Toward a pathway-centered approach for the treatment of adrenocortical carcinoma

Kimberly J. Busseya and Michael J. Demeurea,b

ªClinical Translational Research Division, Translational Genomics Research Institute, Phoenix and Virginia G. Piper Cancer Center, Scottsdale, Arizona, USA

Correspondence to Michael J. Demeure, MD, MBA, Senior Investigator, Translational Genomics Research Institute (TGen), Virginia G. Piper Cancer Center, 10460 N 92nd St, Suite 200, Scottsdale, AZ 85258, USA

E-mail: mdemeure@tgen.org

Current Opinion in Oncology 2011, 23:34-44

Purpose of review

Adrenocortical carcinoma is an aggressive, lethal malignancy of the adrenal cortex. The rarity of the disease has stymied therapeutic development. Recent work toward understanding the molecular pathogenesis of the disease has identified several potential new diagnostic and therapeutic targets.

Recent findings

The molecular characterization of adrenocortical carcinoma has identified dysregulation of the Gap 2/mitosis transition and the insulin-like growth factor 1 receptor signaling cascade as two major pathways for therapeutic development. These studies have also highlighted an unappreciated heterogeneity of the disease at the gene level that nevertheless seems to converge onto common cellular pathways. Additionally, the characterization of Wnt signaling through ß-catenin in adrenal development, the demonstration of the involvement of BMP signaling in adrenocortical carcinoma growth regulation, and the discovery that ERCC1 expression levels can predict therapeutic response to platinum are just a few of the recent advances that promise to shed light on adrenocortical carcinoma biology.

Summary

Short-term, therapeutic development should target the Gap 2/mitosis transition and the downstream signaling of the insulin-like growth factor 1 receptor receptor. Long-term, additional characterization of patient samples, particularly at the sequence level, is required to fully understand adrenocortical carcinoma biology and apply that knowledge to clinical practice.

Keywords

adrenocortical carcinoma, cell cycle, insulin-like growth factor 2, molecular targets, therapeutic targets

Curr Opin Oncol 23:34-44 @ 2010 Wolters Kluwer Health | Lippincott Williams & Wilkins 1040-8746

Introduction

Adrenocortical carcinoma (ACC) is an aggressive, lethal malignancy of the adrenal cortex. Most patients (40- 70%) present with metastases at the time of diagnosis precluding a surgical cure [1°]. The rarity of the disease (the incidence is one to two per million) has stymied efforts to improve treatment, both by making large randomized trials difficult to conduct and by limiting the biological material available for research. To date, most of the research in the field has centered on identify- ing the molecular aberrations involved in ACC patho- genesis, with some success. The postgenomic era has seen a move toward using high-throughput, large-scale profiling techniques to describe the genomes and result- ing transcriptomes and proteomes of multiple tumor types. This review will focus on recent advances in identifying potential therapeutic targets and their appli- cation to the diagnosis and treatment of ACC.

1040-8746 @ 2010 Wolters Kluwer Health | Lippincott Williams & Wilkins

Gene expression and miRNA studies

In the last 7 years, several studies have been published detailing the gene expression profiles of both benign and malignant adrenocortical tumors. All highlight the Gap 2/ mitosis (G2/M) transition and insulin-like growth factor 2 (IGF2)/insulin-like growth factor 1 receptor (IGF-1R) dysregulation as distinguishing ACC from either normal adrenals or benign adrenocortical adenomas (ACA) [2,3°,4°,5,6°,7*,8]. Furthermore, all demonstrate that there are at least two subgroups of ACC that differ with respect to their survival based on hierarchical clustering of expression profiles [2,3°,4°,5,6°,7*,8]. This clustering is dependent upon the expression of genes annotated as G2/M transition and/or G2/M checkpoint (Fig. 1) [3°,4°]. Given that these were populations of asynchronous cells in tumor samples, it suggests that the highlighted genes have additional functions in other aspects of the cell cycle that may be important. It also suggests that compounds

DOI:10.1097/CCO.0b013e328340d879

Figure 1 Model of cell cycle progression and regulation in the presence of overexpressed Cdc2 and G2/M-associated genes

E2F

Cyclin

F21

P53

CdK2

BUBIB

MAD2L1

CCNB1

BUB1

Expressiont

CDC2

CDK6

PTTG1

SKP2

ANAPC5

CDC23

CDC25B

CdK2

CdK4

CdK5

Expression

Cyclin A

Cyclin D

Cell progression from G1-S, through S, G2-M requires the activation of different combinations of kinases and associated cyclin proteins. DNA damage leads to cycle cessation through p53 activated p21 inhibition of Cdk2/cyclin phosphorylation of Rb. In the absence of DNA damage or with dysfunctional p53/p21, Cdk2/cyclin actively phosphorylates Rb, leading to release of the active E2F transcription factor. E2F initiates transcription of G2/M-specific genes. The upregulation of G2/M-associ- ated genes specifically, rather than the entire cell cycle progression program, suggests that the proteins involved in this transition may be particularly important for ACC, making ACC vulnerable to agents that target these proteins.

that target, either directly or indirectly, proteins involved in the G2/M transition may be effective against ACC. No single gene expression signature has emerged to reliably discriminate either ACC from ACA or one group of ACC from another. Differences in platform, experimental design, and analysis methods may be partially responsible. This does not imply the data are, or analysis is, bad or unreliable. Rather, it simply confirms the difficulty in interpreting data across studies. Perhaps more importantly, the differing results at the gene level, but not at the pathway level, may indicate that the genomes and tran- scriptomes of ACC are more heterogeneous than pre- viously appreciated. Moving forward, we should focus on targeting pathways, as opposed to individual genes, as we develop new treatment approaches. Data from our laboratory suggest this is a viable pursuit. In our in-vitro experiments, both ACC cell lines, SW-13 and H295R, are sensitive to treatments that target the G2/M transition [9°].

