Original Article
Oncocytic adrenocortical carcinomas: A pathological and immunohistochemical study of four cases in comparison with conventional adrenocortical carcinomas
Sang Yong Song,1 Sunhoo Park,2 Seong Rim Kim1* and Yeon-Lim Suh1
1Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine and 2Department of Pathology, Korea Cancer Center Hospital, Seoul, Korea
Clinicopathological features of four cases of oncocytic adrenocortical carcinomas were studied. All tumors were large, circumscribed tumors with average size and weight of 11.5 cm and 586 g, respectively. The cut surfaces were yellow or brown and tan with areas of hemorrhage, necro- sis, fibrosis, myxoid and cystic change. The tumor cells were exclusively oncocytic with a diffuse or compact and solid arrangement. Nuclear atypia was identified but mitosis was rare. Capsular invasion was identified in all tumors and vascular invasion was identified in one tumor. All tumors were immunoreactive for vimentin and inhibins. Immunore- activity for pancytokeratin, synaptophysin and S-100 pro- tein was variable and focal. All tumors had low proliferative indices, of less than 1%, and were negative for p53 protein. Ultrastructurally, the cytoplasm of tumor cells showed numerous mitochondria in a compact arrangement. Onco- cytic adrenocortical carcinomas showed a similar sex ratio, slightly older mean age, similar left predilection, slightly smaller size and lighter weight compared with the conven- tional carcinomas. We suggest that most oncocytic adreno- cortical carcinomas might be low-grade malignancies with less aggressive histological features compared with con- ventional carcinomas. However, they should be excised completely because of the likelihood of recurrence and metastasis during the follow-up period.
Key words: adrenal gland neoplasms, immunohistochemistry, pathology
Oncocytic adrenocortical carcinomas are rare, non-functional tumors found in adults. Oncocytic neoplasms are comprised
exclusively of oncocytic tumor cells, which are characterized by abundant granular eosinophilic cytoplasm by light micros- copy.1 To date, 33 cases of oncocytic adrenocortical neoplasms have been reported, which comprise 24 oncocy- tomas, two oncocytic neoplasms of uncertain malignant potential and seven oncocytic carcinomas.2-19 Among the reported cases, two cases of oncocytomas occurred in the heterotopic retroperitoneal adrenal tissue.3,16 In addition, all were non-functional except for one, which caused viriliza- tion.5 Only seven cases of oncocytic carcinomas have been reported with fine needle aspiration findings, immunohisto- chemical and ultrastructural studies.4,7,11,12 Despite a com- prehensive study by Hoang et al.,7 the pathological and immunohistochemical features of oncocytic adrenocortical carcinomas have not been clearly characterized.
Herein, we present the clinical, histopathological, immuno- histochemical and ultrastructural findings of four cases of oncocytic adrenocortical carcinomas. All cases were studied extensively using various immunohistochemical markers and were compared with the histopathological characteristics of conventional adrenocortical carcinomas.
MATERIALS AND METHODS
The computerized pathological records of the Department of Pathology, Samsung Medical Center, between August 1994 and August 2003 and the Department of Pathology, Korean Cancer Center Hospital, between January 1993 and August 2003 were searched for oncocytic and conventional adreno- cortical neoplasms. Four cases of oncocytic adrenocortical carcinomas were selected on the basis of being composed exclusively of oncocytic tumor cells, which were reminiscent of renal oncocytoma. No initial chemotherapy or radio- therapy was performed before tumor excision. Clinical information was obtained by chart review. Hematoxylin- eosin-stained slides were carefully reviewed and evaluated
Correspondence: Yeon-Lim Suh, MD, PHD, Department of Pathology, Samsung Medical Center, 50 Irwon-dong Gangnam-gu, Seoul 135- 710, Korea. Email: yodasong@smc.samsung.co.kr
*Present address: Medplan Pathology Laboratory Center, #401 Technoplaza building, 149-10 Yatab-dong, Bundang-gu, Seongnam, Gyeonggi-do 463-816, Korea.
