Clinicopathological study of a series of 92 adrenocortical carcinomas: from a proposal of simplified diagnostic algorithm to prognostic stratification
Marco Volante, Enrico Bollito, Paola Sperone,1 Veronica Tavaglione, Fulvia Daffara,2 Francesco Porpiglia,3 Massimo Terzolo,2 Alfredo Berruti1 & Mauro Papotti Divisions of Pathology, 1 Oncology, 2Internal Medicine and 3 Urology, Department of Clinical and Biological Sciences, University of Turin at San Luigi Hospital, Turin, Italy
Date of submission 21 January 2009 Accepted for publication 6 May 2009
Volante M, Bollito E, Sperone P, Tavaglione V, Daffara F, Porpiglia F, Terzolo M, Berruti A & Papotti M (2009) Histopathology 55, 535-543
Clinicopathological study of a series of 92 adrenocortical carcinomas: from a proposal of simplified diagnostic algorithm to prognostic stratification
Aims: Pathological diagnosis of adrenocortical car- cinoma relies on several microscopic features commonly used in combination in different scoring systems that are sometimes subjective and/or time consuming. The aim was to investigate the impact of individual pathological parameters in the diagnosis and prognosis of adrenocortical carcinoma.
Methods and results: The series included 92 malignant cases and a control series of 47 adenomas, all classified according to Weiss score criteria. The presence of disruption of the reticular network, as highlighted by histochemical staining, was present in all adrenocor- tical carcinomas and the inclusion of at least one of the three following additional parameters - mitotic index
>5/50 high-power fields (HPF), presence of necrosis and presence of vascular invasion - gave an algorithm with 100% sensitivity and specificity to recognize malignant tumours according to the Weiss system, with easier and more practical applicability. Moreover, on multivariate analysis, stage III/IV and mitotic count
9/50 HPF showed a strong adverse impact on disease-free and overall survival, leading to the iden- tification of three risk groups affected by a significantly different prognosis.
Conclusions: We have defined an easy-to-perform and highly specific and sensitive algorithm for the diagnosis and prognostic categorization of adrenocortical tumours.
Keywords: adrenal cortex, carcinoma, diagnosis, prognosis, reticulin stain
Abbreviation: HPF, high-power field
Introduction
Adrenocortical carcinoma is a rare tumour of the adrenal cortex that accounts for no more than 0.2% of all malignancies, with two peaks of incidence in early childhood and adults >60 years old. It represents an aggressive disease with up to two-thirds of cases having distant metastases at the time of diagnosis and a mean
survival of <30 months,1-3 although combined treat- ment modalities, with special reference to mitotane- based therapy, seem to improve its outcome.4
The differential diagnosis of carcinomas from adeno- mas has been based on several microscopic features, none of them alone being absolutely indicative of malignancy.5,6 Despite the fact that most adrenocorti- cal carcinomas do not represent a diagnostic pitfall in clinical practice, there are some cases in which differentiation from their benign counterpart is not straightforward. These cases are a challenge for the pathologist, since the therapeutic strategy in adreno- cortical carcinomas is radically different from that of
Address for correspondence: M Volante, Department of Clinical and Biological Sciences, University of Turin at San Luigi Hospital, Regione Gonzole 10, 10043 Orbassano, Turin, Italy.
e-mail: marco.volante@unito.it
@ 2009 The Authors. Journal compilation @ 2009 Blackwell Publishing Limited.
adenomas, and accurate diagnosis is mandatory. This issue is also relevant because adrenal tumours are increasingly recognized in clinical practice: the wide- spread use of imaging techniques has resulted in the clinical dilemma of the adrenal ‘incidentaloma’,7 and although the vast majority of adrenal ‘incidentalomas’ are benign lesions, in up to 12% of cases an incidental adrenocortical carcinoma may be discovered.8
Different scoring systems for adrenocortical car- cinoma have therefore been developed, using mathe- matical models or numerical scores based on the association of a given threshold for each considered parameter with malignancy.9-11 The most widely employed scoring system was proposed11 and revisited in a subsequent report12 by Weiss, and includes nine parameters related to tumour structure (loss of clear cytoplasm, presence of diffuse architecture and of necrosis), cytological features (atypia, mitotic count, atypical mitotic figures) and invasive properties (sinu- soidal, venous and capsular invasion).
