ORIGINAL E n d o c r i n e

ARTICLE C a r e

Preclinical Targeting of the Type I Insulin-Like Growth Factor Receptor in Adrenocortical Carcinoma
Ferdous M. Barlaskar, Aaron C. Spalding, Joanne H. Heaton, Rork Kuick, Alex C. Kim, Dafydd G. Thomas, Thomas J. Giordano, Edgar Ben-Josef, and Gary D. Hammer
Department of Internal Medicine-Division of Metabolism, Endocrinology, and Diabetes (F.M.B., J.H.H., A.C.K., G.D.H.), Departments of Radiation Oncology (A.C.S., E.B.-J.) and Department of Pathology (D.G.T., T.J.G.) and Biostatistics Core, Comprehensive Cancer Center (R.K.), University of Michigan Medical School, Ann Arbor, Michigan 48109

Context: Drug therapy for adrenocortical carcinoma (ACC), a rare and lethal malignancy, is largely empirical and ineffective. New treatments directed at molecular targets critical to the pathophysiology of ACC may prove more efficacious. Objective: The objective of the study was to profile human adrenal tumors and ACC cell lines to assess activated IGF signaling and determine the efficacy of two IGF receptor (IGF-1R) antagonists alone and in combination with mitotane. Experimental Design: ACC cell lines that display or lack activated IGF signaling are used to assess the effects of two IGF-1R antagonists in cultured cells and ACC xenograft tumors. Results: Transcriptional profiling data derived from DNA microarray analysis of human adrenal tumors implicate IGF2 as the single highest up-regulated transcript in the vast majority of carcinomas. We show that the majority of ACC cell lines tested display constitutive IGF ligand production and activation of downstream effector pathways. Both IGF-1R antagonists cause significant dosedependent growth inhibition in ACC cell lines. Furthermore, we observe that mitotane, the firstline adrenolytic drug used in patients with ACC, results in enhanced growth inhibition when used in combination with the IGF-1R antagonists. We next examined the activity of IGF-1R antagonists against ACC xenografts in athymic nude mice. IGF inhibition markedly reduced tumor growth greater than that observed with mitotane treatment, and combination therapy with mitotane significantly enhanced tumor growth suppression. Conclusion: These findings establish a critical role of IGF signaling in ACC pathophysiology and provide rationale for use of targeted IGF-1R antagonists to treat adrenocortical carcinoma in future clinical trials. (J Clin Endocrinol Metab 94: 204 212, 2009)

drenocortical carcinoma (ACC) is a rare endocrine malignancy characterized by a limited understanding of its development and pathophysiology, dismal clinical prognosis, and lack of efficacious therapeutic regimens. The annual incidence of ACC ranges from 0.5 to 2 cases per million (1). Whereas complete operative resection remains the only potentially curative option for ACC, approximately half of all patients present with metastatic disease (1, 2). This results in a 5-yr survival rate of less than 10% (1, 3). A better understanding of the etiology and pathogenesis of this devastating disease could lead to more efISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright 2009 by The Endocrine Society doi: 10.1210/jc.2008-1456 Received July 8, 2008. Accepted October 3, 2008. First Published Online October 14, 2008

fective drug designs and the development of molecularly targeted treatments. ACCs association with a select number of genetic syndromes such as Beckwith-Wiedemann syndrome (BWS) has provided insights into its pathophysiology. BWS arises from a loss of heterozygosity and/or a loss of imprinting of the 11p15.5 chromosomal region. This locus includes the mitogenic hormone, IGF-2 gene (IGF2), and locus dysregulation results in significant overexpression of this gene. Transcriptional profiling of sporadic ACC tissues provides additional support for this hormones
Abbreviations: ACA, Adrenocortical adenoma; ACC, adrenocortical carcinoma; BWS, Beckwith-Wiedemann syndrome; FITC, fluorescein isothiocyanate; IGF-1R, IGF receptor; MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium; VEGF, vascular endothelial growth factor.

204

jcem.endojournals.org

J Clin Endocrinol Metab. January 2009, 94(1):204 212

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009

J Clin Endocrinol Metab, January 2009, 94(1):204 212

jcem.endojournals.org

205

pathogenic role. We and others have shown IGF2 as the single most up-regulated transcript in 80 90% of ACCs (4 6). IGF-II mainly elicits its cellular effects through the ubiquitously expressed type 1 IGF receptor (IGF-1R). Importantly, human ACCs also exhibit elevated levels of IGF-1R mRNA and protein (7). Taken together, these observations suggest that activation of the IGF pathway is a common pathological mechanism used by tumor cells during adrenocortical tumorigenesis. In this study, we analyzed a large series of benign and malignant human adrenal tumors and a panel of ACC cell lines to confirm enhanced IGF signaling in ACCs. We used a small molecule inhibitor (NVP-AEW541) and a fully human monoclonal antibody (IMC-A12), both targeting IGF-1R, to demonstrate specific abrogation of IGF-mediated signaling and concomitant inhibition of proliferation. Only ACC lines with increased IGF signaling responded to both agents. Synergistic antiproliferative effects were observed when IGF-1R inhibition was combined with mitotane in culture. In vivo, both IGF-1R antagonists markedly attenuated human ACC xenograft growth in athymic nude mice. Moreover, IGF inhibition combined with mitotane significantly enhanced single agent tumor growth inhibition. Our results validate IGF-1R as an important target in ACC and provide rationale for the testing of IGF-1R antagonists as a promising therapeutic agent in clinical trials.

data on The Endocrine Societys Journals Online Web site at http:// jcem.endojournals.org.