Explorations of miRNA expression are still few but are yielding a similar picture of heterogeneity. miRNA tends

to be underexpressed in all ACC regardless of age of presentation [10°-12°]. miR-503 is overexpressed in ACC relative to ACA in all studies to date. However, it is the only miRNA alteration that has replicated across studies [10°-12°]. Predicted targets of miR-503 included components of both the mitogen-activated protein (MAP) kinase and phosphoinositide-3 kinase (PI3K) signaling cascades as well as some T-cell factor (TCF) transcription factor family members and several proteins involved in cell cycle control.

Established targets

Certain genes have been implicated in ACC pathogenesis through the association of an increased risk of ACC with a given genetic syndrome. These genes are by far the best characterized in terms of their role in ACC development and represent the immediate focus for new therapeutic development.

Igf-2 and Igf-1 receptor

The association of ACC with Beckwith-Wiedemann syndrome (BWS) has generated an interest in the role of IGF2 overexpression in ACC pathogenesis. BWS is an overgrowth syndrome that arises from the misexpression of genes from an imprinted domain at 11p15. Two regions in the imprinted domain have been implicated in the phenotypic variation of the disease, particularly tumor spectrum. The telomeric region encompasses paternally expressed IGF2 and maternally expressed H19. Abnorm- alities that lead to the lack of H19 expression and over- expression of IGF2, such as paternal uniparental disomy or aberrant methylation of H19, give rise to BWS with a preponderance of Wilms tumor [13]. It is the absence of H19 expression that is thought to be responsible for tumor development as studies have shown that over- expression of IGF2 alone fails to result in tumor for- mation in mice [1°]. However, IGF2 is a potent tumor- promoting factor. The centromeric region includes the LIT1 transcript of KvLQT1 encoded by KCNQ1 and CDKN1C, the gene that encodes p57kip. Hypomethyla- tion of LIT1 or mutations in CDKN1C have been implicated in BWS with a tumor spectrum that is skewed toward embryonal tumors such as rhabdomyosarcoma, hepatoblastoma, gondadoblastoma, and ACC [13]. p57kip is a known negative regulator of cell cycle progression, and as such is a prime candidate for a tumor suppressor in this region.

Approximately 90% of ACC demonstrate overexpression of IGF2 [14]. In microarray studies for gene expression, IGF2 and associated binding proteins are consistently identified as differentially expressed in ACC as compared to normal adrenal tissue, with most studies identifying IGF2 as upregulated [2,3°,4°,5,6°,7*,8]. Because of this, the Igf-1 receptor has become a therapeutic target.

Figure 2 IGF-1R signaling pathway with therapeutic targets highlighted

IBP

IGF-1

IGF-1 receptor

IGF-2

GRB2

Long-term depression in cerebellum

14-3-3 epsilon

14-3-3 beta/alpha

Sho

ASK1 (MAP3K5)

+ F

c-Raf-1

+ P

Inhibition of ASK- induced apoptosis

14-3-3 zeta/delta

SOS

PI3K reg class IA

IRS-1

BAD

Bcl-XL

2.7.1.137

+ P

Survival

PI3K cat class IA

FOXO3A

Caspase-9

Ptdins(4,5)P2

+ P

AKT (PKB)

H-Ras

- P

3.1.3.67

PTEN

Ptdins(3,4,5)P3

GSK3 alpha/beta

Cyclin D

+ F

c-Raf-1

p90Rsk

PDK (PDPK1)

Tuberin

GYS1

c-Myc

Cell cycle regulation

1-kb

Glycogen

MEK1 (MAP2K1)

MEK2 (MAP2K2)

IKK (cat)

IKK-alpha

RHEB2

Hamartin

RPS6

Erk (MAPK1/3)

CREB1

NF-KB

p70 S6 kinase1

mTOR

4E-BP1

2.4.1.11

Elk-1

Cell proliferation

Translation

UDP-D-glucose

In ACC, signaling through IGF-1R occurs via IGF2 overexpression. This results in a strong proliferative signal through both PI3K and MAP-kinase cascades. In addition, there is a strong antiapoptotic signal transduced through both PI3K and 14-3-3. This figure was generated in GeneGo using the show therapeutic targets visualization on the ‘Development_IGF-1 Receptor Signaling’ map (GeneGo, Inc.).

Recent work demonstrates that antibody or small-mole- cule inhibition of the Igf-1 receptor results in a reduction in viability of adrenocortical cancer cells in vitro and in vivo [15 ** ]. In addition, modulation of the Igf-2/Igf-1 receptor signaling pathway by targeting downstream effector molecules, such as PI3K and mammalian target of rapamycin (mTOR), are being explored. Figure 2 shows the potential therapeutic targets in the Igf-1 receptor signaling cascade. One study has shown that the use of the peroxisome proliferator-activator receptor gamma (PPAR-y) agonist, rosiglitazone, resulted in a loss of Akt phosphorylation [16]. In addition, downregulation of the Igf signaling system through the induction of PPAR-y has been shown to inhibit ACC cell growth in vivo [17 ** ]. Doghman et al. [10°] have shown that the mTOR inhibitor everolimus reduces the growth of both

H295R, which is dependent on Igf-2, as well as Igf-2- independent SW-13, in xenograft models. Currently, there are three clinical trials investigating the effective- ness of inhibitors of Igf-1 receptor alone or in combi- nation with other agents (see Table 1).

p53

ACC is one of the tumor types associated with Li- Fraumeni syndrome [18]. This familial cancer syndrome is most often a result of heterozygous germline mutations in the tumor suppressor p53. Mutations in families with ACC lie outside of the traditional hot spots of the DNA binding domain, instead residing in the loops opposite the DNA binding domain or in tetramerization domain [19]. This distribution is most evident in southern Brazil where the R337H germline mutation results not in overt

