Adrenocortical cancer and IGF2: is the game over or our experimental models limited?

Abstract

Adrenocortical cancer (ACC) is a rare (0.5–2 cases per million per year) but aggressive disease with low survival rates of 16 to 47%, depending on the clinical series. Although mitotane alone or in combination with chemotherapy can improve patients' survival, there is no efficient treatment of metastatic ACC, which is found in almost half of the patients at diagnosis. Identification of the molecular underpinnings of ACC may open the way to targeted therapy of this aggressive cancer. Over the last 20 years, considerable concerted effort has led to the identification of molecular alterations in patients' samples that may drive the tumorigenic process. One of the breakthroughs in this long process was the discovery by Gicquel et al (1) in 1994 that the growth factor IGF2 was markedly overexpressed in about 80–90% of adrenocortical carcinomas compared with adenomas or healthy adrenals (1). This seminal report was followed by a number of larger-scale studies that confirmed that IGF2 was the most up-regulated gene in ACC. These studies also showed overexpression of the IGF2 and IGF-I receptor (IGF-IR) and of one IGF binding protein (IGFBP2) in adrenal carcinomas (reviewed in Ref. 2). IGF2 is a growth factor that signals through the MAPK and phosphatidylinositol-3-kinase pathways, which stimulate proliferation and inhibit apoptosis in a number of cell types. IGF2 is expressed from the paternal allele at the maternally imprinted 11p15 locus. A possible involvement of IGF2 in adrenal tumorigenesis was initially postulated because of the presence of adrenocortical adenomas and carcinomas in patients with Beckwith-Wiedemann syndrome, a disease resulting from structural alterations at 11p15 that cause overexpression of IGF2 and deregulation of other genes in the locus (reviewed in Ref. 3). In sporadic adrenocortical carcinomas, overexpression of IGF2 is frequently associated with loss of the maternal allele and duplication of the paternal allele, resulting in biallelic expression of the gene (1). Overexpression of IGF2 is a strong independent predictor of shorter disease-free survival (4), and immunohistochemistry for IGF2 has been proposed as an efficient diagnostic tool. Consistent with a potential role of IGF2 in adrenal tumorigenesis, inhibition of IGF-IR signaling by a tyrosine kinase inhibitor (NVP-AEW541) or a specific blocking antibody (IMC-A12) reduced ACC cell proliferation and tumor formation in nude mice (reviewed in Ref. 2). This suggested that IGF signaling could be an interesting therapeutic target in the context of ACC. However, phase I trials aiming at evaluating the feasibility of IGF-IR (Figitumumab and OSI-906) and insulin receptor (OSI-906) inhibition in ACC have only shown one partial response by Response Evaluation Criteria In Solid Tumors (RECIST) criteria (OSI-906 protocol) out of a total of 25 patients (5, 6). The observation that some patients displayed stable disease after 3 months on the 2 protocols could be considered encouraging. This will have to be further evaluated in light of the ongoing advanced clinical trials. Although the robust expression of IGF2 in ACC suggests a possible involvement in tumor initiation and/or progression, this is generally accompanied by deregulation of the whole 11p15 locus, which encompasses potential tumor suppressors such as H19 and P57KIP2 and potential oncogenes such as the recently described micro-RNA hsa-miR483. To elucidate the specific role of IGF2 in adrenocortical tumorigenesis, our team (7) and Hammer's group (8) have used distinct genetic strategies to induce Igf2 overexpression alone or in combination with constitutive Wnt pathway activation in transgenic mice. We achieved overexpression of Igf2 in the adrenal cortex through the regulatory regions of Akr1b7 or CYP11A1. Both promoters allow expression of transgenes in all steroidogenic cells of the adrenal cortex. Hammer's group (8) used a mouse model in which the endogenous Igf2/H19 imprinting control region was specifically deleted in some Sf-1-positive cells in the adrenal cortex, which resulted in stochastic biallelic expression of Igf2. These 3 different genetic models allowed levels of Igf2 expression ranging from 2- to 87-fold of that found in control mice, which was in the range of IGF2 overexpression in patients (7- to 582-fold) (8, 9). However, there was no adrenal tumor development or even hyperplasia in any of the transgenic models over a time course of 14 months, although the IGF signaling pathway was activated (7, 8). These observations are in line with previous reports in mouse models of Beckwith-Wiedemann syndrome and confirm analyses relying on incidental phosphoenolpyruvate carboxykinase-driven overexpression of IGF2 in the adrenal cortex of transgenic mice (10). Interestingly, we showed increased proliferation in the adrenal cortices of Igf2-overexpressing mice, which was in line with the effect of IGF signaling inhibition in human adrenocortical tumor cell lines. However, this