ELSEVIER
Original contribution
Diagnostic and prognostic role of steroidogenic factor 1 in adrenocortical carcinoma: a validation study focusing on clinical and pathologic correlates
Eleonora Duregon MDª, Marco Volante MDa,*, Jessica Giorcelli BSc (Med) a, Massimo Terzolo MDb, Enzo Lalli MDC, Mauro Papotti MDa
ªDivision of Pathology, University of Turin at San Luigi Hospital, 10043 Orbassano, Torino, Italy
Division of Internal Medicine 1 of the Department of Clinical and Biological Sciences, University of Turin at San Luigi Hospital, 10043 Orbassano, Torino, Italy
“Institut de Pharmacologie Moléculaire et Cellulaire CNRS and Université de Nice-Sophia Antipolis, 06560 Valbonne, France
Received 10 May 2012; revised 30 July 2012; accepted 30 July 2012
Keywords:
Adrenal cancer; SF-1;
Monoclonal antibody; Prognosis
Summary The pathologic characterization of adrenocortical cancer is still problematic for several reasons, including the identification of novel markers of diagnostic or prognostic relevance. Among them, steroidogenic factor 1 deserves major interest because of its potential usefulness as a marker of adrenocortical derivation and of biological aggressiveness. Our aim was to validate its prognostic relevance in a large series of adrenocortical cancer, comparing the performance of 2 different commercial antibodies and investigating its expression in adrenocortical cancer variants and in comparison with clinical and pathologic features. Seventy-five (including 53 classical, 10 myxoid, and 12 oncocytic) adrenocortical cancer cases were included in tissue microarrays and analyzed for the immunohistochemical expression of steroidogenic factor 1 using 2 commercial antibodies, 1 polyclonal and 1 monoclonal (N1665). Nuclear steroidogenic factor 1 staining was assessed using a semiquantitative score and correlated with adrenocortical cancer type and clinical pathologic characteristics. A weak but significant correlation was found comparing the 2 antibodies with a positive rate of 88% and 58% using the monoclonal and polyclonal antibodies, respectively. High steroidogenic factor 1 expression with the N1665 antibody was positively correlated with high mitotic count, high Ki-67 index, and high European Network for the Study of Adrenal Tumors (ENSAT) stage and negatively associated with loss of functionality and presence of oncocytic features. Moreover, high steroidogenic factor 1 expression with this same antibody was significantly associated at univariate analysis with a decreased survival, together with high Ki-67 and mitotic indexes, with a trend to significance confirmed by multivariate analysis, thus supporting the detection of steroidogenic factor 1 using the N1665 antibody as a novel prognostic marker in adrenocortical cancer. @ 2013 Elsevier Inc. All rights reserved.
# Conflict of interest statement: All authors declare the absence of any potential conflict of interest.
Source of support: Work supported by grants from the Italian Association for Cancer Research (AIRC, Milan, grant no. IG/10795/2010 to M. P.); Institut National du Cancer, FP7 ENS@T-Cancer under grant agreement 259735 and Association pour la Recherche sur le Cancer (Projets ARC SFI20111203563) to E. L.
* Corresponding author. Department of Clinical & Biological Sciences, University of Turin at San Luigi Hospital, Regione Gonzole 10, 10043 Orbassano, Torino, Italy.