Table 1 Current clinical trials enrolling patients with ACC
NCT ID	Title	Interventions	Mechanism of action	Sponsors	Phases	Enrollment	Funded by	Study types	Outcome measures	URL
NCT00324012	Trial with Taxotere and	Drug: cisplatin, taxotere	DNA alkylating agent, tubulin binding agent	Rigshospitalet, Denmark	Phase II	39	Other	Interventional	Response rate: Survival, time to	http://ClinicalTrials. gov/show/
	Cisplatin in Nonoperable								Response rate: Survival, time to	http://ClinicalTrials. gov/show/
	Adrenocortical Carcinoma								progression, best overall	NCT00324012
NCT00453674 NCT00924989 NCT00848016 NCT00778817	German Adrenocortical Carcinoma Registry Study of OSI-906 in Patients With Locally Advanced or Metastatic Adrenocortical Carcinoma Gossypol Acetic Acid in Treating Patients With Recurrent, Metastatic, or Primary Adrenocortical Cancer that Cannot be Removed by Surgery Mitotane with or without IMC-A12 in Treating Patients with Recurrent, Metastatic, or Primary Adrenocortical Cancer that Cannot be Removed by Surgery	Drug: OSI-906; other: placebo Drug: R-(-)- gossypol acetic acid Biological: cixutumumab; drug: mitotane	None IGF-1R inhibitor	University of Wuerzburg/ Deutsche Krebshilfe e.V., Bonn (Germany) OSI Pharmaceuticals	Phase III	1000 135	Other Industry	Observational Interventional	response rate, and duration of response Overall survival of single agent OSI-906 vs. placebo;	http://ClinicalTrials. gov/show/ NCT00453674 http://ClinicalTrials. gov/show/ NCT00924989
NCT00453674 NCT00924989 NCT00848016 NCT00778817			pan-Bcl2 inhibitor Anti-IGF-1R antibody, adrenolytic	Mayo Clinic/National Cancer Institute (NCI) University of Chicago/NCI	Phase II Phase II	44 122	Other; NIH Other; NIH	Interventional Interventional	Progression-free survival; Disease control rate; Best overall response rate; Duration of response; Quality of life; Safety profile; Pharmacokinetics, Pharmacodynamics Proportion of patients who achieve a confirmed objective response (complete or partial response by RECIST criteria) to treatment; Toxicity as assessed by NCI CTCAE v3.0; Time to progression and overall survival; Time to treatment failure and time to subsequent therapy; Duration of response Progression-free survival; Objective response rates; Changes in tumor size over time; Overall survival	http://ClinicalTrials. gov/show/ NCT00848016 http://ClinicalTrials. gov/show/ NCT00778817
NCT00777244	Efficacy of Adjuvant Mitotane Treatment (ADIUVO)	Drug: mitotane	Adrenolytic	University of Turin, Italy	Phase III	200	Other	Interventional	Disease-free survival; Overall survival	http://ClinicalTrials. gov/show/ NCT00777244
NCT00568139	Evaluation of Side Effects of Mitotane			University of Wuerzburg		200	Other	Observational	Documentation of side effects	http://ClinicalTrials. gov/show/ NCT00568139

(continued overleaf)

Table 1 (continued)
NCT ID	Title	Interventions	Mechanism of action	Sponsors	Phases	Enrollment	Funded by	Study types	Outcome measures	URL
NCT00304070	Cisplatin-Based	Biological: filgrastim;	Granulocytic-colony	Children's	Phase III	79	Network; NIH	Interventional	Primary tumor	http://ClinicalTrials.
	Chemotherapy and/or	biological: pegfilgrastim;	stimulating factor	Oncology					measurement;	gov/show/
	Surgery in Treating	drug: cisplatin; drug:	analog, DNA	Group/NCI					Measurable	NCT00304070
	Young Patients	doxorubicin hydrochloride;	akylating agent,						metastatic disease;
	with Adrenocortical	drug: etoposide; drug:	topoisomerase II						Response of
	Tumor	mitotane; procedure:	inhibitor, adrenolytic						metastatic target
		conventional surgery							lesions;
									Nonmeasurable metastatic disease;
NCT00678223	Monoclonal Antibody	Biological: cixutumumab;	Anti-IGF-1R antibody,	M.D. Anderson	Phase I	99	Other; NIH	Interventional	Overall response Maximum tolerated	http://ClinicalTrials.
	IMC-A12 and	drug: temsirolimus	mTOR inhibitor	Cancer					dose and dose-	gov/show/
	Temsirolimus in			Center \|NCI					limiting toxicity of	NCT00678223
	Treating Patients								anti-IGF-1R
	with Locally								recombinant
	Advanced or								monoclonal
	Metastatic								antibody IMC-A12
	Cancer								and temsirolimus;
									Change in phosphorylation levels of AKT and
									other biomarkers
									(i.e., IGF-1R,
									pIGF-1R, IRS-1, PTEN, VEGFR-1,
									VEGFR-2, and
									CD31) before and after treatment;
									Tumor metabolism
									as assessed by
									PET scan before
									and after treatment;
									Tumor response
									rate (complete
									response and partial response)
NCT00003453	Antineoplaston Therapy	Drug: antineoplaston A10;	DNA intercalation,	Burzynski	Phase II	40	Other	Interventional	Response rate	http://ClinicalTrials.
NCT00003453	in Treating Patients	drug: antineoplaston	inhibition of	Research					based on tumor	gov/show/
	with Stage IV Adrenal	AS2-1	L-gluatmine	Institute					measurements	NCT00003453
	Gland Cancer		incorporation						at 12 weeks;
									Survival at 1, 2, and 5 years from
									the start of
									treatment