Received 4 November 2003. Accepted for publication 3 April 2004.
for features associated with the well-known criteria of malignancy.20,21
Immunohistochemical studies were performed using formalin-fixed, paraffin-embedded tissue sections and a microwave-induced antigen retrieval method with citrate buffer (pH 6.0). Avidin-polymer peroxidase complex tech- nique (EnVison, Dako, Carpinteria, CA, USA) was performed using an automated immunostainer (TechMate 1000, Dako, Glostrup, Denmark). Detailed information of the primary anti- bodies is listed in Table 1. The immunostaining was devel- oped using 3,3’-diaminobenzidine tetrahydrochloride as the chromogen. The slides were counterstained with Mayer’s hematoxylin. The negative control was processed in the same manner as the test staining, with omission of the pri- mary antibody.
A positive result for p53 protein was defined as more than 5% cells with distinct nuclear staining. A positive result for E-cadherin was defined as cells with distinct membra- nous staining.
For electron microscopy examination, the tissue samples of two out of four oncocytic tumors were fixed in 3% glutaral- dehyde and postfixed in 1% osmium tetroxide, and the thin sections were stained with lead citrate and uranyl acetate. All grids were examined with a transmission electron micro- scope (H-7100, Hitachi, Tokyo, Japan).
RESULTS
Clinical data
Case 1
A 64-year-old woman presented with an incidentally found left abdominal mass. An abdominal computed tomography (CT) scan revealed a huge left adrenal mass without evi- dence of invasion to the adjacent structures. There was no clinical evidence that the neoplasm was functional. The patient underwent left radical nephrectomy including adrena- lectomy. An abdominal CT scan performed every 6 months post-surgery showed no evidence of residual or recurrent disease for 11 years after diagnosis.
Case 2
A 47-year-old woman presented with a 4 week history of left flank pain. Abdominal CT and magnetic resonance imaging (MRI) revealed a huge left adrenal mass without evidence of invasion to the adjacent structures. There was no clinical evidence that the neoplasm was functional. The patient underwent left adrenalectomy. She developed a retroperito- neal mass and celiac lymph node metastases 10 months later. Wide excision of the retroperitoneal mass, splenectomy, partial pancreatectomy and celiac lymph node dissection were performed. The patient developed a suprarenal mass and a peritoneal nodule 2 years after diagnosis. Radical nephrectomy and excision of the peritoneal nodule were per- formed. The patient is alive with no evidence of residual or metastatic disease for 2 years and 10 months after diagnosis.
Case 3
A 37-year-old woman presented with a 10 day history of left flank pain. An abdominal CT revealed a large left adrenal mass without evidence of invasion to the adjacent structures. There was no clinical evidence that the neoplasm was func- tional. The patient underwent left adrenalectomy. Abdominal CT performed every year post-surgery showed no evidence of residual or recurrent disease for 2 years and 11 months after diagnosis.
Case 4
A 35-year-old man presented with an incidentally found right abdominal mass. An abdominal CT revealed a large right adrenal mass without evidence of invasion to the adjacent structures. There was no clinical evidence that the neoplasm was functional. The patient underwent right adrenalectomy. Abdominal CT performed every year post-surgery showed no evidence of residual or recurrent disease for 5 years after diagnosis.