1. Introduction
The pathologic characterization of adrenocortical carcinoma (ACC) still remains challenging for the pathologist for several reasons. First of all, it is a rare neoplasm (approximately 1 case per million) for which clear-cut morphological parameters to define malignancy are lacking [1]. Many diagnostic methods have been proposed, either based on scoring systems including the most frequently used Weiss system [2,3] or using alternative algorithmic approaches [4,5]. In addition, rare morphological variants exist, such as myxoid [6] and oncocytic [7,8] ACC, for which the Weiss system is not adequate due to the risk of under- or overestimating malignancy, respectively. Moreover, even when malignancy is easily assessable, differential diagnosis of ACC might be uncertain from other primary adrenal (pheochromocytoma and PEComa) or extra- adrenal (renal cell carcinoma, melanoma, poorly differentiated metastatic carcinomas, and retroperitoneal sarcomas) neo- plasms, and immunophenotyping is not always conclusive. Finally, ACC is a highly aggressive malignancy, and a more precise prognostic estimate is required. As a consequence, many complementary tools to the classical morphological diagnosis were proposed, generally based on specific marker immunodetection and gene, micro-RNA, or gene methylation profiling [9-14]. Among the most promising novel tools, steroidogenic factor 1 (SF-1) has been recently proposed as an adrenocortical-specific marker as well as a relevant prognostic indicator in ACC [15]. In fact, SF-1, which is expressed by adrenocortical cells during fetal and adult life, mostly in the zona glomerulosa and fasciculata, appeared as a highly sensitive (98.6%) and specific (100%) diagnostic marker of adrenocortical derivation, and its high protein expression levels were also correlated with a poor prognosis. The diagnostic role of SF-1 was later confirmed by Sangoi et al [16] on a series of 63 adrenocortical lesions, whereas the prognostic role still needs to be validated. The aim of the present study was therefore (a) to compare the performance of 2 different SF-1 commercial antibodies in a large series of ACC, (b) to investigate if ACC variants differ in SF-1 expression in comparison with the classical ACC, and (c) to compare SF-1 expression with several clinical and pathologic features and validate its prognostic role in patients with ACC.
2. Materials and methods
2.1. Tissue collection
One hundred ninety-seven consecutive adrenocortical tumors having a Weiss score of 3 or higher [2,3] were collected between 1990 and 2011 from the pathology files of the University of Turin, including 138 consultation cases. Among them, 75 cases had paraffin-embedded tissue suitable for tissue microarray (TMA) construction and formed the basis of the current study (Table 1). Most of these patients were
| Table 1 Clinical and pathologic features of 75 ACC cases included in the TMA | |
| Parameter | |
| M/F ratio | 32/43 |
| Age, mean (y) | 46 (20-79) |
| Functional status (not known, 15) | Not functioning: 23 Functioning: 37ª |
| Size, mean (cm) | 11.5 (2-30) |
| Weight, mean (g) | 481 (8-3200) |
| Histologic type | Conventional: 53 Myxoid: 10 |
| Oncocytic: 12 | |
| Weiss score distribution | 3-4: 15 |
| 5-6: 24 | |
| 7-8-9: 36 | |
| ENSAT stage (not known, 20) | 1-2: 20 |
| 3-4: 35 | |
| FU (lost to FU, 6) | NED/DOC: 21 AWD: 15 DOD: 33 |
| Median overall survival (mo) | 53 |
Abbreviations: M, male; F, female; FU, follow-up; NED, no evidence of disease; DOC, died of other cause; AWD, alive with disease; DOD, died of disease.
a Functionality included the presence of predominant cortisol (28 cases, including 1 oncocytic and 5 mixoid tumors), aldosterone (6 cases, including 1 oncocytic tumor), and androgens (3 cases, including 1 oncocytic tumor) secretion.
treated at our institution, which serves as a referral center for ACC in Italy. The histopathologic features of 22 classical, 10 myxoid, and 12 oncocytic ACCs of this data set have already been reported [5-7]. For all cases, the clinical, pathologic, and follow-up data were collected. The study received ethical approval from the local review board of our institution.
2.2. TMA construction
Hematoxylin and eosin-stained slides of the 75 ACCs were reviewed by 1 of us (E. D.) to identify 3 representative areas of well-preserved morphology and no significant necrotic tissue. The corresponding areas on the paraffin block were marked for tissue punching. TMAs were assembled using a commercially available manual tissue puncher/arrayer (Quick-RAY tissue arrayer; Bio-Optica, Milan, Italy) according to the manufac- turer’s instructions. To minimize the number of inadequate cases and to increase concordance rates among different cores, 3 cores with a diameter of 2 mm were punched from each tissue block and arrayed into the recipient paraffin block. TMAs also included 2 cores of different control tissues (adrenal cortex, adrenal medulla, kidney).
