Abstract P2-11-20: Evaluation of folate receptor alpha expression in breast cancer by immunohistochemistry: Analysis with three different cut-off points

Abstract

Abstract Background: Folate receptor alpha (FRA) is a newly recognized molecule that is selectively expressed in certain types of solid tumors, and several targeted therapies against FRA have shown promise in clinical trials for treating FRA-expressing ovarian and lung cancer. FRA has been shown to be selectively expressed in ER/PR negative and triple negative (TN) breast cancer; tumor types which would benefit from FRA targeted therapy. With these FRA-targeted therapies becoming available, correctly identifying eligible patients is very important. Currently, the cut off points for FAR expression by immunochemical (IHC) analysis have not been uniform. The commonly used cut points for expression that have been reported included 5%, 10% and the M score (using both staining intensity (3+, 2+, and 1+) and percentage of cells 0-100% (X, Y, Z for each staining intensity) in the equation (M-score = [3X+2Y+1Z]/ 6). The aim of the current study is to investigate these 3 most frequently used cut off points for FRA IHC analysis, in an effort to identify the one that is the most clinically relevant. Methods: 430 cases of infiltrating ductal carcinoma diagnosed between 1997 and 2008 in our institution were identified, reviewed, and 25 blocks of tissue microassays (TMA) were constructed. The association between the expression of FRA by IHC analysis, the clinicopathologic features, expression of ER, PR, HER2 and Ki-67, and the clinical outcome of these tumors were evaluated. Three cut off points (≥5%, ≥10% and M score of ≥10) were compared. 5% and 10% were scored as percent of membrane/cytoplasmic staining of the tumor cells regardless of the staining intensity while the M score takes into consideration both percentage and intensity of the staining of tumor cells.Results: Among the 430 cases studied, 399 cases had informative data from IHC analysis. The positive rates for FRA were 33%, 29% and 17%, respectively, for 5%, 10% and M score 10 as the positive cut off. All three cut off points were associated with significantly higher tumor grades and higher nodal status; they were also associated significantly with ER negativity, PR negativity or high proliferation rates (Ki-67≥15%) (P-values <0.0001 in all), and ER-/PR-, TN, or Basal-like tumors (p-values <0.0001 in all). Also, they were all significantly associated with worse disease-free survival with the p-values of 0.027, 0.026, and 0.009, for 5%, 10% and M score 10, respectively. FRA positive rates with three different cut off points FRA ≥5% (%)FRA ≥10% (%)FRA M-score ≥10 (%)Age >50 (year)32%28%17%Tumor size <2 (cm)28%26%16%Histologic grade 351%47%32%LN status >041%34%20%ER negative77%74%56%PR negative66%63%46%HER2 3+48%37%26%Ki67>15%60%56%41%ER-/PR-77%74%54%Triple Negative82%80%60%Basal-like82%80%59% Conclusion: Our data demonstrates that all three cut off points (5%, 10% and M score 10) could effectively separate breast tumors into subgroups where FRA expression correlated with specific pathologic features and a worse disease-free survival compared with FRA negative tumors. Using an M score of >10 as the cut off for FRA may be most specific method for determining the prognostic significance of FRA expression in breast cancer. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-11-20.