ELSEVIER
d
SCIENCE
DIRECT ®
& BIOMEDICINE PHARMACOTHERAPY
Original article
http://france.elsevier.com/direct/BIOPHA/
Gene expression profiling of adrenal cortical tumors by cDNA macroarray analysis. Results of a preliminary study
C.P. Lombardi ª, M. Raffaelli a,*, G. Pani b, A. Maffione b, P. Princi ª, E. Traini ª, T. Galeotti b, E.D. Rossi , G. Fadda , R. Bellantone ª
ª Division of Endocrine Surgery, Department of Surgery, Università Cattolica del Sacro Cuore, L.go A. Gemelli 8, 00168 Rome, Italy b Institute of Pathology, Università Cattolica del Sacro Cuore, L.go A. Gemelli 8, 00168 Rome, Italy
” Institute of Histopathology, Università Cattolica del Sacro Cuore, L.go A. Gemelli 8, 00168 Rome, Italy
Received 24 November 2005; accepted 10 March 2006 Available online 31 March 2006
Abstract
Adrenocortical carcinoma (ACC) are highly malignant tumors with poor prognosis. To verify if it is possible to assess their differential gene expression by a cDNA macroarray analysis using RNA extracted from paraffin sections, we analyzed two different cohorts of adrenal cortical adenoma (ACA) and ACC. Paraffin sections of seven ACC and seven ACA were analyzed. Transcriptional profiles were generated by commer- cially available c-DNA arrays testing 82 genes. Hybridization signals were quantified by densitometry and the intensity signal was compared for each gene between ACA and ACC cohorts. RNA was successfully extracted in only four out of 14 cases. Four genes displayed a significantly different expression (ACC/ACA ratio > 1.5 or <0.6). Heat shock protein 60 (HSP-60) (ratio >2), Ciclin D1 and topoisomerase I (ratio >1.5) were overexpressed in the ACC cohort, while jun proto-oncogene was down-regulated.
cDNA macroarray analysis from paraffin sections of adrenal tumors is feasible, despite with a low success rate.
The different expression of HSP-60, Ciclin D1, jun proto-oncogene and topoisomerase I indicates that these genes may play a role in ACC pathogenesis and could represent potential diagnostic/prognostic/therapeutic target markers.
Larger series of patients are necessary to confirm the biologic, diagnostic, prognostic and therapeutic implications of these findings. @ 2006 Elsevier SAS. All rights reserved.
Keywords: Adrenal tumors; Gene profiling; cDNA array; Macroarrays analysis, Adrenocortical carcinoma
1. Introduction
Adrenocortical carcinoma (ACC) is a rare and highly malig- nant neoplasm with poor prognosis [1]. Its annual incidence is approximately one to two patients per million population [1,2].
The pathological diagnosis of ACC is straightforward in most cases and different scores for the diagnosis of malignant adrenal-cortical tumors have been developed [3,4]. Typical his- topathological features of adrenocortical malignancy include
Abbreviations: ACA, adrenal cortical adenoma; ACC, adrenal cortical carcinoma; Ciclin D1, jun proto-oncogene; HSP-60, heat shock protein 60; TOP2A, topoisomerase II a.
* This paper has been presented at the ESES 2004 Meeting, Pisa, Italy, May 13-15, 2004.
Corresponding author.
E-mail address: marcoraffaelli@rm.unicatt.it (M. Raffaelli).
0753-3322/$ - see front matter @ 2006 Elsevier SAS. All rights reserved. doi:10.1016/j.biopha.2006.03.006
large tumor size and weight, solid growth pattern, extensive tumor necrosis, fibrous band, lipid poor cells, abundant mi- toses, atypical mitoses, nuclear pleomorphism, capsular and/ or vascular invasion [1,2].
However, differential diagnosis between adrenal cortical adenoma (ACA) and ACC is sometimes difficult, in absence of any clear evidence of local invasion and distant metastases [1,5]. Nuclear DNA content by flow cytometry correlates poorly with histological and mitotic indices and cannot reliably differentiate between ACA and ACC [6]. P53 tumor suppres- sor gene alterations have been demonstrated to be variably ex- pressed in ACC and to be of little prognostic value [6,7].
