High hopes for RANKL: will the mouse model live up to its promise?

The steroid hormones, estrogens and progesterone are key drivers of postnatal breast development and are linked to breast carcinogenesis. Experiments in the mouse mammary gland have revealed that they rely on paracrine factors to relegate their signal locally and to amplify it. In particular, RANKL is a key mediator of progesterone action. Systemic inhibition of RANKL blocked proliferation in the mammary epithelium with potential clinical implications: a RANKL-inhibiting antibody, Denosumab (Amgen), has been approved by the US Food and Drug Administration for osteoporosis treatment. Two publications now provide evidence that progestin-driven mouse mammary tumorigenesis can be blocked by ablating RANK signaling. Can the osteoporosis drug help breast cancer patients? The burning question now is whether the role of this pathway is conserved in the human breast and whether RANKL signaling has a role in the pathogenesis of one or more subtypes of breast cancer.

subsequently shown to control bone remodeling by inducing osteoclast diff erentiation [4]. Analysis of RANK-and RANKL-defi cient mice pointed to a role for this pathway in the mammary gland during pregnancy [5] and in mammary carcinoma metastasis to the bone [6].
RANKL expression in the mammary epithelium is controlled by progesterone [7,8]. Recent work identifi ed RANKL as a key paracrine mediator of progesteroneinduced mouse mammary epithelial cell proliferation [9,10] and implicated the cytokine in stem cell control [11,12]. Systemic inhibition of RANKL signaling by intravenous injection of recombinant osteoprotegerin, its decoy recep tor, blocked progesterone-induced prolifera tion in the mammary epithelium, suggesting RANKL may be used as a therapeutic target in the mammary gland [9].
Schramek and colleagues [13] and Gonzalez-Suarez and colleagues [14] now induced mammary carcinomas in mice using the progestin medroxyprogesterone acetate (MPA) and the mutagenic agent 7,12-dimethylbenz(a) anthracene (DMBA) and demonstrate that RANKL is a key factor in this process. Over-expression of RANK by means of a mouse mammary tumor virus (MMTV)driven transgene accelerated hyperplasia and tumor forma tion [14] whereas pharmacological inhibition of RANKL [13,14] and genetic inactivation of RANK in the mammary epithelium [14] decreased incidence and delayed onset of tumorigenesis in this system.
To investigate whether RANKL is similarly important in the pathogenesis of other tumor types that arise independently of exogenous hormones, Gonzalez-Suarez and colleagues used the MMTV-neu transgenic mouse model. Interestingly, in ErbB2-driven carcinogenesis, pharmacological inhibition of RANKL also reduces tumor growth and lung metastasis, suggesting that the RANK signaling pathway may be of functional relevance in a wider tumor spectrum. Th is is of great clinical interest given the heterogeneity of human breast cancer.
Schramek and colleagues observe that RANK deletion does not alter the incidence of mammary cancer in MMTV-neuT transgenic mice. It is conceivable that the apparent discrepancy results from Keratin5-cre RANK deleted cells being outgrown by MMTV-neuT cells that have kept the wild-type RANK allele.

Abstract
The steroid hormones, estrogens and progesterone are key drivers of postnatal breast development and are linked to breast carcinogenesis. Experiments in the mouse mammary gland have revealed that they rely on paracrine factors to relegate their signal locally and to amplify it. In particular, RANKL is a key mediator of progesterone action. Systemic inhibition of RANKL blocked proliferation in the mammary epithelium with potential clinical implications: a RANKL-inhibiting antibody, Denosumab (Amgen), has been approved by the US Food and Drug Administration for osteoporosis treatment. Two publications now provide evidence that progestin-driven mouse mammary tumorigenesis can be blocked by ablating RANK signaling. Can the osteoporosis drug help breast cancer patients? The burning question now is whether the role of this pathway is conserved in the human breast and whether RANKL signaling has a role in the pathogenesis of one or more subtypes of breast cancer. Importantly, Gonzalez-Suarez and colleagues demonstrate that Fc-RANK inhibits proliferation of normal mammary epithelium and hyperplasias, but not in carcinomas, suggesting RANKL inhibition would be useful in early stages of tumorigenesis and/or as a preventive drug.
Can these fi ndings be extrapolated to humans? Dissociation of hormone receptor expression and proliferation in the adult mammary gland is a phenomenon that is conserved between human and rodents (reviewed in [15]), suggesting that paracrine signals are also of relevance to the human breast. Caution is required as it is unclear whether the paracrine circuitry is shared between the two species or whether the human breast may rely on distinct factors.
Immunostainings of human breast carcinomas reveal that RANKL is expressed in 11% of human tumors. Th ere is no evidence for colocalization within the epithelium for receptor and ligand but expression is found in some stromal cells [14]. Th is raises the possibility that the scenario is more complex in human breast cancer than in the present mouse tumor models with infi ltrating immune cells providing the ligand. In any case, this drug seems too palatable not to try! Th e challenge lies in designing clinical trials with short endpoints and identifying adequate preclinical models to identify swiftly which particular patients may benefi t from the drug.