9158680

THE AMERICAN JOURNAL OF SURGICAL PATHOLOGY

D. Wolters Kluwer

Adrenocortical Neoplasms: Role of Prognostic Markers MIB-1, P53, and RB

Vargas M. P. M.D .; Vargas, H. I. M.D .; Kleiner, D. E. M.D., Ph.D .; Merino, M. J. M.D.

The American Journal of Surgical PathologyAmerican Journal of Surgical Pathology. 21:p 556-562, May 1997.

Author Information

From the Laboratory of Pathology (M.P.V., D.E.K., M.J.M.) and Surgery Branch (H.I.V.), National Cancer Institute, National Institutes of Health, Bethesda, Maryland, U.S.A.

Address correspondence and reprint requests to Dr. M.J. Merino, Laboratory of Pathology, National Cancer Institute, 9000 Rockville Pike, Bethesda, MD 20892, U.S.A.

Abstract

Differentiation between benign and malignant adrenocortical neoplasms is made using a combination of clinical and pathologic parameters. Despite these parameters, it is still difficult to predict the biologic potential of some tumors. Forty adrenocortical lesions, including 10 hyperplasias, 10 adenomas, 12 carcinomas and eight metastatic/recurrent adrenocortical carcinomas were studied for the expression of MiB-1, p53, and the retinoblastoma gene product (RB) utilizing immunohistochemical techniques. The mean tumor proliferating fraction (TPF), expressed as the number of MiB-1-positive nuclei per 1,000 tumor cells, was 14.9 in adenomas, 31.5 in hyperplasias, 208.1 in carcinomas and 166.1 in recurrent or metastatic disease. None of the 20 benign lesions had a TPF of >80, and only one of the 20 malignancies had a TPF of <80. Nine of the 20 carcinomas were positive for p53. None of the benign lesions were p53 positive. Thirty-nine cases, including benign and malignant ones, were RB positive, and one was uninterpretable. We conclude that prognostic markers can be of great assistance in recognizing adrenocortical carcinomas. High TPF (>80), as measured by staining with MiB-1, and positive p53 strongly correlate with malignant behavior and therefore may be useful in distinguishing benign from malignant adrenal lesions. Staining for RB does not appear to be a helpful technique.

Adrenocortical carcinoma is a rare and highly aggressive neoplasm with an incidence of approximately one case per 1,700,000 population, accounting for 0.05-0.2% of all cancers (8). The 5-year survival rate varies from 16% to 32% (8). The poor prognosis associated with these tumors is due in part to the fact that in most instances the disease is well advanced at the time of diagnosis (10).

Diagnosis is usually made by a combination of clinical and anatomical parameters, such as weight loss, urinary 17-ketosteroid excretion, response to adrenocorticotropic hormone (ACTH) stimulation, tumor mass (>100 g), and established histologic criteria (including mitotic rate of>5/50 high-power fields [hpf], presence of atypical mitotic figures, necrosis, and vascular or capsular invasion) (7,18,20).

The availability and increased use of high-resolution computed tomography (CT) has resulted in detection of smaller or asymptomatic adrenal masses in which the malignant potential may be more difficult to assess. In an effort to resolve this diagnostic problem, Goldblum et al. studied a series of adrenal tumors and showed that proliferative activity, measured by immunostaining with MiB-1, can aid in the distinction between benign and malignant adrenocortical tumors (5). MiB-1 is an antibody that reacts in formalin-fixed paraffin embedded tissue with a human nuclear proliferation-associated antigen (Ki-67 molecule), which is expressed in all cells that are not in the G0 phase of the cell cycle.

It is widely accepted that both activation of protooncogenes and inactivation of tumor suppressor genes are involved in the tumorigenesis of various human tumors. Well- characterized examples of tumor suppressor genes include the retinoblastoma susceptibility (RB) and p53 genes, which are implicated in a variety of cancers (19). Ohgaki et al. have implicated p53 in the development of adrenocortical carcinomas at the molecular level (16). The products of these tumor suppressor genes, the RB and p53 proteins, act at the G1/S transition to restrain cell cycle progression.

Immunohistochemical evaluation of these proteins has been conducted in a variety of endocrine and nonendocrine tumors but has not been fully evaluated in adrenal cortical lesions.

