IGFII and MIB1 immunohistochemistry is helpful for the differentiation of benign from malignant adrenocortical tumours
A Schmitt, P Saremaslani, S Schmid, V Rousson,2 M Montani, D M Schmid,3 Ph U Heitz, P Komminoth & A Perren
Institute for Surgical Pathology, Department of Pathology, University Hospital Zurich, 1Kantonsspital Baden, Baden, 2 Department of Biostatistics University of Zurich and 3 Clinic for Urology, University Hospital Zurich, Zurich, Switzerland
Date of submission 15 August 2005 Accepted for publication 7 December 2005
Schmitt A, Saremaslani P, Schmid S, Rousson V, Montani M, Schmid D M, Heitz Ph U, Komminoth P & Perren A (2006) Histopathology 49, 298-307
IGFII and MIB1 immunohistochemistry is helpful for the differentiation of benign from malignant adrenocortical tumours
Aims: The differentiation of adrenocortical carcinomas from adenomas may be difficult based on morphology alone. Differential expression of insulin-like growth factor (IGF) II and cyclin-dependent kinase (CDK) 4 has recently been described in these tumours. The aim of this study was to investigate the diagnostic usefulness of these markers immunohistochemically.
Methods and results: We examined 22 benign and 17 malignant adrenocortical tumours and compared IGFII and CDK4 expression with known immunohistochem- ical as well as morphological criteria of malignancy. Thirteen of 17 carcinomas showed immunohistochem- ical reactivity for IGFII, whereas all adenomas but one
were negative. Intense CDK4 expression was detected in 11 of 17 carcinomas but was present in only three of 22 adenomas. The MIB1 index was > 5% in 14 of 16 carcinomas and was < 5% in all adenomas but one. The combination of IGFII immunohistochemistry with MIB1 index led to high sensitivity and specificity in detecting adrenocortical carcinomas.
Conclusions: IGFII and MIB1 are helpful immunohisto- chemical markers to predict malignancy in adreno- cortical neoplasms. These markers can be used in addition to clinical, gross and morphological features to establish a diagnosis in difficult cases.
Keywords: adrenocortical adenoma, adrenocortical carcinoma, IGFII, immunohistochemistry, MIB1, p53 Abbreviations: ACC, adrenocortical carcinoma; CDK, cyclin-dependent kinase; FGF, fibroblast growth factor; IGF, insulin-like growth factor
Introduction
Adrenocortical carcinoma (ACC) is an infrequent malignant tumour with an incidence of 0.5-2 patients per million population per year.1-3 At the time of first presentation, metastatic spread has occurred in a significant proportion (30-85%) of patients.4 As dis- ease stage and completeness of resection are primary determinants of outcome for these patients,5,6 it is
important to recognize malignancy at the initial histological examination of the specimen. In the absence of metastases, the morphological criteria for malignancy as described by Weiss,7,8 Hough9 and van Slooten10 are used to differentiate adenomas from carcinomas. However, there remains a group of tumours in which application of those criteria do not lead to a straightforward diagnosis and their biological behaviour and aggressiveness are difficult to predict. Therefore, additional markers of malignancy are needed to achieve more precise diagnosis. There is a consensus that proliferative activity measured by MIB1 staining is significantly higher in adrenocortical
Address for correspondence: Anja Schmitt, Department of Pathology, University Hospital Zurich, Schmelzbergstr 12, CH-8091 Zurich, Switzerland. e-mail: anja.schmitt@usz.ch
carcinomas than it is in adenomas,11-16 but it is uncertain where to set a significant threshold. The tumour suppressor gene p53 also seems to be import- ant in adrenocortical carcinogenesis. A p53 mutation frequency of 20%17 and 27%18, respectively, has been described in adrenocortical carcinomas. On a molecu- lar basis, Zhao et al.19 were able to show co-amplifica- tion of the cell cycle genes SAS/CDK4 and MDM2 in adrenocortical carcinomas. Furthermore, several in vitro studies have demonstrated that the insulin-like growth factor II gene IGFII is up-regulated and that its gene product plays an autocrine role in adrenocortical carcinomas.2º Recently IGFII was shown to be over- expressed in adrenocortical carcinomas by RNA-array analysis21 and it was possible to discriminate benign from malignant adrenocortical tumours at the mRNA level.
The aim of this study was to investigate protein expression of IGFII, cyclin-dependent kinase (CDK) 4, MIB1 and p53 in 39 benign and malignant adreno- cortical tumours and to evaluate their potential to differentiate adrenocortical adenomas from carcinomas in routine histopathological diagnosis.