Mice
Animal studies were performed in accordance to an institutionally approved protocol under the auspice of the University Committee on Use and Care of Animals at the University of Michigan. Tumor xenografts were established by sc injection of 1 106 RL251 or 2.5 106 H295 cells into flank regions of 4- to 6-wk-old female athymic mice. Animals were treated with IMC-A12 (8) or NVP-AEW541 (9). Mitotane was reformulated in 10% Tween 80/0.9% normal saline and administered by ip injection once daily at 300 mg/kg.

DNA microarray analysis

The full details of the DNA microarray analysis will be reported separately (manuscript in press). The Affymetrix array data have already been deposited into Gene Expression Omnibus as series GSE10927.

Human adrenal tissue microarray

An adrenal-specific tissue array was constructed using human formalin-fixed, paraffin-embedded tissues that correspond to the tissues used for DNA microarray analysis and was stained with either phosphoAktSer473 (Invitrogen Life Technologies, Carlsbad, CA) or phospho-IGF1RTyr1158/62/63 (Millipore). Immunoreactivity was scored blindly by a four-tier [negative, low (1 ), medium (2 ), and high (3 ) positive] grading scheme.

Statistical analysis

Materials and Methods

Reagents
IMC-A12 was provided by ImClone Systems (New York, NY) (8). NVP-AEW541 was provided by Novartis (Basel, Switzerland) (9). Recombinant human IGF-I and IGF-II ligands were from Peprotech (Rocky Hill, NJ). Mitotane, lectin-fluorescein isothiocyanate (FITC), and anti-actin were from Sigma (St. Louis, MO). Anti-IGF-1R was obtained from Santa Cruz Biotechnology (Santa Cruz, CA), whereas anti-Akt and anti-phospho-AktSer473 were from Cell Signaling Technology (Danvers, MA). Anti-phospho-tyrosine (4G10) was from Millipore Bioscience (Billerica, MA).

Tests between pairs of groups were performed with two-sided, twosample t tests. For in vivo and in vitro data testing combinations of agents, two-way ANOVA models were used to test differences in cell viability or tumor size between difference combinations of agents and test for interactions. We also used Calcusyn software to determine combination indices with in vitro mitotane and NVP-AEW541 MTS assay.

Results
Expression profile of IGF2 and downstream signaling in human ACC tissues Using DNA microarray technology, we analyzed human tissues derived from normal adrenal cortex, adrenocortical adenomas (ACAs), and ACCs to reveal gene expression profiles (manuscript in press). From these data, we specifically examined the 11p15.5 chromosomal region in which locus dysregulation has been associated with adrenocortical cancers (Fig. 1A). The vast majority of ACCs display overexpression of IGF2 gene transcripts, whereas the H19 [a micro-RNA negatively regulating IGF2 expression (16, 17)] and CDKN1C (encoding the cell cycle dependent kinase inhibitor, p57kip2) genes are down-regulated, suggesting an imprinting defect or loss of heterozygosity of this chromosomal region, similar to that commonly observed in BWS. To validate these microarray results, quantitative RT-PCR was performed on RNA isolated from three randomly selected ACAs and three ACCs (Fig. 1B). We found a greater than 60-fold increase of IGF2 transcripts in all three ACC samples when compared with IGF2 levels in ACA samples. Further analysis of active IGF signaling with these six human tumor samples was performed by immunoblotting for levels of total IGF-1R protein and phosphorylated AktSer473, a downstream mediator of active IGF signaling (Fig. 1C). Expression of IGF-1R was observed in all six

Cell lines and cell culture

All standard cell culture reagents were purchased from Invitrogen Life Technologies (Carlsbad, CA). The cell lines NCI-H295 (10), Y1 (11), and SW13 (12) were obtained from American Type Culture Collection (Manassas, VA). The RL251 (13) cell line was previously described and generously provided by Dr. David Schteingart (University of Michigan, Ann Arbor, MI), whereas the ST5 (14) cell line was generated and provided by Dr. Synthia Mellon (University of California, San Francisco, San Francisco, CA).

Cell proliferation assay

Cells were seeded on 96-well plates with increasing concentrations of the compounds added in triplicate wells. Incubation continued for 72 h (for NVP-AEW541 treatment) or 7 d (for IMC-A12 treatment) and then an 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay (Promega Corp., Madison, WI) was performed according to the manufacturers protocol.

Immunoblotting, immunohistochemistry, and RT-PCR

Assays were performed as described previously (15) with primer sequences located in supplementary Table 1, published as supplemental

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009

206

Barlaskar et al.

IGF-IR Inhibition for Adrenal Cancer

J Clin Endocrinol Metab, January 2009, 94(1):204 212

Normal

ACA

ACC
H19_224646_x_at H19_224997_x_at IGF2_210881_s_at IGF2_202410_x_at CDKN1C_219534_x_at CDKN1C_213182_x_at

