific for HCC. The membranous or cytoplasmic immunopositivity could be observed in many other kinds of neoplasms. If the criteria of the canalicular pattern were stringently followed, we never observed this pattern in other neoplasms. In comparison with HPA, these markers were inferior in their sensitivities. Consequently, we propose using HPA, pCEA, and CD10 as a panel. We do not think AFP is very useful because its sensitivity is too low, and adding it to the panel does not improve the overall sensitivity and specificity.
The sensitivity of RCC Ma for RCC in this series was lower than in past reports.6,7 The immunoreactivity for this marker was greatly influenced by the antigen retrieval pretreatment. After testing different methods, we found that trypsin digestion gave the best result. Different conditions of fixation and block preservation probably also affect the antigenicity. The positive rates varied among different types of RCC. The clear cell and papillary RCC are more likely to express this antigen than the chromophobe RCC.
Since RCC Ma was not as sensitive as we expected, we resorted to other panels of antibodies. CD10 is a candidate for its higher sensitivity.22 However, due to the lack of specificity, “negative” markers, such as negative staining for HPA and pCEA, should be used in parallel. A combination of positive RCC Ma and an immunoprofile of positive CD10 and negative HPA and pCEA would recognize around three fourths of RCC with a specificity up to 99% among the major differential diagnoses. The traditional profile of double immunopositivity for vimentin and cytokeratin23 was not superior to the above ones in terms of the sensitivity and specificity. We recommend substituting CD10 for vimentin as a positive marker for RCC.
Antibodies against «-inhibin and antibody A103 have been advocated as being useful in distinguishing adrenocor- tical tumors.9-12 Previously 73% to 100% of ACCs were reported as positive for «-inhibin and 50% to 100% for A103.12 Our study showed that combining the two antibodies is more effective than using only one. This immunostaining is rather specific given that other epithelial tumors rarely express the two markers. However, certain types of nonepithelial neoplasms, including sex cord stromal tumors, melanocytic tumors, and angiomyolipoma-related tumors, may reveal immunoreactivity.9,24-26 They can be discriminated from adrenocortical tumors by clinical features, histology, and IHC such as HMB-45 and S-100 protein.
We also tested a series of epithelial markers, including cytokeratins and non-keratin antigens recognized by anti- bodies MOC31 and BG8. These markers have been suggested as valuable in discerning HCC from cholangiocarcinoma and metastatic adenocarcinoma, given the higher positive rates for MOC31 in latter two compared with HCC.3 However, our study showed that these markers were not absolutely discriminatory: a small proportion of HCC was also positive. The difference is more quantitative than qualitative, so its diagnostic utility is hampered with limited biopsy specimens.
The tissue array technique is advantageous to the present comprehensive study. One may question whether tumor heterogeneity was fully addressed. Past reports in this regard did not disclose significant discrepancies between the results obtained from conventional slides and tissue array slides.13 In
this study, the 2-mm-diameter tissue cylinders in our manually constructed tissue array blocks were significantly larger than the 0.6-mm-diameter cylinders produced by the commercial tissue arrayer. Furthermore, we took multiple samples to overcome the problems raised by the tumor heterogeneity.
In conclusion, we have shown an effective model in which we systemically evaluated the efficacy of different panels for the diagnoses of HCC, RCC, and ACC. The tissue array technique can be adopted by ordinary pathology laboratories for setting up similar protocols according to their own resources and requirements.
ACKNOWLEDGMENTS
The authors thank I-Chung Wang for his technical assistance and Gerald W. Kamp for editing the manuscript.
REFERENCES
1. Chu PG, Ishizawa S, Wu E, et al. Hepatocyte antigen as a marker of hepatocellular carcinoma: an immunohistochemical comparison to car- cinoembryonic antigen, CD10, and alpha-fetoprotein. Am J Surg Pathol. 2002;26:978-988.
2. Fan Z, Van De RM, Montgomery K, et al. Hep Par 1 antibody stain for the differential diagnosis of hepatocellular carcinoma: 676 tumors tested using tissue microarrays and conventional tissue sections. Mod Pathol. 2003;16:137-144.
3. Morrison C, Marsh W Jr, Frankel WL. A comparison of CD10 to pCEA, MOC-31, and hepatocyte for the distinction of malignant tumors in the liver. Mod Pathol. 2002;15:1279-1287.
4. Murakata LA, Nzeako UC. Clear cell carcinoma of the liver: a comparative immunohistochemical study with renal clear cell carcinoma. Mod Pathol. 2000;13:874-881.