NCT00786110	Sorafenib Plus Paclitaxel in Adreno-Cortical- Cancer Patients	Drug: sorafenib; drug: paclitaxel	RAF inhibitor, tubulin binding antimitotic	University of Turin, Italy	Phase II	30	Other	Interventional	Disease-free survival; Overall survival	http://ClinicalTrials. gov/show/ NCT00786110
NCT00831844 NCT01048892	Cixutumumab in Treating	Biological: cixutumumab; other: laboratory	Anti-IGF-1R antibody	Children's	Phase II	140	Network; NIH	Interventional	Response rate;	http://ClinicalTrials.
	Patients with	Biological: cixutumumab; other: laboratory	Anti-IGF-1R antibody	Oncology	Phase II	140		Interventional	Toxicity; Relationship	gov/show/
	Relapsed or Refractory Solid Tumors Seneca Valley Virus-001 in Treating Young	biomarker analysis Biological: Seneca Valley virus-001; other: laboratory biomarker analysis; other: pharmacological study	Oncolytic virus selectively targeting	Group/NCI Children's Oncology	Phase I	34	Network; NIH	Interventional	between tumor expression of IGF-I, IGF-II, and IGF-IR and response; Human antihuman antibody response; Effect of cixutumumab on circulating levels of proteins involved in linear growth and glucose homeostasis, including IGF-I, IGF-II, IGF-BP3, growth hormone, insulin, and C-peptide Safety and tolerability;	NCT00831844 http://ClinicalTrials. gov/show/
	Patients with Relapsed or Refractory		tumors with neuroendocrine features	Group/NCI					Recommended phase II dose	NCT01048892
	Neuroblastoma,								of Seneca Valley virus-001 (NTX-010); Tumor response; Viral titers in blood and stool; Development of antibodies to NTX-010
	Rhabdomyosarcoma, or Rare Tumors
	with Neuroendocrine Features
	with Neuroendocrine Features
NCT00669266	Adrenal Tumors:		None	University of		500	Other Other_Gov	Observational		http://ClinicalTrials. gov/show/
	Pathogenesis and		None	Wuerzburg			Other Other_Gov			http://ClinicalTrials. gov/show/
	Therapy			Deutsche Krebshilfe e.V.,						NCT00669266
				Bonn (Germany); German Federal
				Bonn (Germany); German Federal
				Ministry of
NCT00538850	Fentanyl Sublingual Spray in Treating Patients with Breakthrough Cancer Pain	Drug: fentanyl sublingual spray; other: questionnaire administration	Analgesic	Education and Research Insys Therapeutics, Inc.	Phase III	130	Industry	Interventional	Pain relief by 30 min after dosing; Pain relief at various time points; Safety,	http://ClinicalTrials. gov/show/ NCT00538850
NCT00538850			Analgesic	Education and Research Insys Therapeutics, Inc.	Phase III	130	Industry	Interventional	tolerability, and acceptability	http://ClinicalTrials. gov/show/ NCT00538850
NCT00005927	Study of Adrenal Gland Tumors		None	Eunice Kennedy Shriver National Institute of		2000	NIH	Observational		http://ClinicalTrials. gov/show/ NCT00005927
NCT00005927	Study of Adrenal Gland Tumors		None	Child Health and Human Development (NICHD)		2000	NIH	Observational		http://ClinicalTrials. gov/show/ NCT00005927

(continued overleaf)

Table 1 (continued)
NCT ID	Title	Interventions	Mechanism of action	Sponsors	Phases	Enrollment	Funded by	Study types	Outcome measures	URL
NCT00637637	External-Beam Radiation Therapy with or without Indinavir and Ritonavir in Treating Patients with Brain Metastases	Drug: indinavir sulfate; drug: ritonavir; radiation: radiation therapy	Antiviral protease inhibitor	Oncology Institute of Southern Switzerland	Phase II	60	Other	Interventional	Time to treatment failure in the brain (TTF) as determined by the radiological response rate; Overall survival (OS); Radiological volumetric response to treatment; Local intracranial disease progression at 4 months; Progression-free survival at 6 months; Improvement of symptoms; Time to symptom relapse or symptom progression; Duration of use of steroids; Duration of use of anticonvulsive	http://ClinicalTrials. gov/show/ NCT00637637
NCT00637637			Antiviral protease inhibitor						drugs	http://ClinicalTrials. gov/show/ NCT00637637

Li-Fraumeni, but a 1 in 10 chance of developing ACC [20,21]. In sporadic ACC, the actual rate of p53 mutation remains unclear, with evidence for mutation present in 20-70% of tumors, depending on the study [14]. How- ever, even if p53 is not mutated, the behavior of ACC suggests that the pathway is underactive or inactive. Loss of heterozygosity has been demonstrated in the region

[22], and recent work suggests that p53 activity may be compromised by haploinsufficiency [23]. Differences in p53 activity based on single nucleotide polymorphisms (SNPs) may also be important. One such SNP, R72P, has been demonstrated to affect the ability of p53 to induce apoptosis [24°]. The proline allele, which is less active, has been associated with the risk of developing ACC [25].

Emerging therapeutic leads

Work from other tumor types and developmental biology, combined with the results of genome and transcriptome profiling in ACC, has generated interest in examining in depth a wide variety of targets. Each potential target is usually implicated by one or two studies, but all represent intriguing preliminary data suggesting at least some potential for translational development.

B-Catenin, Wnt signaling, and adrenal development

B-Catenin is one component of adherins junctions and plays a role in the growth and adherence of cells. It also anchors the actin cytoskeleton and may be involved in contact inhibition. ß-Catenin is involved in Wnt signal- ing. Glycogen synthase kinase-3 beta (GSK-3B) phos- phorylates ß-catenin at Ser45 while it is bound to axin and adenomatous polyposis coli protein (APC). This phos- phorylation results in the degradation of ß-catenin. Wnt signaling results in the inactivation of GSK-3B. This stabilizes ß-catenin in cytoplasm where it is free to translocate to the nucleus and bind with TCF/LEF family members, activating the Wnt transcriptional program.

In ACC, 11 of 13 (77%) tumors show diffuse cytoplasmic and nuclear staining of ß-catenin, suggesting activation of B-catenin’s transcriptional functions [26]. Of these tumors, four of 13 had mutation altering exon 3 leading to constitutive activation of ß-catenin. The ACC cell line, NCI-H295R, has activated ß-catenin due to a hetero- zygous mutation, S45P [27].

The understanding of ß-catenin’s role in adrenal devel- opment and tumorigenesis has advanced with the creation of a transgenic mouse with adrenal-specific constitutive expression of ß-catenin [28 ** ]. Such mice demonstrate cortical hyperplasia with ectopic expansion of zona glomerulosa differentiation at the expense of the zona fasciculata. There is also ectopic expansion of spongiocytes and subcapsular cell populations resulting

in dysplasia [28 ** ]. Tumors form over 17-month time course. The tumors are highly proliferative, but cells lack B-catenin accumulation or SF-1 expression, suggesting an indirect effect on tumor formation that may include the requirement for additional genomic alterations. It further suggests that the tumors potentially arise from adrenal progenitors [28 ** ]. The authors did not see upregulation of Igf-2 expression, raising the question of whether Igf-2 overexpression is specific to human ACC. However, the tumors were low grade. Human expression data suggest that prior to overexpression of Igf-2, ACC loses expres- sion of NOV [3°,4°,29°]. This finding was recapitulated in the mice [28 ** ].