PATHOLOGICAL FINDINGS
The clinicopathological findings are summarized in Table 2. Macroscopically, all four oncocytic adrenocortical carcinomas
| Antibody | Clone | Dilution | Source |
|---|---|---|---|
| Pancytokeratin | AE1/AE3 | 1:80 | Zymed, South San Francisco, CA, USA |
| Vimentin | VIM | 1:400 | Dako, Carpinteria, CA, USA |
| Chromogranin | DAK-A3 | 1:2000 | Dako, Carpinteria, CA, USA |
| Synaptophysin | SY38 | 1:40 | Dako, Carpinteria, CA, USA |
| S-100 protein | Polyclonal | 1:200 | Dako, Carpinteria, CA, USA |
| Inhibin a | R1 | 1:50 | Serotec, Oxford, UK |
| Inhibin ß | E4 | 1:40 | Serotec, Oxford, UK |
| p53 protein | BP53.12 | 1:400 | Zymed, South San Francisco, CA, USA |
| Ki-67 | MIB1 | 1:50 | Dako, Carpinteria, CA, USA |
| E-cadherin | 4A2C7 | 1:80 | Dako, Carpinteria, CA, USA |
| Case 1 | Case 2 | Case 3 | Case 4 | |
|---|---|---|---|---|
| Sex | Female | Female | Female | Male |
| Age (years) | 64 | 47 | 37 | 35 |
| Presentation | Incidental mass | Flank pain | Flank pain | Incidental mass |
| Hormonal manifestation | No | No | No | No |
| Side | Left | Left | Left | Right |
| Size (cm) | 11 | 15 | 11.5 | 8.5 |
| Weight (kg) | 0.42 | 1.22 | 0.41 | 0.29 |
| Nuclear grade | Focally high | Focally high | Focally high | Focally high |
| Mitosis/10 HPF | ~1 | ~1 | ~1 | ~1 |
| Growth pattern | Diffuse nesting | Diffuse | Diffuse | Diffuse |
| Necrosis | Present | Present | Present | Absent |
| Vessel invasion | Absent | Present | Absent | Absent |
| Capsule invasion | Present | Present | Present | Present |
| Fibrous band | Absent | Present | Absent | Absent |
| Positive numbers of Weiss criteriat | 5 | 6 | 5 | 4 |
| Van Slooten index# | 15.3 | 15.3 | 15.3 | 15.3 |
| Follow up | 1 year NED | 2 years REC/MET 2 years 10 months NED | 2 years NED | 5 years NED |
HPF, high-power fields; MET, metastasis; NED, no evidence of disease; REC, recurrence. tPresence of three or more criteria highly correlates with subsequent malignant behavior. Histological index of more than 8 correlates with subsequent malignant behavior.
showed large, circumscribed tumors with a thin rim of normal adrenal gland along one edge. The tumors measured 8.5, 11.0, 11.5 and 15.0 cm in the greatest dimension, with a mean of 11.5 cm, and weighed 290, 414, 420 and 1220 g, with a mean of 586 g. The cut surfaces were yellow (cases 3 and 4) or brown and tan (cases 1 and 2) with areas of hemorrhage (cases 1, 3 and 4), necrosis (cases 1, 2 and 3), central fibrosis, myxoid degeneration and cystic change (case 2) (Fig. 1).
Microscopically, the adrenal gland was almost completely replaced by a well-circumscribed, partially encapsulated neo- plasm that compressed the residual normal adrenal gland. The neoplastic cells were exclusively oncocytes containing abundant, eosinophilic, and granular cytoplasm (Fig. 2). In all four cases, the tumor cells were arranged in a diffuse or
compact and solid pattern. Additional patterns including a small nesting (case 1) or trabecular (cases 2 and 4) arrange- ment of tumor cells were also identified. Foci of edematous or myxoid stroma were present (cases 1 and 2). One of these foci was associated with a central fibrous band and cystic change (case 2).
Nuclear atypia was identified in all four cases, with varying degrees and focal extent. In cases 1 and 3, nuclear atypism was mild to moderate and the extent was less than 1% of tumor cells. Some tumor cells showed hyperchromatic nuclei and binucleation. In cases 2 and 4, bizarre tumor cells with marked nuclear atypia and giant cytoplasm were found
focally in less than 10% of tumor cells (Fig. 3). Mitoses of all four cases were extremely rare. Only one mitosis per 50 high power fields (HPF) was identified after extensive searching of all four cases. Vascular invasion was identified in case 2 (Fig. 4) and capsular invasion was identified in all four. Foci of foamy tumor cell aggregation were noted in case 4.
Evidence of malignant criteria for adrenocortical carcinoma was based on the counting system of Weiss21 and the index scoring of Van Slooten et al.20 (Table 2).
Immunohistochemical findings
All tumors were immunoreactive for vimentin (Fig. 5A). Immu- noreactivity for high and low molecular weight cytokeratin (25%) (Fig. 5B), synaptophysin (75%) (Fig. 5C) and S-100 protein (25%) was variable and focal. Inhibin & and ß were focally (5-20%) and weakly to moderately positive in all four tumors (Fig. 5D). Inhibin ß showed stronger immunoreactivity than inhibin a. Using the anti-Ki-67 antibody (MIB1), the proliferative index was calculated for each tumor. All four adrenocortical carcinomas showed low proliferative indices with less than 1% of neoplastic cells. All four tumors were negative for p53 protein. Some tumors were totally negative for p53 protein while some tumors showed a few positive cells with less than 5% of tumor cells. The tumors were negative for E-cadherin.