The purpose of this study is to evaluate the expression of MiB-1, RB, and p53 and to correlate them with histologic diagnosis and clinical outcome.

Back to Top

MATERIALS AND METHODS

Forty cases of adrenocortical lesions, including 10 hyperplasias, 10 adenomas and 20 carcinomas (12 primary, three recurrent, and five metastatic) were obtained from the files of the Laboratory of Pathology at the National Cancer Institute.

Hematoxylin and eosin-stained sections were reviewed, and the diagnosis of adrenocortical hyperplasia was confirmed by standard pathologic criteria (12). The distinction between adenomas and carcinomas was made using the gross and microscopic criteria previously defined (7,18,20,21). The weight was taken from the surgical pathology report, except in four cases (two adenomas and two carcinomas) in which it was not recorded. Mitoses were evaluated by counting 50 random high-power fields in the areas of greatest numbers of mitotic fugures.

Back to Top

Immunohistochemistry

Formalin-fixed, paraffin embedded 5-um sections were mounted on charged glass slides, deparaffinized, and rehydrated through graded alcohols. For MiB-1 (Immunotech), p53 (DAKO), and RB (Novocastra) immunohistochemistry slides were subjected to microwaving in 10 mM of citrate buffer at pH 6 in a 750-W oven for 40 min in a pressure cooker. Primary antibody was applied for 32 min at 41°℃ at a concentration of 1:10 for MiB-1, 1:50 for p53, and 1:10 for RB in an automated immunostainer (Ventana ES). Formalin-fixed paraffin-embedded sections of tonsils, colonic adenocarcinoma, and parathyroid adenoma were used as positive controls for MiB-1, p53, and RB antibodies, respectively.

The count of MiB-1-positive nuclei was made in the areas of highest labeling density of each individual tumor. The tumor proliferative fraction (TPF) was defined as the number of MiB-1-positive tumor nuclei per 1,000 tumor cells. A tumor was scored as positive for p53 if more than 1% of cells showed nuclear staining. Tumors were considered to be RB positive if they had a heterogeneous pattern of nuclear staining for RB protein, and to be RB negative if>99% of cells lacked nuclear staining.

Statistical significance was determined using the Mann-Whitney U test.

Back to Top

RESULTS

Back to Top

Clinical Data

The clinical data are summarized in Tables 1 and 2. The patients’ ages ranged from 4 to 79 years, with a mean of 44, 51, and 33 years for hyperplasias, adenomas, and carcinomas, respectively. There were 15 males and 25 females. The follow-up ranged from 1 to 83 months, with a median of 10, 18, and 21 months for hyperplasias, adenomas, and carcinomas, respectively. All the adenomas had their clinical syndrome resolved after surgery and did not show evidence of recurrent disease at the time of the last visit (range 6-31 months). Four of the hyperplasias were secondary to a tumor producing ACTH, and four were secondary to a pituitary adenoma. In these eight cases, Cushing’s syndrome resolved after adrenalectomy, but patients continued to have symptoms related to their primary tumor. The remaining two hyperplasias were primary, and the patients were considered to be cured after surgery. Nine of the 20 carcinomas (45%) were functional. Six patients presented with Cushing’s syndrome, one with primary hyperaldosteronism, one with virilism, and one with feminization. The median disease-free interval for the entire carcinoma group was 7 months. All 12 primary carcinomas metastasized or recurred within a median time of 4 months (range 2-77 months). Three cases had metastases to the lung, soft tissue, and peritoneum at the time of presentation. All the recurrent carcinomas developed distant metastases (range 6-33 months). The most common metastatic sites were lung and liver, with at least one of these sites present 73% of the time. Two patients with carcinoma died of disease 12 and 76 months after diagnosis, and two had no evidence of disease 10 and 36 months after metastasectomy. The remaining 16 were alive with disease at the time of last follow-up, which was less than 12 months in 8 cases.

	Hyperplasia	Adenoma	Carcinoma
Mean age (yr)	44	51	33
Sex (M/F)	5/5	4/6	6/14
Presentation
Cushing	9	4	6
Hyperaldosteronism	1	6	1
Feminization			1
Virilization			1
Status	10 NED	10 NED	2 NED
			16 AWD 2 DOD
Follow-up (median mo)	10	18	21

NED, no evidence of disease; AWD, alive with disease; DOD, died of disease.