2.3. Immunohistochemistry
Serial 5-um-thick paraffin sections from the constructed TMA blocks were processed for immunohistochemistry. Two commercially available antibodies against SF-1 were used, the
first polyclonal (diluted 1:500; Upstate, Millipore, Billerica, MA) and the second monoclonal (clone N1665, diluted 1:100; Perseus Proteomics, Tokyo, Japan). The proliferative index was evaluated in parallel with MIB-1 antibody to Ki-67 (diluted 1:150; Dako Cytomation, Glostrup, Denmark). Sections were deparaffinized and rehydrated in ethanol. For antigen retrieval, a pressure cooker treatment for 5 minutes at 125°C, followed by a quick 10-second step at 90℃, was performed in EDTA buffer (pH 8.0) for all antibodies. Samples were allowed to cool to room temperature before incubation with 3% hydrogen peroxide in Tris-buffered saline for 10 minutes to quench endogenous peroxidase activity. Subse- quently, sections were incubated for 1 hour at 37℃ with SF-1 antibodies or for 1 hour at room temperature for Ki-67. A biotin-free, dextran chain-based detection system (EnVysion; Dako) was used according to a standard protocol and using diaminobenzidine as the chromogen. Finally, slides were
counterstained with hematoxylin, dehydrated, and mounted. TMA cores of normal tissue including adrenal cortex, adrenal medulla, and kidney served as appropriate internal positive (the former) and negative (the others) controls.
2.4. Staining interpretation and scoring system
All slides were analyzed independently by 2 of us (E. D. and M. V.), who included in the analysis only samples with 2 or 3 evaluable cores after the immunostaining. Staining was assessed for each core using a semiquantitative scoring system based on the evaluation of nuclear staining, as follows: score 0, no staining; score 1, focal (<50% of the core) staining; score 2, diffuse (≥50% of the core) staining. An overall score for each case was calculated as the mean of the assessed cores. In case of discrepancies, slides were reviewed at a multihead microscope, and a consensus was reached. For
Monoclonal SF-1 antibody
Polyclonal SF-1 antibody
Normal adrenal cortex
Adrenocortical carcinoma
descriptive purposes, all cases with an overall score different from 0 were considered positive. Moreover, for subsequent statistical analysis, cases were then categorized as “negative/ low” or “high” expression according to the mean score value of the overall population (corresponding to 1.11).
2.5. Statistical analysis
Correlation between the 2 SF-1 antibodies was made using a 2-tailed Spearman test. Clinical and pathologic characteristics were compared with staining patterns of the 2 SF-1 antibodies by Fisher exact or x2 and the Student t tests, for categorical and continuous variables, respectively. To analyze the prognostic impact of all clinical and pathologic variables considered, univariate overall survival analysis was based on the Kaplan- Meier product limit estimate of survival distribution. Clinical pathologic parameters considered in overall survival analysis included sex, age, hormonal secretion, weight, size, Weiss score, Ki-67 index, mitotic index, ENSAT stage, and the reactivity to SF-1 antibodies. Unadjusted differences between survival curves were tested using the log-rank test. All parameters with a significant impact on survival at univariate
analysis were considered for multivariate analysis using the Cox proportional hazards model. Statistical significance was set at P < . 05. All tests were performed using GraphPad Prism version 5.0 for Windows (GraphPad Software, La Jolla, USA; www.graphpad.com) and STATISTICA version 7.0 (StatSoft Italia, Vigonza, Padova, Italy) softwares.
3. Results
3.1. Clinical and pathologic data and SF-1 protein expression
Clinical and pathologic data of the ACC series analyzed are summarized in Table 1. All tumors had a Weiss score of 3 or higher and were defined as malignant also using the “reticulin algorithm” [5]. All tumors had reticulin framework disruption in the presence of at least 1 of either high mitotic count or necrosis or vascular invasion.
SF-1 protein expression (including all cases with a mean case score different from 0) was detected in 64 (88%) of 73
| Table 2 SF-1 immunohistochemistry: clinical and pathologic correlates in ACC | |||||||
|---|---|---|---|---|---|---|---|
| Parameter | SF-1 polyclonalª | SF-1 monoclonalª | |||||
| Negative/low | High | P | Negative/low | High | P | ||
| Sex | M | 24 | 8 | .016 | 25 | 6 | .19 |
| F | 18 | 22 | 27 | 15 | |||
| Age | ≤45 y | 21 | 17 | .64 | 29 | 10 | .61 |
| >45 y | 21 | 13 | 23 | 11 | |||
| Histologic type | Conventional | 31 | 19 | .44 | 37 | 14 | .026 |
| Myxoid | 4 | 6 | 4 | 6 | |||
| Oncocytic | 7 | 5 | 11 | 1 | |||
| Functional status | Functioning | 21 | 14 | .40 | 22 | 14 | .015 |
| Not functioning | 16 | 6 | 21 | 2 | |||
| Type of hormone | Cortisol | 16 | 10 | 1.00 | 16 | 11 | 1.00 |
| Aldosterone/androgens | 5 | 4 | 6 | 3 | |||
| Size b | <11 | 15 | 11 | .77 | 21 | 9 | .59 |
| ≥11 | 15 | 8 | 17 | 10 | |||
| Weight b | ≤260 | 19 | 11 | .59 | 16 | 10 | .38 |
| >260 | 24 | 11 | 17 | 6 | |||
| Mitotic index ×50 HPF b | <11 | 29 | 11 | .017 | 33 | 7 | .036 |
| ≥11 | 14 | 18 | 19 | 14 | |||
| Ki-67 index b | <20 | 18 | 10 | .43 | 26 | 2 | .0002 |
| ≥20 | 16 | 15 | 15 | 17 | |||
| Weiss score | 3-4 | 8 | 6 | .99 | 14 | 0 | .024 |
| 5-6 | 14 | 10 | 16 | 7 | |||
| 7-8-9 | 20 | 14 | 22 | 14 | |||
| ENSAT stage | 1-2 | 12 | 7 | .77 | 18 | 2 | .029 |
| 3-4 | 18 | 14 | 20 | 14 | |||
| Disease status | NED/DOC | 15 | 6 | .24 | 17 | 3 | .21 |
| AWD | 6 | 8 | 10 | 5 | |||
| DOD | 19 | 12 | 20 | 12 | |||
NOTE. SF-1 polyclonal: Upstate, Millipore; SF-1 monoclonal: clone N1665, Perseus. Abbreviation: HPF, high-power field.
a SF-1 polyclonal, 72 cases evaluable; SF-1 monoclonal, 73 cases evaluable.
b Median value.
evaluable ACC cases by the monoclonal antibody and in 42 (58%) of 72 by the polyclonal antibody. Both antibodies tested positive at the nuclear level in normal adrenocortical tissue (Fig. 1) and negative in control tissues from adrenal medulla and kidney. A general more intense immunoreactivity, both in normal and neoplastic adrenocortical tissues, was observed using the SF-1 monoclonal antibody. An additional weak and occasional cytoplasmic immunoreactivity was observed when the SF-1 polyclonal antibody was used only. A significant, although weak, positive correlation was found comparing the scoring results obtained with the 2 antibodies (Spearman r, 0.4786; 95% confidence interval, 0.2696-0.6446; P < . 0001).