Materials and methods
TUMOUR SPECIMENS AND MORPHOLOGY
A total of 39 adrenocortical neoplasms were drawn from the files of the Institutes of Clinical Pathology at the University Hospitals of Zurich and Basel, Switzer- land. The series comprised 22 adrenocortical adeno- mas and 17 carcinomas. Surgery for an adrenal mass had been performed in the years between 1974 and 2002. Specimens from children and oncocytic adre- nal neoplasms were excluded from this study. Follow- up data from 11 carcinomas and 13 adenomas were obtained by consulting patients’ clinical files. Six patients with carcinoma by morphology and nine adenoma patients were lost to follow-up. The tissue had been fixed by immersion in buffered formalin and embedded in paraffin according to standard proto- cols. Haematoxylin and eosin (H&E) sections were reviewed using the criteria of malignancy proposed by Weiss,7,8 Hough9 and van Slooten,10 including assessment of atypia, nuclei, mitotic activity, archi- tecture, venous, sinusoidal and capsular invasion, necrosis, regressive changes and fibrous bands. The study was performed according to local ethical guidelines and all patient data were anonymized. The morphological features as well as the clinical details and the immunohistochemical results are summarized in Table 1.
IMMUNOHISTOCHEMISTRY
Immunohistochemistry for IGFII, CDK4, MIB1 and p53 was performed as follows. Formalin-fixed paraffin- embedded tissue was sectioned at 4 um. Immuno- histochemistry for IGFII (clone S1F1; Upstate Biotechnology, Lake Placid, NY, USA, dilution 1 : 400, pretreatment with 1% trypsin for 2.5 min at 37°℃) and CDK4 (C-22; Santa Cruz Biotechnology, Santa Cruz, CA, USA, dilution 1 : 100, antigen retrieval by 10 min cooking in citrate buffer, pH 6.0) was performed using the Vectastain Elite Kit (Vector Laboratories, Burlingame, CA, USA). The chromogenic reaction was carried out with 3-3’diaminobenzidine using nickel cobalt amplification22 which gives a black staining product. Immunohistochemistry for MIB1 and p53 was performed on an automated staining system (Ventana BenchMark; Ventana Medical Systems, Tuc- son, AZ, USA). Pretreatment was performed by heating (CC1 (cell conditioning solution, Ventana Medical Systems) mild for MIB1 and p53). The primary antibodies were incubated for 30 min with a dilution of 1 : 20 for MIB1 (Dako A/S, Glostrup, Denmark) and 1 : 80 for p53 (IK ImmunoKontact; AMS Biotechno- logy Ltd, Abingdon, UK). Visualization was performed using the avidin-biotin complex (ABC) method leading to a brown staining signal.
The immunohistochemical preparations for MIB1 were evaluated by computer-assisted morphometry using AnalySIS Pro® software (Soft Imaging System, Münster, Germany). The number of positive nuclei per 1000 total tumour nuclei were counted for every case and the percentage of positive nuclei determined. IGFII immunoreactivity was scored as positive when a perinuclear dot-like signal was observed. CDK4 was assessed semiquantitatively, resulting in three groups: strong positive staining (++), weak positive staining (+) and negative staining (-). p53 was considered to be positive when > 5% of the tumour nuclei stained dark brown.
Results
MORPHOLOGY
Nineteen tumours had originally been diagnosed as carcinomas and 21 as adenomas. No uniform diag- nostic criteria for malignancy were used, as the tumours were collected over a period of 31 years. Reviewing the H&E sections according to the criteria of Weiss,7,8 Hough9 and van Slooten10 allowed a classification as follows: 16 out of 18 tumours were classified as carcinoma by all three scoring systems.