In recent years molecular markers of adrenocortical malig- nancy have been quite widely investigated [8-14]. A Ki-67 staining index of more than 5% in adrenocortical tumors has been reported to be suggestive of an ACC [8]. Other marker of
cell proliferation, as PCNA, epidermal growth factor receptor (EGF-r) and topoisomerase IIa (TOP2A), have shown in- creased expression in ACC [9,10]; however, no evidence of correlation between these proteins and clinical outcome has been demonstrated [9,11,12].
Furthermore, the apoptosis regulation protein (bcl2 and bax gene) have been investigated in adrenocortical tumors [11,13, 14]. Despite they could represent diagnostic marker of malig- nancy the results are not univocal in published studies [11,13, 14].
With the emerging technology of cDNA array hybridization it is now possible to screen the expression of several genes simultaneously in order to detect the genes that might have potential clinical relevance as tumor marker for early diagnosis, prognostic significance and specific target for selection of post- operative drug therapy [5,15].
Extraction of RNA required for gene profiling is usually obtained from specimens frozen stored immediately after resec- tion [5,15-21]. However, the prospective collection of speci- mens can take a considerable length of time. And this is parti- cularly true for rarer diseases, as ACC is.
Tissue specimens have been preserved over years in all clin- ical settings as paraffin blocks. This archive could represent a valuable source of material from which expressed genes can be investigated. This is the reason why the possibility to extract RNA from paraffin sections has been evaluated. Indeed the possibility to have an access to these archives has several the- oretical advantages related to the relative amplitude of the ser- ies, to the possibility to conduct retrospective studies and, thus, to easily correlate the clinical outcomes with the histological and biological tumor characteristics. Moreover, the paraffin blocks tissue archives could be very helpful for investigating and understanding the changes that occur in gene function dur- ing the progression of the disease. However, extraction of RNA from paraffin sections has proven to be problematic [16]. Indeed, even if many researchers have attempted to over- come this kind of problem and RNA can be successfully ex- tracted from paraffin sections, this procedure results in low yields of a highly degraded product [16].
The aim of this study was to evaluate if it is possible to assess the different gene expression profiles of benign and ma- lignant adrenocortical tumors and to identify new markers of malignant behavior by cDNA macroarray analysis, using RNA extracted from paraffin section of adrenocortical lesions.
2. Materials and methods
Paraffin sections of adrenal surgical specimens from 14 pa- tients who underwent adrenalectomy for primary adrenal corti- cal tumors (seven unequivocal adenomas and seven unequivo- cal carcinomas) were analyzed. There were five men and nine women with a mean age of 48.5 + 17.5 years. Transcriptional profiles were generated by commercially available oligonu- cleotide arrays (CLONTECH, Palo Alto, CA, USA) testing 82 genes in each sample.
2.1. Tissue samples
Tissue samples were stained with haematoxylin and eosin. Diagnosis of adrenocortical adenoma and carcinoma was estab- lished on the basis of conventional histological criteria [1,2]. Seven paired tissue samples (ACA vs. ACC) were used for array studies. RNA isolation, array production, hybridization and image analysis were performed as previously described according to manufacturer’s protocols (Atlas cDNA expression arrays, Atlas technology, BD Biosciences Clontech, Palo Alto, CA, USA).
2.2. RNA extraction/labeled cDNA preparation/hybridization
Tissue samples were washed and homogenized in Trizol re- agent (Life Technologies Inc., Gaithersburg, MD, USA) with a TURRAX homogenizer. Total RNA was extracted and purified according to the manufacturer’s manual (Atlas Pure Total RNA Labeling system user, Atlas technology, BD Biosciences Clon- tech, Palo Alto, CA, USA). After digestion with DNAse I, RNA samples were quantified by spectrophotometry and sub- jected to 1% agarose gel electrophoresis and ethydium bromide staining, in order to determine the extent of degradation. The quality of RNA was further evaluated by RT-PCR of a 15 bp exonic fragment of actin. Samples which scored positive for RT-PCR (presence of amplifiable RNA) and negative for con- ventional PCR (absence of genomic DNA contamination) were further processed, to detect eventual sample degradation. In- stead of Poly-A RNA (because of the scarcity and large degra- dation of mRNA), for cDNA preparation it has been used total RNA according to an alternative procedure suggested by man- ufacture manual. Cell line HEK-293 has been used to detect the adequacy of the sample confirmed by ß-actin RT-PCR am- plification with specific primers in all specimens.
cDNA preparation and labeling from total RNA samples was performed according to the manufacturer’s manual (“AT- LAS cDNA expression arrays”, Atlas Technology TM, BD Biosciences Clontech, Palo Alto, CA, USA). Total RNA from the human embryonic kidney cell line HEK-293 was used as positive control throughout the procedure. A chromatographic purification to separate the radioactive probe from the radioac- tive labeled cDNA from each sample was purified from unin- corporated P32ATP by chromatography and hybridized onto ATLAS membranes in ExpressHyb TM hybridization solution (Clontech, Palo Alto, CA, USA).