Ultrastructural findings
The cytoplasm of the tumor cells showed compactly arranged numerous mitochondria (Fig. 6). In some cells, the mitochon- dria were more loosely arranged. The pattern of cristae was
variable: flat, lamellar, vesicular and tubular. Golgi apparatus, lysosomes and rough endoplasmic reticulums were noted in the perinuclear area (Fig. 6). Small dense bodies, lipid drop- lets and smooth endoplasmic reticulum were rarely identified.
Comparison of clinicopathological features between oncocytic and conventional adrenocortical carcinoma
The clinicopathological findings of the oncocytic adrenocor- tical carcinomas of the four cases in the present study and seven previously reported cases were compared to the 12 conventional adrenocortical carcinomas and summarized in Table 3. There was no predilection for male patients in oncocytic carcinoma (6:5), although conventional carcinoma showed a male predilection (2:1). The mean age of the patients with oncocytic carcinoma was 52.8 years, which was slightly older than that of patients with the conventional car- cinoma of 47.7 years. The left adrenal gland was involved more frequently in both oncocytic (8:3) and conventional (5:1) carcinoma. The oncocytic carcinoma (12.3 cm in mean great- est dimension and 626.3 g in mean weight) was slightly smaller and lighter than the conventional carcinoma (13.3 cm in greatest dimension and 692.8 g in weight).
Histologically, the oncocytic carcinomas differed somewhat from the conventional carcinomas. The oncocytic carcinomas showed extremely rare mitoses including no atypical ones, and no sinusoidal invasion, whereas the conventional carci- nomas showed frequent mitoses with more than 5/50 HPF (66.7%), atypical mitoses (41.7%) and sinusoidal invasion (66.7%). Other histopathological parameters related with malignancy were less frequent in the oncocytic carcinomas than in the conventional carcinomas: necrosis, 80.0% versus 91.7%; venous invasion, 40% versus 75%; capsular invasion,
a
b
c
d
70% versus 100%; and fibrosis, 11.1% versus 83.3%. The incidence of a diffuse growth pattern was the same in the oncocytic and conventional carcinomas. There was a higher amount of bizarre tumor cells in the oncocytic carcinomas (50% of the cases in the present study) than in the conven- tional carcinomas (25%), but the proportion of these cells was always focal in the oncocytic carcinomas and diffuse in the conventional carcinomas. The main pattern of tumor cell arrangement in the oncocytic carcinomas was compact and solid growth, whereas the conventional carcinomas were trabecular and compact, and solid. Additional arrangement patterns in the conventional carcinomas were tubular and hemangiopericytomatous.
Immunohistochemically, there was some difference between the oncocytic and conventional carcinomas (Table 4). The oncocytic carcinomas were frequently positive for synaptophysin (50%), although their intensity was weak and the proportion focal. Only 33% of conventional carcino- mas showed focal and weak immunoreactivity only for syn-
aptophysin. There was no immunoreactivity for chromogranin in both the oncocytic and conventional carcinomas. Two out of 12 conventional carcinomas showed immunoreactivity for p53 protein, although there was only one p53-positive case in the oncocytic carcinomas. Two out of 12 conventional carcinomas showed 5-10% in the proliferative index. There were no oncocytic tumors with MIB1-positive tumor cells more than 5%. Inhibin was more frequently expressed in the conventional carcinomas (100%) than in the oncocytic carci- nomas (62.5%). Furthermore, the intensity and proportion of inhibin expression was weak and focal in the oncocytic carcinomas.