Unfortunately, Weiss and other scoring systems are sometimes difficult to apply, subjective and/or time consuming, despite several re-visitations and imple- mentations occurring over the years to simplify the diagnostic procedure and provide clinically relevant information.13-15 Moreover, in paediatric adrenocorti- cal tumours some features included in the Weiss system (i.e. eosinophilic cytoplasm, diffuse architecture or sinusoidal invasion) have been found to be less relevant compared with adult tumours,16 indicating that the diagnostic work-up of childhood tumours should prob- ably employ a different panel of pathological factors.
Recently, a different approach based on the definition of an algorithm centred on mitotic count [>5/50 high- power fields (HPF)], followed by the recognition of other parameters (including nuclear grade, diffuse growth pattern and tissue reaction), has been pro- posed,17 and although based on a relatively small series of malignant cases (defined by the presence of metas- tases), it has the advantage of simplifying the current diagnostic approach based on numerical scores.
A second major issue is the distinction between morphologically and clinically malignant adrenocorti- cal tumours. Not all adrenal tumors having unequiv- ocal microscopic features of malignancy progress to a fatal outcome, therefore a better definition of the clinical and pathological features that influence prog- nosis and a stratification of adrenal cancer patients into prognostic categories are needed. Among the most commonly used pathological parameters for the diag- nosis of adrenocortical carcinoma, mitotic activity and/or proliferative index show a significant associa- tion with survival,10,12,18 as do age and overall Weiss
score did.19 In a very large study on metastatic tumours aimed at recognizing prognostic rather than diagnostic parameters, multivariate analysis identified the presence of a mitotic index >20/50 HPF and the number of organs involved as the major factors that influence prognosis in advanced adrenal cancer patients.20
The aim of the present study was to reconsider the clinical pathological parameters currently used in the diagnosis of adrenocortical carcinoma in a large retrospective series of 92 cases diagnosed and treated at one institution, with the aim of revising the classical diagnostic scoring systems in an algorithmic approach and to provide a pathologically based prognostic stratification of adrenal cancer patients.
Materials and methods
CASE SELECTION AND PATHOLOGICAL CLASSIFICATION
From a total of 153 cases of primary adrenocortical carcinomas in an adult population included in the clinical databases (from 1995 to 2007) of the Divisions of Oncology and Internal Medicine at San Luigi Hospital (University of Turin, Italy) that serves as a national referral centre for this neoplasm, 92 cases had patho- logical material retrievable from the files of the Pathol- ogy Departments of the University of Turin (including 44 consultation cases, operated on elsewhere). All cases were independently classified by three of us (M.V., E.B., M.P.) re-assessing the parameters commonly considered in the Weiss system, using 2-10 slides for each case. All parameters included in the Weiss system12 were recorded independently, together with clinicopatholog- ical information such as sex, age, location, clinical syndrome, weight and size. For comparison, all cases were also scored using the Weiss modified (according to Aubert et al.)14 and Van Slooten et al.10 scoring sys- tems. Baseline demographic and clinical data were complete and follow-up information was available for survival analysis in all but 15 cases, either recent (n = 7) or lost to follow-up (n = 8). Tumour staging was assessed following recommendations by Sullivan et al. and the Armed Forces Institute of Pathology.
Seventy-nine patients with stage I to III disease underwent radical surgery followed by adjuvant mito- tane in 14 of them. The remaining 13 patients, who were not amenable to surgery due to locally advanced or metastatic disease, received mitotane alone (n = 2) or mitotane plus chemotherapy (n = 11) as up-front treatment. One of them was radically operated on after six cycles of chemotherapy plus mitotane.
| Diagnosis (n) | F/M ratio | Age, mean [range] | Location | Size (mm), mean [range] | Weight (g) mean, [range] | Functional status | Outcome |
|---|---|---|---|---|---|---|---|
| Adrenocortical carcinoma (92) | 1.5 | 46 [20-85] | LA 48* | 114 [16-250] | 428 [8-3100] | NF: 50 | NED: 32 |
| RA 45* | Cortisol: 13 | AWD: 15 | |||||
| Aldosterone: 6 Androgens: 8 Not known: 15 | DOD: 28 Recent cases: 7 Lost: 8 | ||||||
| Adrenocortical adenoma (47) | 3.2 | 53 [15-77] | LA 23 | 40 [10-100] | 39 [4-150] | NF: 13 | NED: 47 |
| RA 24 | Cortisol: 18 Aldosterone: 15 Not known: 1 |
LA, Left adrenal; RA, right adrenal; NF, not functioning; NED, no evidence of disease; AWD, alive with disease; DOD, died of disease.