To complicate matters, recent work demonstrates that telomerase colocalizes via interaction with BRG1 to Wnt-responsive promoters, suggesting a potential link between telomere dysfunction and Wnt-ß-catenin sig- naling [30 ** ]. Most ACC exhibit telomere elongation through activation of TERT [31], the enzymatic com- ponent of telomerase. The acd mouse, which has a defect in telomere function due to a mutation in acd/Tpp1, part of the telosome/shelterin complex, exhibits adrenal agen- esis in a wild-type p53 background. When the acd mutation is crossed into p53-knockout mice, there is a 5% incidence of ACC [32 ** ]. How telomere biology, Wnt signaling through ß-catenin, and p53 interact to promote ACC remains to be elucidated.

Bone morphogenetic protein signaling

Bone morphogenetic proteins (BMPs) are cytokines belonging to the transforming growth factor beta (TGF-ß) superfamily. BMPs have been implicated in a wide variety of cellular processes including organ devel- opment, lineage specification, tumorigenesis, and the regulation of steroidogenesis in the adrenal cortex [33°]. In ACC, BMP2 and BMP5 are downregulated when compared to normal adrenal glands, while the receptor levels remain unchanged. This was seen at both the mRNA and the protein levels [34°]. Downstream activation of SMAD 1/5/8, part of the TGF-ß signaling cascade, was also absent in ACC when compared to normal adrenal gland [34°]. When the ACC cell lines SW-13 and NCI-H295R or primary ACC cell cultures were exposed to either BMP2 or BMP5, growth inhi- bition was observed and was accompanied by down- stream activation of SMAD signaling [34°]. This inhi- bition antagonizes the potent growth stimulation signal of Igf-1. Because the loss of BMP2 and BMP5 expression could be traced to a reduction in transcripts, Johnsen et al. [34°] looked at the transcriptional regulation of both genes in ACC. Treatment of cells with retinoid acid or phorbol ester or overexpression of GATA6 results in increased BMP2 or BMP5 expression. Although the promoter of BMP2 did not show an increase in DNA methylation in ACC samples, treatment of NCI-H295R

with the demethylation agent 5-aza-2’deoxycytidine induced BMP2 expression [34°]. This compound is avail- able clinically, and this study supports the investigation of such therapy in ACC.

Epidermal growth factor receptor and RAS-RAF-MEK signaling

The role of the epidermal growth factor (EGF) signaling cascade has been documented in numerous tumor types, including breast, lung, and pancreatic tumors. Several therapies that target multiple points along the signaling cascade are in clinical use or in clinical trial. A recent study by Kotoula et al. [35°] looked at the mutation frequencies of epidermal growth factor receptor (EGFR) and com- ponents immediately downstream of the receptor in ACC. In a series of 35 ACC, they identified mutations in the pathway in seven samples (20%). These included two samples with mutations in BRAF, four samples with mutations in EGFR, one of which came from a sample with a BRAF mutation, one sample with a mutations KRAS, and one sample with two different mutations in NRAS [35°]. Neither SW-13 nor NCI-H295 (the parental line to NCI- H295R) harbors mutations in the genes studied [27]. Immunohistological staining for the phosphorylated forms of mitogen-activated protein kinase (ERK) and dual-speci- ficity mitogen-activated protein kinase kinase (MEK) demonstrated an increase in downstream activation of the EGFR pathway in those samples with mutations [35°]. Although useful, this study may underestimate the number of mutations in this pathway. The authors only sequenced regions identified as mutation hot spots in other tumor types. Recent work in colorectal carcinoma examining the incidence of KRAS mutations suggests that mutational hot spots only identify two-thirds of the tumors harboring mutations [36°]. The mutational status of the various members of the EGFR pathway impacts the response to treatment with EGFR inhibitors. Mutant EGFR confers a better response to inhibitors, while mutations in KRAS yield resistance [37°-39°]. Given the demonstrated clinical utility of agents targeting the EGFR signaling pathway, the results of this study support targeted sequencing of the genes in their entirety in a larger set of ACC samples to clarify both the incidence and the mutational spectrum.

Paracrine/autocrine systems

The identification of paracrine and/or autocrine factors that promote ACC tumorigenesis represents another potential area for diagnostic and/or therapeutic develop- ment. Two secreted proteins, parathyroid hormone- related protein (PTHrP) and osteopontin, have been examined in adrenocortical tumors [40,41°]. PTHrP mRNA in adrenocortical tumors was increased relative to ACA and correlated with markers of malignancy including stage, Weiss score (a histologic grading system specific to adrenal cortical tumors), and the presence of

metastases. Additionally, PTHrP can stimulate prolifer- ation of H295R that can be blocked with either antibody or PTHrP antagonist. This leads to accumulation of cells in S and G2/M followed by increase in apoptosis [40]. Osteo- pontin stimulates invasion but not proliferation in NCI- H295 cells, and integrin «VB3 enhances osteopontin’s effects [41°]. It is increased in expression by both IHC and western blotting in ACC as compared to normal adrenal, but overexpression is not associated with survival, making it a candidate marker for the diagnosis of malig- nancy [41°].

Another promising marker for diagnosis is the glucocorti- coid receptor. In a study looking at two cohorts of adreno- cortical tumors, nuclear staining for glucocorticoid receptor was seen in 78-94% of ACC, but was absent in 94-98% of ACA [42 ** ]. This suggests that glucocorticoid receptor may be an excellent diagnostic marker of malignancy. The therapeutic potential of this dichotomy remains to be explored. Finally, two studies have looked at the role of estrogen receptor (ER) in ACC. In a study by Shen et al. [43°], eight of 17 ACC tumor samples were positive for nuclear ER staining. ER staining was correlated with low stage and associated with poor prognosis in a univariate model [43°], but was not an independent predictor in a multivariate model. Barzon et al. [44] have demonstrated that relative to normal adrenal cortex and adrenal adeno- mas, ACC samples demonstrate a reduction in ERß levels with an increase in ER& and aromatase levels. Taken together, these studies suggest that antiestrogens and aromatase inhibitors warrant further study in the treatment of ACC.