2.3. Visualization and data analysis
After extensive washing, hybridized membranes were ex- posed either to autoradiography films (KODAK SPA, Cinisello Balsamo-MI, Italy) or to the Cyclone Gene Array system (Per- kinElmer, Meriden, CT, USA) in order to reduce the time of exposure. Spots were quantified by the Quantitative.One soft- ware (BIORAD 1000, Hercules, CA, USA). After background subtraction, absorbance values were normalized for the mean value of the nine housekeeping genes present on the mem- brane. The normalized intensity was used to calculate malig-
nant/benign ratio. Only genes with a ratio higher than 1.5 or lower than 0.6 were considered as significantly different be- tween malignant and benign tumor samples.
3. Results
The ACA cohort included tissue specimens from one man and six women, of the mean age of 45.0 ± 18.0 years (range: 33-83). Mean lesion size for ACA was 3.2 ± 1.7 cm (range: 1.5-6). The ACC cohort included tissue specimens from four men and three women, of the mean age of 52.0 years (range: 27-72). Mean lesion size for ACC was 10.1 ±5.4 cm (range: 6-18).
RT-PCR for ß-actin gave positive results in four out of 14 samples: two ACCs and two ACAs. Only these cases were used for further macroarray hybridization. The majority of “housekeeping” genes were well visible (H1-H9) in the array reaction, while no signal was detected in the negative controls (H10-H12), demonstrating a specific reaction of the radioac- tive probes with cDNA (Fig. 1).
Four genes displayed a significantly different expression (ACC/ACA ratio>1.5 or<0.6) (Fig. 2). The ACC cohort showed overexpression of the heat shock protein 60 (HSP-60) (ratio > 2), of the Ciclin D1 and topoisomerase I (ratio> 1.5) compared with ACA. The expression of jun proto-oncogene was down-regulated in the ACC cohort (ratio < 0.6).
4. Discussion
The biological characteristics of various cancers are cur- rently under investigation using a wide range of molecular bio-
Increase
Decrease
A
B
C
D
*
E
F
*
*
G
*
*
0
0,5
1
1,5
2
2,5
Rate
[ACC]/[ACA]
logical techniques [17], with the main objectives to identify genes implicated in tumorigenesis. cDNA array technology gives the opportunity to comprehensively examine the tran- scriptional profile of tumors. It has been employed in different histological type of tumors to clarify the relation between the gene expression and the clinicopathological factors [5,17-21].
Gene profiling requires RNA extraction which is usually obtained from surgical specimens which are frozen stored im- mediately after resection [5,15-21]. However, for rare diseases,
1 2 3 4 5 6 7 8 9 10 11 12
A
ACA
B
C
D
E
F
G
H
1 2 3 4 5 6 7 8 9 10 11 12
A
B
C
ACC
D
E
..
F
G
..
H
Atlas™ Array Orientation Grid
the prospective collection of this kind of specimens may be quite long. For this reason the possibility to extract RNA from paraffin sections has been explored. Indeed, paraffin embedded tissues archives represent a vast well-characterized source of specimens that could be available for molecular biological in- vestigation.
However, RNA extraction from paraffin sections is proble- matic. In spite of several attempts to overcome technical pro- blems successful extraction usually results in low yields of highly degraded products [16]. Nonetheless, even if degraded it could serve as template for a large number of applications [16]. Our experience confirms the technical difficulties related to RNA extraction from paraffin fixed surgical specimens. The extraction of RNA was adequate in only four cases out of 14.
Limitations related to RNA extraction from paraffin sections are attributable to three main reasons: frequent degradation of nucleic acids (especially RNA), not homogeneous fixation and inclusion processes, length of the extraction procedure.
