THE AMERICAN JOURNAL OF SURGICAL PATHOLOGY
D. Wolters Kluwer
Oncocytic Adrenocortical Neoplasms: A Report of Seven Cases and Review of the Literature
Lin Bryan T .- Y. M.D. Ph.D .; Bonsib, Stephen M. M.D .; Mierau, Gary W. Ph.D .; Weiss, Lawrence M. M.D .; Medeiros, L. Jeffrey M.D.
The American Journal of Surgical PathologyAmerican Journal of Surgical Pathology. 22:p 603-614, May 1998.
Author Information
From the Division of Pathology (B.T-Y.L., L.M.W., L.J.M.), City of Hope National Medical Center, Duarte, California, USA; Department of Pathology, University of Arkansas for Medical Sciences (S.M.B.), Little Rock, Arkansas, USA; and Department of Pathology, The Children’s Hospital (G.W.M.), Denver, Colorado, USA.
Address correspondence and reprint requests to Dr. L.J. Medeiros, Division of Pathology, City of Hope National Medical Center, 1500 East Duarte Road, Duarte, CA 91010-0269, USA.
Abstract
Oncocytic neoplasms of the adrenal gland are rare. We describe the clinicopathologic and immunohistochemical findings of seven oncocytic adrenocortical neoplasms, five oncocytomas, and two oncocytic neoplasms of uncertain malignant potential. Three tumors were studied using electron microscopy. These neoplasms occurred in five women and two men (median age, 55 years) with no clinical evidence that the neoplasms were functional. The size of the neoplasms varied from 5.0 cm to 13.5 cm. Histologically, each neoplasm was composed exclusively of oncocytes. The oncocytomas had very low or absent mitotic activity and no evidence of necrosis. The two oncocytic neoplasms of uncertain malignant potential had increased mitotic activity and necrosis but no evidence of invasion or metastases. Nuclear atypia, either focal or generalized, was found in all neoplasms. Immunohistochemical studies performed using fixed, paraffin-embedded sections showed strong reactivity with the mitochondrial antibody mES-13 in all neoplasms. Four of five oncocytomas and one oncocytic neoplasm of uncertain malignant potential expressed keratin, predominantly keratin 18, as shown using the CAM 5.2 and AE3 antibodies. Two neuroendocrine-associated markers, neuron specific enolase and synaptophysin, were positive in seven and five
neoplasms, respectively. However, all neoplasms were negative for the other neuroendocrine markers tested, including chromogranin A, tyrosine hydroxylase, and dopamine ß-hydroxylase, as well as for epithelial membrane antigen, S100, and p53. Using the MIB-1 (Ki-67) antibody, proliferative activity was increased in both oncocytic neoplasms of uncertain malignant potential. All six patients with available clinical follow-up data are alive without evidence disease, although the follow-up interval is relatively short (<2 years) for the two patients with oncocytic neoplasms of uncertain malignant potential. We conclude that oncocytic adrenocortical neoplasms are nonfunctional tumors that can become large before they are detected by radiologic studies. The majority of neoplasms are benign and should not be misdiagnosed as carcinoma.
Oncocytomas are neoplasms that histologically are composed of epithelial cells with abundant acidophilic, granular cytoplasm that can be arranged in alveolar, tubular, or solid patterns.4 Electron microscopic studies of oncocytomas have shown that the cytoplasm of oncocytes is packed with mitochondria.4,8 Oncocytomas most commonly occur in the kidney, the salivary gland, and the endocrine organs, but these tumors also have been reported in a variety of anatomic sites.4
Nine oncocytomas and one oncocytic carcinoma that occurred in the adrenal gland have been reported in the English literature.2,5-7,11,21,25 In addition, five oncocytomas have been reported in abstract form, 18 and one reported case of adrenocortical oncocytoma occurred in a heterotopic site.19 For the cases with available clinical information, all were nonfunctional except one neoplasm, reported by Erlandson and Reuter,6 which caused virilization. The one case of oncocytic carcinoma locally invaded the inferior vena cava and liver, but clinical follow-up information is not available. 5 The immunohistochemical workup of these tumors is limited. 2,5,6,18,19
In this report, seven oncocytic neoplasms that occurred in the adrenal gland are described (five oncocytomas and two oncocytic neoplasms with increased proliferative activity of uncertain malignant potential). The clinicopathologic findings are presented and all cases were studied extensively using immunohistochemical methods.
Back to Top
MATERIALS AND METHODS
The seven cases in this study were selected on the basis of their being composed exclusively of oncocytic cells. Adrenocortical neoplasms with focal oncocytic features were not included. Clinical information was obtained by chart review or was provided
by the patient’s physician. Two cases (2 and 4) were mentioned in the discussion of a prior case report.6
Hematoxylin and eosin-stained slides were reviewed, and each tumor was evaluated carefully for features associated with malignancy, as described by others. 16,26,27 Mitotic activity was assessed in well-fixed, thin sections of tumor. In areas with the greatest mitotic activity, 4 sets of 50 consecutive high-power fields (HPFs), assessed using a microscope with a 40× objective and a 10x ocular, were counted, and the results were averaged.