3.2. Correlation of SF-1 expression with clinical and pathologic variables
High SF-1 nuclear expression, as detected by using the polyclonal antibody, was associated with female sex and high mitotic rate (Table 2). In contrast, high SF-1 nuclear expression, as detected by using the monoclonal antibody, was positively correlated with high mitotic count, high Ki-67 index, and advanced ENSAT stage. Interestingly, SF-1 protein using the monoclonal antibody was also differential- ly expressed in ACC variants, being negative or low in most cases of the oncocytic type, and a higher rate of reactivity was observed in functioning tumors, too, although not significantly associated with the type of hormone secreted.
3.3. Correlation of SF-1 expression with disease outcome
Univariate analysis showed that high Ki-67 proliferation and high mitotic indexes were associated with shorter overall
| Table le 3 Univariate overall survival analysis | |||
|---|---|---|---|
| Parameter | HR | 95% CI | P |
| Female sex | 1.064 | 0.5317-2.129 | .8608 |
| Age ≤45 y | 0.9039 | 0.4529-1.804 | .7744 |
| Presence of hormonal activity | 1.406 | 0.6741-2.933 | .3635 |
| Weight >260 g | 0.8803 | 0.3990-1.942 | .7992 |
| Size >11 cm | 1.257 | 0.5852-2.699 | .5577 |
| Weiss score ≥7 | 1.333 | 0.6639-2.676 | .4189 |
| Ki-67 ≥20%ª | 2.366 | 1.095-5.112 | .0285 * |
| Mitotic index ×50 HPF ≥11 ª | 3.482 | 1.608-7.538 | .0015* |
| ENSAT stages 3-4 | 1.611 | 0.7724-3.360 | .2036 |
| High SF-1 levels | 1.162 | 0.5522-2.443 | .6931 |
| (polyclonal) | |||
| High SF-1 levels | 3.924 | 1.561-9.860 | .0036 * |
| (monoclonal) ª | |||
NOTE. SF-1 polyclonal: Upstate, Millipore; SF-1 monoclonal: clone N1665, Perseus.
Abbreviations: HR, hazard ratio; CI, confidence interval.
a Parameters included in multivariate analysis: mitotic index, P =
.028; Ki-67, P = . 342; SF-1 monoclonal, P = . 117.
* Significant P value.
Overall Survival
100
-L. SF-1 high
80
SF-1 negative/low
60
40
20
0
0
100
200
300
400
500
months
survival (Table 3). Moreover, high SF-1 expression, as detected by means of the monoclonal antibody, was associated with an increased risk of disease-related death (hazard ratio, 3.924; P = . 0036) and decreased survival (median survival of 27 versus 82 months in cases with high and negative/low expression, respectively) (Fig. 2). A trend toward significance was maintained at multivariate analysis (P = . 117) that showed mitotic index as the most powerful prognostic indicator in our series (P= . 028). By contrast, SF- 1 expression as detected by the polyclonal antibody did not show any prognostic relevance in the current ACC series.
4. Discussion
In the present report, we investigated the expression of SF-1 in a large series of ACC. The aim of our study was to validate the utility of SF-1 as a prognostic marker, testing the immunohistochemical performance of 2 different commer- cially available antibodies and comparing its expression with several clinical and pathologic parameters. The first antibody that we used is a monoclonal reagent, which has been previously demonstrated to be highly specific and sensitive for the recognition of adrenocortical tissues as well as prognostic in patients with ACC [15,16]. The second antibody is a polyclonal antibody never tested in ACC samples, so far, to the best of our knowledge.
Among several immunohistochemical and molecular markers recently proposed in ACC, SF-1 deserves major interest for the dual potential implications in the diagnostic and prognostic characterization of ACC. In fact, as already mentioned above, SF-1 expression is a strong tool to determine the adrenocortical derivation of a given tumoral lesion, irrespective of its benign or malignant nature, being expressed by few other steroidogenic tissues only (such as ovarian stroma or steroid hormone-producing testicular cells). Although we did not design this study to validate this
specific diagnostic utility, we could confirm that SF-1 is expressed in a relevant percentage of ACC, thus being a highly sensitive tool to differentiate ACC from other mimickers in the adrenal gland. However, the expression rates were different when considering the 2 different antibodies; in fact, it increased from 58% positive cases only with the polyclonal serum to 88% with the monoclonal antibody. This latter figure is slightly lower than that originally reported by Sbiera et al (98%) [15] but comparable with what reported recently on TMA material (86%) [16], using the same monoclonal antibody herein used.