*One case had bilateral tumours.
An additional 47 adrenocortical tumours, including all cases observed in the same time period having Weiss scores 1 or 2 (16 cases) or adenomas of large size, were collected as a benign tumour control group.
The main clinical and pathological features of the whole series of 139 adrenocortical tumours are sum- marized in Table 1.
IMMUNOHISTOCHEMISTRY AND HISTOCHEMISTRY
In selected cases, to confirm the adrenocortical nature of the lesion, immunohistochemistry for chromogranin A (clone LK2H10, diluted 1:800; Novocastra, Newcas- tle, UK), synaptophysin (clone SY38, diluted 1:100; DakoCytomation, Glostrup, Denmark), Melan A (clone A103, diluted 1:50; DakoCytomation), alpha-inhibin (clone R1, diluted 1:75; Diamedix, Miami, FL, USA) and CD10 (clone 56C6, diluted 1:30; Santa Cruz Biotechnology, Santa Cruz, CA, USA) was performed. A standard automated (Dako Autostainer, Glostrup, Denmark) immunoperoxidase procedure was employed, and immunoreactions were revealed by a biotin-free dextran-chain detection system (Envision; Dako) and developed using diaminobenzidine as the chromogen.
To better evaluate the growth pattern, which is sometimes hard to assess on pure haematoxylin and eosin staining, in all cases of adrenal carcinoma with non-diffuse growth according to the original definition of Weiss12 and in all adrenocortical adenomas, reticu- lin staining was performed to define the status of the reticulin framework, using a commercially available silver impregnation-based kit (Bio Optica, Milan, Italy). Moreover, basal membrane material was highlighted by immunohistochemical detection of laminin (clone
4C7, diluted 1:50; Novocastra) and collagen type IV (clone CIV22, diluted 1:100; DakoCytomation), as detailed above. The presence of a disruption of the reticulin/basal membrane network, similar to that described in the setting of other endocrine diseases, 23,24 was defined by the loss of continuity of reticular fibres or basal membrane network evaluated in one HPF (single high power field = 0.2 mm2) (400x), extended to at least one-third of the lesion.
STATISTICAL ANALYSIS
All data were analysed with STATISTICA for Windows software version 6.1 (Stat Soft, Padova, Italy). A level of P < 0.05 was considered statistically significant.
The distribution of individual Weiss’s criteria in the overall group of 139 adrenocortical neoplasms was evaluated using contingency tables to calculate sensi- tivity and specificity in relation to the diagnosis of adrenocortical carcinoma. To analyse the prognostic impact of all clinicopathological variables considered, univariate disease-free and overall survival analysis was based on the Kaplan-Meier product limit estimate of survival distribution. Clinicopathological parameters considered in survival analysis of adrenocortical carci- noma group included: sex, age, location, size, weight, stage, Weiss, Weiss modified and van Slooten overall scores, and all nine individual histological parameters considered in the Weiss scoring system. Unadjusted differences between survival curves were tested using the log rank test. Multivariate analysis of the relative impact on survival of each parameter included in the univariate analysis was estimated using the Cox proportional hazards regression model.
Results
DISTRIBUTION OF WEISS’S PARAMETERS IN ADRENOCORTICAL TUMOURS
The summary of the Weiss’s parameter(s) distribution in the whole series of 139 adrenocortical neoplasms is shown in Table 2. None of the parameters reached both 100% sensitivity and specificity for the diagnosis of adrenocortical carcinoma. All but three parameters (presence of diffuse architecture, nuclear atypia and eosinophilic cytoplasm) were 100% specific for a diagnosis of adrenocortical carcinoma. The most sen- sitive feature was necrosis, which was recognizable in 84% of cases, followed by mitotic count exceeding 5 in 50 HPF (71%) and venous invasion (64%). The presence of atypical mitoses was highly specific, but was observed in only 34% of cases.