| Tumour no. | Tumour type | Age, years | Sex Clinical symptoms | Tumour weight, g | Tumour size, mm | Weiss* | Hought | van Slooten# | IGFII CDK4 | MIB1 (%) | p53 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Carcinomas with metastasis or local recurrence | |||||||||||||
| C1 | Carcinoma | UN | UN Cushing's syndrome; relapse | UN | 8 | 4.61 | 28.4 | + | ++ | 33.00 | – | ||
| C2 | Carcinoma | 66 | M Cushing's syndrome; relapse after 4 years, liver and lung metastasis after 6 years | 65 | 75 | 5 | 2.37 | 15.2 | + | ++ | 12.60 | – | |
| C3 | Carcinoma | 54 | F Cushing's syndrome; liver metastasis | UN | 90 | 7 | 4.06 | 22.7 | + | ++ | 10.40 | – | |
| C4 | Carcinoma | 69 | F Cushing's syndrome; liver metastasis | 515 | 132 | 7 | 4.29 | 16.8 | + | ++ | 9.70 | – | |
| C5 | Carcinoma | 43 | M Inactive; spine and lung metastasis, DOD after 11 years | 256 | 100 | 5 | 3.92 | 16.8 | + | ++ | 10.80 | – | |
| C6 | Carcinoma | 44 | F inactive; liver, omental and gut metastasis | 1008 | 160 | 9 | 4.98 | 28.4 | + | ++ | 49.40 | – | |
| C7 | Carcinoma | 30 | F Inactive; multiple metastases, DOD after 3 years | 73 | UN | 8 | 4.98 | 25.8 | + | ++ | 46.00 | – | |
| C8 | Carcinoma | 69 | M Relapse retroperitoneal/ colon | 4200 | 340 | 3 | 3 | 12 | + | ++ | 3.90 | + | |
| C9 | Carcinoma | 63 | F Relapse | UN | UN | 7 | 4.06 | 28.4 | – | + | 19.00 | – | |
| C10 | Carcinoma | 39 | M | Cushing's syndrome; relapse, metastases, DOD after 2 years | 295 | 90 | 9 | 3.98 | 28.4 | + | ++ | 39.00 | – |
| Carcinomas without metastasis or local recurrence/lost to follow-up | |||||||||||||
| C11 | Carcinoma | 29 | F Inactive; LTFU | 700 | 120 | 7 | 4.98 | 24.3 | + | ++ | – | ||
| C12 | Carcinoma | 75 | F Inactive; LTFU | UN | 60 | 5 | 3.69 | 25.1 | – | + | 51.30 | + | |
| C13 | Carcinoma | 71 | M Inactive, LTFU | UN | 70 | 8 | 4.98 | 24.3 | – | + | 21.00 | – | |
| C14 | Carcinoma | 27 | F Inactive; NED after 14 years | 800 | 140 | 4 | 3 | 11.1 | + | + | 1.90 | – | |
| C15 | Carcinoma | 48 | F UN, LTFU | 896 | 130 | 7 | 4.22 | 24.3 | + | ++ | 20.20 | – | |
| C16 | Carcinoma | 62 | F UN, LTFU | 108.5 | 50 | 9 | 4.98 | 25.8 | – | + | 49.10 | + | |
| C17 | Carcinoma | 51 | F UN, LTFU | UN | 100 | 9 | 4.98 | 25.8 | + | + | 29.00 | – | |
| Adenomas without metastasis or local recurrence | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A1 | Adenoma | 58 | F | Conn's syndrome; NED after 8 years | 16.8 | 23 | 0 | 0 | 0 | – | + | 1.00 | – |
| A2 | Adenoma | 44 | F | Conn's syndrome; NED after 7 years | 15.9 | 36 | 2 | 0.39 | 8.8 | + | ++ | 10.70 | – |
| A3 | Adenoma | 63 | M | Conn's syndrome; NED after 6 years | 27.5 | 26 | 0 | 0 | 0 | – | + | 0.70 | – |
| A4 | Adenoma | 45 | M | Conn's syndrome; NED after 5 years | 44 | 55 | 1 | 0.69 | 0 | – | + | 0.90 | – |
| A5 | Adenoma | 39 | F | Conn's syndrome; NED after4 years | 2.1 | 23 | 0 | 0 | 0 | – | ++ | 0.30 | – |
| A6 | Adenoma | 61 | M | Conn's syndrome; NED after 5 years | 23.2 | 19 | 0 | 0 | 0 | – | + | 1.40 | – |
| A7 | Adenoma | 43 | M | Conn's syndrome; NED after 3 years | 160 | 100 | 0 | 0 | 5.7 | – | + | 2.30 | – |
| A8 | Adenoma | 37 | F | Conn's syndrome; NED after 11 years | 5.5 | 12 | 0 | 0 | 0 | – | + | 0.00 | – |
| A9 | Adenoma | 35 | F | Cushing's syndrome; NED after 13 years | 15 | 42 | 0 | 0 | 0 | – | ++ | 1.40 | – |