Moreover, we found that total RNA is highly preferable to Poly A+ mRNA as template for radioactive probe generation. The higher efficiency of total RNA may be related to the pre- sence of degraded mRNA fragments which, even if they have lost the Poly A+ sequence, are still capable of specific hybri- dization with the cDNA array probes. In addition, further total RNA manipulation, which is necessary for Poly A+ RNA iso- lation, could be responsible for a critical waste of nucleic acid.
However, the presence of degraded RNA otherwise does not obstruct its use for gene expression analysis, since the de- gradation process could be similar in all cellular RNA and the quantity of nucleic acid could be sufficient to define the gene profiling [16].
On the basis of the results obtained with ß-actin RT-PCR amplification, only four specimens (two ACA and two ACC) were used for cDNA hybridization.
As expected because of the common origin from adrenocor- tical cells, the ACC and ACA cohort showed a similar genetic pattern. The similarity of the genetic profiles obtained concurs in demonstrating that degradation processes involving paraffin sections slightly interferes with array analysis.
Another potential pitfall affecting analyses like the one here described, is represented by the confounding presence of non- neoplastic tissue in the tumor samples. It is certainly advisable that this risk be minimized by means of micro- or macro-dis- section [22] procedures. Laser microdissection is certainly ad- visable to isolate cancer from normal tissue. Nonetheless, it also accelerates RNA degradation and often requires cDNA amplification steps in order to provide sufficient biological ma- terial for gene expression profiling [23,24].
Major gene expression changes are expected to be scored also in non-purified samples. Indeed, the data here presented, indicating a general consistency in the gene expression profiles from different tumors, again suggest that gross heterogeneities in sample composition can be ruled out. Potentially confound- ing information deriving from the presence of non-cancer tis- sue or from individual variability have limited impact on the study. On the other hand, relatively small differences in gene expression levels, as those here reported (see below), are likely
to become more dramatic once exclusively neoplastic tissue were included in the molecular analysis.
In spite of the overall similarity of the genetic profile, the expression of specific genes resulted selectively modified in the ACC cohort. In particular the ACC cohort showed over- expression of the heat shock protein 60 (HSP-60) (ratio > 2), of the Ciclin D1 and topoisomerase I (ratio > 1.5) compared with ACA, while the expression of jun proto-oncogene was down-regulated in the ACC. Despite these differences should be verified in larger series and eventually validated by specific RT-PCR, Northern and Western blotting analyses, these find- ings have a quite relevant biologic significance.
Heat shock proteins (HSPs) or stress proteins are a set of highly conserved proteins expressed in all major compartments of eukaryotic cells and oncologic research has shown that HSPs may be crucial for tumorigenesis [25]. Enhanced HSP60 expression has been reported in myeloid leukemia, breast and prostate cancer [26]. HSP 60 is known to help the induction of apoptosis and seems increased in adrenal Cushing tumors [25]. The results of our study suggested a possible role of this chaperonin in ACC pathogenesis.
Cyclin D1 is a regulator of the G1 to S transition phase of the cell cycle. Overexpression of nuclear cyclin D1 has been demonstrated to occur at a high frequency in a variety of car- cinomas [27] and cell-cycle targeted therapies using cyclin de- pendent kinase inhibitors have been proposed [28]. Despite previous studies have excluded the expression of cyclin D1 in both normal and malignant adrenal tumors [12], in our study cyclin D1 was highly overexpressed in malignant cells. Thus a possible role of cyclin D1 in ACC pathogenesis could be speculated.
The jun proto-oncogene, implicated cell cycle regulation and oncogenic transformation [29], has been demonstrated to be induced by ACTH and down-regulated by dexamethasone administration [30]. The results of our study in association with these findings could led to speculate a possible role of this oncoprotein in the pathogenesis of hormone secreting adrenal tumors. The overexpression of topoisomerase I could suggest a possible role of topoisomerase inhibitors in the treatment of ACC. In an experimental study on ACC human tumor cell line (H295), topoisomerase I inhibitor, SN38 (the active metabolite of irinotecan, CPT-11), has been demonstrated to achieve a higher IC50 (IC50 < 1 µM) than the reference drug, cisplatin [31]. However, this encouraging result was not confirmed by a clinical prospective study.