21

29 40

18 Fold change from the median

1/8 1/4 1/2 0 2 4 8

13

14

ACC

ACA
21 29 40

B
Relative IGF-II Expression

C
1000 100 10 1 0.1 ACC13 ACC14 ACC18 ACA21 ACA29 ACA40

13 14 18

IGF-IR p-Akt Akt -actin

D
p-IGF-IR

ACC
1 2 3 3

ACA
0 1 2

Normal
2 1

p-Akt

0 1

2 0

1 0 1

FIG. 1. Up-regulation of IGF2, overexpression of IGF-1R, and active IGF signaling in human ACCs in comparison with normal and adenoma tissues. A, Snapshot heat map of the 11p15.5 chromosomal region generated from Affymetrix U133A 2.0 Plus oligonucleotide array. Shown are the imprinted genes IGF2, H19, and CDKN1C of 65 patient samples consisting of 10 normal adrenal tissues, 22 adrenal adenomas, and 33 sporadic ACCs. Shades of red indicate increased expression of gene indicated on the right, whereas shades of green reveal decreased expression levels. Numbers below heat map represent location of tumor samples used for quantitative RT-PCR and immunoblotting. B, Quantitative RT-PCR of three ACCs and three adenomas for human IGF2 transcript. The y-axis represents relative IGF2 message levels normalized to glyceraldehyde 3-phosphate dehydrogenase transcript levels. Data shown represent the mean SD of triplicate samples of one representative experiment. C, Immunoblot analysis of tumor samples described above for IGF-1R expression, Akt, and activated phosho-AktSer473 as a readout for active IGF signaling. Immunoblotting of -actin is shown as a protein loading control. D, Histologically graded human tissue microarray staining for phospho-IGF-1R and phospho-Akt with results depicted as percentage pie charts. Immunoreactivity was scored blindly by a four-tier [negative, low (1 ), medium (2 ), and high (3 ) positive] grading scheme of 24 ACCs, 22 ACAs, and four normal adrenal tissue.

tissues, whereas two ACC samples possessed far greater levels of the receptor. Immunoblotting for phospho-AktSer473 suggested active IGF signaling in all three ACC samples and in only one ACA sample. To further validate the observation that IGF-mediated signaling was specifically increased in ACC compared with normal and adenomatous samples, tissue microarray slides containing 24 ACC, 22 ACA, and four normal adrenals were stained for phospho-IGF-1R and phospho-AktSer473 (Fig. 1D). A marked increase in signal intensity of phospho-IGF-1R and phospho-AktSer473 was observed in ACC samples compared with ACA and normal adrenal tissue, represented here as a shift in the percentage of ACC samples with high intensity staining for these activated proteins. In summary, molecular profiling of human adrenal tumors demonstrated overexpression of two critical components (IGF2 and IGF-1R) of the IGF signaling cascade and concomitant activation of the downstream effector, Akt. These results are consistent with the IGF pathway playing a critical role in ACC pathogenesis.

Expression of IGF pathway members and downstream signaling in ACC cell lines To begin to determine the biologic relevance of IGF-1R in ACC, we first examined the endogenous expression profiles of IGF ligands, IGF-1R, and downstream effectors of IGF-mediated signaling in five ACC cell lines (Fig. 2). We chose two mouse lines (Y1 and ST5) and three human lines derived from invasive primary adrenocortical carcinomas (NCI-H295, SW13, and RL251). Using RNA purified from cells maintained in serum-free or serum-containing media, RT-PCR was performed to detect IGF1 and IGF2 gene expression. The two mouse cell lines expressed IGF1, whereas the H295 cells expressed IGF2 transcript, in agreement with a previous study describing an autocrine role of IGF-II in H295 cells (18). To assess protein expression, IGF1R was immunoprecipitated from whole-cell lysates followed by immunoblot analysis. As shown in Fig. 2, we observed only minimal levels of IGF-1R protein in RL251 cells compared with the four other cell lines. However, there was also variable protein

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009

J Clin Endocrinol Metab, January 2009, 94(1):204 212

jcem.endojournals.org

207

Y1 Serum IGF-I IGF-II IP: IGF-IR IB: IGF-IR IP: IGF-IR IB: p -Y p-Akt Akt + -

ST5 +

H295 +

SW13 +

RL251 +

FIG. 2. Endogenous IGF signaling in a panel of ACC cell lines. A panel of mouse (Y1 and ST5) and human (H295, SW13, and RL251) ACC cell lines cultured in serum-free ( ) or serum-containing ( ) media assessed for endogenous transcript expression of IGF1 or IGF2 by gel-based RT-PCR (top two panels). Whole-cell lysates were immunoprecipitated (IP) for the IGF-1R and subsequently immunoblotted (IB) for IGF-1R and phospho-tyrosine (middle two panels). Lower two panels show immunoblots from whole-cell lysates for phospho-AktSer473 and total Akt for protein loading control.

expression between the four IGF-1R-positive cell lines. Immunoprecipitates were also probed for phospho-tyrosine and demonstrated active signaling in all cell lines except RL251. Pathway activation was additionally confirmed by immunoblotting for phospho-AktSer473. Immunoblots for total protein levels of Akt were used as a loading control. The H295 cell line emerged as the most efficient in vitro recapitulation of ACC, as assessed by its constitutive overexpression of IGF-II and active IGF signaling. Therefore, this was the cell line of choice for further experiments, although other ACC cell lines were used to confirm all initial results obtained from H295 experiments.

Inhibition of IGF signaling by IGF-1R antagonists To assess the ability of both IGF antagonists (IMC-A12 and NVP-AEW541) to specifically inhibit IGF-II/IGF-1R-mediated signaling, log-phase H295 cells were subjected to increasing amounts of IMC-A12 or NVP-AEW541 in the presence of 10 nM IGF-I/ IGF-II (Fig. 3). Immunoblotting of cell lysates incubated with increasing concentrations of IMC-A12 resulted in a dosedependent reduction of IGF-1R levels, with an approximately 80% receptor decrease achieved at 100 nM concentrations of antibody (Fig. 3A). This decrease may be attributed to antibodymediated receptor internalization and degradation, a phenomenon not seen when IGF ligands are incubated alone (8). IMCA12 attenuated phospho-AktSer473 in a dose-dependent manner (100 nM treatment resulted in 70% decrease in the ratio of p-Akt to Akt band intensities over controls), indicating targeted decrease of IGF signaling. We also found NVP-AEW541 inhibited IGF-1R signaling in H295 cells (Fig. 3B). At 5 M NVPAEW541 concentrations, receptor phospho-tyrosine levels were undetectable and phospho-AktSer473 levels were also completely abrogated. No change in total IGF-1R or Akt levels was observed with treatment. Taken together, these results indicate that both pharmacologic agents are able to specifically target and inhibit IGF-mediated signaling in ACC cell lines. IGF-1R antagonists inhibit proliferation of ACC cell lines IGF signaling regulates several aspects of cellular function including the enhancement of proliferation. To test the functional consequence of IGF-1R inhibition, MTS proliferation assays were performed on H295 cells treated with IMC-A12 or NVP-AEW541 (Fig. 4). We used the RL251 cell line as a negative control in these assays because it lacks expression of IGF-1R

A
IMC-A12 [nM] IGFI/II 0 0 + 3 + 10 + 30 + 100 +

B
NVP AEW-541[M] IGFI/II IP: IGF-IR IB: p-Y 0 0 + 0.1 + 0.5 + 1 + 5 + 10 +

IGF-IR

IGF-IR Akt Akt p-Akt p-Akt

-actin

-actin

FIG. 3. IGF-1R antagonist treatments decrease IGF-mediated signaling. A, H295 cells were pretreated with increasing nanomolar concentrations of IMC-A12 for 1 h before addition of 10 nM IGF-I/II ligand mix for 10 min. Cells were subsequently harvested and immunoblotted (IB) for IGF-1R , Akt, phospho-AktSer473, or -actin. B, H295 cells were pretreated with increasing micromolar concentrations of NVP-AEW541 for 30 min before addition of 10 nM IGF-I/II ligand mix for 10 min. Cells were then harvested and either immunopreciptated (IP) with an anti-IGF-1R antibody and immunoblotted for phospho-tyrosine residues or directly immunoblotted for IGF-1R , Akt, phospho-AktSer473, or -actin.

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009

208

Barlaskar et al.

IGF-IR Inhibition for Adrenal Cancer

J Clin Endocrinol Metab, January 2009, 94(1):204 212

100 Proliferation (%) 80 60 40 20 0 0 25 50 75

RL251

100 Proliferation (%) 80 60 40 20 0 100 0 10 20 30 H295 RL251

*
H295

IMC-A12 Concentration [nM]

NVP AEW-541 Concentration [M]

FIG. 4. Antiproliferative effects of IGF-1R antagonist treatments in vitro. H295 and RL251 cells were incubated with increasing micromolar concentrations of NVPAEW541 (right panel) or nanomolar concentrations of IMC-A12 (left panel), and proliferation was assessed with MTS reagent. Squares (f) represent the mean of quadruplicate wells of RL251 cells, whereas diamonds ( ) represent quadruplicate wells of H295 cells with error bars indicating SD. Data are representative of at least four independent experiments. *, P 0.05.

protein in comparison with other ACC lines. As predicted, increasing doses of IMC-A12 had no substantial effect on the RL251 cells, whereas high concentrations of the antibody exhibited antiproliferative effects on H295 cells (Fig. 4, left panel). The IMC-A12 IC50 value for the H295 cells was 90 nM, but 50% inhibition was not achieved with RL251 cells, even at the highest IMC-A12 dose (P 0.001 comparing RL251 vs. H295 at 100 nM A12). The NVP-AEW541 IC50 value for H295 cells was 2.7 M, whereas for RL251 cells, the IC50 was 36 M (P 0.001 comparing RL251 vs. H295 at 3 M NVP-AEW541), again demonstrating targeted suppression of the H295 cell lines ability to proliferate due to the presence of IGF-1R. Targeting of IGF-1R delays tumor xenograft growth Stemming from our data revealing IGF-1R inhibition results in a significant decrease of downstream signaling and proliferation in vitro, we assessed its in vivo activity using human ACC cell lines. Using the rationale discussed above, we used the RL251 cell line as an additional control in these assays. H295 or RL251 cells were inoculated sc into athymic nude mice, and the mice were subsequently randomized into three treatment groups (n 8 tumors per group for H295 and n 10 for RL251): placebo, IMC-A12 treatment, and NVP-AEW541 treatment. Tumor volume measurements were taken three times a week for 21 d. The data were plotted as the log ratios of tumor size over initial tumor size to more accurately reflect the percent reduction in tumor size and, at the same time, account for the variable initial tumor sizes in each mouse (Fig. 5A). RL251 xenografts had a higher tumor engraftment rate than H295 xenografts (94 vs. 75%, data not shown). IMC-A12 was well tolerated in treated mice with no substantial adverse effects or weight changes observed between groups (data not shown). The treatment of H295 tumors with IMC-A12 over a 21-d span (Fig. 5A, left panel) resulted in a significant reduction in tumor size over untreated controls (P 0.001 from day 7 onwards). Treatment in RL251 xenografts (Fig. 5A, right panel) resulted in a statistically insignificant reduction in tumor volume at each time point, underscoring the specificity of IMC-A12. Tests comparing the tumor growth in both cell lines showed that IMC-A12 reduced H295 tumor volumes (compared with tumors receiving no treatment) signifi-

cantly more than for RL251 xenografts (P 0.017). Whereas H295 xenografts treated with NVP-AEW541 were similar in histological appearance to those treated with IMC-A12 and exhibited a 34% reduction in tumor burden over controls, this decrease fell short of being statistically significant (P 0.086) (supplementary Fig. 2, published as supplemental data on The Endocrine Societys Journals Online Web site at http://jcem. endojournals.org). NVP-AEW541 treatment of mice harboring RL251 xenografts also had an insignificant reduction in tumor volume (P 0.57). NVP-AEW541 was well tolerated in treated mice with no substantial adverse effects or weight changes observed during treatment (data not shown). To determine whether blocking IGF-1R resulted in decreased IGF signaling in vivo, H295 xenografts were subjected to immunoblot analysis (Fig. 5B). Lysates prepared from two separate tumors from control and IMC-A12-treated mice were immunoblotted for Akt and phospho-AktSer473 levels. Although total Akt expression levels were similar in both treatment arms, phosphoAktSer473 levels were decreased in the tumors of mice treated with IMC-A12, thus revealing a targeted decrease in IGF signaling. Taken together, these results indicate in vivo IGF-1R inhibition caused significant and targeted tumor growth delay of human ACC xenografts and underscores the importance of IGF signaling in ACC pathophysiology. Inhibition of IGF-1R enhances the inhibitory effects of mitotane Mitotane is the standard of care for treatment of ACC because of its specific cytotoxic effects in cortical cells. To evaluate whether this response could be enhanced by additional IGF inhibition, we investigated the effect of combining mitotane with IGF antagonists in culture and in vivo tumor growth. First, MTS proliferation assays were performed on H295 and RL251 cells with increasing concentrations of NVP-AEW541 and mitotane (Fig. 6A). Whereas both H295 and RL251 cells exhibited a dose-dependent reduction in proliferation when incubated with mitotane (no NVP-AEW541), only H295 cells demonstrated a further decrease in growth when incubated with mitotane and NVP-AEW541. Individually, 30 M mitotane and 3 M NVP-AEW541 reduced cell viability to 85 and 77% of control levels, respectively. An additive model for the com-

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009

J Clin Endocrinol Metab, January 2009, 94(1):204 212

jcem.endojournals.org

209

A
log2(volume/initial volume)

4 3 2 1 0 -1 0 5 10 15 20 25 Days Treatment Control IMC-A12

4 3 2 1 0 -1 0 5 10 15 20 25 Days Treatment Control IMC-A12

B
1
IGF-IR Akt p-Akt -Actin

Control 2

IMC-A12 1 2

FIG. 5. Targeted inhibition of tumor growth in vivo. A, H295 (left panel) and RL251 (right panel) cells were injected sc into athymic nude mice and mice were randomized into treatment groups (n 8 for H295 and n 10 for RL251). Groups were treated with vehicle or IMC-A12 every other day. Tumor dimensions on control ( ) mice or IMC-A12-treated (f) mice were measured three times a week for the duration of the 21-d study, and data are shown as the log ratios of tumor size to initial tumor size means SE. B, To confirm in vivo targeting of IGF inhibition, harvested H295 tumors were lysed and two tumors from each treatment arm were subjected to immunoblotting for IGF-1R , Akt, phospho-AktSer473, and -actin.

binatorial effects of the two agents would predict about 65% cell viability compared with untreated cells. However, we observed just 9% cell viability for this combination compared with no treatment, which was significantly less than expected under the additive model (P 4 10 15), indicating significant synergy (interaction) between the two drugs. Analysis of these data using Calcusyn software to graphically quantitate combined drug effects also confirmed a synergistic effect of mitotane and NVP-AEW541 in vitro (supplementary Fig. 3). To evaluate this effect in vivo, H295 xenografted nude mice were treated with mitotane or IMC-A12 either as single agents or in combination (n 20 tumors per treatment group) (Fig. 6B). Treatment with antibody alone resulted in a 51% reduction in tumor size (P 0.001), comparable with the 55% reduction observed in the monotherapy experiment detailed in Fig. 5A. Mitotane treatment as a single agent was not as effective as IMCA12 but resulted in a 25% decrease in tumor volume (P 0.035). The combination of IMC-A12 and mitotane resulted in a 70% decrease in tumor burden over untreated controls (P 0.001). Although the combined treatment induced more than an additive effect predicted with mitotane and IMC-A12 (as calculated at d 21 of treatment), this difference was not sufficient to demonstrate synergy between these two compounds (P 0.22, two sided F test). IMC-A12, mitotane, and their combination were well tolerated in treated mice with no substantial adverse effects or weight changes observed between groups (data not shown). After 21 d, xenograft specimens were collected for histochemical

analysis (Fig. 6C). Interestingly, hematoxylin and eosin staining revealed treatment results in a clear decrease in vascularity compared with control sections. We investigated this further by immunohistochemically evaluating microvessel density using a FITC-conjugated lectin molecule that binds specifically to endothelial cells. Dense areas of lectin-positive endothelial cells are observed throughout control xenograft sections, but IMC-A12-treated xenograft sections display a moderate decrease in endothelial cell density. Moreover, the greatest histologic decrease was observed in xenografts treated with the combination of IMC-A12 and mitotane. We hypothesized IGF inhibition may reduce the angiogenic potential imparted by the expression of IGF-II via induction of vascular endothelial growth factor (VEGF) expression, seen in other systems (19, 20). Therefore, we generated cDNA from xenografts for quantitative RT-PCR to assess human VEGF expression using PCR primers designed to recognize all four isoforms of VEGF transcripts (21) (Fig. 6D). Results demonstrated a 1.6-fold decrease in human VEGF expression in IMC-A12-treated samples (P 0.044) and a 2.2-fold decrease when antibody was combined with mitotane (P 0.008). Mitotane treatment alone resulted in a small and statistically insignificant decrease (P 0.46). Thus, inhibition of IGF signaling is the major factor causing the decreased VEGF transcript levels in these tumors. These results, taken together, demonstrate that IGF signaling inhibition with IMC-A12 leads to a larger suppression of ACC tumor growth in comparison with mitotane and that combining both agents can

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009

log2(volume/initial volume)

RL251 Tumor Size

H295 Tumor Size

210

Barlaskar et al.

IGF-IR Inhibition for Adrenal Cancer

J Clin Endocrinol Metab, January 2009, 94(1):204 212

A
100 RL251 Proliferation (%) 80 60 40 20 0 0 2 4 6 8 10 NVP AEW-541 Concentration (M) H295 Proliferation (%) 100 80 60 40 20 0 0 2 4 6 8 10 NVP AEW-541 Concentration (M) Mitotane 0 M 3 M 30 M 100

2.5 2.0
log2(volume/initial volume)

H295 Tumor Size

Control Mitotane IMC-A12 IMC-A12+Mitotane

H&E

Lectin - FITC

1.5

Control

1.0

0.5

Mitotane

0 0 2 4 6 8 10 12 14 16 18 20 22 Days Treatment

D
hVEGF Expression (% of control) 100 80 60 40 20 0 Control Mitotane IMC-A12 IMC-A12 + Mitotane

IMC-A12

IMC-A12 + Mitotane

FIG. 6. IGF-1R antagonists enhance the inhibitory effects of mitotane. A, RL251 (left panel) and H295 (right panel) cells were incubated in triplicate with a combination of mitotane and increasing concentrations of NVP-AEW541. Proliferation was assessed with MTS reagent. Data are representative of three independent experiments and displayed as mean SD. B, Mice harboring H295 xenografts were randomized into four groups (n 20 per treatment arm) and treated with vehicle or IMC-A12 every other day and/or mitotane once daily for the duration of the experiment. Tumor volumes were measured three times a week. Data are presented as log ratios of tumor size over initial tumor size means SE. C, Hematoxylin and eosin (H & E)-stained section of H295 tumor xenografts (left panels) at 40 magnification. LectinFITC immunohistochemical analysis (right panels) was performed to detect relative levels of endothelial cells. D, Quantitative RT-PCR of three or four tumors from each treatment arm with primers detecting all four isoforms of human VEGF. The y-axis represents relative VEGF levels normalized to glyceraldehyde 3-phosphate dehydrogenase transcript levels followed by normalization to control tissue values. Each value represents the average of triplicates of two independent experiments and data are presented as the mean SE. *, P 0.05.

specifically decrease ACC xenograft growth greater than either agent alone.

Discussion
Current therapy for adrenocortical cancer consists of surgical resection and adjuvant mitotane treatment or, in nonresectable

disease, mitotane in combination with other cytotoxic chemotherapies (22). A recent retrospective study evaluating the efficacy of mitotane after radical resection in adrenocortical cancer revealed that mitotane may indeed prolong recurrence-free survival to 23 times that of untreated patients (23, 24). However, mitotane, a structural isomer of the pesticide dichlorodiphenyltrichloroethane, is associated with significant toxicity due to the high doses required for its adrenolytic effects. Unfortunately, a

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009

J Clin Endocrinol Metab, January 2009, 94(1):204 212

jcem.endojournals.org

211

large number of ACC patients present with metastatic disease, which typically precludes surgery and carries a dismal prognosis (25). These outcomes emphasize the need for new treatment strategies for this fatal disease with therapies predicated on identification of specific pathways critical for ACC pathogenesis. The adrenal gland requires dense vascularization to allow access of hormonal control from higher-order structures and for secretion of its hormone products to systemic blood. ACTH has been shown to increase the production and secretion of VEGF in the fetal adrenal, indicating coordinated growth and vascular programming (26). Patients with ACC have significantly greater serum VEGF levels than patients with benign adrenal tumors, and the highest levels are observed in ACC patients with cortisol hypersecretion (27). Additionally, our adrenal tumor gene expression profiling also confirms increased VEGF transcript expression in ACCs (data not shown). Therefore, it would be rational to assess the effectiveness of angiogenesis inhibitors as single agents or in combination with mitotane and/or IGF inhibition in future preclinical and clinical studies. Imprinting is an epigenetic mechanism by which cells regulate gene expression in a parent-of-origin-specific manner, whether maternally or paternally inherited, so that one allele is expressed, whereas the other is stably and efficiently silenced through DNA methylation. In pathologic situations, as in the case in the majority of patients with BWS, this process goes awry and results in aberrant expression of imprinted genes (28). Our gene expression profiling of sporadic human adrenal tumors revealed dysregulation of the 11p15.5 expression profile that is reminiscent of what is commonly observed in BWS, suggesting that similar imprinting defects may underlie the pathogenesis of nonhereditary ACC. Interestingly, during organ development in the human fetus, the adrenal gland has one of the highest levels of IGF2 transcript expression, which subsequently declines after birth (29). The reactivation of IGF-II expression in sporadic ACCs lends credence to the hypothesis that this cancer may represent an aberrant epigenetic reprogramming in the repopulating cells (tissue specific stem/progenitor cells) of the adult adrenal. In the normal colon epithelium, loss of imprinting of IGF2 is frequently observed in patients with a higher risk of developing colorectal cancer. Transgenic mouse models overexpressing IGF through loss of imprinting expands the progenitor cell population of the colon (30). These proposed cancer stem cells may represent a bona fide malignant population of cells that both initiate tumorigenesis and evade conventional chemotherapies. Whether initiation of ACC is mediated through similar mechanisms is an area of active investigation. In this report, we have demonstrated overexpression of IGF-1R and its ligand and activated IGF signaling in human ACC samples and ACC cell lines. Antagonizing this pathway with two pharmacological agents resulted in inhibition of growth in vitro and in vivo. Importantly, this targeted inhibition was more potent than mitotane treatment in decreasing xenograft growth and the combination of IGF inhibition and mitotane resulted in greater antiproliferative effects in vitro and greater xenograft growth inhibition in vivo over single agent treatment. These data raise the prospect of using targeted disruption of IGF-1R signaling to attain a

