Objectives: Previous studies have sugested that early changes in Ki-67 expression following treatment with aromatase inhibitors (AIs) may be better predictors of long term benefits than tumour volume reduction. Previous work with RNA microarrays analyzing AI-treated post-menopausal ER(+) breast cancers identified novel candidate markers that could improve Ki-67 performance. Here we aimed to validate these markers in a breast cancer mouse model.

Methodology: Following 6 weeks growth of a subcutaneous implant of MCF-7 cells transfected with aromatase (AROM-1 cells) in ovariectomized athymic female mice supplemented with androstenedione we divided mice into a control group and two Letrozole treatment groups with sub-optimal (T1, 2mg/kg/day) and optimal (T2, 20mg/kg/day) oral dosage. We extracted RNA from Fine Needle Aspirates of established tumors at the beginning and and 1 week after treatment. Using qRT-PCR we determined changes in the expression of ten genes: Ki-67, CyclinD1, pS2, TFF3, TOP2α, UBE2C, PDZK1, STC2, MAN1A1 and FAS.

Results: In Arom-1 breast tumor xenografts four genes: CCND1, pS2, TFF3 and SCT2 showed significant downregulation after Letrozol treatment (Table 1). We found modest differences in Ki-67, FAS, UBE2C and PDZK1 expression levels and no changes were observed in the remaining markers. The Mann-Whitney U test showed significant differences only for pS2 expression levels between sub-optimal and optimal oral dosage.

Table 1. Expression levels of candidate markers after Letrozole (T1 and T2) treatment in FNA of established tumors.

Conclusion: Cyclin D1, pS2, TFF3 and Stanniocalcin-2 expression levels merit clinical validation as predictors of long term benefit with AI.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA