Within a multicentric, double-blind, placebo-controlled phase II trial, postmenopausal women with primary breast cancer were randomized to either 60 or 600 mg daily of raloxifene or placebo administered for 2 weeks prior to surgery. Circulating levels of sex-hormone binding globulin (SHBG), dehydroepiandrosterone-sulfate (DHEA-S), estrone (E1), estrone-sulfate (E1-S), estradiol (E2), and C-terminal telopeptide (CTX), were determined at baseline and the day before surgery in 37 women enrolled at the European Institute of Oncology in Milan. Raloxifene had a statistically significant different effect compared with placebo on SHBG (p<0.001) and E2 (p=0.03), without any dose-response relationship. Women on raloxifene (n=26) showed an increase from baseline of 10.7% (inter-quartile range: 5.9; 24.2) in SHBG and of 1.3% (inter-quartile range: -8.0; 24.5) in E2, whereas women on placebo (n=11) showed a decrease by 15.5% (inter-quartile range: 7.9; 18.1) and 17.3% (inter-quartile range: 6.5; 40.0), respectively. No significant differences were found on the other variables. A positive correlation was observed between DHEA-S and E1-S (p=0.001) or E2 (p<0.001), while SHBG correlated negatively with E1-S (p=0.024) and E2 (p=0.01), and positively with DHEA-S (p=0.016) and CTX (p=0.040). Our results provide evidence for an early endocrine effect of raloxifene which further suggests a favorable impact on breast cancer prevention.