DEFINITION OF DISRUPTED RETICULIN FRAMEWORK
A diffuse growth pattern was detected in 79% of adrenocortical carcinomas, and in 9% of adrenocor- tical tumours having a Weiss score < 3. Diffuse architecture was considered according to Weiss12 as the presence of patternless sheets of cells in more than one-third of the tumour, with trabecular, cord-like, columnar, alveolar, or nesting patterns considered as non-diffuse patterns. In order to define the pattern of growth better, in all adrenocortical carcinoma cases with a non-diffuse architecture at histology, as well as in all 47 adrenocortical adenoma samples, histochem- ical staining for reticulin and immunohistochemistry
for laminin and/or collagen IV were performed. Disruption of the reticulin/basal membrane network was detected in all adrenocortical carcinoma samples (usually in more than half of the tumour) (Figure 1). In adrenocortical tumours with a borderline Weiss value (score 2), two cases only, having diffuse and trabecular growth rather than the alveolar architec- ture typical of adrenocortical adenoma, showed the loss of a regularly distributed reticulin network. Both of these tumours were large (90 and 60 mm) and in addition to a diffuse pattern had either nuclear atypia or eosinophilic cytoplasm; the respective patients are alive and well at 2 and 3 years’ follow-up in the absence of any adjuvant therapy after surgery. Indeed, the morphological (Weiss definition of diffuse growth) and/or histochemical/immunohistochemical demon- stration of the lack of a well-preserved reticulin framework was 100% sensitive and 96% specific in the identification of malignant adrenocortical lesions (Weiss score ≥ 3).
DESIGN OF A SIMPLIFIED DIAGNOSTIC ALGORITHM
Analysis of the frequency of any single Weiss param- eter in this large series of adrenocortical tumours led to the design of a diagnostic algorithm based on the most sensitive and specific features, replacing the classical numerical scoring systems.
The algorithm takes into consideration the presence of a disrupted reticulin framework, which was the single most sensitive feature of malignancy, followed by at least one of the three parameters with 100%
| Parameter | Frequency | SP (%) | NPV (%) | |||
|---|---|---|---|---|---|---|
| Weiss < 3 | Weiss ≥ 3 | SN (%) | PPV (%) | |||
| Mitoses >5/50 HPF | 0/47 | 65/92 | 71 | 100 | 100 | 64 |
| Atypical mitoses | 0/47 | 31/92 | 34 | 100 | 100 | 43 |
| Necrosis | 0/47 | 77/92 | 84 | 100 | 100 | 76 |
| Venous invasion | 0/47 | 59/92 | 64 | 100 | 100 | 59 |
| Sinusal invasion | 0/47 | 52/92 | 57 | 100 | 100 | 54 |
| Capsular invasion | 0/47 | 50/92 | 54 | 100 | 100 | 53 |
| Diffuse growth | 4/47 | 73/92 | 79 | 91 | 95 | 69 |
| Nuclear atypia | 9/47 | 78/92 | 85 | 81 | 90 | 73 |
| Clear cells <25% | 6/47 | 60/92 | 65 | 87 | 91 | 56 |
SN, sensitivity; SP, specificity; PPV, positive predictive value; NPV, negative predictive value.
Table 2. Distribution of Weiss scoring system path- ological parameters in 139 adrenocortical tumours, including 92 carcinomas and 47 adenomas
A
B
C
;Ď
Disrupted reticulin framework
No
ACA
Yes
Presence of any of the following: Mitosis >5 x 50 HPF
No
Necrosis
ACA
Venous invasion
Yes
Low risk group: Stage 1-2 and Mitosis ≤9 × 50HPF
ACC
Intermediate risk group:
Stage 1-2/Mitosis >9 × 50 HPF or Stage 3-4/Mitosis ≤9 × 50 HPF
High risk group: Stage 3-4 and Mitosis >9 × 50 HPF
specificity and highest sensitivity and specificity, namely mitotic count >5/50 HPF, presence of necrosis (irrespective of its extent) and/or venous invasion (Figure 2). The retrospective evaluation of the whole series of 139 adrenocortical lesions classified according to the Weiss system demonstrated that this algorithm was 100% sensitive and specific for recognizing all cases having a Weiss score ≥ 3, thus considered malignant.