Indeed, we mainly focused our analysis of SF-1 expression in a cohort of ACC with the specific aim to validate its prognostic value and investigate different SF-1 expression profiles as compared with clinical and pathologic features. In the original article by Sbiera et al [15], the subanalysis of SF-1 distribution was limited to sex, age, tumor stage, and hormone secretion, apparently with no specific correlation. In our series, the 2 different antibodies gave heterogeneous results as a possible consequence of their poor reciprocal correlation. In fact, although the polyclonal SF-1 antibody was associated with sex and mitotic index only, the monoclonal antibody showed a wider association with “high-grade” clinical and pathologic parameters that is to say characteristics predictive of worse prognosis. High SF- 1 expression, as detected with the monoclonal antibody, was positively correlated with advanced ENSAT stage, high proliferation and mitotic indexes, and high Weiss score values. These data are consistent with previous studies showing that SF-1 overexpression triggers proliferation of ACC cells and induces adrenocortical tumor formation in mice [17] and with in vitro observations that specific SF-1 antagonists are capable of decreasing proliferation in ACC cell models [18]. Moreover, we originally found that SF-1 was differentially expressed in ACC histologic types because it was slightly more expressed in myxoid ACC (6/10 cases) compared with the conventional ones (14/51 cases), and it was very infrequently reactive in the oncocytic variant of ACC (1/12 cases). This observation might possibly reflect the generally lower growth potential in this latter group as compared with classical ACC [7], but alternative mecha- nisms linked to the specific metabolic/functional properties of oncocytic cell and tumors might be also active. Interestingly, we also observed a striking difference in the prevalence of SF-1 overexpression according to the functional status of the tumors analyzed: SF-1 was detected at a high expression level in nearly 40% of functioning but in 9% only of nonfunctioning tumors. Although this is apparently in contrast with the previous observation by Sbiera et al [15], it is quite reasonable to assume that the expression levels of SF-1, whose properties are related to steroidogenesis [19], might be proportional to hormonal production and secretion by tumor cells. It is worth noticing that the prevalence of functional tumors in the oncocytic and myxoid cases was opposite because 3 of 7 and 5 of 6 were hormone secreting in the oncocytic and myxoid group of
tumors, respectively. Therefore, it should be considered that this finding might be, at least partially, responsible for the correlation between SF-1 expression and histologic variant. In practical terms, the present data show that SF-1 expression may be a useful diagnostic tool, given that the results are integrated in an appropriate morphological and clinical context because its sensitivity might be lower in specific ACC variants and in nonfunctioning tumors.
Finally, we validated the impact of SF-1 expression levels on prognosis of the current ACC series. At univariate analysis, high expression levels of SF-1, as detected by the N1665 monoclonal antibody, were closely correlated with survival, together with mitotic and proliferation indices. At multivariate analysis, mitotic count was the only significant independent parameter in our series, but a trend toward significance was maintained for SF-1, too. Although methodological differences (SF-1 scoring system and cutoff value selection, size of the case series, patients’ character- istics) are possibly responsible for the slightly lower statistical power of SF-1 as compared with that reported by Sbiera et al [15], our data are supportive of the potential role of SF-1 as a prognostic marker in ACC. For this reason, its inclusion in a minimal set of markers for the characterization of ACC samples is advisable.
In conclusion, SF-1 is detectable in a high percentage of ACC and its high expression levels correlate with high-grade features, histologic type, functional status, and adverse outcome. The N1665 monoclonal antibody proved superior to commercial polyclonal antisera, and its use is therefore currently advisable in the clinical practice for both diagnostic and prognostic purposes.