Immunohistochemical studies were performed using fixed, paraffin-embedded sections and a heat-induced epitope retrieval method, as previously described. Cited Here … The source and dilutions of the primary antibodies used (monoclonal unless otherwise specified) were as follows: vimentin (V9, 1:100), S100 (polyclonal, 1:200), and epithelial membrane antigen (E29, 1:50), (DAKO, Carpinteria, CA); p53 (DO7, 1:100, Novocastra, New-castle on Tyne, United Kingdom); Ki-67 (MIB-1, 1:100, Immunotech, Westbrook, ME); keratin (a cocktail of antibodies AE1, AE3, CAM 5.2, and 35;H11, prediluted, Ventana, Tucson, AZ); chromogranin A (LK2H10, 1:100) and synaptophysin (Sy38, 1:10) (Boehringer Mannheim, Indianapolis, IN); antityrosine hydroxylase (1:500) and antidopamine ß-hydroxylase (polyclonal, 1:1000) (Incstar, Stillwater, MN); and mES-13 (1:1000) (reactive with mitochondria, courtesy of Dr. Eugenio Santos, National Institutes of Health, Bethesda, MD).
All tumors positive for keratin using the antibody cocktail (cases 1, 2, 3, 4 and 7) were also assessed with the following monoclonal antikeratin antibodies individually: keratin 7 (OV-TL/30, 1:50) and keratin 8 (35BH11, 1;50) (DAKO); keratin 14 (LL002, 1:10, (Neomarkers, Fremont, CA); CAM 5.2 (neat, Becton-Dickinson, San Jose, CA), AE1 (1:50) and AE3 (1:50) (Boehringer Mannheim); and keratin 20 (IT-KS-20.8, neat, American Research Products, Belmont, MA).
In two neoplasms with lymphoid aggregates, the aggregates were assessed with antibodies specific for the B-cell antigen CD20 (L26, monoclonal, 1:50) and the T-cell antigen CD3 (polyclonal, 1:50) (both from DAKO).
Using the MIB-1 antibody, cell proliferative activity was quantified by selecting microscopic fields with the greatest number of positive nuclei, counting 100 cells in 20 HPFs, and calculating the mean score. The tumor proliferative fraction (TPF), as defined by others9,23 as the number of MIB-1-positive nuclei per 1000 cells counted, was calculated.
DNA content was determined using paraffin-embedded tissue and the method of Hedley et al.10 Cell nuclei isolated from paraffin-embedded tissue were analyzed using the
FACSCalibur instrument, the CellQuest acquisition system, and the ModFit DNA analysis program (Becton-Dickinson). Single-parameter histograms were used to display the data. Comparison of the histograms with normal paraffin-embedded tissue allowed determination of the G0/G1 peak. The DNA aneuploidy was defined by the presence of a distinct second peak, with a lower or higher channel number, or by the presence of a tetraploid population more than 15% of all nuclei evaluated.
Electron microscopy was performed using fresh tissue samples of three neoplasms (cases 1, 2, and 4). Tissue was fixed in glutaraldehyde, postfixed in osmium tetroxide, dehydrated in a series of alcohols, and embedded in epoxy resin. Sections, approximately 80 nm in thickness, were stained with uranyl acetate and lead citrate and evaluated with a transmission electron microscope.
Back to Top
RESULTS
Back to Top
Clinical Data
Clinical information is summarized in Table 1. The median age of the seven patients was 55 years (range, 29 to 68 years). There were two men and five women. Four patients (cases, 1, 3, 4, and 5) had abdominal pain, massive hematuria developed in one patient (cases 6), and one patient (case 7) had episodic vomiting. One patient (case 2) had a history of breast carcinoma, and 9 years later her physician palpated an abdominal mass during an annual examination. In each patient, the adrenal tumor was detected by one or more abdominal radiologic studies: computerized tomography in six, magnetic resonance imaging in one, intravenous renal pyelography in one, and ultrasonography in one. Six neoplasms occurred in the left adrenal gland, and one involved the right (case 4).
| Case | Age (yrs) | Gender | Symptoms/ findings at diagnosis* | Side | Size (cm) | Weight (g) | Mitotic activity (per 50 hpf) | Necrosis | Histologic diagnosis | Clinical follow-up |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 68 | F | Abdominal pain | L | 8 | NA | <1 | - | Oncocytoma | NED, 13 months |
| 2 | 55 | F | Abdominal pain | L | 6.7 | 85 | <1 | - | Oncocytoma | NED, 64 months |
| 3 | 41 | F | Annual physical examination | L | 7.5 | 140 | 0 | - | Oncocytoma | NED, 99 months |
| 4 | 52 | M | Abdominal pain | R | 9 | 294 | 0 | - | Oncocytoma | NED, 96 months |
| 5 | 63 | F | Abdominal pain | L | 5 | 75 | 0 | - | Oncocytoma | Recently diagnosed |
| 6 | 67 | M | Hematuria | L | 9 | NA | 6 | + | Oncocytic neoplasm of UMP | NED, 19 months |
| 7 | 29 | F | Epsiodic vomiting | L | 13.5 | 620 | 2 | + ☒ | Oncocytic neoplasm of UMP | NED, 12 months |
NA, not available; hpf, high power field (400x); NED, no evidence of disease; m, months; UMP, uncertain malignant potential
* All patients had no clinical or laboratory evidence of adrenal tumor function.
