Adrenal Cortical Carcinoma with a Liver Metastasis in a Child
Report of a Case with Fine Needle Aspiration Findings and Image Analysis of Nuclear DNA
Krzysztof Moroz, M.D., William H. Robichaux, M.D., Mary K. Cheles, H.T.L., and Nina Dhurandhar, M.D.
BACKGROUND: Adrenal cortical carcinoma is a rare and highly malignant neoplasm. Liver metastases from it may require special differential techniques in addition to cytologic findings.
measures of the biologic potential of adrenal cortical car- cinoma. The validity of these tests is enhanced with the use of image-based quantitative approaches that provide greater reproducibility and objectivity of their results. (Acta Cytol 1997;41:569-576)
CASE: A 14-month-old child had adrenal cortical carci- noma and a liver metastasis confirmed by fine needle as- piration biopsy (FNAB). Histologic and clinical crite- ria of malignancy only par- tially applied to this tumor. Immunohistochemical analysis of both the primary tumor and metastasis sup- ported an adrenal cortical origin. Other ancillary tests- DNA ploidy, proliferation index (proliferating cell nuclear antigen, Ki-67) and p53 protein immunolocalization-were utilized in a diagnostic se- quence. Flow cytometric and image analysis of DNA ploidy of the primary tumor gave similar results. Assess- ment of DNA content of the adrenal cortical carcinoma and liver metastasis by image analysis showed identical patterns.
CONCLUSION: The applicability of new diagnostic techniques to FNAB material may provide new objective
More studies are needed to correlate p53 protein overexpression in adrenal cortical carcinoma with ploidy and proliferation marker assessment.
Keywords: adrenal cortex neoplasms; aspiration biopsy; image analysis, computer-assisted.
Adrenal cortical carcino- ma is rare in both children and adults, constituting <1% of all malignant neoplasms.32,38 The criteria for malignancy in adrenal cortical tumors in adults are based on a combination of histologic and clini- cal features.16,34,36 In children, however, the prog- nostic value of the same criteria is less well estab- lished,5,6,19 In particular, tumor size and weight as predictors of aggressive behavior are question- able.5,6 The value of DNA ploidy and proliferation index results in the assessment of malignancy is controversial as well.4,5,7,13 With newer radiologic imaging techniques, smaller tumors are being de-
Ms. Cheles is Anatomic Pathology Manager.
Dr. Dhurandhar is Associate Professor.
Address reprint requests to: Krzysztof Moroz, M.D., Department of Pathology SL79, Tulane University School of Medicine, 1430 Tulane Avenue, New Orleans, Louisiana, 70112-7389.
Received for publication May 31, 1995.
Accepted for publication August 10, 1995.
0001-5547 /97 /4102-0569/$02.00/0 @ The International Academy of Cytology Acta Cytologica
tected,18 and with the recent advent of fine needle aspiration biopsy (FNAB), a smaller amount of ma- terial is available for diagnosis.18,35 Furthermore, a histologic scoring system may not be applicable or
Reproducibility and consistency of the results may be greatly enhanced by use of a quantitative approach, image analysis
technically feasible for cytologic specimens, and other ways to establish the diagnosis should be con- sidered.
We describe a case of adrenal cortical carcinoma in a 14-month-old child who developed a liver metastasis. FNAB findings from the liver lesion and corresponding histology from the adrenal primary were compared.
Case Report
The mother of a 14-month-old, black girl first noted enlargement of the clitoris and development of fine, sparse, dark public hair (Tanner stage III) two months earlier. There was no hypertension, intesti- nal dysfunction, edema, excessive weight gain or behavioral changes. Complete blood count and electrolytes were normal. Total testosterone (12 ng/dL, normal 3-10), free testosterone (1.6 pg/mL, 0.15-0.6), percent free testosterone (1.3%, 0.4-0.9%), dehydroepiandrosterone (DHEA) (10,880 ng/dL) and DHEA-sulfate (1,820 µg/dL) were elevated, and sex hormone binding globulin was low (1.0 ug/dL, 1.8-5.5). Radiographic studies showed a 7-8-cm right adrenal mass and a separate, noncon- tiguous 1-2-cm central hepatic nodule in the right lobe. At surgery, the adrenal mass was adherent to the inferior vena cava and could not be completely excised. Computed tomography-guided fine nee- dle aspiration of the liver nodule (inaccessible at surgery) was subsequently performed.