PROGNOSTIC STRATIFICATION OF ADRENOCORTICAL CARCINOMA PATIENTS
A secondary aim of the present study was the prognostic stratification of adrenocortical cancer pa- tients. Among the clinical and pathological parameters investigated by means of univariate analysis, advanced stage (stage 3 and 4; P = 0.007), mitotic count higher than the median value of 9 (P = 0.02), presence of atypical mitoses (P = 0.047), Weiss score > 6 (P = 0.02), and Weiss modified score > 5 (P = 0.02) were significantly associated with a worse overall survival, whereas age over the median of 45 years (P = 0.04), stage (stage 3 and 4; P = 0.008) and
| Hazard ratio | 95% CI | P | |
|---|---|---|---|
| Disease-free survival | |||
| Stage | |||
| 1-2 | |||
| 3-4 | 2.064 | 1.283, 3.319 | 0.003 |
| Mitoses ≤9 | |||
| >9 | 1.021 | 1.004, 1.038 | 0.016 |
| Diffuse architecture Absent | |||
| Present | 0.402 | 0.196, 0.822 | 0.013 |
| Overall survival | |||
| Stage | |||
| 1-2 | |||
| 3-4 | 2.349 | 1.285, 4.293 | 0.006 |
| Mitoses ≤9 | |||
| >9 | 1.033 | 1.012, 1.053 | 0.001 |
| Diffuse architecture Absent | |||
| Present | 0.363 | 0.145, 0.905 | 0.03 |
Age, sex and atypical mitoses failed to enter both Cox models.
mitotic count over the median value of 9 (P = 0.04) were associated with a shorter disease-free survival.
The multivariate Cox regression model showed an independent prognostic role, on both disease-free and overall survival analysis, for stage, mitotic count and presence of diffuse growth pattern (Table 3).
Based on multivariate analysis, adrenocortical carcinomas were re-classified into three risk groups according to the status of the independent prognostic factors (diffuse growth pattern was excluded, being part of the diagnostic algorithm as the first step criterion for adrenal carcinoma). Survival analysis demonstrated that the stratification of adrenocortical cancer patients according to the scheme proposed was strongly corre- lated with disease-free (P < 0.007) and overall (P < 0.005) outcome status (Figure 3). Mean disease- free survival was 61.9, 17.3 and 12.2 months, and overall survival was not attained, 66.2 and
Cumulative proportion disease free surviving
1.0
(a)
· Completed
+ Censored
0.8
- Stage I-II and mitosis ≤9
- - Stage I-II and mitosis >9
0.6
or Stage III-IV and mitosis ≤9
Stage III-IV and mitosis >9
0.4
+
0.2
P < 0.003
0.0
0
50
100
150
200
250
300
Months
1.0
(b)
·Completed
Cumulative proportion surviving
0.9
+Censored
0.8
OF
0.7
0.6
¿
Ô
&
0.5
ar-
.
0.4
&
- Stage I-II and mitosis ≤9
0.3
- - Stage I-II and mitosis >9
or Stage III-IV and mitosis ≤9
0.2
¿
—· Stage III-IV and mitosis >9
0.1
P < 0.005
0.0
0
50
100
150
200
250
300
350
400
Months
26.4 months in low-, intermediate- and high-risk groups, respectively.
Discussion
In the present study, a novel diagnostic approach and the prognostic stratification of adrenal cancer was defined by extensive analysis of clinicopathological features in a large series of tumours. The major relevant findings of our study were: (i) the definition of the presence of a disrupted reticulin framework as a hallmark for the diagnosis of adrenocortical carcinoma; (ii) the proposal of a simplified diagnostic algorithm based on the recognition of unequivocal malignancy- related features rather than on numerical scores; and (iii) the prognostic categorization of adrenal cancer patients into three groups characterized by significantly different outcomes.
Our first observation stemmed from the re-evaluation of the concept of diffuse growth pattern, which in both Weiss’ original papers11,12 and also in a recent study17
was highlighted as a major diagnostic feature for adrenal cancer, although less relevant for others.14 On morphological analysis and appropriate histochemical and immunohistochemical reactions, all adrenocortical carcinomas were found to have disruption of the reticulin framework, irrespective of the growth pattern. In other terms, not only diffuse growth but also trabecular, cord-like, nesting or other patterns were affected by the loss of the reticulin/basal membrane network typical of normal or adenomatous adrenal cortex, thus throwing new light on the concept of growth pattern as a primary feature in adrenocortical carcinoma diagnosis. Interestingly, such an observa- tion fits with some recent findings from our group on the expression by adrenal cancer cells of proteins involved in matrix digestion, such as matrix metallo- proteinase type 2,25 and on the deregulation of gene transcription of laminin isoforms in malignant com- pared with non-neoplastic adrenal tissue (Volante M et al., unpublished observation).