therapeutic advantage when used as an adjuvant in mitotane therapy or possibly other chemotherapeutics in ACC patients.

Note Added in Proof

During the review of this manuscript, Almeida et al. reported similar findings in JCEM that IGF inhibition in ACC cell lines results in significant reduction in proliferation and concomitant activation of adoptosis, in vitro (31).

Acknowledgments
We thank ImClone Systems and Novartis for generously providing their respective targeted reagents. We thank Dr. David E. Schteingart for helpful advice and Julie Pepera for technical support. Address all correspondence and requests for reprints to: Dr. Gary D. Hammer, Department of Internal Medicine-Division of Metabolism, Endocrinology, and Diabetes, University of Michigan Medical School, 1502 BSRB, 109 Zina Pitcher Place, Ann Arbor, Michigan 48109-2200. E-mail: ghammer@umich.edu. This work was supported by National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases Grant DK 062027 (to G.D.H.) and the American Cancer Society Grant RSG-04236 (to G.D.H.). F.M.B. is supported through a predoctoral fellowship from National Institutes of Health/Training Program for Organogenesis and Grant T-32-HD007505 and is a fellow in the Medical Scientist Training Program. A.C.S. is supported by a Resident Seed Grant from ASTRO. Disclosure Statement: The authors have nothing to disclose.

References
1. Wajchenberg BL, Albergaria Pereira MA, Medonca BB, Latronico AC, Campos Carneiro P, Alves VA, Zerbini MC, Liberman B, Carlos Gomes G, Kirschner MA 2000 Adrenocortical carcinoma: clinical and laboratory observations. Cancer 88: 711736 2. Allolio B, Fassnacht M 2006 Clinical review: adrenocortical carcinoma: clinical update. J Clin Endocrinol Metab 91:20272037 3. Icard P, Goudet P, Charpenay C, Andreassian B, Carnaille B, Chapuis Y, Cougard P, Henry JF, Proye C 2001 Adrenocortical carcinomas: surgical trends and results of a 253-patient series from the French Association of Endocrine Surgeons Study Group. World J Surg 25:891 897 4. Giordano TJ, Thomas DG, Kuick R, Lizyness M, Misek DE, Smith AL, Sanders D, Aljundi RT, Gauger PG, Thompson NW, Taylor JM, Hanash SM 2003 Distinct transcriptional profiles of adrenocortical tumors uncovered by DNA microarray analysis. Am J Pathol 162:521531 5. Velazquez-Fernandez D, Laurell C, Geli J, Hoog A, Odeberg J, Kjellman M, Lundeberg J, Hamberger B, Nilsson P, Backdahl M 2005 Expression profiling of adrenocortical neoplasms suggests a molecular signature of malignancy. Surgery 138:10871094 6. West AN, Neale GA, Pounds S, Figueredo BC, Rodriguez Galindo C, Pianovski MA, Oliveira Filho AG, Malkin D, Lalli E, Ribeiro R, Zambetti GP 2007 Gene expression profiling of childhood adrenocortical tumors. Cancer Res 67:600 608 7. Weber MM, Auernhammer CJ, Kiess W, Engelhardt D 1997 Insulin-like growth factor receptors in normal and tumorous adult human adrenocortical glands. European J Endocrinol 136:296 303 8. Burtrum D, Zhu Z, Lu D, Anderson DM, Prewett M, Pereira DS, Bassi R, Abdullah R, Hooper AT, Koo H, Jimenez X, Johnson D, Apblett R, Kussie P, Bohlen P, Witte L, Hicklin DJ, Ludwig DL 2003 A fully human monoclonal antibody to the insulin-like growth factor I receptor blocks ligand-dependent signaling and inhibits human tumor growth in vivo. Cancer Res 63:8912 8921 9. Garcia-Echeverria C, Pearson MA, Marti A, Meyer T, Mestan J, Zimmermann

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009

212

Barlaskar et al.

IGF-IR Inhibition for Adrenal Cancer

J Clin Endocrinol Metab, January 2009, 94(1):204 212

10.

11.

12.

13.

14.

15.

16. 17.

18.

19.

20.