Histopathology
The adrenal specimen consisted of a partially en- capsulated, 56-g soft tissue mass (Figure 1) with a variegated, tan-yellow-maroon cut surface. Micro- scopic examination showed sheets of mitotically ac- tive, pleomorphic tumor cells with abundant, vac- uolated, eosinophilic cytoplasm and prominent
nucleoli (Figure 2). Mitotic figures were multipolar and ranged from one in one to one in three high- power fields. Multiple zones of necrosis and calcifi- cation were admixed with the tumor cells. Except for rare fibrous bands within the tumor, nests of tumor cells were invested in thin, fibrovascular stroma. Focally, normal adrenal cortical cells were compressed in the attenuated fibrous capsule. Areas suspicious for vascular invasion were identi- fied.
Cytopathology
The aspirate was cellular and revealed numerous large, single polygonal cells with abundant cyto- plasm (Figure 3A). Cell outlines were irregular, with cytoplasmic protrusions that focally had a sharp, pink boundary on Diff-Quik and were remi- niscent of secretory vacuoles seen in “flare” cells in thyroid aspirates. The cytoplasm was dense, with fine, blue granules concentrated around the nuclei with a paler periphery. On Papanicolaou stain, this differential staining of the cytoplasm was lost. Sev- eral cells had minute peripheral cytoplasmic vac- uoles. Nuclei were large, eccentric, and round to oval, with smooth outlines. Nuclear pleomorphism and hyperchromasia were striking. The chromatin was coarsely granular, with irregular clumping. Single and multiple round and angular nucleoli were conspicuous. Mitotic figures were easily iden- tifiable (Figure 3B). There were many binucleate cells and occasional trinucleate and quadrinucleate cells. Rare intranuclear vacuoles were seen. Tumor
cells could be clearly differentiated from hepatic cells, which were much smaller, had amphophilic cytoplasm and finer blue granules, and lacked the dense perinuclear staining in the tumor cells. Hepa- tocyte nuclei were small, with dense chromatin. Differences between tumor cells and normal hepa- tocytes were accentuated on the cell block, which demonstrated the two cell types in close apposition to each other (Figure 4). Tumor cells were gigantic and polygonal, with well-defined cell borders, and lacked the sinusoidal pattern seen in the hepatic fragments. The cytoplasmic granularity of the
tumor cells was prominent. Nuclear pleomor- phism, multinucleation and irregular chromatin distribution with multiple nucleoli in the cell block were similar to those features in the aspirate smears.
Special Studies and Results
Immunohistochemical staining was performed on 4-um sections of resected adrenal tumor and the cell block of the liver aspirate by the streptavidin-biotin complex method, as described previously.11,29 Ap- propriate negative and positive controls were ex- amined. Differences in staining pattern and intensi- ty in the primary lesion, liver metastasis and normal hepatocytes were assessed (Table I). The primary and metastatic tumors showed similar staining patterns. Keratins (AE1/AE3 and CAM 5.2) showed positive staining in hepatocytes but were negative in tumor cells (Figure 5A). Normal adrenal tissue expressed staining with CAM 5.2 only. Staining of bile canaliculi with polyclonal car- cinoembryonic antigen further helped to separate normal liver from tumor cells. In reverse fashion, vimentin, synaptophysin and neuron-specific eno- lase (NSE) were expressed by tumor cells (Figure 5B) but not by normal hepatocytes.
Tumor proliferative activity was evaluated using monoclonal antibodies to proliferating cell nuclear antigen (PCNA), clone PC10 (Dako Corporation, Carpinteria, California, U.S.A.) and Ki-67 (MIB-1, Immunotech, Westbrook, Maine, U.S.A.) both at 1:50 dilution (Figure 5C). Normal adrenal and liver
A
B
tissues present on the same slide were used as in- ternal controls for baseline proliferative activity. The percentage of cell nuclei staining positively was quantitated by the CAS 200 Image Analysis System (Becton Dickinson Cellular Imaging Systems, San Jose, California, U.S.A.) with Quantitative Prolifer- ative Index software (Cell Analysis System, Inc., Lombard, Illinois, U.S.A.). At least 25,000 um2 of the nuclear area (a mean of 24 fields; minimum 14, max- imum 36) were analyzed for each run. The results are presented in Table II.