| A10 | Adenoma | 50 | F | Cushing's syndrome; NED after 6 years | UN | 25 | 1 | 0.37 | 3.3 | – | + | 1.30 | – |
| A11 | Adenoma | 65 | M | Cushing's syndrome; NED after 2 years | 15 | 30 | 0 | 0 | 0 | – | + | 0.40 | – |
| A12 | Adenoma | 30 | F | Cushing's syndrome; NED after 2 months | 20 | 35 | 0 | 0 | 0 | – | + | 0.50 | – |
| A13 | Adenoma | 52 | F | Cushing's syndrome; NED after 1.5 years | 13.4 | 40 | 0 | 0 | 0 | – | + | 0.60 | – |
| Adenomas by morphology alone/lost to follow-up | |||||||||||||
| A14 | Adenoma | 36 | F | Conn's syndrome | UN | 25 | 0 | 0 | 0 | – | + | 2.80 | – |
| A15 | Adenoma | 52 | F | Conn's syndrome | UN | 20 | 1 | 0.37 | 3.3 | – | + | 0.30 | – |
| A16 | Adenoma | 46 | F | Conn's syndrome | 2.7 | 14 | 1 | 0.37 | 3.3 | – | – | 0.00 | – |
| A17 | Adenoma | 49 | F | Conn's syndrome | 13.8 | 28 | 0 | 0 | 0 | – | + | 0.00 | – |
| A18 | Adenoma | 58 | M | Conn's syndrome | UN | 11 | 2 | 0.37 | 0 | – | + | 0.30 | – |
| Tumour no. | Tumour type | Age, years | Sex | Clinical symptoms | Tumour weight, g | Tumour size, mm | Weiss* | Hought | van Slooten# | IGFII | CDK4 | MIB1 (%) | p53 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A19 | Adenoma | 32 | F | Cushing's syndrome | 20 | 30 | 0 | 0 | 0 | – | + | 0.00 | – |
| A20 | Adenoma | 47 | F | Cushing's syndrome | UN | 40 | 0 | 0 | 0 | – | + | 1.10 | – |
| A21 | Adenoma | 28 | F | Cushing's syndrome | UN | 35 | 0 | 0 | 0 | – | + | 0.40 | – |
| A22 | Adenoma | 53 | F | Hypertensive attacks, elevated catecholamine serum levels | 31 | UN | 0 | 0 | 0 | – | + | 0.00 | – |
NED, No evidence of disease; DOD, dead of disease; LTFU, lost to follow-up; UN, unknown.
*Weiss et al .: 5 1-3 benign, ≥ malignant.
+Hough et al .: 6 0.17 ± 0.26 benign, ± 0.58 indeterminate, 2.91 + 0.9 malignant.
van Slooten et al .: 7 <8 benign, ≥ 8 malignant.
One tumour was reclassified as oncocytoma and excluded from the study. Another tumour was reclassified as adenoma according to all scoring sys- tems (case A7). One carcinoma which recurred intra- abdominally (Table 1, case C8) reached scores required for malignancy using the criteria of Hough and van Slooten, but not using the criteria of Weiss.
Nineteen out of 21 adenoma diagnoses were con- firmed by all three scoring systems. One adenoma (A4) was classified as having indeterminate biological beha- viour when applying the scoring system of Hough and was scored as benign when applying the Weiss and van Slooten systems. Adenoma A2 displayed heterogeneous morphology, attaining the criteria for malignancy with the scoring system of van Slooten, but not with the systems of Weiss and Hough. Follow-up data revealed the benign behaviour of both adenoma A4 (60 months’ follow-up) and A2 (84 months’ follow-up) (Table 1).
IGFII IMMUNOHISTOCHEMISTRY
Twenty-two adenomas and 17 carcinomas were eval- uated. Twenty-one adenomas (12 without evidence of disease in the follow-up, nine lost to follow-up) out of 22 were negative for IGFII (specificity 95.5%), one adenoma (by morphology and clinical behaviour) (4.5%) was focally positive for IGFII. Nine of 10 carcinomas with known metastases/relapses were immunoreactive for IGFII and four of seven carcino- mas according to all three morphological scoring systems were IGFII+. In total, 13 carcinomas out of 17 were positive for IGFII (sensitivity 76,5%), four carcinomas were negative for IGFII (Table 1).