Clinical data

Site	Cases	%
Lung	9	47
Liver	9	47
Lung and/or liver	14	73
Lymph node	2	10
Bone	1	5
Kidney	2	10
Other	5	26

Adrenocortical carcinomas: metastatic sites (n = 19 cases)

Back to Top

Pathologic Findings

The pathologic findings are presented in Table 3. The hyperplastic adrenal glands had a mean weight of 21.4 g (range 5.5-86.9). They had a mean mitotic count of 0.9/50 hpf, with a range of 0-4/50 hpf. The adenomas had a mean weight of 16.2 g (range 10-20), with a mean mitotic count of 0.2/50 hpf (range 0-1/50 hpf). The carcinomas had a mean

weight of 761.8 g (range 43.5-4,700 g), with three cases weighing <100 g. Carcinomas had higher mitotic counts, with a mean of 54/50 hpf (Fig. 1). Only in one case were there <5/50 hpf. This case was classified as malignant because of the weight (4,700 g) and diffuse areas of liver and soft-tissue invasion.

Case	Diagnosis	Weight (g)	Mit/50HPF	MiB-1 (TPF)	p53	RB	Disease free interval	Follow-up	Status
1	H	5.5	2	30.3		+		N/A
2	H	8.1	4	77.5		+		24	NED
3	H	9.6	0	16.8		+		10	NED
4	H	21.1	0	18.8		+		28	NED
5	H	9.9	0	25.6		+		10	NED
6	H	15.7	2	29.6		+		12	NED
7	H	86.9	0	12.3		+		6	NED
8	H	24.3	0	26.4		+		6	NED
9	H	17.7	1	24.1		+		2	NED
10	H	15	0	53.4		+		14	NED
11	AD	20	1	15.7		+		18	NED
12	AD	17.5	0	8.3		+		15	NED
13	AD	13.3	1	19.8		+		31	NED
14	AD	11.6	0	11.4		+		6	NED
15	AD	17.6	0	34.2		+		18	NED
16	AD	14*	0	13.2		+		22	NED
17	AD	16.8	0	9.9		+		23	NED
18	AD	30	0	7.8				N/A
19	AD	11ª	0	20.2				41	NED
20	AD	10	0	8.7		+		5	NED
21	CA	683	104	123.9		+	77	78	AWD
22	CA	53ª	81	220	+		0	2	AWD
23	CA	4,700	1	16.6		+	8	21	AWD
24	CA	1,000	49	137.6		+	3	9	AWD
25	CA	43.5	51	342.6	+	+	12	12	AWD
26	CA	190	178	210.4	+(f)	+	2	2	AWD
27	CA	1,180	5	105.7		+	8	10	AWD
28	CA	140ª	147	413.3	+	+	3	11	AWD
29	CA/SA	119	38	291.9	+	+	0	1	AWD
30	CA	665	32	261.6	+(f)	+	4	4	AWD
31	CA	270	36	126	+(f)	UI	29	33	AWD
32	CA	99	18	247	+	+	0	10	NED
33	RCA		6	97.2		+	33	34	AWD
34	RCA		70	97.2		+	7	12	DOD
35	RCA		46	210.4		+	6	30	AWD
36	MCA		108	142.6	+	+	12	17	AWD
37	MCA		11	156		+	72	76	DOD
38	MCA		40	127.3		+	46	83	AWD
39	MCA		55	208.3		+	3	9	AWD
40	MCA		5	290		+	35	82	NED

H, hyperplasia; AD, adenoma; CA, carcinoma; CA/SA, carcinosarcoma; RCA, recurrent CA; MCA, metastatic CA; UI, uninterpretable. a Estimated weight by measurements.

Adrenocortical carcinoma with broad trabeculae separated by delicate vascular channels. Tumor cells have nonclear cytoplasm, increased mitotic activity with atypical figures (arrowhead) and high nuclear grade.