Expression of p53 protein was studied using mouse monoclonal antibody DO7 (Signet Laborato- ries, Dedham, Massachusetts, U.S.A.) specific for the human p53 oncoprotein (wild and mutant forms). Quantitation of the number of nuclei ex- pressing p53 protein was evaluated by the CAS 200 with the Cell Measurement Program software (ver- sion 3.0, Cell Analysis System). In both the adrenal cortical carcinoma and liver metastasis, 53% of the nuclei showed p53-specific immunostaining (Fig- ure 5D). Normal adrenal and liver tissue adjacent to the tumor had only weak expression of stain in sin- gle, isolated nuclei (<1%).
The DNA content of tumor cell nuclei from paraffin-embedded material was retrospectively analyzed by flow cytometry and image analysis. The nuclear suspension for flow cytometry from the adrenal tumor was prepared using Hedley’s tech- nique.15 At least 20,000 events per sample were an- alyzed with a Coulter EPICS XL (Coulter Corpora- tion, Hialeah, Florida, U.S.A.) flow cytometer. DNA content of normal cells (stromal, inflammatory) was used as a built-in internal control. The DNA index (DI) was determined by calculating the ratio of the mean fluorescence of each population divided by the mean fluorescence of the diploid standard. Tumor with DI >1.1 was considered aneuploid. Analysis of a suspension of tumor cell nuclei demonstrated an aneuploid population of cells (60% of total) with DI = 1.79 and coefficient of vari- ation of 6.0. Approximately 26% of the cells were in G2 of the aneuploid cycle. The percentage of cells in
| Antibody | Dilution | Source | Adrenal carcinoma | Normal adrenal | Liver metastasis | Normal liver |
|---|---|---|---|---|---|---|
| AFP | 1:20,000 | Dako, Carpinteria, California, U.S.A. | 1 | 1 | 1 | |
| Factor XIllaª | 1:250 | Calbiochem, La Jolla, California, U.S.A. | 1 | 1 | 1 | |
| CAM 5.2 | Prediluted | Becton Dickinson, San Jose, California, U.S.A. | 1 | + | 1 | + |
| AE1/AE3 | 1:40 | Signet Labs, Dedham, Massachusetts, U.S.A. | 1 | 1 | 1 | +b |
| CEAm | 1:200 | Dako | 1 | 1 | 1 | |
| CEAp | 1:400 | Dako | 1 | 1 | +c | |
| EMA | 1:200 | Dako | 1 | 1 | 1 | |
| Vimentin | 1:40 | Dako | + | 1 | + | 1 |
| Chromogranin | 1:50 | Dako | 1 | 1 | 1 | |
| NSE | 1:50 | Dako | + | + | 1 | |
| Synaptophysin | 1:200 | Dako | + | 1 | + | 1 |
| Leu-7 | 1:10 | Becton Dickinson | 1 | 1 | 1 |
AFP= c-fetoprotein, CAM 5.2 = low molecular weight keratins, AE1/AE3 = mixture of low and high molecular weight keratins, CEAm = monoclonal carcinoem- bryonic antigen, CEAp = polyclonal CEA, EMA = epithelial membrane antigen. aPositive in endothelial cells.
bFocally positive.
cPositive in bile canaliculi.
A
C
B
D
S-phase of the aneuploid cycle was 5.3%, and the proliferation fraction (percentage of cells in S-phase + G2/M phases of the aneuploid peak) was 31.6%.