5. Kakar S, Muir T, Murphy LM, et al. Immunoreactivity of Hep Par 1 in hepatic and extrahepatic tumors and its correlation with albumin in situ hybridization in hepatocellular carcinoma. Am J Clin Pathol. 2003; 119:361-366.
6. Avery AK, Beckstead J, Renshaw AA, et al. Use of antibodies to RCC and CD10 in the differential diagnosis of renal neoplasms. Am J Surg Pathol. 2000;24:203-210.
7. McGregor DK, Khurana KK, Cao C, et al. Diagnosing primary and metastatic renal cell carcinoma: the use of the monoclonal antibody “Renal Cell Carcinoma Marker.” Am J Surg Pathol. 2001;25:1485-1492.
8. Yoshida SO, Imam A. Monoclonal antibody to a proximal nephrogenic renal antigen: immunohistochemical analysis of formalin-fixed, paraffin- embedded human renal cell carcinomas. Cancer Res. 1989;49:1802-1809.
9. Busam KJ, Iversen K, Coplan KA, et al. Immunoreactivity for A103, an antibody to melan-A (Mart-1), in adrenocortical and other steroid tumors. Am J Surg Pathol. 1998;22:57-63.
10. Renshaw AA, Granter SR. A comparison of A103 and inhibin reactivity in adrenal cortical tumors: distinction from hepatocellular carcinoma and renal tumors. Mod Pathol. 1998;11:1160-1164.
11. Ghorab Z, Jorda M, Ganjei P, et al. Melan A (A103) is expressed in adrenocortical neoplasms but not in renal cell and hepatocellular carcinomas. Appl Immunohistochem Mol Morphol. 2003;11:330-333.
12. Zhang PJ, Genega EM, Tomaszewski JE, et al. The role of calretinin, inhibin, melan-A, BCL-2, and C-kit in differentiating adrenal cortical and medullary tumors: an immunohistochemical study. Mod Pathol. 2003; 16:591-597.
13. Bubendorf L, Nocito A, Moch H, et al. Tissue microarray (TMA) technology: miniaturized pathology archives for high-throughput in situ studies. J Pathol. 2001;195:72-79.
14. Kononen J, Bubendorf L, Kallioniemi A, et al. Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat Med. 1998;4:844-847.
15. Hirohashi S, Ishak KG, Kojiro M, et al. Hepatocellular carcinoma. In: Hamilton SR, Aaltonen LA, eds. Pathology and Genetics of Tumours of the Digestive System. Lyon: IARC Press, 2000.
16. Kovacs G, Akhtar M, Beckwith BJ, et al. The Heidelberg classification of renal cell tumours. J Pathol. 1997;183:131-133.
17. Storkel S, Eble JN, Adlakha K, et al. Classification of renal cell carcinoma: Workgroup No. 1. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer. 1997;80:987-989.
18. Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol. 1984;8:163- 169.
19. Weiss LM, Medeiros LJ, Vickery AL Jr. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol. 1989;13:202- 206.
20. Pan CC, Chen PC, Chiang H. An easy method for manual construction of high-density tissue arrays. Appl Immunohistochem Mol Morphol. 2004; 12:370-372.
21. Chu P, Wu E, Weiss LM. Cytokeratin 7 and cytokeratin 20 expression in epithelial neoplasms: a survey of 435 cases. Mod Pathol. 2000;13:962- 972.
22. Chu P, Arber DA. Paraffin-section detection of CD10 in 505 non- hematopoietic neoplasms. Frequent expression in renal cell carcinoma and endometrial stromal sarcoma. Am J Clin Pathol. 2000;113:374-382.
23. Medeiros LJ, Michie SA, Johnson DE, et al. An immunoperoxidase study of renal cell carcinomas: correlation with nuclear grade, cell type, and histologic pattern. Hum Pathol. 1988;19:980-987.
24. Deavers MT, Malpica A, Liu J, et al. Ovarian sex cord-stromal tumors: an immunohistochemical study including a comparison of calretinin and inhibin. Mod Pathol. 2003;16:584-590.
25. Busam KJ, Jungbluth AA. Melan-A, a new melanocytic differentiation marker. Adv Anat Pathol. 1999;6:12-18.
26. Tuna EB, Lebe B, Yorukoglu K. H MB 45 and Melan-A expression in renal angiomyolipoma and their significance for the diagnosis. Tumori. 2003;89:46-48.