CDK4 IMMUNOHISTOCHEMISTRY
Twenty-two adenomas and 18 carcinomas were eval- uated immunohistochemically for CDK4. One adenoma (according to all three morphological scoring systems) out of 22 was negative for CDK4, 18 adenomas (10 with no evidence of disease on follow-up, eight lost to
Figure 1. Immunohistochemistry for insulin-like growth factor (IGF) II, MIB1, cyclin-dependent kinase (CDK) 4 and p53 in representative examples of adrenocortical adenomas and carcinomas. Adenomas in the left, carcinomas in the right column. Negativitiy for IGFII in adenoma A19 (A) and positivity for IGFII in carcinoma C13, note perinuclear staining (B): Low proliferation index (< 5%) in MIB1 immunohistochemistry in adenoma A21 (C) and frankly elevated
proliferation (51.3%) in carcinoma C12 (D). Weak nuclear staining
(+) for CDK4 in adenoma A20 (E) and strong nuclear staining (++) for CDK4 in carcinoma C4 (F): Negativity for p53 in adenoma A14
(G) and positivity for p53 in carcinoma C12 (H).
A
B
C
D
E
F
G
H
A
B
C
D
E
F
G
H
Figure 2. Morphology and immunohistochemistry of case A2. The tumour in most areas displays inconspicuous morphological features (A). There are other, patchy areas with hyperchromasia, polymorphism and prominent nucleoli (B). This heterogeneous morphology is reflected by immunohistochemistry: Areas with negativity for insulin-like growth factor (IGF) II (C) and areas with strong positivity for IGFII (D). Most parts of the tumour show a proliferation index < 5% (E), focally, however, the MIB1 index is up to 10.7% (F). Most of the tumour cells show only weak (+) nuclear staining for cyclin-dependent kinase (CDK) 4 (G), whereas other groups of tumour cells exhibit a strong (++) nuclear staining for CDK4 (H).
follow-up) (81.9%) showed weak positivity for CDK4 and three adenomas without evidence of disease on follow-up (13.6%) were strongly positive for CDK4. Nine of the 10 carcinomas with metastases/relapse were strongly CDK4+, as were two of the seven carcinomas without known recurrences. In total, 11 out of 17 carcinomas (64.7%) were strongly positive for CDK4 and six carcinomas (35.3%) were weakly positive. None of the 17 carcinomas was negative for CDK4 (Table 1).
MIB1 IMMUNOHISTOCHEMISTRY
Twenty-two adenomas and 16 carcinomas were eval- uated immunohistochemically for MIB1. According to previous studies,16 a proliferation index < 5% was considered to be negative and a proliferation index > 5% was considered to be positive. Out of 22 adenomas, 21 were negative for MIB1 (12 without evidence of disease on follow-up, nine lost to follow-up) (specificity 95.5%), one adenoma (by morphology and clinical behaviour) showed a proliferation index of 10% (Table 1; A2). Of 10 carcinomas with metasta- sis/recurrence, nine had a proliferation index > 5%, as did five of six carcinomas with unknown recurrence status. In total, 14 out of 16 carcinomas were positive for MIB1 (sensitivity 87.5%). However, two carcinomas did not show an increased proliferation index, one of which was scored as malignant according to all three morphological scoring systems but had no evidence of disease 14 years after surgery.
p53 IMMUNOHISTOCHEMISTRY
Twenty-two adenomas and 17 carcinomas were eval- uated immunohistochemically for p53. All adenomas were negative for p53 (specificity 100%). Only three out of 17 carcinomas were positive for p53 (sensitivity 17.6%), defined as more than 5% positive nuclei. The p53 index was > 50% in two of these tumours.
Discussion
We have investigated the potential of IGFII, CDK4, MIB1 and p53 antibodies to differentiate malignant
from benign adrenocortical tumours. The analysis was performed on 39 adrenocortical neoplasms classified according to development of metastases/recurrences and the morphological criteria of Weiss, Hough and van Slooten.7-10 We found that both IGFII and MIB1 on their own are good markers of malignancy with a sensitivity of 76.5% and 87.5% and a specificity of 95.5% and 95.5%, respectively. An even higher sensitivity of 100% with a specificity of 95.5% was reached when both markers were combined. CDK4 was expressed at higher levels in carcinomas; however, a scoring of immunohistochemical intensities is imprac- ticable in daily routine practice. Overexpression of p53 is a highly specific marker, which, however, is only of use in a small proportion (17.6%) of positive carcinomas.