Image analysis was performed with the CAS 200 analyzer on 5-um sections from formalin-fixed, paraffin-embedded blocks of adrenal tumor and liver aspirate cell block. These were stained with the CAS DNA-staining kit (Cell Analysis System). Tetraploid rat hepatocyte nuclei (Cell Analysis Sys- tem) were used for instrument calibration. The pa- tient’s lymphocytes and hepatocytes were used as internal controls for the adrenal and liver lesions, respectively. To correct for cut nuclei, the tissue cor- rection feature of CAS version 3.0 ploidy software
(Cell Analysis System) was used. Histograms were considered diploid if the mean DNA indices of the Go/G1 peak were within an established diploid range based on the internal control (DI=0.9-1.1). DNA analysis of tumor cell nuclei from the prima-
| Specimen | PCNA (%) | Ki-67 (%) |
|---|---|---|
| Normal adrenal | 4.36 | 3.53 |
| Adrenal cortical carcinoma | 41.93 | 45.16 |
| Normal liver | 2.67 | 2.24 |
| Liver metastasis | 48.98 | 29.23 |
| Specimen | Main peak Diploid cells Aneuploid cells Total cells DI Mean DI CV DI Mean DI CV evaluated | ||||||
|---|---|---|---|---|---|---|---|
| Patient's lymphocytes | 0.97 | 0.96 | 2.69 | — | — | — | 150 |
| Adrenal carcinoma | 0.98 | 1.00 | 6.05 | 1.73 | 1.74 | 5.34 | 300 |
| Normal liver | 0.95 | 1.00 | 6.05 | — | — | — | 150 |
| Liver metastasis | 1.02 | 1.05 | 2.67 | 1.74 | 1.74 | 5.89 | 300 |
CV = coefficient of variation.
ry and metastatic lesion yielded aneuploid popula- tions in both cases (DI = 1.73 and 1.74, respectively). DNA ploidy data are summarized in Table III.
Discussion
There are only a few descriptions of fine needle as- piration cytologic findings of primary adrenal corti- cal carcinoma in the literature.17,21,26,38 The most frequent metastatic sites are lung and liver.19
According to Cagle et al,6 the size of the tumor in pediatric adrenal cortical carcinomas is the only predictor of malignancy. In their study of 23 cases, all tumors weighing >500 g were malignant. Fea- tures including mitoses; necrosis; broad, fibrous bands; and cellular pleomorphism were present in both benign and malignant tumors. These features were more frequent in benign pediatric tumors than in benign adult tumors. Bugg et al5 used four histo- logic criteria-mitotic index, confluent necrosis, atypical mitoses and nuclear grade-in their assess- ment of malignant behavior. Histologic classifica- tion and tumor weight > 100 g were the most pre- dictive of malignancy in their study of 54 cases. The proliferative index was marginally related, and ploidy was not significantly predictive of outcome.
Application of immunohistochemical stains may be useful in supporting the diagnosis of adrenal cortical carcinoma. Cytokeratins (AE1/AE3 and CAM 5.2) are usually negative in this tumor in con- trast to normal adrenal cortex.8,37 An opposite pat- tern is seen with vimentin, which is present in the majority of adrenal cortical carcinomas, whereas normal adrenal cortex stains negatively.8,37 Adeno- mas show variable staining for vimentin.
Differentiation of liver metastases of adrenal cor- tical carcinoma from primary hepatic lesions may be a diagnostic challenge. Immunohistochemistry in such cases may be very helpful. a-Fetoprotein
and a-1-antitrypsin may show variable staining in hepatocellular carcinoma12 but are negative in adrenal cortical tumors. Keratins, however, are typ- ically expressed in primary liver carcinomas20 but are negative in adrenal cortical carcinomas, as noted above. In addition, positive immunohisto- chemical staining for neuron-specific enolase (NSE) and synaptophysin has been reported in adrenocor- tical carcinomas,28 but not in benign or malignant hepatocytes.
Aneuploidy is an expression of genomic instabil- ity. Its presence in adrenocortical carcinoma has been documented by cytogenetic studies.23,24 How- ever, data on the predictive value of nuclear DNA content in adrenal cortical tumors are controver- sial.1,4,5,7,25,31 Adrenocortical carcinomas may be diploid, and adenomas may be aneuploid.7,13 Pa- tients with diploid and aneuploid carcinomas have been clinically free of metastases.5,7,25,31 However, in published series, most patients who developed metastases had aneuploid tumors.1,4,31 Therefore, aneuploidy may help to select a group of adrenal cortical tumors with an increased potential for metastasis. Recently, overexpression of p53 in colo- rectal carcinoma,3 breast carcinoma33 and lung car- cinoma27 has been shown to provide additional in- formation on the metastatic potential of these tumors. Immunohistochemical detection of p53 protein overexpression is considered to be an indi- cator of p53 gene mutation.2,9,14 There is strong ev- idence that abnormalities of p53 represent the most common molecular change in human malignan- cies22 and are the most common genetic abnormali- ty associated with malignancy.30 Deb et al10 showed that the expression of mutant p53 corre- lates significantly with an increase in PCNA pro- moter activity. The latter may be assessed immuno- histochemically with antibodies to PCNA/cyclin.
