Silver binding nucleolar organizer regions in adrenocortical neoplasia
B. ÖZ, S. DERVISOGLU, M. DERVISOGLU* AND Ö. ÖZ
Department of Pathology, Istanbul University Cerrahpasa Medical School, Istanbul,
*Department of Chemical Engineering, Bogazici University, Istanbul, and } Department of Biomedical Statistics, Istanbul University, Cerrahpasa Medical School, Istanbul, Turkey
Accepted for publication 6 January 1992
ÖZ B., DERVISOGLU S., DER VISOGLU M. AND ÖZ Ö. (1992) Cytopathology 3, 93-99
Silver binding nucleolar organizer regions in adrenocortical neoplasia
The possible contribution of the silver colloid technique for staining nucleolar organizer regions in the distinction between benign and malignant adrenocortical neoplasms was investigated. Nine cases of adenoma, eight cases of carcinoma, nine cases of hyperplasia and four normal adrenal cortex specimens were examined. The mean silver binding nucleolar organizer region (Ag-NOR) value in adenoma was 4.29, and in carcinoma 7.16 (P <0.001). Adenomas with diameters >3 cm had significantly higher Ag-NOR counts than smaller adenomas. For normal cortex, the mean Ag-NOR value was 2.05 and in hyperplasia, 3.62. The results indicate that the Ag-NOR technique can help in differential diagnosis between benign and malignant adrenocortical lesions and thus may have a prognostic value.
Keywords: nucleolar organizer region, silver staining, adrenocortical neoplasia
INTRODUCTION
Nucleolar organizer regions (NORs) are loops of ribosomal DNA which are present in the nucleoli of cells and which transcribe to ribosomal RNA1-4. NORs have been demonstrated in tissue sections by a silver colloid technique whereby acidic proteins associated with ribo- somal RNA can be visualized as black dots known as silver binding NORs1-5. It has been suggested that the number of Ag-NORs per cell correlates with cellular activity and may be an indicator of malignancy2,6,7.
It is difficult to distinguish histologically between benign and malignant adrenocortical neoplasms and the course of the disease is sometimes the only distinguishing parameter8-11. Although a tumour mass of ≥ 50 g is considered an important indication of malignancy, low- weight malignant tumours with metastases have been described10,12-14. Recent reports
Correspondence: Dr Büge Öz, Istanbul Üniversitesi, Cerrahpasa Tip Fakültesi, Patoloji Anabilim Dali Cerrahpasa, Istanbul, Turkey.
suggest that flow cytometry could be a useful method for separating malignant and benign adrenal tumours10,15. Here we describe the identification of Ag-NORs in adrenocortical neoplasms, in order to try to distinguish between benign and malignant cases.
MATERIALS AND METHODS
Nine cases of adenoma and eight cases of carcinoma were examined. Nine cases of adreno- cortical hyperplasia (diffuse or nodular) and four fragments of normal adrenal cortical tissues were included as control. These were diagnosed through conventional histological techniques. All adenomas weighed less than 70 g and all carcinomas weighed more than 70 g. All carcinomas had capsular invasion mostly as large solid masses. Some carcinomas had vascular invasion and necrosis. Paraffin-embedded sections (3 um thick) were cut and rehydrated through graded alcohols. Ag-NOR staining was performed as described16-18. Briefly, the silver colloid solution was prepared by dissolving gelatin in 1% aqueous formic acid at a concentration of 2%; one volume of this solution was mixed with two volumes of 50% aqueous silver nitrate to obtain the final working solution. This was dropped onto the sections and left for 30 min at room temperature, in a dark room. The sections were taken to xylene and were mounted in a synthetic medium.
Ag-NORs were visualized as distinct intranuclear dots. We aimed to examine 100 cells for each case using an x 100 oil immersion lens. However, in some cases there were less than 100 countable cells due to necrosis and some staining artefacts. The mean number of Ag-NORs per cell was calculated taking the average of at least 80 countable cells. In order to test the reproducibility of data, determinations were made by two observers for each case. Previous studies had shown that more than 50 cell samples can be representative of the whole and this number of cells were examined from randomly selected fields1,19. Student’s t-test was used to compare mean Ag-NOR countings.
RESULTS
Nucleoli stained as a single ring structure by the colloid silver nitrate were counted as one Ag- NOR. A single nucleolus may contain many well-defined dots. Smaller, distinct, brownish black dots may also be present in the nuclei, lying away from the main nucleolar structures. However, these were not scored.
Ag-NOR scores recorded by two observers are shown as a scattergram in Fig. 1. Both observers found significant differences between the mean Ag-NOR counts in benign and malignant adrenocortical neoplasia. The mean numbers of Ag-NOR counts for the four case groups determined by each observer are shown in Table 1. The trends obtained by both observers show a steady increase in the number of mean Ag-NOR counts from normal adrenal cortex to carcinoma. The mean number of Ag-NORs per cell as determined by combining data from both observers was 2.05+0.034 for normal adrenal cortex and 3.62±0.034 for hyperplasia. The Ag-NOR dots were relatively regular in size and shape in both groups (Fig. 2). The mean Ag-NOR values for both normal cortex and hyperplastic cortex were significantly smaller than the mean Ag-NOR values obtained for adrenocortical adenomas (P<0.001).