References
[1] Papotti M, Libè R, Duregon E, Volante M, Bertherat J, Tissier F. The Weiss score and beyond-histopathology for adrenocortical carcino- ma. Horm Cancer 2011;2:333-40.
[2] Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984;8: 163-9.
[3] Weiss LM, Medeiros LJ, Vickery Jr AL. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13:202-6.
[4] Blanes A, Diaz-Cano SJ. Histologic criteria for adrenocortical proliferative lesions: value of mitotic figure variability. Am J Clin Pathol 2007;127:398-408.
[5] Volante M, Bollito E, Sperone P, et al. Clinicopathological study of a series of 92 adrenocortical carcinomas: from a proposal of simplified diagnostic algorithm to prognostic stratification. Histopathology 2009; 55:535-43.
[6] Papotti M, Volante M, Duregon E, et al. Adrenocortical tumors with myxoid features: a distinct morphologic and phenotypical variant exhibiting malignant behavior. Am J Surg Pathol 2010;34:973-83.
[7] Duregon E, Volante M, Cappia S, et al. Oncocytic adrenocortical tumors: diagnostic algorithm and mitochondrial DNA profile in 27 cases. Am J Surg Pathol 2011;35:1882-93.
[8] Wong DD, Spagnolo DV, Bisceglia M, Havlat M, McCallum D, Platten MA. Oncocytic adrenocortical neoplasms-a clinico- pathologic study of 13 new cases emphasizing the importance of their recognition. HUM PATHOL 2011;42:489-99.
[9] de Reyniès A, Assié G, Rickman DS, et al. Gene expression profiling reveals a new classification of adrenocortical tumors and identifies molecular predictors of malignancy and survival. J Clin Oncol 2009; 27:1108-15.
[10] Ragazzon B, Libé R, Gaujoux S, et al. Transcriptome analysis reveals that p53 and {beta}-catenin alterations occur in a group of aggressive adrenocortical cancers. Cancer Res 2010;70:8276-81.
[11] Ragazzon B, Assié G, Bertherat J. Transcriptome analysis of adrenocortical cancers: from molecular classification to the identifi- cation of new treatments. Endocr Relat Cancer 2011;18:R15-27.
[12] Patterson EE, Holloway AK, Weng J, Fojo T, Kebebew E. MicroRNA profiling of adrenocortical tumors reveals miR-483 as a marker of malignancy. Cancer 2011;117:1630-9.
[13] Özata DM, Caramuta S, Velázquez-Fernández D, et al. The role of microRNA deregulation in the pathogenesis of adrenocortical carcinoma. Endocr Relat Cancer 2011;18:643-55.
[14] Rechache NS, Wang Y, Stevenson HS, et al. DNA methylation profiling identifies global methylation differences and markers of
adrenocortical tumors. J Clin Endocrinol Metab 2012 [Epub ahead of print].
[15] Sbiera S, Schmull S, Assie G, et al. High diagnostic and prognostic value of steroidogenic factor-1 expression in adrenal tumors. J Clin Endocrinol Metab 2010;95:E161-71.
[16] Sangoi AR, Fujiwara M, West RB, et al. Immunohistochemical distinction of primary adrenal cortical lesions from metastatic clear cell renal cell carcinoma: a study of 248 cases. Am J Surg Pathol 2011;35: 678-86.
[17] Doghman M, Karpova T, Rodrigues GA, et al. Increased steroidogenic factor-1 dosage triggers adrenocortical cell proliferation and cancer. Mol Endocrinol 2007;21:2968-87.
[18] Doghman M, Cazareth J, Douguet D, Madoux F, Hodder P, Lalli E. Inhibition of adrenocortical carcinoma cell proliferation by steroido- genic factor-1 inverse agonists. J Clin Endocrinol Metab 2009;94: 2178-83.
[19] Lalli E. Adrenocortical development and cancer: focus on SF-1. J Mol Endocrinol 2010;44:301-7.