In six patients, there was no clinical evidence that the adrenal gland neoplasms were functional. One patient (case 3) initially sought medical attention for hirsutism, leading to the clinical suspicion of an androgen-producing adrenal neoplasm. However, the hirsutism did not resolve after surgical excision of the adrenal mass, and it persists at last clinical follow-up.
The results of preoperative laboratory studies are available in five patients (cases 1, 2, 3, 5, and 7). Urine studies to assess for the possibility of pheochromocytoma, performed in four patients (cases 1, 2, 5, and 7) were negative. These studies included epinephrine, norepinephrine, and dopamine (cases 1 and 7), metanephrines (cases 1, 2, and 5), normetanephrines (case 1), and vanillylmandelic acid (case 2). In case 7, the urine normetanephrines were elevated slightly at 509 mg/dL (normal, 82-500 mg/dL). In case 5, results of urine studies for cortisol and aldosterone were normal. Results of serum studies, performed in five patients (cases 1, 2, 3, 5, and 7), also were normal and included cortisol (cases 1, 2, 3, and 5), corticotropin (ACTH, case 2), renin (case 7), and plasma catecholamines, including epinephrine, norepinephrine, and dopamine (case 7). In case 3, serum dehydroepiandrosterone (DHEA) sulfate and free testosterone were analyzed four separate times before surgery. The serum dehydroepiandrosterone (6.6 µg/mL; normal, 0.5-4.1 µg/mL) and free testosterone (4.8 pg/mL; normal, 0.6-2.5 pg/mL) were slightly elevated 10 months before surgical excision of the adrenal mass. However, results of subsequent analyses 4 months and 2 months before surgical excision were normal.
All patients underwent surgical excision: five underwent adrenalectomy and two underwent adrenalectomy with nephrectomy (cases 1 and 6). None of the patients were treated with adjuvant radiation therapy or chemotherapy. Case 5 is recent. Clinical follow-up for cases 1, 2, 3, 4, 6, and 7 ranged from 12 to 99 months (mean, 50.2 months).
Back to Top
Pathologic Findings
The pathologic findings are summarized in Table 1. Grossly, the five benign oncocytomas (cases 1, 2, 3, 4, and 5) ranged from 5.0 cm to 9 cm in greatest dimension. Tumor weight was available for four tumors and ranged from 75 g to 294 g. The two oncocytic neoplasms of uncertain malignant potential (cases 6 and 7) were 9 cm and 13.5 cm, respectively. Tumor weight was available for case 7, which was 620 g.
Histologically, the five oncocytomas and the two oncocytic neoplasms of uncertain malignant potential were very similar. In all seven cases, the adrenal gland was almost
completely replaced by a well-circumscribed, partially encapsulated neoplasm that compressed the residual normal adrenal gland. The neoplastic cells were exclusively oncocytes, with abundant, eosinophilic, and granular cytoplasm (Fig. 1), predominantly arranged in a diffuse or solid pattern (Fig. 2A). However, foci also were present where the cells were arranged in nests that formed an alveolar or acinarlike configuration (Fig. 2B). These acini were surrounded by an abundant capillary network. Foci of edematous stroma were present in every tumor. These foci were associated with degenerative changes that included hyalinized stroma (all cases), recent hemorrhage (cases 1, 4, and 6), hemosiderin deposition (cases 1 and 4), and cystic spaces filled with eosinophilic clear fluid (cases 1, 4, 5) (Fig. 3). Hyalinized fibrous bands were present in two tumors (cases 1 and 7) with foci of dystrophic calcification in one tumor (case 1).
Each neoplasm was exclusively composed of oncocytes with abundant eosinophilic cytoplasm. Nuclear atypia also is present in this field (case 7).
B
Cystic degeneration was observed in three neoplasms (case 1).
Electron microscopic studies, performed on three neoplasms (cases 1, 2, and 4), confirmed the oncocytic nature of these tumors by establishing that their light microscopic appearance was the result of an abnormal accumulation of mitochondria (Fig. 4A). In cases 1 and 2, the mitochondria were round to oval without tubular or elongate forms (the type usually encountered in normal zona glomerulosa cells). The mitochondria contained circular or lamellar arrays of cristae with infrequent tubular forms. An occasional stack of rough endoplasmic reticulum was present, and no lipid vacuoles were identified. Scattered lysosomes and rare lipochrome pigment inclusions also were noted. The cells were invested by basal lamina and attached by infrequent, poorly formed cell junctions. In case 4, frequent mitochondria displayed tubular cristae, and a subset contained large, round matrical densities. Additional features that were observed included prominent smooth endoplasmic reticulum and short stacks of rough endoplasmic reticulum. Some cells contained substantial amounts of dense-core cytoplasmic granules. An unexpected finding, in some cells, was the presence of
membrane-bound crystalline cytoplasmic inclusions of varying size (Fig. 4B). One such inclusion was observed in an intranuclear location.
A
8
Nuclear atypia was identified in both the oncocytomas and the oncocytic neoplasms of uncertain malignant potential. The extent of atypia was variable. In case 1, nuclear atypia was relatively mild and characterized by cells with moderately enlarged nuclei and prominent nucleoli. Some cells were multinucleated. The atypia was found in less than 1% of all oncocytes. In case 2, distinctive islands of cells (5-10% of the neoplasm) with atypia were present. The majority of cells in these islands had marked nuclear atypia characterized by enlarged nuclei with prominent nucleoli, multinucleation, irregular nuclear contours, nuclear pseudoinclusions, and hyperchromatic chromatin. In the other five neoplasms (cases 3, 4, 5, 6, and 7), the nuclear atypia was similar to case 2, but was present diffusely, affecting approximately 25% (and focally >50%) of the neoplastic cells (Fig. 5).