DISCUSSION
The majority of oncocytic neoplasms of the adrenal gland were considered to be benign and were diagnosed as oncocytoma.10,13 In some cases, the decision of malignancy
of adrenal oncocytic neoplasm was difficult. Kurek et al. reported a case of oncocytic adrenocortical carcinoma that had initially been diagnosed as an oncocytic adrenocortical carcinoma of uncertain malignant potential.11 In fact, case 2 in our series was initially diagnosed as an oncocytic adreno- cortical neoplasm of uncertain malignant potential because the histological features of capsular and vascular invasion were not typical in the first biopsy. Furthermore, there was no mitosis despite an extensive search on enough samples of the tumor.
Various systems have been devised to distinguish benign from malignant adrenal cortical tumors.22 According to the so-called Weiss system, nine pathological findings that were most common in adrenocortical carcinomas were: high nuclear grade or atypia; more than 75% of eosinophilic tumor cell cytoplasm; more than 33% of diffuse architecture; and the presence of necrosis, mitotic figures of more than 5/50 HPF, atypical mitotic figures, capsular invasion, venous inva- sion and sinusoidal invasion.21,23 The presence of three or more of the above findings highly correlates with subsequent malignant behavior. Van Slooten et al. proposed a system of seven histological criteria with different numeric values: mitotic activity of more than 2/10 HPF, 9.0; extensive regres- sive changes such as necrosis, hemorrhage, fibrosis and calcification, 5.7; abnormal nucleoli, 4.1; vascular or capsular invasion, 3.3; moderate to marked hyperchromasia, 2.6; moderate to marked nuclear atypia, 2.1; and loss of normal structure, 1.6.20 In Van Slooten et al’s system, patients with a histological index (i.e. a summation of each numeric value) less than 8 more often had smaller neoplasms and remained free of metastases during the 10 year clinical follow-up period.
It is hard to decide whether the Weiss system is applicable to oncocytic adrenocortical neoplasm or not because every oncocytic adrenocortical tumor contains at least two histolog- ical findings associated with malignancy in the Weiss system: eosinophilic tumor cell cytoplasm and diffuse architecture.
| Oncocytic adrenocortical carcinoma (n= 11) | Conventional adrenocortical carcinoma (n= 12) | |
|---|---|---|
| Male : Female | 6:5 | 8:4 |
| Mean age (range) | 52.8 (35-74) years | 47.7 (32-75) years |
| Left : Right | 8:3 | 10:2 |
| Mean size (range) | 12.3 (8-17) cm | 13.3 (6.5-24) cm |
| Weight (range) | 626.3 (100-1220) g | 692.8 (109-2200) g |
| Mitosis >5/50 HPF | 0% | 66.7% |
| Atypical mitosis | 0% | 41.7% |
| Eosinophilic tumor cellst | 100% | 33.3% |
| Diffuse growth pattern | 100% | 100% |
| Necrosis | 80.0% | 91.7% |
| Venous invasion | 40.0% | 75% |
| Sinusoidal invasion | 0% | 66.7% |
| Capsular invasion | 70.0% | 100% |
| Fibrosis | 11.1% | 83.3% |
HPF, high-power fields. tEosinophilic tumor cells more than 75% of tumor cells.