5. Conclusion
On the basis of the results of this study it is possible to conclude that gene expression profiling by cDNA macroarray analysis from paraffin embedded adrenal tumors is feasible, despite with a low success rate. The possibility to use archival specimens for the evaluation of genetic profile of adrenal tu- mors is of critical importance, because of the rarity of malig- nant adrenocortical disorders. Indeed, it gives the opportunity to examine a large collection of specimens and to correlate the molecular findings with biological behavior and clinical out-
come, basing on the availability of long follow-up. However, this is only a pilot study based on a very low number of speci- mens and mainly aimed to set up and validate the methodol- ogy. Indications here obtained will be certainly applied to lar- ger scale analyses.
Even if data presented have a limited statistical power, they are in line with recent literature on alterations of stress pro- teins and cell cycle regulators in cancer tissues, of endocrine and non-endocrine origin. The different expression in ACC and ACA of HSP-60, a stress protein reportedly involved in tumor progression, and of Ciclin D1 and jun proto-oncogene, implicated in cell cycle regulation, indicates that these genes may play a role in ACC pathogenesis and could represent po- tential diagnostic/prognostic markers. The overexpression of topoisomerase I could suggest a possible role of topoisomerase inhibitors in the treatment of ACC. Despite the analyzed series is very limited, and the results should be confirmed by larger studies and eventually validated by specific RT-PCR, Northern and Western blotting analyses, these findings have a quite re- levant potential biologic, diagnostic, prognostic and therapeutic significance. As said above, results are not meant to convey definitive information at the moment, in the absence of larger numbers of samples. Again, preliminary indications on mole- cular changes in adrenal cancer here obtained appear to be coherent with previous literature and biologically sound. Moreover, In a larger scale study result validation by RT- PCR and, where possible, by immunohistochemistry should be certainly carried out.
References
[1] Allolio B, Hahner S, Weismann D, Fassnacht M. Management of adre- nocortical carcinoma. Clin Endocrinol (Oxf) 2004;60:273-87.
[2] Dackiw APB, Lee JE, Gagel RF, Evans DB. Adrenal cortical carcinoma. World J Surg 2001;25:914-26.
[3] Weiss LM, Medeiros LJ, Vickery AL. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13: 202-6.
[4] Harrison LE, Gaudin PB, Brennan MF. Pathologic features of prognostic significance for adrenocortical carcinoma after curative resection. Arch Surg 1999;134:181-5.
[5] Giordano TJ, Thomas DG, Kuick R, Lizyness M, Misek DE, Smith AL, et al. Distinct transcriptional profiles of adrenocortical tumors uncovered by DNA microarray analysis. Am J Pathol 2003;162:521-31.
6] Haak HR, Cornelisse CJ, Hermans J, Cobben L, Fleuren GJ. Nuclear DNA content and morphological characteristics in the prognosis of adre- nocortical carcinoma. Br J Cancer 1993;68:151-5.
[7] Reincke M, Karl M, Travis VH, Mastorakos G, Allolio B, Linehan HM. P53 mutations in human adrenocortical neoplasm: immmunohistochem- ical and molecular studies. J Clin Endocrinol Metab 1994;78:790-4.
[8] Wachenfeld C, Beushlein F, Zwermann O, Zwermann O, Mora P, Fas- snacht M, et al. Discerning malignancy in adrenocortical tumors: are mo- lecular markers useful? Eur J Endocrinol 2001;145:335-41.
[9] Goldblum JR, Shannon R, Kaldjan EP, Thiny M, Davenport R, Thomp- son N, et al. Immunohistochemical assessment of proliferative activity in adrenocortical neoplasms. Mod Pathol 1993;6:663-8.
[10] Iino K, Sasano H, Yaboki N, Oki Y, Kikuchi A, Yoshimi T, 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;10:901-7.
[11] Stojadinovic A, Ghossein RA, Hoos A, Nissan A, Marshall D, Dudas ME, et al. Adrenocortical carcinoma: clinical, morphologic and molecular characterization. J Clin Oncol 2002;20:941-50.
[12] Stojadinovic A, Brennan M, Hoos A, Omeroglu A, Leung DHY, Dudas ME, et al. Adrenocortical adenoma and carcinoma: histopatholo- gical and molecular comparative analysis. Mod Pathol 2003;16:742-51.
[13] Bernini GP, Moretti A, Viacava P, Bonadio AG, Iacconi P, Miccoli P, et al. Apoptosis control and proliferation marker in human normal and neoplastic tissues. Br J Cancer 2002;86:1561-5.