A recent cDNA expression array study demonstrated a substantially increased IGFII RNA level in 91% of adrenocortical carcinomas when compared with aden- omas.21 We investigated the immunohistochemical expression of IGFII in a tumour set composed of 17 adrenocortical carcinomas and 22 adenomas. We found that 76.5% of carcinomas but only 4.5% of adenomas were immunoreactive for IGFII. In an earlier study, Erickson et al. investigated 64 adenomas and 67 carcinomas and found that 92.5% of carcinomas but also 54.7% of adenomas were immunoreactive for IGFII.23 Ilvesmäki et al. found immunoreactivity for IGFII in two functional carcinomas but not in normal adrenal gland tissue.25
Of note in our study was the immunohistochemical staining pattern: all positive carcinomas exhibited a reproducible perinuclear dot-like or Golgi pattern immunoreactivity for IGFII (Figure 1B). This pattern was not described by Erickson et al.,23 who scored any cytoplasmic staining as positive and who used a different antibody at a higher concentration. The aforementioned points may explain why IGFII positiv- ity was also reported in 55% of the adenomas that they studied. The perinuclear pattern reported here can be explained by the molecular findings in adrenocortical carcinomas reported by Boulle et al.24 In in vitro experiments they showed a 10-fold increase in IGFII protein content in nine malignant adrenocortical tumours, in contrast to nine control normal adrenal
glands and benign tumours, suggesting not only the existence but also the effective translation of IGFII mRNA in the malignant tumours.24 Interestingly, the translation products included not only the mature 7.5-kDa IGFII,26 but also precursor forms of higher molecular weight.24 This suggests that IGFII processing is altered in malignant adrenocortical tumours. Indeed, under in vitro conditions Boulle et al. showed a marked increase in levels of intracellular IGFII and a marked decrease in mature IGFII secretion in favour of secre- tion of higher molecular weight forms of the protein if cells were treated with FGF-2,27 which seems to cooperate mitogenically with IGFII in the developing adrenal.28 All these data point towards abnormal IGFII processing, which occurs in the cis- and trans- Golgi apparatus,29 thus resulting in an accumulation of the protein in the Golgi apparatus.
Only one adenoma in our study (case A2, Table 1) was positive for IGFII and showed the characteristic perinuclear Golgi pattern in a patchy distribution (Figure 2F). Some of those areas were conspicuous already by means of H&E morphology (atypia, atypi- cal mitoses, hyperchromasia, nucleoli), leading to a classification as ‘malignant’ using the criteria proposed by van Slooten. This adenoma was the only adenoma with a focal proliferation index > 5%, i.e. 10.7% in the above described regions (Figure 2D). In addition, it showed circumscribed areas with strong nuclear stain- ing for CDK4 (Figure 2H). Interestingly, this tumour was the only one of the whole subset which displayed frank heterogeneity by means of H&E staining and immunohistochemistry (Figure 2). Those findings could be interpreted as small foci of malignant trans- formation within an adenoma. The concept of an adenoma-carcinoma sequence in adrenocortical tu- mours, however, is controversial, although there is some evidence that adrenocortical tumorigenesis might be a multistep process via clonal proliferation.30-32
In addition to IGFII, we investigated MIB1 as a marker for malignancy in adrenocortical tumours. Using a cut-off level of 5% according to Wachenfeld et al.,16 our study generated a sensitivity of 87.5% and a specificity of 95.5%, thus making MIB1 an equally good marker for malignancy as IGFII. These findings are in agreement with a series of other reports.11-16 Remarkably, those carcinomas in our study which were immunohistochemically negative for IGFII exhib- ited obvious positivity for MIB1, namely 19%, 21%, 49.1% and 51.3%. Vice versa, all carcinomas with a MIB1 proliferation index < 5% showed clear positivity for IGFII. In our series the sensitivity for the combined use of IGFII and MIB1 was 100% with a specificity of 95.5%. Thus, it appears that immunohistochemical
positivity for IGFII and/or MIB1 index > 5% can be used to discriminate reliably benign from malignant adrenocortical neoplasms.
Of our adrenocortical carcinomas, 17.6% were p53+ (> 5% of tumour cells) and all of them showed a high MIB1 index. p53 positivity has been reported in 0.7- 93.8%16,18,33-36 of adrenocortical carcinomas. This wide range might be explained by the use of different antibodies and pretreatment methods;35 however, another reason might be a different interpretation of p53 immunoreactivity. Comparison of immunohisto- chemical with molecular data (e.g. p53 mutation analysis) has revealed a poor concordance of both methods. Thus, immunohistochemistry detected only 10% of adrenocortical carcinomas with p53 muta- tions.33 At the molecular level, somatic p53 mutations were found in 18.2%,18 20%,17 47.6%35 and 70%33 of adrenocortical carcinomas. So the conclusion must be that p53 is often mutated but only seldom accumulates in these tumours and therefore is not reliably detectable by routinely performed immunohistochemistry.