| Antibody | Present | Hoang et al.7 | Kurek et al.11 | El-Naggar et al.4 | Oncocytic carcinoma | Oncocytic neoplasms8 |
|---|---|---|---|---|---|---|
| Pancytokeratin | 1/4 | 3/4 | Not done | Not done | 4/8 (50%) | 4/5 (80%) |
| Vimentin | 4/4 | 4/4 | 1/1 | 1/1 | 10/10 (100%) | 5/7 (41.4%) |
| Chromogranin | 0/4 | 0/2 | 0/1 | Not done | 0/7 (0%) | 0/7 (0%) |
| Synaptophysin | 3/4 | 1/3 | 0/1 | Not done | 4/8 (50%) | 5/7 (41.4%) |
| S-100 protein | 1/4 | 0/2 | Not done | Not done | 1/6 (16.7%) | 0/7 (0%) |
| Inhibin (a/B) | 4/4 | 1/4 | Not done | Not done | 5/8 (62.5%) | Not done |
| p53 protein | 0/4 | 1/2 | Not done | Not done | 1/6 (16.7%) | 0/7 (0%) |
| Ki-67 (MIB1) | <1% | ~5% | Not done | Not done | 0-5% | 0.1-9.5% |
| E-cadherin | 0/4 | Not done | Not done | Not done | 0/4 (0%) | Not done |
Only one more histological feature will justify the diagnosis of malignancy. Thus, Lin et al. suggested that oncocytic adrenocortical neoplasms should be assessed conserva- tively in the absence of mitotic activity, necrosis or invasion.13 This suggestion contributed to the initial diagnosis of uncer- tain malignant potential in case 2 in the present series. Because of the possible recurrence or metastasis in bland- looking oncocytic neoplasm, Kurek et al. recommended that all oncocytic adrenocortical tumors of uncertain malignant potential should be considered to be carcinomas.11
Based on the results of the present study, mitotic figures of more than 5/50 HPF (0%), atypical mitotic figures (0%), sinusoidal invasion (0%) and fibrosis (11.1%) were uncom- mon in oncocytic adrenocortical carcinomas. Among the his- tological criteria of the Weiss system, capsular invasion (70%) and necrosis (80%) are easily identified in oncocytic adrenocortical carcinomas. Cellular proliferation in oncocytic carcinomas seems very low because the mitotic count was always less than 5/50 HPF and the proliferative index in the immunohistochemical study was always less than 1% of tumor cells in the present study. Therefore, we suggest that a modified system be applied to the assessment of malig- nancy of oncocytic adrenocortical carcinomas, with less emphasis on the presence of mitosis. These tumors were usually large in size and had a low mitotic count, diffuse growth pattern, frequent necrosis, hemorrhage and capsular invasion, and rare sinusoidal invasion and fibrosis. Among the seven cases that were followed for at least 1 year, three showed recurrence or metastasis. To date, there have been no cancer-related deaths of oncocytic adrenocortical carci- noma. Thus, we suggest that oncocytic adrenocortical carci- noma might be a low-grade malignancy.
The immunohistochemical profile of oncocytic adrenocor- tical carcinoma was not strikingly different from that of non-malignant oncocytic adrenocortical neoplasms. In the present study, the expression rates of pancytokeratins were higher in non-malignant neoplasms than in carcinoma. How- ever, the expression rates of vimentin, S-100 protein and p53 protein were reversed. Chromogranin was always negative in oncocytic adrenocortical neoplasms. Inhibin, which has recently been shown to stain the majority of adrenocortical tumors,24 was also shown in all cases in the present study.
However, the proportion and intensity was focal (less than 20%) and weak to moderate, which was different from the diffuse or strong staining patterns of conventional adrenocor- tical neoplasms described by our colleagues.25
p53 protein and the proliferative index have been reported to be potential prognostic indicators of adrenocor- tical neoplasms.26 To date, there has only been one case of oncocytic adrenocortical carcinoma with p53 expression reported in English literature.7 There has been controversy as to whether or not the proliferative index is useful for the prediction of oncocytic adrenocortical carcinomas.7,10,13 In the results in the present study, there were no remarkable data to contribute the significance of p53 and the prolifera- tive index as a prognostic indicator in oncocytic adrenocor- tical carcinomas. Thus, the role of either p53 or the proliferative index as biomarkers still remains uncertain due to the limitation of accumulated case numbers and the fol- low-up period.
We described four cases of oncocytic adrenocortical carcinomas with detailed clinicopathological, immunohis- tochemical and ultrastructural findings. The follow-up data and histological and immunohistochemical comparisons with conventional carcinoma were also described with a literature review. We suggest that oncocytic adrenocortical carcinoma may be a low-grade malignancy and should be excised completely.
REFERENCES
1 Chang A, Harawi SJ. Oncocytes, oncocytosis and oncocytic tumors. Pathol Annu 1992; 27: 263-304.
2 Begin LR. Adrenocortical oncocytoma: case report with immu- nohistochemical and ultrastructural study. Virchows Arch 1992; 421: 533-7.
3 Corsi A, Riminucci M, Petrozza V et al. Incidentally detected giant oncocytoma arising in retroperitoneal heterotopic adrenal tissue. Arch Pathol Lab Med 2002; 126: 1118-22.