The mean Ag-NOR count per cell revealed by combined data on adenoma was 4.29±0.024 (Table 1) and the Ag-NOR dots were of regular shape (Fig. 3). When the relationship between size of adenoma and mean Ag-NOR values was studied, it was found
12
(a)
10
8
6
4
2
Mean Ag-NOR value
0
I
Normal
Hyperplasia
Adenoma
Carcinoma
0
12
( b)
10
8
6
4
2
0
1
Normal
Hyperplasia
Adenoma
Carcinoma
0
| Observer 1 | Observer 2 | Combined | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Mean | n | s.d. | s.e.m. | Mean | n | s.d. | s.e.m. | Mean | s.e.m. | |
| Adrenal cortex | 2.00 | 404 | 0.81 | 0.040 | 2.10 | 346 | 0.79 | 0.042 | 2.05 | 0.034 |
| Hyperplasia | 3.81 | 792 | 1.34 | 0.047 | 3.35 | 804 | 1.46 | 0.051 | 3.62 | 0.034 |
| Adenoma | 4.74 | 891 | 1.90 | 0.064 | 3.83 | 871 | 1.70 | 0.059 | 4.29 | 0.024 |
| Carcinoma | 7.23 | 793 | 2.83 | 0.010 | 7.01 | 763 | 3.09 | 0.112 | 7.16 | 0.079 |
n, number of cells examined.
that adenomas with larger diameters (> 3 cm) had high Ag-NOR contents per cell, compared with smaller ones (<3 cm). This difference was statistically significant (P<0.01) as shown in Table 2.
Adrenocortical carcinomas had a mean Ag-NOR count of 7.16±0.079. The dots were relatively smaller in size and were usually dispersed and some differences in the number of counts from one field to another were observed (Fig. 4).
There was a statistically significant difference between Ag-NOR counts of adenomas and adrenocortical carcinomas (P<0.001). However, the counts in two cases of carcinomas overlapped with adenomas in the region of four to six Ag-NOR dots per nuclear profile.
DISCUSSION
NORs are loops of DNA which occur in the nuclei of cells which possess ribosomal RNA. In humans, five chromosomes have NORs and these are the sites that hybridize with rRNA and are important for protein synthesis4,16,20. The function of NOR-associated proteins is rather uncertain, but they probably act as regulators of rDNA transcription. Whatever their precise nature, NORs have been observed for over a decade and are most readily visualized using silver colloid technique1,5-7,17,18,20. Crocker and colleagues have pioneered the application of the Ag-NOR technique to diagnostic tumour pathology, and correlation between Ag-NOR counts and cell proliferation indices, using DNA flow cytometry20 and Ki675 have also been reported. It is also likely that correlation with ploidy can be substantiated. Thus Ag-NOR counts may assist in the grading of neoplasms2,4,5,20.
| Diameter > 3 cm (n1 = 5) | Diameter <3 cm (n) = 3) | |||||||
|---|---|---|---|---|---|---|---|---|
| Mean | n | s.d. | s.e.m. | Mean | n | s.d. | s.e.m. | |
| Observer 1 | 4.80 | 494 | 2.15 | 0.09 | 4.44 | 293 | 1.51 | 0.09 |
| Observer 2 | 4.09 | 475 | 1.93 | 0.09 | 3.31 | 297 | 1.30 | 0.08 |
n, number of cells examined
n1, number of cases.
In histopathology, the differential diagnosis of adrenocortical adenoma and carcinoma frequently causes diagnostic difficulties. The weight of tumour seems to be important in differentiating between malignant and benign tumours12,14, but there is evidence that even a few low-weight tumours may metastasize13. In some studies8,10,11 complete separation
of metastatic neoplasms from those that had not metastasized required application of histological and clinical indexes of malignancy. In one study8 the histological features evalu- ated for the distinction between benign and malignant adrenocortical tumours included fibrous bands, diffuse growth pattern, vascular invasion and the presence of tumour necrosis. In another study11 the mitotic index had the highest discriminatory value. However, both reports were unable to find any histological feature that distinguished carcinomas from adenomas consistently. Recently, rapid analysis of nuclear DNA content by flow cytometry has been advocated as an objective prognostic factor in adrenocortical neoplasia10,15.
We have investigated Ag-NOR staining as a method of distinguishing between malignant and benign forms of adrenocortical neoplasia. We found that benign cells (in normal adrenal cortex cells, in hyperplasia and in adenoma cells) tend to have a regular nucleolus containing tightly clustered Ag-NORs. In contrast, malignant cells (in carcinomas) often showed dis- persal of Ag-NORs throughout the nucleolus as well as having multiple nucleoli containing clustered Ag-NORs.