0
?
A
B
We have made the distinction between oncocytoma (cases 1, 2, 3, 4, and 5) and oncocytic neoplasm of uncertain malignant potential (cases 6 and 7) on the basis of mitotic activity and necrosis. In the five oncocytomas, mitoses were rare. After extensive searching, less than 1 mitotic figure per 50 HPFs was identified in cases 1 and 2. No mitotic figures were found in cases 3, 4 and 5. In contrast, in the oncocytic neoplasms of uncertain malignant potential, small foci of coagulative necrosis were present, and mitotic figures were more easily identified: 6 per 50 HPFs in case 6 and 2 per 50 HPFs in case 7.
One additional finding was the presence of discrete, often relatively large foci of adipose tissue in three of the five oncocytomas (cases 2, 3, and 4) (Fig. 6). In two tumors (cases 3 and 4), the foci of adipose tissue were associated with small aggregates of small lymphocytes, plasma cells, and histiocytes. Small clusters of lymphocytes not associated with adipose tissue also were found in every tumor (Fig. 7). There was no evidence of capsular, sinusoidal, or vascular invasion in either the oncocytomas or the oncocytic neoplasms of uncertain malignant potential.
Aggregates of lymphocytes were present in every neoplasm (case 1).
Back to Top
Immunohistochemical Findings
The results of the immunohistochemical findings are summarized in Tables 2 and 3. Four of five oncocytomas (cases 1, 2, 3, and 4) and one oncocytic neoplasm of uncertain malignant potential (case 7) were positive for keratin, as shown using the antikeratin cocktail (Table 2)(Fig. 8A). Using individual antikeratin antibodies on the five tumors that were positive with the cocktail, all were strongly positive with the CAM 5.2 and AE3 antibodies. Four tumors (cases 2, 3, 4, and 7) showed focal positivity with the 35BH11 antibody (keratin 8). One tumor (case 2) was focally positive with the AE1 antibody. All five were negative for antibodies reactive with keratins 7, 14, and 20.
| Case no. | Keratin cocktail* | Keratin 7 | 35BH11 (keratin 8) | CAM 5.2 (keratins 8, 18) | Keratin 14 | AE1 | AE3 | Keratin 20 |
|---|---|---|---|---|---|---|---|---|
| 1 | + ☒ | - | - | ++ | - | - | ++ | - |
| 2 | ++ | ☐ - | + | ++ | ☐ | + | ++ | - |
| 3 | ++ | ☐ - | + | ++ | ☐ - | - ☐ | + ☒ | ☐ |
| 4 | ++ | - | + | ++ | ☐ - | - | ++ | ☐ - |
| 5 | - | ND | ND | ND | ND | ND | ND | ND |
| 6 | - | ND | ND | ND | ND | ND | ND | ND |
| 7 | ++ | - | + ☒ | ++ | - | - | + ☒ | - |
+, subset of cells positive; ++, diffusely positive; - , negative; ND, not done.
* The keratin cocktail includes four antibodies: AE1, AE3, CAM 5.2, and 35BH11
| Case | EMA | Vimentin | mES-13 | NSE | Synaptophysin | Chromogranin A | Tyrosine hydroxylase | Dopamine ß-hydroxylase | S100 | p53 | Ki-67 (TPF) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | ++ | ++ | + ☒ | + | 65.0 | ||||||
| 2 | ☐ | ++ | ++ | + ☒ | ☐ | 7.5 | |||||
| 3 | ++ ☒ | ++ | + ☒ | + ☒ | 2.5 | ||||||
| 4 | ++ | + ☒ | + ☒ | 1.0 | |||||||
| 5 | ☐ | + ☒ | ++ | + ☒ | 54.0 | ||||||
| 6 | ++ | ++ | ++ | 85.0 | |||||||
| 7 | + ☒ | ++ | ++ | + | ☐ | ☐ | 95.0 |
EMA, epithelial membrane antigen; TPF, tumor proliferative fraction (number of positive nuclei/1000 cells); +, subset of cells positive; ++, diffusely positive; - , negative.
Summary of immunohistochemical findings (except keratin) in adrenocortical oncocytic neoplasms
00
B
C
D
(A) Five neoplasms were positive for keratin using the antikeratin cocktail (case 2). (B) Neuron-specific enolase was positive, either focally or diffuse, in all neoplasms (case 2). (C) Synaptophysin was focally positive in five neoplasms (case 7). (D) Intense and diffusely cytoplasmic reactivity with the antimitochondrial antibody mES-13 antibody was characteristic of every neoplasm (case 2). (A,B,C,D) Immunoperoxidase with hematoxylin counterstain, x 400)
The majority of neoplasms were positive for two neuroendocrine-associated markers: neuron-specific enolase (NSE) and synaptophysin. All seven neoplasms were positive for NSE (Fig. 8B), and five tumors were positive for synaptophysin (Fig. 8C), (cases 5 and 6 were negative). Staining for NSE and synaptophysin was clearly above the background staining that can be seen in oncocytic neoplasms. The staining for NSE was diffuse in four tumors and focal in three. The staining for synaptophysin was focal in all five positive neoplasms. No tumors were positive for three additional neuroendocrine markers tested (chromogranin A, tyrosine hydroxylase, and dopamine ß-hydroxylase).
