Defining how oestrogen influences the anti-tumour effects of adjuvant bisphosphonates using in vivo models of breast cancer

Bone is the most common site for breast cancer to metastasise to. Clinical trials have demonstrated that adding adjuvant zoledronic acid (Zol) to standard of care therapy reduces the risk of breast cancer relapse in bone and other organs in post-menopausal women. Whereas pre- and peri-menopausal women experience reduced bone metastasis but increased recurrence outside of bone. Serum analysis of the AZURE clinical trial has found a correlation between the concentration of oestradiol and the possible outcome following adjuvant therapy with Zol.
I hypothesise that Zol reduces bone metastasis via inhibition of bone resorption and further oestrogen driven changes to the bone microenvironment drives tumour cells to other sites. I aimed to establish the mechanism by which Zol interacts with oestrogen to affect tumour growth in bone and non-bone sites.
Methods: 12-week-old female BALB/c and BALB/c nude mice were ovariectomised (OVX) before receiving 0, 1.38 or 12.5 mg/L oestradiol to model pre-, peri- and post-menopausal conditions. All animals received 40ug/day Goserelin to prevent OVX induced FSH and animals were randomised to 100 µg/kg/week Zol or control. For tumour studies, MDA-MB-231-Luc2 or 4T1-Luc2 cells were administered via intra-cardiac injection, 3 and 4-days after OVX respectively for metastatic models or directly into the mammary fat pad for a model of sprontanesous metastasis from the primary site. A clinically relevant dose of 100 µg/Kg Zol was administered weekly until end of experiment. Bone volume was measured by µCT, bone turnover and oestradiol/FSH were measured by ELISA. Histomorphometric analysis was carried out using Osteomeasure. Numbers of tumour and tumour growth was monitored by IVIS-Luc2 system and analysed using LUMINA II software.
Results/conclusions: Administration of 0, 1.38 or 12.5 mg/L oestradiol, to OVX mice, for 2-weeks resulted in serum oestradiol concentrations of 13±10 pg/ml, 49±18 pg/ml and 153±18 pg/ml, which mimic those found in pre-, peri- and post-menopausal women respectively. Oestradiol leads to increased bone volume, which is further increased in the presence of Zol (p=0.001 and p=0.036, respectively, observed in pre-menopausal concentration of oestradiol). In accordance with increased bone mass, increased numbers of osteoblasts and simultaneously decreased osteoclasts were observed. These effects were reversed when oestradiol levels were decreased due to OVX; Effects of oestradiol on the bone microenvironment are strain-dependant. Similarly, Zol on osteoblasts varies between mouse model whereas it consistently decreases number and activity of osteoclasts across the different studies. Oestradiol did not alter the overall number of metastatic tumours in both immunocompromised and immunocompetent mouse model (p=0.595 & p=0.488, respectively), whereas Zol had a tendency towards a decrease in number of tumours in the immunocompromised model. In immunocompetent BALB/c mice, oestradiol did not alter tumour burden in the primary tumour model but in the metastatic model, post-menopausal concentrations of oestradiol increased tumour burden (p=0.012). Addition of Zol caused an increase of tumour burden in the primary tumour site under pre-menopausal concentrations of oestradiol (p=0.0002), and Zol decreased tumour burden in post-menopausal concentration of oestradiol in a metastatic model (p=0.080), indicating that the protective role of Zol is only observed under post-menopausal concentrations of oestradiol as observed in clinical studies.
Reduction of oestradiol as observed during the menopausal transition alters the bone cellular landscape by increasing bone resorption. Despite Zol affecting osteoclastic activity and cell number, bone mass did not return to pre-menopausal levels. Post-menopausal levels of oestradiol increased tumour burden in immunocompetent mice only. Zol reduced tumour burden in the post-menopausal group but not pre-menopausal group in the immunocompetent model. These results suggests that more complex cellular interactions apart from the osteoblast/osteoclast/tumour cells may be involved under the menopausal transition to produce the differential effect of Zol.