Back to Top

Immunohistochemical Data

The immunohistochemical data are shown in Table 4. The TPF was 31.5 in hyperplasias (range 12.3-77.5), 14.9 in adenomas (range 7.8-34.2), 208.1 in carcinomas (range 16.6-413.3), and 166.1 in metastatic or recurrent tumors (range 97.2- 290) (Fig. 2). None of the 20 benign lesions had a TPF of>80, and only one of the 20 malignancies had a TPF of <80 (Fig.3). This one case showed osseous metaplasia, and the tissue was subjected to decalcification before processing, which may have resulted in reduced staining.

Case	MIB-1 (TPF)	p53	RB	Disease-free interval	Follow-up (mo)	Status
1	30.3		+		N/A
2	77.5		+		24	NEDª
3	16.8				10	NEDª
4	18.8				28	NEDª
5	25.6				10	NEDª
6	29.6				12	NEDª
7	12.3				6	NEDª
8	26.4				6	NEDª
9	24.1				2	NEDª
10	53.4				14	NEDª
11	15.7				18	NED
12	8.3				15	NED
13	19.8				31	NED
14	11.4				6	NED
15	34.2				18	NED
16	13.2				22	NED
17	9.9				23	NED
18	7.8				N/A
19	20.2				41	NED
20	8.7				5	NED
21	123.9			77	78	AWD
22	220	+		0	2	AWD
23	16.6			8	21	AWD
24	137.6			3	9	AWD
25	342.6	+		12	12	AWD
26	210.4	+(f)		2	2	AWD
27	105.7			8	10	AWD
28	413.3	+		3	11	AWD
29	291.9	+		0	1	AWD
30	261.6	+(f)		4	4	AWD
31	97.2			33	34	AWD
32	97.2			7	12	DOD
33	210.4			6	30	AWD
34	142.6	+		12	17	AWD
35	156			72	76	DOD
36	127.3			46	83	AWD
37	208.3			3	9	AWD

ª With respect to adrenal disease.

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

☒

Distribution of tumor proliferative fraction (TPF) measured with MiB-1 immunostaining in adrenocortical neoplasms. Horizontal bar indicates mean value.

500

TPF

400

300

200

100

0

H

AD

CA

R/MCA

Eight of the 12 (66.6%) primary carcinomas were p53 positive (Fig. 4), whereas only one of the seven recurrent or metastatic tumors was positive. The pattern of p53 staining

was diffuse in six cases and focal in three of the nine positive cases. Among the malignant cases, the mean TPF was 251 for p53 positive cases, whereas p53-negative cases had a mean TPF of 143 (p = 0.008). None of the benign lesions were p53 positive. Using the literature criteria (7,18,20,21), as the standard for distinguishing benign from malignant adrenal tumors, p53 staining had a sensitivity of 67%, specificity of 100%, and positive predictive value of 82%.

Thirty-nine cases, all benign and 19 malignant, were RB positive. One malignant case was considered uninterpretable because of lack of internal control. In general, the pattern of staining was diffuse.

The follow-up in the primary carcinoma group was not long enough to be able to correlate any parameter with the aggressiveness of the tumor in terms of survival. p53 immunostaining appeared to correlate with a shorter disease-free interval (DFI). If one assumes that the p53 staining pattern remains unaltered between primaries, recurrences, and metastases, patients with p53-negative tumors had a longer DFI (median 8 months) than did patients with p53-positive tumors (median 3 months; p = 0.02). However, if the recurrent and metastatic cases are removed from the analysis, the result is no longer

statistically significant, despite the median DFI remaining essentially unchanged for p53-positive and -negative tumors.

Back to Top

DISCUSSION

Differentiation between malignant and benign adrenal tumors may be difficult. Until now, the most important criteria of malignancy was the presence of metastasis. Even though the weight of the tumor was traditionally considered an important parameter (11), malignant tumors of low weight with metastases also have been described (3,4). Histologic criteria proposed by several investigators (7,18,20), include nuclear grade, mitotic rate, atypical mitoses, growth pattern, tumor necrosis, fibrous bands, cellular pleomorphism, and invasion of the capsule or venous or sinusoidal structures. It is generally accepted that no single criterion is useful alone and that combination of them is the most helpful approach to make the diagnosis of malignancy. However, there still remains a small number of cases in which it is extremely difficult to predict behavior, and the term “adrenal cortical tumor of undetermined malignant potential” has been proposed (12).