4 El-Naggar AK, Evans DB, Mackay B. Oncocytic adrenal cortical carcinoma. Ultrastruct Pathol 1991; 15: 549-56.
5 Erlandson RA, Reuter VE. Oncocytic adrenal cortical ade- noma. Ultrastruct Pathol 1991; 15: 549-56.
6 Gandras EJ, Schwartz LH, Panicek DM et al. Adrenocortical oncocytoma: CT and MRI findings. J Comput Assist Tomogr 1996; 20: 407-9.
7 Hoang MP, Ayala GA, Albores-Saavedra J. Oncocytic adreno- cortical carcinoma: a morphologic, immunohistochemical and ultrastructural study of four cases. Mod Pathol 2002; 15: 973-8.
8 Kakimoto S, Yushita Y, Sanefuji T et al. Non-hormonal adreno- cortical adenoma with oncocytoma-like appearances. Acta Urol Jpn 1986; 32: 757-63.
9 Kitching PA, Patel V, Harach R. Adrenocortical oncocytoma. J ☒ Clin Pathol 1999; 52: 151-3.
10 Krishnamurthy S, Ordonez NG, Shelton TO et al. Fine-needle aspiration cytology of a case of oncocytic adrenocortical carci- noma. Diagn Cytopathol 2000; 22: 299-303.
11 Kurek R, Knobloch VR, Feek U et al. Local recurrence of an oncocytic adrenocortical carcinoma with ovary metastasis. J ☒ Urol 2001; 166: 985.
12 Lazaro Santander R, Andres Gozalbo C, Cortes Vizcaino V et al. Low-grade oncocytic adrenal carcinoma. Arch Esp Urol 2001; 54: 1123-6.
13 Lin BTY, Bonsib SM, Mierau GW et al. Oncocytic adrenocorti- cal neoplasms: a report of seven cases and review of the literature. Am J Surg Pathol 1998; 22: 603-14.
14 Macchi C, Rebuffat P, Blandamura S et al. Adrenal oncocy- toma: case report and review of the literature. Tumori 1998; 84: 403-7.
15 Muir TE, Ferreiro JA, Carney JA. Oncocytoma of the adrenal gland (Abstract). Mod Pathol 1996; 9: 50A.
16 Nguyen GK, Vriend R, Ronaghan D et al. Heterotopic adreno- cortical oncocytoma. A case report with light and electron microscopic studies. Cancer 1992; 70: 2681-4.
17 Sasano H, Suzuki T, Sano T et al. Adrenocortical oncocytoma. A true nonfunctioning adrenocortical tumor. Am J Surg Pathol 1991; 15: 949-56.
18 Segal S, Cytron S, Shenhav S et al. Adrenocortical oncocytoma in pregnancy. Obstet Gynecol 2001; 98: 916-18.
19 Walters PR, Haselhuhn GD, Gunning WT et al. Adrenocortical oncocytoma: two case reports and review of the literature. Urology 1997; 49: 624-8.
20 Van Slooten H, Schaberg A, Smeenk D et al. Morphologic characteristics of benign and malignant adrenocortical tumors. Cancer 1985; 55: 766-73.
21 Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984; 8: 163-9.
22 Medeiros LJ, Weiss LM. New developments in the pathologic diagnosis of adrenal cortical neoplasms: a review. Am J Clin Pathol 1992; 97: 73-83.
23 Weiss LM, Medeiros LJ, Vickery AL. Pathologic features of prognostic significance in adrenocortical tumors. Am J Surg Pathol 1989; 13: 202-6.
24 Cho EY, Ahn GH. Immunoexpression of inhibin o-subunit in adrenal neoplasms. Appl Immunohistochem Mol Morph 2001; 9: 222-8.
25 Bostwick DG, Qian J, Ramnani DM. Immunohistochemistry of the prostate and bladder, testis, and renal tumors. In: Dabbs DJ, ed. Diagnostic Immunohistochemistry. Philadelphia: Churchill Livingstone, 2002; 407-85.
26 Vargas MP, Vargas HI, Kleiner DE et al. Adrenocortical neo- plasms: role of prognostic markers MIB-1, p53, and RB. Am J Surg Pathol 1997; 21: 556-62.