[14] Kanauchi H, Wada N, Clark OH, Duh QY. Apoptosis regulating genes, bel-2 and bax, and human telomerase reverse transcriptase messenger RNA expression in adrenal tumors: possible diagnostic and prognostic importance. Surgery 2002;132:1021-7.
[15] Brown PO, Botstein D. Exploring the new world of the genome with DNA microarrays. Nat Gen 1999;21:33-7.
[16] Lewis F, Maughan NJ, Smith V, Hillan K, Quirke P. Unlocking the ar- chive-gene expression in paraffin-embedded tissue. J Pathol 2001;195: 66-71.
[17] Tsunoda T, Koh Y, Koizumi F, Tsukiyama S, Ueda H, Taguchi F, et al. Differential gene expression profiles and identification of the genes rele- vant to clinicopathologic factors in colorectal cancer selected by cDNA array method in combination with principal component analysis. Intern J Oncol 2003;3:49-59.
[18] Delpuech O, Trabut JB, Carnet F, Feuilard J, Brechot C, Kremsdorf D. Identification, using cDNA macroarray analysis, of distinct gene expres- sion profiles associated with pathological and virological features of he- patocellular carcinoma. Oncogene 2002;21:2926-37.
[19] Zhao J, Roth J, Bode-Llesniewska B, Pfaltz M, Heitz PU, Komminoth P. Combined comparative genomic hybridization and genomic microarray for detection of gene amplifications in pulmonary artery intimal sarcomas and adrenocortical tumors. Genes Chrom Cancer 2002;34:48-57.
[20] Kakajima F, Nishimori H, Hata F, Yasoshima T, Nomura H, Tanaka H, et al. Gene expression screening using a cDNA macroarray to clarify the mechanism of peritoneal dissemination of pancreatic cancer. Surg Today 2003;33:190-5.
[21] Musholt PB, Imkamp F, von Wasielewski R, Schmid KW, Musholt TJ. RET rearrangements in archival oxyphilic thyroid tumors: new insights in tumorigenesis and classification of Hurtle cell carcinomas? Surgery 2003;134:881-9.
[22] Hu YC, Lam KY, Law S, Wong J, Srivastava G. Identification of differ- entially expressed genes in esophageal squamous cell carcinoma (ESCC) by cDNA expression array: overexpression of Fra-1, Neogenin, Id-1 and CDC25B genes in ESCC. Clin Cancer Res 2001;7:2213-21.
[23] Kim JO, Kim HN, Hwang MH, Shin HI, Kim SY, Park RW, et al. Dif- ferential gene expression analysis using paraffin-embedded tissues after laser microdissection. J Cell Biochem 2003;90:998-1006.
[24] Soonmyung P, Chung-yeul K, Yong-kuk S, Wan-seop K. Technology Insight: application of molecular techniques to formalin-fixed paraffin- embedded tissues from breast cancer. Nat Clin Pract Oncol 2005;2:246- 54.
[25] Pignatelli D, Ferreira J, Soares P, Costa MJ, Magalhaes MC. Immunohis- tochemical study of Heat Shock Proteins 27, 60 and 70 in normal human adrenal and in adrenal tumors with suppressed ACTH production. Mi- crosc Res Tech 2003;61:315-23.
[26] Cornford PA, Dodson AR, Parsons KF, Desmonds AD, Woolfenden A, Fordham M, et al. Heat shock protein expression independently predicts clinical outcome in prostate cancer. Cancer Res 2000;60:7099-105.
[27] Sherr CJ. Cancer cell cycles. Science 1996;274:1672-7.
[28] Swanson C. Cell-cycle targeted therapies. Lancet Oncol 2004;5:27-36.
[29] Vogt PK. Jun, the oncoprotein. Oncogene 2001;20:2365-77.
[30] Armelin HA, Lotfi CF, Lepique AP. Regulation of growth by ACTH in the Y-1 line of mouse adrenocortical cells. Endocr Res 1996;22:373-83.
[31] Baudin E, Docao C, Gicquel C, Vassal G, Bachelot A, Penfornis A, et al. Use of a topoisomerase I inhibitor (irinotecan, CPT-11) in metastatic adrenocortical carcinoma. Ann Oncol 2002;13:1806-9.