J, Gao J, Brueggen J, Capraro HG, Cozens R, Evans DB, Fabbro D, Furet P, Porta DG, Liebetanz J, Martiny-Baron G, Ruetz S, Hofmann F 2004 In vivo antitumor activity of NVP-AEW541-A novel, potent, and selective inhibitor of the IGF-IR kinase. Cancer Cell 5:231239 Gazdar AF, Oie HK, Shackleton CH, Chen TR, Triche TJ, Myers CE, Chrousos GP, Brennan MF, Stein CA, La Rocca RV 1990 Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res 50:5488 5496 Yasumura Y, Buonassisi V, Sato G 1966 Clonal analysis of differentiated function in animal cell cultures. I. Possible correlated maintenance of differentiated function and the diploid karyotype. Cancer Res 26:529 535 Leibovitz A, McCombs 3rd WM, Johnston D, McCoy CE, Stinson JC 1973 New human cancer cell culture lines. I. SW-13, small-cell carcinoma of the adrenal cortex. J Natl Cancer Inst 51:691 697 Schteingart DE, Giordano TJ, Benitez RS, Burdick MD, Starkman MN, Arenberg DA, Strieter RM 2001 Overexpression of CXC chemokines by an adrenocortical carcinoma: a novel clinical syndrome. J Clin Endocrinol Metab 86:3968 3974 Mellon SH, Miller WL, Bair SR, Moore CC, Vigne JL, Weiner RI 1994 Steroidogenic adrenocortical cell lines produced by genetically targeted tumorigenesis in transgenic mice. Mol Endocrinol 8:97108 Looyenga BD, Hammer GD 2007 Genetic removal of Smad3 from inhibin-null mice attenuates tumor progression by uncoupling extracellular mitogenic signals from the cell cycle machinery. Mol Endocrinol 21:2440 2457 Cai X, Cullen BR 2007 The imprinted H19 noncoding RNA is a primary microRNA precursor. RNA 13:313316 Gabory A, Ripoche MA, Yoshimizu T, Dandolo L 2006 The H19 gene: regulation and function of a non-coding RNA. Cytogenet Genome Res 113:188 193 Logie A, Boulle N, Gaston V, Perin L, Boudou P, Le Bouc Y, Gicquel C 1999 Autocrine role of IGF-II in proliferation of human adrenocortical carcinoma NCI H295R cell line. J Mol Endocrinol 23:2332 Kwon YW, Kwon KS, Moon HE, Park JA, Choi KS, Kim YS, Jang HS, Oh CK, Lee YM, Kwon YG, Lee YS, Kim KW 2004 Insulin-like growth factor-II regulates the expression of vascular endothelial growth factor by the human keratinocyte cell line HaCaT. J Invest Dermatol 123:152158 Beppu K, Nakamura K, Linehan WM, Rapisarda A, Thiele CJ 2005 Topotecan blocks hypoxia-inducible factor-1alpha and vascular endothelial growth fac-

21.

22. 23.

24. 25. 26.

27.

28. 29.

30. 31.

tor expression induced by insulin-like growth factor-I in neuroblastoma cells. Cancer Res 65:4775 4781 Hata K, Osaki M, Dhar DK, Nakayama K, Fujiwaki R, Ito H, Nagasue N, Miyazaki K 2004 Evaluation of the antiangiogenic effect of Taxol in a human epithelial ovarian carcinoma cell line. Cancer Chemother Pharmacol 53:68 74 Kirschner LS 2006 Emerging treatment strategies for adrenocortical carcinoma: a new hope. J Clin Endocrinol Metab 91:14 21 Terzolo M, Angeli A, Fassnacht M, Daffara F, Tauchmanova L, Conton PA, Rossetto R, Buci L, Sperone P, Grossrubatscher E, Reimondo G, Bollito E, Papotti M, Saeger W, Hahner S, Koschker AC, Arvat E, Ambrosi B, Loli P, Lombardi G, Mannelli M, Bruzzi P, Mantero F, Allolio B, Dogliotti L, Berruti A 2007 Adjuvant mitotane treatment for adrenocortical carcinoma. N Engl J Med 356:23722380 Schteingart DE 2007 Adjuvant mitotane therapy of adrenal cancer-use and controversy. N Engl J Med 356:24152418 Ng L, Libertino JM 2003 Adrenocortical carcinoma: diagnosis, evaluation and treatment. J Urol 169:511 Shifren JL, Mesiano S, Taylor RN, Ferrara N, Jaffe RB 1998 Corticotropin regulates vascular endothelial growth factor expression in human fetal adrenal cortical cells. J Clin Endocrinol Metab 83:13421347 Zacharieva S, Atanassova I, Orbetzova M, Nachev E, Kalinov K, Kirilov G, Shigarminova R, Ivanova R, Dashev G 2004 Circulating vascular endothelial growth factor and active renin concentrations and prostaglandin E2 urinary excretion in patients with adrenal tumours. Eur J Endocrinol 150:345349 Feinberg AP 2007 Phenotypic plasticity and the epigenetics of human disease. Nature 447:433 440 Han VK, Lund PK, Lee DC, DErcole AJ 1988 Expression of somatomedin/ insulin-like growth factor messenger ribonucleic acids in the human fetus: identification, characterization, and tissue distribution. J Clin Endocrinol Metab 66:422 429 Feinberg AP, Ohlsson R, Henikoff S 2006 The epigenetic progenitor origin of human cancer. Nat Rev 7:2133 Almeida MQ, Fragoso MC, Lotfi CF, Santos MG, Nishi MY, Costa MH, Lerario AM, Maciel CC, Mattos GE, Jorge AA, Mendonca BB, Latronico AC 2008 Expression of insulin-like growth factor-II and its receptor in pediatric and adult adrenocortical tumors. J Clin Endocrinol Metab 93:3524 3531

Downloaded from jcem.endojournals.org at Biblioteca Area Biomedica Universita Tor Vergata on July 23, 2009