Excision repair cross-complementing 1

Gene expression studies have shown that genes involved in various aspects of DNA repair are overexpressed in patient samples [2,3°,4°,5,6°,7*,8]. Platinum containing therapies, especially the Italian regimen, are more effec- tive than mitotane alone, but still have response rates of less than 50%. In other tumor types, the expression of excision repair cross-complementing 1 (ERCC1) inversely correlates with response to platinum. Ronchi et al. [45 ** ] looked at the expression of ERCC1 and response to platinum in ACC. They demonstrate that ERCC1 levels were strongly inversely correlated to overall survival after platinum-based therapy. Furthermore, high levels of ERCC1 were independently predictive of poor prognosis in platinum-treated patients [45 ** ]. These results suggest that ERCC1 expression in the tumor should be evaluated prior to the inclusion of platinum in the treatment of ACC.

Epigenetic alterations and DNA methylation

Many tumor types show alterations in DNA methylation that lead to changes in gene expression, some of which are likely pathogenic. The ability to restore the expres- sion of silenced genes has proved to be a useful treatment

in tumor types such as myelodysplastic syndromes and acute myeloid leukemia. Although the role of epigenetic perturbations remains poorly defined in clinical ACC samples, work in vitro using ACC cell lines suggests that such studies should be considered. As mentioned above, the use of the demethylating agent, 5-aza-2’deoxycytidine, restores the expression of BMP-2 [34°], which the same study showed to be a negative regulator of ACC prolifer- ation. Another demethylation agent, decitibine, reduces the proliferation and migration of H295R at low doses [46°]. Such studies point to the need to define the epige- netic profiles of ACC patient samples in order to determine whether such treatments will be worthwhile.

Conclusion

The rarity of ACC has adversely impacted the pace of therapeutic development for the disease. Recent works into the molecular alterations that characterize ACC have identified a previously unappreciated heterogeneity of alterations that nevertheless appear to converge on specific pathways such as the G2/M transition and signaling by Igf-2 through Igf-1 receptor. Currently, these two path- ways represent the best targets for development in part because there are several agents already approved for clinical use or in clinical trials that target these pathways. Additional pathways, such as the role of ß-catenin in ACC, require further exploration.

In the long term, we need to focus our efforts on the comprehensive identification of the molecular alterations underlying ACC as well as the development of both in- vitro and in-vivo models of the disease. With the dramatic reduction in the costs of whole-genome sequencing, it will soon be feasible to consider sequencing the genomes of ACC patient samples. Such a project would yield a wealth of information with the potential to answer the following questions: Are there specific recurrent genomic aberrations in ACC and, if so, what are they? Are benign adrenal adenomas and ACC a continuum of disease development or do they represent two different entities? If they are two different entities, what alterations represent targets for further diagnostic development? If they are a continuum, what early events are indicative of benign disease and what further events create malignancy? What does the mutation spectrum of ACC suggest about types of exposure leading to DNA damage as well as the efficiency of DNA repair pathways? Does the mutational spectrum suggest avenues for prevention or therapeutic intervention? Given the rapid expansion of ‘targeted’ cancer therapies, what currently available therapeutic agents make the most sense as next steps in ACC treatment based on the genomic and tran- scriptomic profiles? How does ACC compare to more common cancers in terms of the genes and/or the pathways altered? What does this suggest about why rare tumors are rare? The answers to these questions will affect how

adrenal tumors are managed clinically, have the potential to dramatically improve the outcome for patients with ACC, and contribute to our understanding of cancer biology in general.

Acknowledgements

The authors have no conflicts of interest to disclose. This work was supported by the Advancing Treatments for Adrenocortical Carcinoma Research Fund.

References and recommended reading

Papers of particular interest, published within the annual period of review, have been highlighted as:

· of special interest

·· of outstanding interest

1 Bussey KJ, Demeure MJ. Genomic and expression profiling of adrenocortical

· carcinoma: application to diagnosis, prognosis and treatment. Future Oncol 2009; 5:641-655.

This review of ACC highlights the recent findings from both genomic surveys and gene expression profiling and how such results might be applied to the diagnosis, prognosis, and treatment of ACC.

2 de Fraipont F, El Atifi M, Cherradi N, et al. Gene expression profiling of human adrenocortical tumors using complementary deoxyribonucleic acid microar- rays identifies several candidate genes as markers of malignancy. J Clin Endocrinol Metab 2005; 90:1819-1829.

3 de Reynies A, Assie 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- 1115.

This is one of four recent papers describing the gene expression profiling of ACC. The authors were able to derive two signatures, one to distinguish ACC from ACA, and one to distinguish subgroups of ACC based on survival.

4 Giordano TJ, Kuick R, Else T, et al. Molecular classification and prognostica- · tion of adrenocortical tumors by transcriptome profiling. Clin Cancer Res 2009; 15:668-676.

This is one of four recent papers describing the gene expression profiling of ACC. It convincingly identifies two subclasses of ACC with differential expression and defined by the expression of genes annotated as involved in the G2/M phase of the cell cycle.

5 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.

6 Laurell C, Velazquez-Fernandez D, Lindsten K, et al. Transcriptional profiling

· enables molecular classification of adrenocortical tumours. Eur J Endocrinol 2009; 161:141-152.

This is one of four recent papers describing the gene expression profiling of ACC.

7 Soon PS, Gill AJ, Benn DE, et al. Microarray gene expression and immuno-

. histochemistry analyses of adrenocortical tumors identify IGF2 and Ki-67 as useful in differentiating carcinomas from adenomas. Endocr Relat Cancer 2009; 16:573-583.