CDK4 was strongly positive in 11 out of 17 (64.7%) carcinomas and three out of 22 (13.6%) of adenomas; however, this finding is difficult to apply in daily practice, as 18 out of 22 (81.8%) of the adenomas were also weakly positive and the interpretation of staining intensity is difficult without a direct comparison.
In conclusion, our results suggest that the combined use of IGFII and MIB1 immunohistochemistry reliably distinguishes benign and malignant adrenocortical neoplasms and provides a helpful additional tool to the morphological scoring systems in difficult cases. IGFII staining is exclusively found in a perinuclear Golgi pattern.
Conflict of interest
None declared.
Acknowledgements
This work has been supported in part by the Swiss National Foundation (SNF 31-10825/1).
References
1. Brennan MF. Adrenocortical carcinoma. CA Cancer J. Clin. 1987; 37; 348-365.
2. Khorram-Manesh A, Ahlman H, Jansson S et al. Adrenocortical carcinoma: surgery and mitotane for treatment and steroid profiles for follow-up. World J. Surg. 1998; 22; 605-611; discussion 611-612.
3. Lipsett MB, Hertz R, Ross GT. Clinical and pathophysiologic aspects of adrenocortical carcinoma. Am. J. Med. 1963; 35; 374-383.
4. Luton JP, Cerdas S, Billaud L et al. Clinical features of adreno- cortical carcinoma, prognostic factors, and the effect of mitotane therapy. N. Engl. J. Med. 1990; 322; 1195-1201.
5. Icard P, Chapuis Y, Andreassian B et al. Adrenocortical carcin- oma in surgically treated patients: a retrospective study on 156 cases by the French Association of Endocrine Surgery. Surgery 1992; 112; 972-979; discussion 979-980.
6. Schulick RD, Brennan MF. Long-term survival after complete resection and repeat resection in patients with adrenocortical carcinoma. Ann. Surg. Oncol. 1999; 6; 719-726.
7. Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am. J. Surg. Pathol. 1984; 8; 163-169.
8. Weiss LM, Medeiros LJ, Vickery AL Jr. Pathologic features of prognostic significance in adrenocortical carcinoma. Am. J. Surg. Pathol. 1989; 13; 202-206.
9. Hough AJ, Hollifield JW, Page DL et al. Prognostic factors in adrenal cortical tumors. A mathematical analysis of clinical and morphologic data. Am. J. Clin. Pathol. 1979; 72; 390-399.
10. van Slooten H, Schaberg A, Smeenk D et al. Morphologic characteristics of benign and malignant adrenocortical tumors. Cancer 1985; 55; 766-773.
11. Goldblum JR, Shannon R, Kaldjian EP et al. Immunohistochem- ical assessment of proliferative activity in adrenocortical neo- plasms. Mod. Pathol. 1993; 6; 663-668.
12. Gupta D, Shidham V, Holden J et al. Value of topoisomerase II alpha, MIB-1, p53, E-cadherin, retinoblastoma gene protein product, and HER-2/neu immunohistochemical expression for the prediction of biologic behavior in adrenocortical neoplasms. Appl. Immunohistochem. Mol. Morph. 2001; 9; 215-221.
13. Iino K, Sasano H, Yabuki N et al. DNA topoisomerase II alpha and Ki-67 in human adrenocortical neoplasms: a possible marker of differentiation between adenomas and carcinomas. Mod. Pathol. 1997; 10; 901-907.
14. Nakazumi H, Sasano H, Iino K et al. Expression of cell cycle inhibitor p27 and Ki-67 in human adrenocortical neoplasms. Mod. Pathol. 1998; 11; 1165-1170.
15. Terzolo M, Boccuzzi A, Bovio S et al. Immunohistochemical assessment of Ki-67 in the differential diagnosis of adrenocortical tumors. Urology 2001; 57; 176-182.
16. Wachenfeld C, Beuschlein F, Zwermann O et al. Discerning malignancy in adrenocortical tumors: are molecular markers useful? Eur. J. Endocrinol. 2001; 145; 335-341.
