Ki67 modulation by low dose tamoxifen in normal epithelial and myoepithelial cells in women with breast cancer.

A143

Background: A phase II chemoprevention trial showed that tamoxifen 20, 5 and 1 mg per day reduced Ki67 expression in estrogen receptor positive (ER+ve) tumor cells with no difference among doses. Since emerging studies implicate myoepithelium and epithelial-mesenchymal interactions as potential regulators of tumorigenesis, we investigated in the same population the modulation of Ki67 in normal tissue (adjacent and distant from the tumor) to provide further insight into the chemopreventive effect of tamoxifen. Methods and Results : 66 ER+ve breast cancer patients were randomized to 1, 5, or 20 mg of tamoxifen daily for 4 weeks before primary surgery. Concomitant non randomized subjects, 19 ER-ve and 8 ER+ve breast cancer, were used as controls. Ki67 staining was measured in adjacent (within 5 mm from the tumor) and distant (separate paraffin block) epithelium (luminal cells) and myoepithelium. The adjacent tissue was analyzed before and after treatment. Distant tissue, collected only at surgery, was compared with the adjacent tissue at biopsy. There was a general reduction of Ki67 from biopsy to surgery in luminal cells, irrespective of treatment and dose of tamoxifen. There was a statistically significant 2% median reduction for adjacent epithelial cells (Wilcoxon p<0.001), but not for adjacent myoepithelial cells (p=0.17). The only effect in distant tissue was observed in the ER-ve control group, where. Ki67 declined significantly compared to the untreated ER+ve group (OR=0.17, 95%IC, 0.05-0.59, p=0.005). Ki67 expression was always higher in luminal cells as compared to myoepithelial cells (p=<0.001 at biopsy), with a direct correlation between epithelium and myoepithelium both in the adjacent and distant tissue (epithelial at biopsy, r=0.4; epithelial and myoepithelial at surgery, r=0.41 and r=0.45, p<0.001). Conclusions Four weeks after biopsy, Ki67 expression was reduced irrespective of the groups in adjacent epithelium, but not in myoepithelial cells. Within the tamoxifen arms there was no dose- response effect. This data support the concept that low dose tamoxifen is effective as standard dose to reduce cell proliferation, only in luminal cells. The Ki67 staining was higher in epithelial cells compared to the myoepithelial compartment, sustaining the hypothesis that luminal cells are physiologically a proliferative site while mioepithelial cells may have a regulatory function.