More studies are needed to correlate p53 protein overexpression in adrenal cortical carcinoma with ploidy and proliferation marker assessment. The significance of a possible correlation with biologic behavior and prognosis remains to be established. Reproducibility and consistency of the results may be greatly enhanced by use of a quantitative ap- proach, image analysis. The applicability of all of these techniques to FNAB material is important.
References
1. Amberson JB, Vaughan ED Jr, Gray GF, Naus GJ: Flow cyto- metric analysis of nuclear DNA from adrenocortical neo- plasms: A retrospective study using paraffin-embedded tis- sue. Cancer 1987;59:2091-2095
2. Baas IO, Mulder J-WR, Offerhaus GJA, Vogelstein B, Hamil- ton SR: An evaluation of six antibodies for immunohisto- chemistry of mutant p53 gene product in archival colorectal neoplasms. J Pathol 1994;172:5-12
3. Bertorelle R, Esposito G, Del Mistro A, Belluco C, Nitti D, Lise M, Chieco-Bianchi L: Association of p53 gene and pro- tein alterations with metastases in colorectal cancer. Am J Surg Pathol 1995;19:463-471
4. Bowlby LS, DeBault LE, Abraham SR: Flow cytometric analysis of adrenal cortical tumor DNA. Cancer 1986;58: 1499-1505
5. Bugg MF, Ribeiro RC, Roberson PK, Lloyd RV, Sandrini R, Silva JB, Epelman S, Shapiro DN, Parham DM: Correlation of pathologic features with clinical outcome in pediatric adrenocortical neoplasia. Am J Clin Pathol 1994;101:625-629
6. Cagle PT, Hough AJ, Pysher TJ, Page DL, Johnson EH, Kirk- land RT, Holcombe JH, Hawkins EP: Comparison of adrenal cortical tumors in children and adults. Cancer 1986;57: 2235-2237
7. Cibas ES, Medeiros LJ, Weinberg DS, Gelb AB, Weiss LM: Cellular DNA profiles of benign and malignant adrenocorti- cal tumors. Am J Surg Pathol 1990;14:948-955
8. Cote RJ, Cordon-Cardo C, Reuter VE, Rosen PP: Im- munopathology of adrenal and renal cortical tumors: Coor- dinated change in antigen expression is associated with neo- plastic conversion in the adrenal cortex. Am J Pathol 1990;136:1077-1084
9. Davidoff AM, Kerns BJM, Iglehart JD, Marks J: Maintenance of p53 alterations throughout breast cancer progression. Cancer Res 1991;51:2605-2610
10. Deb S, Jackson CT, Subler MA, Martin DW: Modulation of cellular and viral promoters by mutant human p53 proteins found in tumor cells. J Virol 1992;66:6154-6170
11. Fucich LF, Cheles MK, Thung SW, Gerber MA, Marrogi AJ: Primary vs metastatic hepatic carcinoma: An immunohisto- chemical study of 34 cases. Arch Pathol Lab Med 1994; 118:927-930
12. Ganjei P, Nadji M, Albores-Saavedra J, Morales AR: Histo- logic markers in primary and metastatic tumors of the liver. Cancer 1988;62:1994-1998
13. Haak HR, Cornelisse CJ, Hermans J, Cobben L, Fleuren GJ:
Nuclear DNA content and morphological characteristics in the prognosis of adrenocortical carcinoma. Br J Cancer 1993;68:151-155
14. Harris CC, Hollstein M: Clinical implications of the p53 tumor-suppressor gene. N Engl J Med 1993;329:1318-1327
15. Hedley DW, Friedlander ML, Taylor IW, Rugg CA, Mus- grove EA: Method for analysis of cellular DNA content of paraffin-embedded pathological material using flow cytom- etry. J Histochem Cytochem 1983;31:1333-1335
16. Hough JA, Hollifield JW, Page DL, Hartmann WH: Prognos- tic factors in adrenal cortical tumors: A mathematical analy- sis of clinical and morphologic data. Am J Clin Pathol 1979;72:390-399