This is one of four recent papers describing the gene expression profiling of ACC.

8 Velazquez-Fernandez D, Laurell C, Geli J, et al. Expression profiling of adrenocortical neoplasms suggests a molecular signature of malignancy. Surgery 2005; 138:1087-1094.

9 Dastrup E, Demeure MJ, Bussey KJ. Adrenocortical carcinoma cell lines are

· sensitive to compounds targeting the G2/M transition [abstract]. In: Proceed- ings of the 101st Annual Meeting of the American Association for Cancer Research; 17-21 April 2010; Washington, DC. Philadelphia, PA: AACR; 2010. Abstract 3672.

This work provides support for the feasibility of targeting proteins involved in G2/M in the treatment of ACC.

10 Doghman M, El Wakil A, Cardinaud B, et al. Regulation of insulin-like growth · factor-mammalian target of rapamycin signaling by microRNA in childhood adrenocortical tumors. Cancer Res 2010; 70:4666-4675.

This is one of three recent reports looking at the differential expression of micro- RNA in ACC. This report focuses on pediatric ACC.

11 Soon PS, Tacon LJ, Gill AJ, et al. miR-195 and miR-483-5p identified as

· predictors of poor prognosis in adrenocortical cancer. Clin Cancer Res 2009; 15:7684-7692.

This is one of three recent reports examining the differential expression of micro- RNA in ACC.

12 Tombol Z, Szabo PM, Molnar V, et al. Integrative molecular bioinformatics · study of human adrenocortical tumors: microRNA, tissue-specific target prediction, and pathway analysis. Endocr Relat Cancer 2009; 16:895-906. This is one of three recent papers describing the differential expression of micro- RNA in ACC.

13 Weksberg R, Nishikawa J, Caluseriu O, et al. Tumor development in the Beckwith-Wiedemann syndrome is associated with a variety of constitutional molecular 11p15 alterations including imprinting defects of KCNQ1OT1. Hum Mol Genet 2001; 10:2989-3000.

14 Libe R, Fratticci A, Bertherat J. Adrenocortical cancer: pathophysiology and clinical management. Endocr Relat Cancer 2007; 14:13-28.

15 Barlaskar FM, Spalding AC, Heaton JH, et al. Preclinical targeting of the type I .. insulin-like growth factor receptor in adrenocortical carcinoma. J Clin Endocrinol Metab 2009; 94:204-212.

This paper describes the preclinical work leading to the clinical trials of IMC-A12 (cituxumumab) in ACC.

16 Betz MJ, Shapiro I, Fassnacht M, et al. Peroxisome proliferator-activated receptor-gamma agonists suppress adrenocortical tumor cell proliferation and induce differentiation. J Clin Endocrinol Metab 2005; 90:3886-3896.

17 Luconi M, Mangoni M, Gelmini S, et al. Rosiglitazone impairs proliferation of .. human adrenocortical cancer: preclinical study in a xenograft mouse model. Endocr Relat Cancer 2010; 17:169-177.

This paper describes the downstream targeting of the Igf1-receptor via inhibition of PPAR-y in ACC. This inhibition hits both PI3K and ERK signaling, effectively inhibiting two of the three signaling cascades triggered by IGF2 overexpression.

18 Birch JM, Alston RD, McNally RJ, et al. Relative frequency and morphology of cancers in carriers of germline TP53 mutations. Oncogene 2001; 20:4621 - 4628.

19 Olivier M, Goldgar DE, Sodha N, et al. Li-Fraumeni and related syndromes: correlation between tumor type, family structure, and TP53 genotype. Cancer Res 2003; 63:6643-6650.

20 DiGiammarino EL, Lee AS, Cadwell C, et al. A novel mechanism of tumor- igenesis involving pH-dependent destabilization of a mutant p53 tetramer. Nat Struct Biol 2002; 9:12-16.

21 Figueiredo BC, Sandrini R, Zambetti GP, et al. Penetrance of adrenocortical tumours associated with the germline TP53 R337H mutation. J Med Genet 2006; 43:91-96.

22 Libe R, Groussin L, Tissier F, et al. Somatic TP53 mutations are relatively rare among adrenocortical cancers with the frequent 17p13 loss of heterozyg- osity. Clin Cancer Res 2007; 13:844-850.

23 Lynch CJ, Milner J. Loss of one p53 allele results in four-fold reduction of p53 mRNA and protein: a basis for p53 haplo-insufficiency. Oncogene 2006; 25:3463-3470.

24 Whibley C, Pharoah PD, Hollstein M. p53 polymorphisms: cancer implica- · tions. Nat Rev Cancer 2009; 9:95-107.

This is an excellent review of the role of p53 polymorphisms in cancer. As the role of p53 in ACC continues to be defined, the influence of polymorphisms on p53 function will need to be taken into account.

25 Ignaszak-Szczepaniak M, Horst-Sikorska W, Sawicka J, et al. The TP53 codon 72 polymorphism and predisposition to adrenocortical cancer in Polish patients. Oncol Rep 2006; 16:65-71.

26 Tissier F, Cavard C, Groussin L, et al. Mutations of ß-catenin in adrenocortical tumors: activation of the Wnt signaling pathway is a frequent event in both benign and malignant adrenocortical tumors. Cancer Res 2005; 65:7622-7627.

27 Forbes SA, Bhamra G, Bamford S, et al .: The Catalogue of Somatic Mutations in Cancer (COSMIC). Curr Protoc Hum Genet 2008; 10:11.

28 Berthon A, Sahut-Barnola I, Lambert-Langlais S, et al. Constitutive ß-catenin

·· activation induces adrenal hyperplasia and promotes adrenal cancer devel- opment. Hum Mol Genet 2010; 19:1561-1576.

This is the first description of adrenal development in the context of constitutive expression of ß-catenin in mice. The mice develop adrenocortical carcinoma after a long latency, but the tumors do not express B-catenin or SF-1. The long latency suggests that a second hit is necessary, and the lack of B-catenin or SF-1 expression may imply that the tumors are arising from a progenitor cell.