Our study revealed that the mean Ag-NOR value in hyperplasia was high compared with normal adrenal cortex (P<0.001). This may reflect hormonal influence on cellular prolifer- ation and substantiates the findings of other groups who have applied Ag-NOR techniques to other tumour-like proliferative lesions2,4,6,16,20
Mean Ag-NOR values determined in adrenocortical carcinoma are significantly higher than those found in adenomas. Overlap of mean Ag-NOR values was observed between adenomas and two carcinomas. We believe that this reflects the degree of differentiation and low mitotic rate of the tumours. However, the mean Ag-NOR value for adenomas was never found to be as high as the mean value determined for carcinomas. Our findings indicate that mean Ag-NOR values >6 may be considered indicative of malignancy.
These results suggest that the Ag-NOR technique can be easily performed on paraffin- embedded tissue sections of adrenal cortex neoplasms, and can be used for differential diagnosis as an additional tool to morphology.
REFERENCES
1 Egan M, Freeth M, Crocker J. Intraepithelial neop- lasia, human papilloma virus infection and argyro- philic nucleo-protein in cervical epithelium. Histopathology 1988; 13: 561-7.
2 Giri DD, Nottingham JF, Lawry J. Silver binding nucleolar organizer regions (Ag-NORs) in benign and malignant breast legions: correlations with ploidy and growth phase by DNA flow cytometry. J Pathol 1989; 157: 307-13.
3 Smith PJ, Skilbeck N, Harrison A, Crocker J. The effect of a series of fixatives on the Ag-NOR tech- nique. J Pathol 1988; 155: 109-12.
4 Underwood JCE, Giri DD. Nucleolar organizer regions as diagnostic discriminants for malig- nancy. J Pathol 1988; 155: 95-6.
5 Murray PG, Boldy AR, Crocker J, Ayres JG. Sequential demonstration of antigens and Ag- NORs in frozen and paraffin sections. J Pathol 1989; 159: 169-72.
6 Crocker J, Egan MJ. Correlation between NOR sizes and numbers in non-Hodgkin’s lymphomas. J Pathol 1988; 156: 233-9.
7 Howat AJ, Giri DD, Cotton DWK, Slater DN. Nucleolar organizer regions in spitz nevi and malignant melanomas. Cancer 1989; 63: 474-8.
8 Hough AJ, Hollifield JW, Page DL, Hartmann WH. Prognostic factors in adrenal cortical tumors. Am J Clin pathol 1979; 72: 390-9.
9 Nader S, Hickey RC, Sellin RV, Samaan NA. Adrenal cortical carcinoma. A study of 77 cases. Cancer 1983; 52: 707-11.
10 Page DL, Hough AJ, Gray GF. Diagnosis and prognosis of adrenocortical neoplasms. Arch Pathol Lab Med 1986; 110: 993-4.
11 Slooten HV, Schaberg A, Smeenk D, Moolenaar AJ. Morphologic characteristics of benign and
malignant aderenocortical tumors. Cancer 1985; 55: 766-73.
12 Corney AJ. Adrenal glands In: Karcioglu ZA, Someren A, eds. Practical Surgical Pathology. Health and Company. Lexington: The Collemore Press, 1985; 667-78.
13 Gondour MJ, Grizzle WE. A small adrenocortical carcinoma with aggressive behaviour: an evalu- ation of criteria for malignancy. Arch Pathol Lab Med 1986; 110: 1076-79.
14 Rosai J. Adrenal gland and other paraganglia. In: Ackerman’s surgical pathology Vol. 1, 7th ed. St Louis: C. V. Mosby Co., 1989: 789-94.
15 Amberson JB, Vaughan ED, Gray GF, Naus GJ. Flow cytometric analysis of nuclear DNA from adrenocortical neoplasms. Cancer 1987; 59: 2091-5.
16 Egan M, Crocker J. Nucleolar organizer regions in cutaneous tumours. J Pathol 1988; 154: 247-53.
17 Ploton D, Bobichon H, Adnet JJ. Ultrastructural localization of NOR in nucleoli of human breast cancer tissues using a one-step Ag-NOR staining method. Biol Cell 1982; 43: 229-32.
18 Ploton D, Menager M, Jeannesson, P. Improve- ment in the staining and in the visualization of the argyrophilic proteins of the nucleolar organizer region at the optical level. Histochem J 1986; 18: 5-14.
19 Ruschoff J, Plate K, Contractor H. Evaluation of nucleolus organizer regions (NORs) by automatic image analysis: a contribution to standardization. J Pathol 1990; 161: 113-18.
20 Crocker J, Macartney JC, Smith PJ. Correlation between DNA flow cytometric and nucleolar organizer region data in non-Hodgkin’s lym- phomas. J Pathol 1988; 154: 151-6.