As has been previously described in several studies (7,18,20,21), we found that tumor weight and mitotic counts correlated with histologic diagnosis. Weiss et al. (21), found that carcinomas with mitotic counts of>20/50 hpf had an associated significantly lower median survival than did those with mitotic counts of <20/50 hpf, and they proposed that these tumors should be designated as high-grade carcinomas. In our study we cannot confirm their findings because we do not have long-term follow-up. If we consider the disease-free interval, there is no correlation with mitotic counts. One case had >100 mitoses per 50 hpf and had the longest disease-free period of the group (77 months). Tumor weight showed no association with aggressive behavior in the carcinoma group. Three of the four smallest carcinomas with weight of <120 g were already metastatic at the time of presentation.

This study was designed to evaluate the diagnostic utility of some specific immunohistochemical markers and correlate them with prognosis. Our results suggest that immunohistochemistry is a helpful diagnostic aid, particularly the use of MiB-1 and p53. We found statistically significant differences on the TPF values between the benign and malignant lesions (p<0.0001). The distribution of the TPF values (Fig. 3), shows a discrete separation of the hyperplasias and adenomas from the carcinoma cases. In only one instance did a carcinoma value overlap with the benign TPF values. However, the tissues in this case had to be decalcified because the prominent osseous metaplasia of the tumor, possibly altering the antigenicity of the Ki-67 molecule. The TPF values

were lower in the recurrent/metastatic carcinomas than in the primary lesions, but they were also significantly higher than in the benign lesions (p < 0.0001).

These findings are in agreement with the ones found by Goldblum et al. (5), in which they found significant differences between adrenocortical adenomas and carcinomas using what they called Ki-67 score and labeling index. Their mean labeling index, which is the percentage of positive cells after counting 1,000 cells, was 1.3 for the adenomas and 8.3 for the carcinomas. These results correspond to our TPF values of 13 and 83. Our results showed similar values for the adenomas (14.9), whereas our carcinoma values are significantly higher (208.1 and 166.1 for primary and metastatic/recurrent carcinomas, respectively).

Goldblum et al. (5), also concluded that MiB-1 immunoreactivity was helpful in predicting clinical outcome. We are unable to evaluate parameters of survival, but high TPF values (>200) appear to correlate with shorter disease-free interval, but a longer follow-up period is necessary.

The p53 gene acts as a tumor suppressor gene similar to the RB gene, and its mutation is one of the most common genetic abnormalities in cancers (6,13,15). Mutations of p53, in addition to abolishing the tumor suppressor activity of the protein, also stabilize it, resulting in accumulation of levels detectable by immunohistochemistry (22). Ohgaki et al. (16). found missense point mutations in three of 15 primary adrenocortical carcinomas (20%) and one nonmissense mutation in 18 adenomas, suggesting that loss of p53 gene function may play an important role in the development of adrenocortical carcinomas. Despite one study (14), that found positive p53 immunostaining in five of 16 adrenocortical adenomas, we did not find any positive case among the adenomas or hyperplasias. In contrast, we found that eight of the 12 primary carcinomas and one of the seven recurrent/metastatic carcinomas were positive for p53. Our results, then, not only implicate p53 in adrenal cortical tumorigenesis but suggest its association with malignant behavior. Moreover, our data suggest that positive p53 immunostaining may be associated with shorter disease-free interval. If the whole group of primary, recurrent, and metastatic tumors is considered, the difference in disease-free interval is significant. This finding is consistent with observations in other tumors in which p53 abnormalities have been proposed as a marker of more aggressive malignancy (1,2,17).

Lastly, inactivation of the RB gene has been implicated in the pathogenesis of a variety of human cancers (2,19). Our results fail to demonstrate any association between RB immunostaining and tumor diagnosis or prognosis.

In conclusion, we have found that MiB-1 and p53 immunostaining can be of great assistance in recognizing adrenocortical carcinomas. High TPF (>80), as measured by

staining with MiB-1, and positive p53 strongly correlate with malignancy and therefore may be useful in the differential diagnosis between benign and malignant adrenal lesions. There seems to be an association between high TPF (>200) and positive p53 staining with short disease-free interval. Clinical outcome in terms of survival could not be evaluated in this study because none of our patients in the primary carcinoma group died. RB stain does not appear to be helpful in distinguishing benign from malignant adrenocortical lesions.

