Abstract Background: In adjuvant trials bisphosphonates, including zoledronic acid (ZOL), have been shown to reduce recurrence rates in women who are postmenopausal (natural or medically induced) but appear to have no benefit in pre or peri-menopausal women. Activin and TGFβ act as tumour suppressors in early breast cancer, and the ovarian hormone inhibin and the paracrine glycoprotein follistatin, can inhibit their activity. Through the suppression of bone resorption, bisphosphonates also inhibit release of activin and TGFβ from bone. We investigated the effect of ZOL on serum levels of reproductive hormones within the TGFβ superfamily of growth factors, in pre- and postmenopausal breast cancer patients (pts) receiving neo-adjuvant chemotherapy ± ZOL. Methods: 40 patients (premenopausal n=18, postmenopausal n=22) with large operable or locally advanced breast cancer were included in a randomised phase II feasibility study investigating the biological effects of neoadjuvant FEC chemotherapy alone (CT) or with ZOL 4mg administered on Day 2 during cycle 1 (CT+ZOL). Randomisation was stratified for menopausal status. TGFβ1, activin A and the activin inhibitors follistatin and inhibin A were measured on stored serum collected on days 0, 5 and 21 of the first chemotherapy cycle. Statistical analysis was performed using SPSS v13. Results: In postmenopausal pts, median baseline follistatin levels were 1596 and 1454pg/ml in the CT+ZOL and CT groups respectively. Median follistatin levels fell in the CT+ZOL group at both day 5 (1102pg/ml) and day 21 (1085pg/ml), compared with an increase in the CT group to 1544pg/ml on day 5 and 1689pg/ml on day 21. Using a linear regression model, an interaction was demonstrated between menopausal status and treatment group for follistatin (p = 0.051). This differential effect with addition of ZOL was not seen in pre-menopausal patients (baseline 1771pg/ml CT+ZOL, 1484pg/ml CT; day 5 1852pg/ml CT+ZOL, 1594pg/ml CT; day 21 1337pg/ml CT+ZOL, 1344pg/ml CT). No significant interaction between menopausal status and treatment group was demonstrated for activin A, TGFβ1 and inhibin A. As expected, menopausal status significantly influenced inhibin A levels, with post-menopausal women experiencing much lower levels (mean difference in AUC (pre-post): 16.9 pg/ml, 95% CI: 9.1 to 24.6, P < 0.001). Conclusion: In this small pilot study, ZOL in combination with FEC chemotherapy appeared to have a differential effect on the activin inhibitor follistatin, according to menopausal status. Therefore ZOL would be expected to further increase the bioavailability of the tumour suppressor activin, in a postmenopausal population with already established low levels of the activin inhibitor, inhibin A. This could potentially contribute to the beneficial effects on extra osseous recurrence rates seen in the AZURE, ABCSG-12 and ZO-FAST trials of ZOL. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-02-04.