In each neoplasm, the neoplastic cells were strongly positive for the mitochondrial antibody mES-1328(Fig. 8D). Three oncocytomas and both oncocytic neoplasms of uncertain malignant potential were positive for vimentin. All tumors were negative for epithelial membrane antigen (EMA), S100, and p53.
Using the MIB-1 antibody, the TPF was calculated for each tumor. The five oncocytomas had a TPF that ranged from 1 to 65 (median, 7.5; mean, 26). The two oncocytic neoplasms of uncertain malignant potential had a TPF of 85 (case 6) and 95 (case 7), respectively. In two neoplasms with lymphoid aggregates studied, the lymphoid cells were a mixture of CD20-positive B cells and CD3-positive T cells, in approximately equal numbers. In some aggregates, the CD20-positive B cells were distributed in tight clusters.
Back to Top
DNA Ploidy Results
Using paraffin-embedded tissue blocks, DNA content was analyzed in each case. One representative block was assessed in five neoplasms, and two blocks were analyzed in two tumors (cases 3 and 4). Only one neoplasm (case 2) was aneuploid with a DNA index of 1.3.
Back to Top
DISCUSSION
The seven cases we describe (Table 1) are similar to those previously reported in the English literature (summarized in Table 4). We identified 11 oncocytic adrenocortical neoplasms fully reported in the English literature.2,5-7,11,19,21,25 Combining these neoplasms with the seven cases we report, 18 oncocytic adrenocortical neoplasms are now described: 15 oncocytomas, 1 oncocytic carcinoma, and 2 oncocytic neoplasms of uncertain malignant potential. The latter two tumors we report had increased
proliferative activity and necrosis but no evidence of invasion and negative, but a relatively short-duration of clinical follow-up. Gender is known for all patients: 7 men and 11 women. Age ranged from 27 to 68 years (mean, 45.5 years). Seventeen tumors occurred in the adrenal gland, and one occurred in heterotopic right supraadrenal gland retroperitoneal tissue.19 The side was specified in all patients with adrenal gland tumors: 12 left and 5 right. Seventeen patients had no clinical evidence of abnormal adrenal function. In these patients, the adrenocortical tumors were detected by radiologic studies during the clinical workup of other complaints, such as abdominal pain, hematuria, essential hypertension; one patient in our series was being followed routinely after successful treatment of breast carcinoma 9 years before. One woman had virilization.6 Tumor size ranged from 3 cm to 15 cm; the median was 8 cm. Tumor weight, available for 13 neoplasms, ranged from 30 g to 865 g; the median was 205 g. Muir and colleagues 18 also reported five oncocytomas in abstract form. Although complete clinical and pathologic data are not provided for these patients, the findings in the abstract also show a female predominance and extend the upper limit for patient age, tumor size, and tumor weight to 71 years, 20 cm, and 2145 g, respectively.
| Reference | Age (yrs) | Gender | Symptoms/findings at diagnosis | Side | Size (cm) | Weight (g) | Histologic diagnosis | Clinical follow-up |
|---|---|---|---|---|---|---|---|---|
| Kakimoto et al.11 | 41 | F | Incidental | L | 5 | 30 | Oncocytoma | NED, 8 months |
| Erlandson and Reuter6 | 61 | F | Virilization | L | 5 | 55 | Oncocytoma | NED, 12 months |
| El-Naggar et al.5 | 56 | M | Abdominal pain | R | 8 | NA | Oncocytic carcinoma | NA |
| Sasano et al.21 | 41 | M | Incidental | L | 9 | 260 | Oncocytoma | NED, 27 months |
| 44 | F | Abdominal pain | R | 15 | 865 | Oncocytoma | NED, 8 months | |
| and hematuria | ||||||||
| 41 | F | Abdominal pain | L | 5 | 64 | Oncocytoma | NED, 23 months | |
| Nguyen et al. 19 | 44 | M | Back pain | Heterotopic | 3 | NA | Oncocytoma | NED, 54 months |
| Begin2 | 58 | M | Incidental | R | 13 | 315 | Oncocytoma | NED, 21 months |
| Gandras et al.7 | 27 | F | Flank pain | L | 6 | NA | Oncocytoma | NA |
| Waters et al.25 | 47 | M | Hematuria | R | 8 | 205 | Oncocytoma | NED, 30 months |
| 41 | F | Flank pain | L | 10 | 550 | Oncocytoma | NED, 9 months | |
| Muir et al. 18 | 45-71 | 4F, 1M | 3 Incidental | NA | 3.5-20 | 21-2415 | 5 Oncocytoma | 4 NED, 6-84 months |
| 1 Abdominal pain | 1 NA | |||||||
| 1 Uknown |
NA, not available; NED, no evidence of disease.