From a diagnostic point of view, the presence of a disrupted reticulin network was a highly sensitive and specific feature in adrenocortical carcinoma diagnosis. Since its recognition is easy and equivocal cases might be elucidated by reticulin staining and/or laminin and type IV collagen immunohistochemistry, we entered this criterion as the primary step of the diagnostic algorithm proposed here. The additional presence of at least one of the malignancy-related features conven- tionally considered in several endocrine tumours (necrosis, high mitotic index and venous vessel inva- sion) increased by up to 100% the sensitivity and specificity of the algorithm for a diagnosis of malig- nancy. The decision of proposing a two-step algorithm, including at a second level additional microscopic features of malignancy, was related to the finding of disruption of the reticulin framework in two cases of large adenomas, which reached a Weiss score of 2: although clinical data were consistent in both cases with benign disease at the last follow-up, the biological meaning of the observed reticulin alteration remains to be defined.
This simplified scheme has the major advantage of reducing the analysis of other parameters such as nuclear atypia, eosinophilic cytoplasm, sinusoidal and capsular invasion, which, although commonly found in malignant lesions, may be equivocal and sub- jective, especially in pathology departments lacking specific experience in endocrine diseases. It is impor- tant to underline that in the great majority of cases (84%) at least two of the malignancy-related param- eters were easily recognizable together with reticulin framework abnormalities, whereas in only a minority
of cases did the definition of malignancy rely on a single parameter (either necrosis, mitotic activity or venous invasion).
A step forward in our algorithmic approach was the identification of risk groups of adrenal cancer patients. In general terms, data in the literature on prognosis of adrenal cancer indicate a plethora of morphological, clinical or molecular features influencing the clinical behaviour of this neoplasm, although some studies are biased by sample size, unclear diagnostic criteria, and/or patient selection. In agreement with data in the literature, in the present study multivariate survival analysis confirmed that diffuse architecture, stage and mitotic index represent the most relevant prognostic factors in adrenal cancer. Concerning mitotic index, the most significant prognostic cut-off in our series was 9/50 HPF (representing the median value), at variance with other authors who used a level of 20/50 HPF.12,20 However, irrespective of the cut-off values employed, it is noteworthy that the vast majority of published series agree that mitotic activity represents the tumour characteristic with the strongest impact on survival.
The clinical impact of the prognostic factors raised in the present multivariate statistical model was further confirmed by survival analysis of the three risk groups identified according to the presence or absence of relevant adverse factors. The relevance of such findings, compared with other studies in the literature,30 is related to the large number of cases here analysed, diagnosed and treated in one institu- tion. A validation set based on an external cohort of cases would be useful to strengthen the applicability of this algorithm.
Acknowledgements
This work was supported by grants from the Italian Ministry of University and Research (MIUR, Rome, ex 60% to M.P., M.T. and M.V.) and from the Regione Piemonte (Ricerca Sanitaria Finalizzata 2007 to M.V. and M.T.). We are grateful to Professor G. Bussolati (University of Turin) for his helpful suggestions.
References
1. Fulmer BR. Diagnosis and management of adrenal cortical carcinoma. Curr. Urol. Rep. 2007; 8; 77-82.
2. Kirschner LS. Emerging treatment strategies for adrenocortical carcinoma: a new hope. J. Clin. Endocrinol. Metab. 2006; 91; 14- 21.
3. Paton BL, Novitsky YW, Zerey M et al. Outcomes of adrenal cortical carcinoma in the United States. Surgery 2006; 140; 914- 920.
4. Terzolo M, Angeli A, Fassnacht M et al. Adjuvant mitotane treatment for adrenocortical carcinoma. N. Engl. J. Med. 2007; 356; 2372-2380.
5. McNicol AM. Assessment of malignancy in adrenal cortical tumors. Endocr. Pathol. 2006; 17; 131-136.
6. Volante M, Buttigliero C, Greco E, Berruti A, Papotti M. Pathological and molecular features of adrenocortical car- cinoma: an update. J. Clin. Pathol. 2008; 61; 787-793.