17. Katz RL, Patel S, Mackay B, Zornoza J: Fine needle aspiration cytology of the adrenal gland. Acta Cytol 1983;28:269-282
18. Katz RL, Shirkhoda A: Diagnostic approach to incidental adrenal nodules in the cancer patient. Cancer 1985;55:1995- 2000
19. Lack EE, Mulvihill JJ, Travis WD, Kozakewich HPW: Adren- al cortical neoplasms in the pediatric and adolescent age group: A clinicopathologic study of 30 cases with emphasis on epidemiological and prognostic factors. Pathol Annu 1992;27:1-53
20. Lai Y-S, Thung SN, Gerber MA, Chen M-L, Schaffner F: Ex- pression of cytokeratins in normal and diseased livers and in primary liver carcinomas. Arch Pathol Lab Med 1989;113: 134-138
21. Levin NP: Fine needle aspiration and histology of adrenal cortical carcinoma: A case report. Acta Cytol 1980;25:421- 424
22. Levine AJ, Momand J, Finlay CA: The p53 tumor suppressor gene. Nature 1991;351:453-456
23. Limon J, Dal Cin P, Kakati S, Huben RP, Rao U, Sandberg AA: Cytogenetic findings in a primary adrenocortical carci- noma. Cancer Genet Cytogenet 1987;26:271-277
24. Marks JL, Wyandt HE, Beazley RM, Milunsky JF, Sheahan K, Milunsky A: Cytogenetic studies of an adrenal cortical carci- noma. Cancer Genet Cytogenet 1992;61:96-98
25. Moore L, Bramwell H, Byard RW: DNA analysis and clinical outcome in pediatric adrenal cortical tumors. Pathology 1993;25:144-147
26. Nance KV, McLeod DL, Silverman JF: Fine-needle aspiration cytology of spindle cell neoplasms of the adrenal gland. Diagn Cytopathol 1992;8:235-241
27. Quinlan DC, Davidson AG, Summers CL, Warden HE, Doshi HM: Accumulation of p53 protein correlates with a poor prognosis in human lung cancer. Cancer Res 1992; 52:4828-4831
28. Schroder S, Padberg BC, Achilles E, Holl K, Gralle H, Klop- pel G: Immunohistochemistry in adrenocortical tumors: A clinicopathological study of 72 neoplasms. Virchows Arch Pathol Anat 1992;420:65-70
29. Sheibani K, Tubbs RR: Enzyme immunohistochemistry: Technical aspects. Semin Diagn Pathol 1984;1:235-250
30. Stratton MR: The p53 gene in human cancer. Br J Cancer 1992;28:293-295
31. Taylor SR, Roederer M, Murphy RF: Flow cytometric DNA
analysis of adrenocortical tumors in children. Cancer 1987;59:2059-2063
32. Third National Cancer Survey: Incidence Data. National Cancer Institute Monograph 41. Bethesda, Maryland, Na- tional Cancer Institute, 1975
33. Thor AD, Moore DH II, Edgerton SM, Kawasaki ES, Reih- saus E, Lynch HT, Marcus JN, Schwartz L, Chen LC, Mayall BH, Smith HS: Accumulation of p53 tumor suppressor gene protein: An independent marker of prognosis of breast can- cers. J Natl Cancer Inst 1992;84:845-855
34. Van Slooten H, Schaberg A, Smeenk D, Moolenaar AJ: Mor- phologic characteristics of benign and malignant adrenocor- tical tumors. Cancer 1985;55:766-773
35. Wadih GE, Nance KV, Silverman JF: Fine-needle aspiration cytology of the adrenal gland. Arch Pathol Lab Med 1992; 116:841-846
36. Weiss LM: Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984;8:163-169
37. Wick MR, Cherwitz DL, McGlennen RC, Dehner LP: Adre-nocortical carcinoma: An immunohistochemical com- parison with renal cell carcinoma. Am J Pathol 1986;122: 343-352
38. Young JL, Ries L, Silverberg E, Horm JW, Miller RW: Can- cer incidence, survival, and mortality for children younger than age 15 years. Cancer 1986;58:598-602