A number of systems have been devised to distinguish benign from malignant adrenocortical tumors 16 and, of these, we have found the Weiss system26,27 to be simple and easy to use. In this system, nine histologic criteria are assessed. The following findings are more common in carcinoma: high nuclear grade (atypia), eosinophilic cell cytoplasm (>75%), diffuse architecture (>33%), and the presence of necrosis, mitotic figures (>5/50 hpfs), atypical mitotic figures, capsular invasion, venous invasion, and sinusoidal invasion. In the original study,26 the presence of four or more of these findings was found only in carcinomas. In contrast, two or fewer findings were present in adenomas. Subsequently, this system was revised; three or more findings were suggested as adequate for the diagnosis of malignancy.27
Based on our experience with these tumors, we believe that the Weiss system needs to be modified for the assessment of adrenocortical oncocytic neoplasms. Every case in this study had at least three findings designated as associated with malignancy using the Weiss system: eosinophilic tumor cytoplasm, diffuse architecture, and nuclear atypia. However, we interpreted five of the tumors in this study as benign oncocytomas, and recurrence or metastases have not developed in four patients with clinical follow-up. Therefore, we conclude that adrenocortical oncocytic neoplasms should be assessed conservatively in the absence of mitotic activity, necrosis, or invasion.
We designated two tumors (cases 6 and 7) in this study as oncocytic neoplasms of uncertain malignant potential. In both neoplasms, there were small areas of coagulative necrosis and foci in which mitotic activity was increased. Necrosis and increased mitotic activity both are uncommon in benign tumors. 16,26,27 However, there was no evidence of capsular or vascular invasion and, to date, neither tumor has recurred locally or metastasized. Nevertheless, the duration of clinical follow-up for these two patients is short-19 months for case 6 and 12 months for case 7-and low-grade adrenocortical carcinomas may not metastasize or recur for many years.27
Immunostaining for MIB-1 (Ki-67) has been reported by others to be helpful in distinguishing adrenocortical adenomas from carcinomas.2,23,29 For example, Vargas and colleagues 23 stained 40 adrenocortical lesions with the MIB-1 antibody, which included 10 cases of hyperplasia, 10 adenomas, 12 primary carcinomas, and 8 metastatic or recurrent carcinomas, and calculated the TPF for each group. Mean TPF was 14.9 in adenomas, 31.5 in hyperplasia cases, 208.1 in carcinomas, and 166.1 in recurrent or metastatic lesions. None of the benign lesions had a TPF more than 80, and only 1 of 20 carcinomas had a TPF less than 80. In our study, the five oncocytomas had a TPF of 1, 2.5, 7.5, 54, and 65, respectively. The two oncocytic neoplasms of uncertain malignant potential had a TPF of 85 and 95, respectively, further showing increased proliferative activity in these neoplasms. The MIB-1 findings may be interpreted in at least two ways. These findings can be used to support the diagnosis of low-grade carcinoma for these two neoplasms (cases 6 and 7). Alternatively, the threshold TPF value of 80 previously reported by others23 may be too low in the assessment of adrenocortical oncocytic neoplasms, and oncocytomas can have foci of increased proliferative activity.
The electron microscopic findings in the three cases of adrenal oncocytoma we studied are virtually identical to those previously reported.2,5,6 The neoplastic cells displayed the defining feature of oncocytes: an abnormal proliferation of mitochondria that fill the cytoplasm nearly to the exclusion of other organelles.4,8 Additionally, the neoplastic cells (most prominent in case 4) had features that were characteristic of adrenocortical cells, such as tubular mitochondrial cristae, prominent smooth endoplasmic reticulum,
and stacked cisternae of rough endoplasmic reticulum. A feature not previously reported in adrenal oncocytoma is the crystalline inclusion shown in Fig. 4B. The nature of these structures is not known. They are similar in appearance to the crystals of Reinke that are found sometimes in Leydig cells of postpubertal testis, and it is possible, because adrenal cortical cells and testicular interstitial cells are sites of steroid synthesis, that the two are related. Reinke’s crystals also have been observed occasionally as a true intranuclear inclusion. The one intranuclear inclusion that we observed was accompanied by a few small dense-core granules, suggesting that it had been transported into the nucleus through the incorporation of a cytoplasmic pseudoinclusion.
The immunohistochemical characterization of adrenocortical oncocytic neoplasms in the literature is very limited; seven cases were studied previously in a nonsystematic manner.2,5,6,18,19 Therefore, we studied this group of cases with a large panel of antibodies. Every neoplasm was strongly positive with the antimitochondrial antibody mES-13.28 The staining with mES-13 was strong, diffuse, and cytoplasmic. Five neoplasms were keratin positive, best detected using the CAM 5.2 and AE3 antibodies. Four tumors had focal expression of keratin 8, detected with the 35BH11 antibody, and one tumor was focally positive with the AE1 antibody. All five neoplasms were negative with antibodies specific for keratins 7, 14, 20.
We also studied these cases for expression of a number of neuroendocrine-associated markers: NSE, synaptophysin, chromogranin A, tyrosine hydroxylase, and dopamine ß- hydroxylase. All tumors were positive for NSE, and five cases were focally positive for synaptophysin. However, all tumors were negative for chromogranin A, tyrosine hydroxylase, and dopamine ß-hydroxylase. In addition, ultrastructural studies performed on three neoplasms revealed no evidence of membrane-bound, dense-core granules. Therefore, there is no convincing evidence that these tumors exhibited true neuroendocrine differentiation. Neuron-specific enolase and synaptophysin can react with nonneuroendocrine types of tumors, such as adrenocortical tumors. 1,12,20
We assessed DNA ploidy in this small group of cases to determine whether DNA content correlates with histologic findings. In this study, six neoplasms were diploid and one neoplasm (case 2) was aneuploid. The aneuploid tumor histologically was a benign oncocytoma, and the patient had no evidence of local recurrence or metastasis during 64 months of clinical follow-up. Therefore, the presence of aneuploid DNA content does not appear to be a finding that can be used to predict malignancy. These results are similar to those reported previously for nononcocytic adrenocortical neoplasms.3
The differential diagnosis of adrenal gland oncocytic neoplasms includes nononcocytic adrenocortical tumors, pheochromocytoma, and renal cell carcinoma involving the
adrenal gland by metastasis or local invasion.