7. Young WF Jr. Clinical practice. The incidentally discovered adrenal mass. N. Engl. J. Med. 2007; 356; 601-610.
8. Mantero F, Terzolo M, Arnaldi G et al. A survey on adrenal incidentaloma in Italy. Study Group on Adrenal Tumors of the Italian Society of Endocrinology. J. Clin. Endocrinol. Metab. 2000; 85; 637-644.
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. Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am. J. Surg. Pathol. 1984; 8; 163-169.
12. Weiss LM, Medeiros LJ, Vickery AL Jr. Pathologic features of prognostic significance in adrenocortical carcinoma. Am. J. Surg. Pathol. 1989; 13; 202-206.
13. Aiba M, Fujibayashi M. Histopathological diagnosis and prog- nostic factors in adrenocortical carcinoma. Endocr. Pathol. 2005; 16; 13-22.
14. Aubert S, Wacrenier A, Leroy X et al. Weiss system revisited: a clinicopathologic and immunohistochemical study of 49 adre- nocortical tumors. Am. J. Surg. Pathol. 2002; 26; 1612-1619.
15. Stojadinovic A, Brennan MF, Hoos A et al. Adrenocortical adenoma and carcinoma: histopathological and molecular com- parative analysis. Mod. Pathol. 2003; 16; 742-751.
16. Wieneke JA, Thompson LD, Heffess CS. Adrenal cortical neo- plasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. Am. J. Surg. Pathol. 2003; 27; 867-881.
17. Blanes A, Diaz-Cano SJ. Histologic criteria for adrenocortical proliferative lesions: value of mitotic figure variability. Am. J. Clin. Pathol. 2007; 127; 398-408.
18. 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.
19. Gicquel C, Bertagna X, Gaston V et al. Molecular markers and long-term recurrences in a large cohort of patients with sporadic adrenocortical tumors. Cancer Res. 2001; 61; 6762-6767.
20. Assie G, Antoni G, Tissier F et al. Prognostic parameters of metastatic adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 2007; 92; 148-154.
21. Lack EE. Adrenal cortical carcinoma. In Lack EE ed. AFIP Atlas of tumor pathology, series 4; Tumors of the adrenal glands and extraadrenal paraganglia. Washington, DC: American Registry of Pathology Press, 2007; 131-160.
22. Sullivan M, Boileau M, Hodges CV. Adrenal cortical carcinoma. J. Urol. 1978; 120; 660-665.
23. Jarzembowski J, Lloyd R, Mckeever P. Type IV collagen immu- nostaining is a simple, reliable diagnostic tool for distinguishing between adenomatous and normal pituitary glands. Arch. Pathol. Lab. Med. 2007; 131; 931-935.
24. Papotti M, Volante M, Bussolati G. Applied histochemistry in disorders of the endocrine system. In Wick MR ed. Diagnostic
histochemistry. Cambridge: Cambridge University Press, 2008; 280-306.
25. Volante M, Sperone P, Bollito E et al. Matrix metalloproteinase type 2 expression in malignant adrenocortical tumors: diagnostic and prognostic significance in a series of 50 adrenocortical carcinomas. Mod. Pathol. 2006; 19; 1563-1569.
26. deFraipont F, El Atifi M, Cherradi N et al. Gene expression profiling of human adrenocortical tumors using complementary deoxyribonucleic acid microarrays identifies several candidate genes as markers of malignancy. J. Clin. Endocrinol. Metab. 2005; 90; 1819-1829.
27. Evans HL, Vassilopoulou-Sellin R. Adrenal cortical neoplasms. A study of 56 cases. Am. J. Clin. Pathol. 1996; 105; 76-86.
28. Stojadinovic A, Ghossein RA, Hoos A et al. Adrenocortical carcinoma: clinical, morphologic, and molecular characteriza- tion. J. Clin. Oncol. 2002; 20; 941-950.
29. Tauchmanovà L, Colao A, Marzano LA et al. Andrenocortical carcinomas: twelve-year prospective experience. World J. Surg. 2004; 28; 896-903.
30. Harrison LE, Gaudin PB, Brennan MF. Pathologic features of prognostic significance for adrenocortical carcinoma after cura- tive resection. Arch. Surg. 1999; 134; 181-185.