17. Ohgaki H, Kleihues P, Heitz PU. p53 mutations in sporadic adrenocortical tumors. Int. J. Cancer 1993; 54; 408-410.
18. Reincke M, Karl M, Travis WH et al. p53 mutations in human adrenocortical neoplasms: immunohistochemical and molecular studies. J. Clin. Endocrinol. Metab. 1994; 78; 790-794.
19. Zhao J, Roth J, Bode-Lesniewska B et al. Combined comparative genomic hybridization and genomic microarray for detection of gene amplifications in pulmonary artery intimal sarcomas and adrenocortical tumors. Genes Chromosomes Cancer 2002; 34; 48-57.
20. Logie A, Boulle N, Gaston V et al. Autocrine role of IGF-II in proliferation of human adrenocortical carcinoma NCL H295R cell line. J. Mol. Endocrinol. 1999; 23; 23-32.
21. Giordano TJ, Thomas DG, Kuick R et al. Distinct transcriptional profiles of adrenocortical tumors uncovered by DNA microarray analysis. Am. J. Pathol. 2003; 162; 521-531.
22. Werner M, Von Wasielewski R, Komminoth P. Antigen retrieval, signal amplification and intensification in immunohistochemistry. Histochem. Cell Biol. 1996; 105; 253-260.
23. Erickson LA, Jin L, Sebo TJ et al. Pathologic features and expression of insulin-like growth factor-2 in adrenocortical neoplasms. Endocrin. Pathol. 2001; 12; 429-435.
24. Boulle N, Logie A, Gicquel C et al. Increased levels of insulin-like growth factor II (IGF-II) and IGF-binding protein-2 are associated with malignancy in sporadic adrenocortical tumors. J. Clin. Endocrinol. Metab. 1998; 83; 1713-1720.
25. Ilvesmaki V, Kahri AI, Miettinen PJ et al. Insulin-like growth factors (IGFs) and their receptors in adrenal tumors: high IGF-II expression in functional adrenocortical carcinomas. J. Clin. Endocrinol. Metab. 1993; 77; 852-858.
26. Binoux M, Lassarre C, Gourmelen M. Specific assay for insulin- like growth factor (IGF) II using the IGF binding proteins extracted from human cerebrospinal fluid. J. Clin. Endocrinol. Metab. 1986; 63; 1151-1155.
27. Boulle N, Gicquel C, Logie A et al. Fibroblast growth factor-2 inhibits the maturation of pro-insulin-like growth factor-II (Pro- IGF-II) and the expression of insulin-like growth factor binding protein-2 (IGFBP-2) in the human adrenocortical tumor cell line NCI-H295R. Endocrinology 2000; 141; 3127-3136.
28. Mesiano S, Mellon SH, Jaffe RB. Mitogenic action, regulation, and localization of insulin-like growth factors in the human fetal adrenal gland. J. Clin. Endocrinol. Metab. 1993; 76; 968- 976.
29. Duguay SJ, Jin Y, Stein J et al. Post-translational processing of the insulin-like growth factor-2 precursor. Analysis of O-glycosy- lation and endoproteolysis. J. Biol. Chem. 1998; 273; 18443- 18451.
30. Bernard MH, Sidhu S, Berger N et al. A case report in favor of a multistep adrenocortical tumorigenesis. J. Clin. Endocrinol. Metab. 2003; 88; 998-1001.
31. Russell AJ, Sibbald J, Haak H et al. Increasing genome instability in adrenocortical carcinoma progression with involvement of chromosomes 3, 9 and X at the adenoma stage. Br. J. Cancer 1999; 81; 684-689.
32. Stratakis CA. Genetics of adrenocortical tumors: gatekeepers, landscapers and conductors in symphony. Trends Endocrinol. Metab. 2003; 14; 404-410.
33. Barzon L, Chilosi M, Fallo F et al. Molecular analysis of CDKN1C and TP53 in sporadic adrenal tumors. Eur. J. Endocrinol. 2001; 145; 207-212.
34. Bernini GP, Moretti A, Viacava P et al. Apoptosis control and proliferation marker in human normal and neoplastic adreno- cortical tissues. Br. J. Cancer 2002; 86; 1561-1565.
35. McNicol AM, Nolan CE, Struthers AJ et al. Expression of p53 in adrenocortical tumours: clinicopathological correlations. J. Pathol. 1997; 181; 146-152.
36. Stojadinovic A, Ghossein RA, Hoos A et al. Adrenocortical carcinoma: clinical, morphologic, and molecular characterization. J. Clin. Oncol. 2002; 20; 941-950.