Many adrenocortical tumors have foci of neoplastic cells with eosinophilic cytoplasm, which are much more common in carcinomas than in adenomas. In a subset of these cases, the neoplastic cell cytoplasm can be abundant and brightly eosinophilic, resembling oncocytes.13 As suggested by others,4 we required that the oncocytic neoplasms in this study be composed exclusively of oncocytes. The mES-13 antibody also may be helpful because oncocytic neoplasms have intense and diffuse cytoplasmic staining in contrast to less intense and focal staining in adrenal adenomas or carcinomas with focal oncocytic features.28
Adrenal gland pheochromocytomas potentially could be confused with oncocytoma. In the majority of cases, this distinction will not be difficult because pheochromocytomas are functional and commonly cause hypertension and symptoms related to catecholamine excess. However, pheochromocytomas can exhibit a variety of histologic appearances, and a subset of cases may have neoplastic cells with abundant eosinophilic cytoplasm that resemble the cells of oncocytoma. 13,14,17 Also, when the tumor is large, it may be difficult to appreciate the medullary origin of pheochromocytoma. For the unusual and difficult case, immunohistochemical studies are very helpful. Pheochromocytomas typically are positive for chromogranin A and other neuroendocrine markers. 15 If necessary, electron microscopic studies will reveal abundant dense-core membrane- bound granules in the cells of pheochromocytoma.8
Renal cell carcinoma may involve the adrenal gland, and renal cell carcinomas with oncocytic features, such as the eosinophilic variant of chromophobe cell carcinoma,22 superficially may resemble oncocytoma. However, the clinical findings that support the diagnosis of renal cell carcinoma usually are evident, and histologically the renal neoplasm typically shows evidence of cytologic atypia or invasion. Chromophobe cell carcinomas also are positive for Hale’s colloidal iron stain. Immunohistochemical studies will be very helpful. Renal cell carcinomas are commonly positive for epithelial membrane antigen and negative for NSE and synaptophysin; the converse is true for adrenocortical oncocytoma. If necessary, electron microscopic studies will easily differentiate the cells of chromophobe cell carcinoma, with their characteristic cytoplasmic vesicles, from the cells of oncocytoma. 8,22
In summary, we described seven oncocytic neoplasms that occurred in the adrenal gland. No patients had clinical evidence of abnormal adrenal gland function before surgical excision. Six patients with clinical follow-up, have not developed local recurrence or metastasis, despite the large size and the foci of increased proliferative activity detected in two neoplasms. Immunohistochemical studies have shown that these tumors are commonly keratin positive and particularly express keratin 18.
Acknowledgments: The authors thank the following people for providing clinical information or patient materials for this study: Chris S. Jensen, M.D., Fort Collins Consultants in Pathology, and Donna L. Sullivan, M.D., Fort Collins Family Medicine Center, Fort Collins, Colorado (case 1); Robert A. Robinson, M.D., and Charles Platz, M.D., University of Iowa Hospitals and Clinics, Iowa City, Iowa (case 2); Terri L. Cooper, M.D., St. Joseph’s Hospital, Bangor, Maine and Ms. Irene Cote, Department of Pathology, Medical Center of Central Massachusetts, Worcester, Massachusetts (case 3); Marshall Turner, M.D., Department of Pathology, Saint Joseph Hospital, Denver, Colorado (case 4); Don V. Stanton, M.D., Redding Pathologists, Redding, California (case 5); Weber L. Smith, M.D., Indian River Memorial Hospital, Vero Beach, Florida (case 6); James J. Scillian, M.D., Kaiser Permanente, Los Angeles, California (case 7).
Back to Top
REFERENCES
1. Alsabeh R, Mazoujian G, Goates J, Medeiros LJ, Weiss LM. Adrenal cortical tumors clinically mimicking pheochromocytoma. Am J Clin Pathol 1995;104:382-90. Cited Here …
2. Begin LR. Adrenocortical oncocytoma: case report with immunocytochemical and ultrastructural study. Virchows Archiv A Pathol Anat 1992;421:533-7. Cited Here …
3. Cibas ES, Medeiros LJ, Weinberg DS, Gelb AB, Weiss LM. Cellular DNA profiles of benign and malignant adrenocortical tumors. Am J Surg Pathol 1990;14:948-55. Cited Here …
4. Chang A, Harawi SJ. Oncocytes, oncocytosis, and oncocytic tumors. Pathol Annu 1992;27:263-304.
Cited Here …
5. El-Naggar AK, Evans DB, Mackay B. Oncocytic adrenal cortical carcinoma. Ultrastruc Pathol 1991;15:549-56.
Cited Here …
6. Erlandson RA, Reuter VE. Oncocytic adrenal cortical adenoma. Ultrastruc Pathol 1991;15:539-47.
Cited Here …
7. Gandras EJ, Schwartz LH, Panicek DM, Levi G. Adrenocortical oncocytoma: CT and MRI findings. J Comput Assist Tomogr 1996;20:407-9.
