Non-linear relationship between HER-2 expression and response to trastuzumab in breast cancer cells

Abstract

13083 Background: Currently, eligibility for treatment with the mAb trastuzumab (T) in breast cancer patients is assessed by semiquantitative immunohistochemical (IHC) analysis of HER-2 expression, where 3+ score represents overexpression, and/or by Fluorescence In Situ Hybridization (FISH), which detects HER-2 gene amplification. However, both methods do not measure small but potentially significant differences in protein expression. Methods: We modulated HER-2 expression by RNA interference (siRNA) in the human HER-2-overexpressing breast cancer cell line SKBR3. Protein expression was measured by Western Blot analysis. Cells expressing different levels of HER-2 protein were then treated with T, to assess the mAb effect on cell proliferation and apoptosis. Results: Compared with the parental cell line, SKBR3 cells transfected with 10, 20, and 30 nM HER2 specific siRNAs, showed a reduction of HER-2 protein levels of 40%, 55% and 70%, respectively. Despite up to a 70% decrease in HER-2 expression, transfected cells still scored 3+ at IHC (Dako HercepTest®). Interestingly, this reduction rendered cancer cells more sensitive to T treatment. In fact, we found that sensitivity to T follows a bell shaped curve, depending on different amounts of HER-2. When T was administered at the standard dose of 10 μg/mL, cells showing a 40% reduction of the HER-2 resulted more sensitive to the growth inhibitory effect of T, compared with parental SKBR3 cells and with cells showing lower levels of receptor. Furthermore, these cells were sensitive to T concentrations lower than standard (6 vs 10 μg/mL), which were markedly less effective in cells expressing higher or lower levels of HER-2 receptor. Conclusions: The observation that cells expressing significantly different amounts of HER-2 protein are all scored as “overexpressors” by conventional IHC suggests that quantitative analysis of HER-2 receptor in breast cancer might be more appropriate for the selection of candidates to T treatment. Moreover, we showed that the inhibitory effect of T depends in a non-linear manner on the quantity of HER-2 protein and that the level of HER-2 expression can influence the sensitivity to T, which requires an optimal drug/receptor ratio. No significant financial relationships to disclose.