29 Szabo PM, Tamasi V, Molnar V, et al. Meta-analysis of adrenocortical tumour

· genomics data: novel pathogenic pathways revealed. Oncogene 2010; 29: 3163-3172.

This paper conducts a meta-analysis of the available gene expression sets and highlights pathways that have been previously unappreciated and/or unreported in the gene expression data.

30 Park JI, Venteicher AS, Hong JY, et al. Telomerase modulates Wnt signalling .. by association with target gene chromatin. Nature 2009; 460:66-72. This report links WNT signaling with telomerase expression. The overexpression of telomerase in many cancers had been assumed to function only in telomere main- tenance. This paper raises the possibility that telomerase overexpression may be having a direct role on gene expression through interaction with B-catenin and BRG1.

44 Endocrine tumors

31 Else T, Giordano TJ, Hammer GD. Evaluation of telomere length maintenance mechanisms in adrenocortical carcinoma. J Clin Endocrinol Metab 2008; 93:1442-1449.

32 Else T, Trovato A, Kim AC, et al. Genetic p53 deficiency partially rescues the

·· adrenocortical dysplasia phenotype at the expense of increased tumorigen- esis. Cancer Cell 2009; 15:465-476.

This paper implicates telomere biology in the presence of a nonfunctional p53 response in the pathogenesis of ACC.

33 Johnsen IK, Beuschlein F. Role of bone morphogenetic proteins in adrenal · physiology and disease. J Mol Endocrinol 2010; 44:203-211.

This is a good review of the role of BMPs in adrenal development and disease.

34 Johnsen IK, Kappler R, Auernhammer CJ, Beuschlein F. Bone morphogenetic . proteins 2 and 5 are down-regulated in adrenocortical carcinoma and modulate adrenal cell proliferation and steroidogenesis. Cancer Res 2009; 69:5784-5792.

This paper suggests that loss of BMP signaling is partially responsible for ACC proliferation and that restoration of BMP2 or BMP5 expression can inhibit ACC proliferation.

35 Kotoula V, Sozopoulos E, Litsiou H, et al. Mutational analysis of the BRAF,

. RAS and EGFR genes in human adrenocortical carcinomas. Endocr Relat Cancer 2009; 16:565-572.

This paper demonstrates that approximately 20% of ACC tumor sample have mutations in the EGFR signaling cascade and therefore may be candidates for agents targeting this signaling cascade.

36 Smith G, Bounds R, Wolf H, et al. Activating K-Ras mutations outwith

· ‘hotspot’ codons in sporadic colorectal tumours: implications for personalised cancer medicine. Br J Cancer 2010; 102:693-703.

This report highlights the need to go beyond hot spot sequencing in order to appropriately apply targeted therapies.

37 Carlson JJ, Garrison LP, Ramsey SD, Veenstra DL. Epidermal growth factor · receptor genomic variation in NSCLC patients receiving tyrosine kinase inhibitor therapy: a systematic review and meta-analysis. J Cancer Res Clin Oncol 2009; 135:1483-1493.

One of three papers to provide evidence for the association of differential response to EFGR signaling inhibition to the mutational status of various members of the EGFR signaling cascade.

38 Laurent-Puig P, Lievre A, Blons H. Mutations and response to epidermal · growth factor receptor inhibitors. Clin Cancer Res 2009; 15:1133-1139. One of three papers to provide evidence for the association of differential response to EFGR signaling inhibition to the mutational status of various members of the EGFR signaling cascade.

39 Mancl EE, Kolesar JM, Vermeulen LC. Clinical and economic value of screen- · ing for Kras mutations as predictors of response to epidermal growth factor receptor inhibitors. Am J Health Syst Pharm 2009; 66:2105-2112.

One of three papers to provide evidence for the association of differential response to EFGR signaling inhibition to the mutational status of various members of the EGFR signaling cascade.

40 Rizk-Rabin M, Assie G, Rene-Corail F, et al. Differential expression of para- thyroid hormone-related protein in adrenocortical tumors: autocrine/paracrine effects on the growth and signaling pathways in H295R cells. Cancer Epidemiol Biomarkers Prev 2008; 17:2275-2285.

41 Weismann D, Briese J, Niemann J, et al. Osteopontin stimulates invasion of

· NCI-h295 cells but is not associated with survival in adrenocortical carcino- ma. J Pathol 2009; 218:232-240.

This paper identifies osteopontin as a candidate marker for malignancy in adre- nocortical tumors.

42 Tacon LJ, Soon PS, Gill AJ, et al. The glucocorticoid receptor is overexpressed .. in malignant adrenocortical tumors. J Clin Endocrinol Metab 2009; 94:4591 - 4599.

This report demonstrates that the overexpression of the glucocorticoid receptor is an excellent marker of malignancy in adrenocortical tumors.

43 Shen XC, Gu CX, Qiu YQ, et al. Estrogen receptor expression in adreno- · cortical carcinoma. J Zhejiang Univ Sci B 2009; 10:1-6.

This paper provides evidence that estrogen receptor is important in ACC patho- genesis, a finding supported by the gene expression literature.

44 Barzon L, Masi G, Pacenti M, et al. Expression of aromatase and estrogen receptors in human adrenocortical tumors. Virchows Arch 2008; 452:181 - 191.

45 Ronchi CL, Sbiera S, Kraus L, et al. Expression of excision repair cross .. complementing group 1 and prognosis in adrenocortical carcinoma patients treated with platinum-based chemotherapy. Endocr Relat Cancer 2009; 16:907-918.

This paper demonstrates that the inverse correlation of ERCC1 expression with response to platinum therapies holds in ACC. This has implications for the inclusion of cisplatin in the treatment of ACC without first assaying for ERCC1 expression.

46 Suh I, Weng J, Fernandez-Ranvier G, et al. Antineoplastic effects of decita- . bine, an inhibitor of DNA promoter methylation, in adrenocortical carcinoma cells. Arch Surg 2010; 145:226-232.

This paper reports the growth inhibition in vitro of H295R after exposure to decitabine. The results of this paper highlight the need to study the epigenetic changes in ACC, particularly DNA methylation, in patient samples.

Quartz 4

Explorer

21045685