Cited Here …
8. Ghadially FN. Diagnostic Electron Microscopy of Tumours, 3rd ed. London; 1988. Cited Here …
9. Goldblum JR, Shannon R, Kaldjian EP, et al. Immunohistochemical assessment of proliferative activity in adrenocortical neoplasms. Mod Pathol 1993;6:663-8. Cited Here …
10. Hedley DW, Friedlander ML, Taylor IW. Application of DNA flow cytometry to paraffin-embedded archival material for the study of aneuploidy and its clinical significance. Cytometry 1985;6:327-33.
Cited Here …
11. Kakimoto S, Yushita Y, Sanefuji T, et al. Non-hormonal adrenocortical adenoma with oncocytoma-like appearances. Acta Urol Jpn 1986;32:757-63.
Cited Here …
12. Komminoth P, Roth J, Schroder S, Saremaslani P, Heitz PU. Overlapping expression of immunohistochemical markers and synaptophysin mRNA in pheochromocytomas and adrenocortical carcinomas: implications for the differential diagnosis of adrenal gland tumors. Lab Invest 1995;72:424-31.
Cited Here …
13. Lack EE. Tumors of the adrenal gland and extra-adrenal paraganglia. In: Rosai J, ed. Atlas of Tumor Pathology. Series 3. Fasicle 19. Washington, DC: Armed Forces Institute of Pathology; 1995:118-9.
Cited Here …
14. Linnoila RI, Keiser HR, Steinberg SM, Lack EE. Histopathology of benign and malignant sympathoadrenal paragangliomas: clinicopathologic study of 120 cases including unusual histologic features. Hum Pathol 1990;21:1168-80.
Cited Here …
15. Lloyd RV, Blaivas M, Wilson BS. Distribution of chromogranin and S-100 protein in normal and abnormal medullary tissue. Arch Pathol Lab Med 1985;109:633-5. Cited Here …
16. Medeiros LJ, Weiss LM. New developments in the pathologic diagnosis of adrenal cortical neoplasms. Am J Clin Pathol 1992;97:73-83.
Cited Here …
17. Medeiros LJ, Wolf BC, Balogh K, Federman M. Adrenal pheochromocytoma: a clinicopathologic review of 60 cases. Hum Pathol 1985;16:580-9.
Cited Here …
18. Muir TE, Ferreiro JA, Carney JA. Oncocytoma of the adrenal gland. Mod Pathol 1996;9:50A. Abstract.
Cited Here …
19. Nguyen G-K, Vriend R, Ronaghan D, Lakey WH. Heterotopic adrenocortical oncocytoma. A case report with light and electron microscopic studies. Cancer 1992;70:2681-4.
Cited Here …
20. Pahlman S, Esscher T, Nilsson K. Expression of y-subunit of enolase, neuron specific enolase, in human non-neuroendocrine tumors and derived cell lines. Lab
Invest 1986;54:554-60. Cited Here …
21. Sasano H, Suzuki T, Sano T, Kameya T, Sasano N, Nagura H. Adrenocortical oncocytoma. A true nonfunctioning adrenocortical tumor. Am J Surg Pathol 1991;15:949-56.
Cited Here …
22. Thoenes W, Storkel S, Rumpelt HJ, Moll R, Baum HP, Werner S. Chromophobe renal cell carcinoma and its variants. A report of 32 cases. J Pathol 1988;155:277-87. Cited Here …
23. Vargas MP, Vargas HI, Kleiner DE, Merino MJ. Adrenocortical neoplasms: role of prognostic markers MIB-1, p53, and RB. Am J Surg Pathol 1997;21:556-62. Cited Here …
24. Vasef MA, Alsabeh R, Medeiros LJ, Weiss LM. Immunophenotype of Reed- Sternberg and Hodgkin’s cells in sequential biopsy specimens of Hodgkin’s disease. A paraffin-section immunohistochemical study using the heat-induced epitope retrieval method. Am J Clin Pathol 1997;108:54-9. Cited Here …
24. Waters PR, Haselhuhn GD, Gunning WT, Phillips ER, Selman SH. Adrenocortical oncocytoma: two case reports and review of literature. Urology 1997;49:624-8. Cited Here …
25. Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984;8:163-9. Cited Here …
26. Weiss LM, Medeiros LJ, Vickery AL. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13:202-6. Cited Here …
27. Weiss LM, Gaffey MJ, Warhol MJ, et al. Immunocytochemical characterization of a monoclonal antibody directed against mitochondria reactive in paraffin-embedded sections. Mod Pathol 1991;4:596-601.
Cited Here …
28. Winston JS, Mompoint A, Wang H, Crossland D. The utility of Ki-67 in the evaluation of adrenocortical neoplasms. Mod Pathol 1997;10:52A. Abstract. Cited Here …
29. Reference not provided.
Cited Here …
Keywords:
Adrenal gland; Oncocytoma; Immunohistochemistry; Electron microscopy; DNA ploidy
C Lippincott-Raven Publishers