Acknowledgment: We thank Ruby Howard, Cynthia Harris, and Sarah Delay-Brown for technical assistance with histologic sectioning and immunostaining.

Back to Top

REFERENCES

1. Andersen TI, Holm R, Nesland JM, Heimdal KR, Ottestad L, Borresen AL. Prognostic significance of TP53 alterations in breast carcinoma. Br J Cancer 1993;68:540-8.

Cited Here …

2. Benedict W, Xu H, Hu S, Takahashi R. Role of the retinoblastoma gene in the initiation and progression of human cancer. J Clin Invest 1990;85:988-93. Cited Here …

3. Didolkar MS, Bescher RA, Elias EG, Moore RH. Natural history of adrenal cortical carcinoma: a clinicopathologic study of 42 patients. Cancer 1981;47:2153-61. Cited Here …

4. Gandour MJ, Grizzle WE. A small adrenocortical carcinoma with aggressive behavior. Arch Pathol Lab Med 1986;110:1076-9. Cited Here …

5. Goldblum JR, Shannon R, Kaldjian EP, et al. Immunohistochemical assessment of proliferative activity in adrenocortical neoplasms. Mod Pathol 1993;6:663-8. Cited Here …

6. Hollstein M, Sidransky D, Vogelstein B, Harris C. p53 mutations in human cancers. Science 1991;253:49-53.

Cited Here …

7. Hough AJ, Hollifield JW, Page DL, Hartmann WH. Prognostic factors in adrenal cortical tumors. A mathematical analysis of clinical and morphologic data. Am J Clin Pathol 1979;72:390-9.

Cited Here …

8. Incidence Data National Cancer Institute Monograph 41. Bethesda, Maryland: US Department of Health, Education and Welfare, Public Health Service, National Institutes of Health, National Cancer Institute; 1975.

Cited Here …

9. Ito T, Seyama T, Mizuno T, et al. Unique association of p53 mutations with undifferentiated but not with differentiated carcinomas of the thyroid gland. Cancer Res 1992;52:1369-71.

Cited Here …

10. Kannan CR. The adrenal gland. In: Clinical Surveys in Endocrinology. New York: Plenum, 1988.

Cited Here …

11. Karsner H. Tumors of the Adrenal. Washington, DC: Armed Forces Institute of Pathology, 1950.

Cited Here …

12. Lack EE. Pathology of the adrenal glands. In: Contemporary Issues in Surgical Pathology. New York: Churchill Livingstone, 1990.

Cited Here …

13. Levine A, Momand J, Finlay C. The p53 tumour suppressor gene. Nature 1991;351:453-6.

Cited Here …

14. Lin S, Lee Y, Tsai J. Mutations of the p53 gene in functional adrenal neoplasms. J Clin Endocrinol Metab 1994;78:483-91.

Cited Here …

15. Nigro JM, Baker SJ, Preisinger AC, et al. Mutations in the p53 gene occur in diverse human tumour types. Nature 1989;342:705-8.

Cited Here …

16. Ohgaki H, Kleihues P, Heitz PU. p53 mutations in sporadic adrenocortical tumors. Int J Cancer 1993;54:408-10.

Cited Here …

17. Thor AD, Moore DH, II, Edgerton SM, et al. Accumulation of p53 tumor suppressor gene protein: an independent marker of prognosis in breast cancers. J Natl Cancer Inst 1992;84:845-55.

Cited Here …

18. Van Slooten H, Schaberg A, Smeenk D, Moolenaar AJ. Morphologic characteristics of benign and malignant adrenocortical tumors. Cancer 1985;55:766-73.

Cited Here …

19. Weinberg RA. Tumor suppressor genes. Science 1991;254:1138-46.

Cited Here …

20. Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984;8:163-9. Cited Here …

21. Weiss LM, Medeiros LJ, Vickery AL. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13:202-6. Cited Here …

22. Wynford-Thomas D. p53 in tumour pathology. Can we trust immunocytochemistry? J Pathol 1992;166:329-30.

Cited Here …

Keywords:

Adrenal; Adrenocortical neoplasms; MiB-1; p53; Retinoblastoma gene product C Lippincott-Raven Publishers