In situ aromatase expression in primary tumor is associated with estrogen receptor expression but is not predictive of response to endocrine therapy in advanced breast cancer

Abstract

Background: New, third-generation aromatase inhibitors (AIs) have proven comparable or superior to the anti-estrogen tamoxifen for treatment of estrogen receptor (ER) and/or progesterone receptor (PR) positive breast cancer. AIs suppress total body and intratumoral estrogen levels. It is unclear whether in situ carcinoma cell aromatization is the primary source of estrogen production for tumor growth and whether the aromatase expression is predictive of response to endocrine therapy. Due to methodological difficulties in the determination of the aromatase protein, COX-2, an enzyme involved in the synthesis of aromatase, has been suggested as a surrogate marker for aromatase expression.

Methods: Primary tumor material was retrospectively collected from 88 patients who participated in a randomized clinical trial comparing the AI letrozole to the anti-estrogen tamoxifen for first-line treatment of advanced breast cancer. Semi-quantitative immunohistochemical (IHC) analysis was performed for ER, PR, COX-2 and aromatase using Tissue Microarrays (TMAs). Aromatase was also analyzed using whole sections (WS). Kappa analysis was applied to compare association of protein expression levels. Univariate Wilcoxon analysis and the Cox-analysis were performed to evaluate time to progression (TTP) in relation to marker expression.

Results: Aromatase expression was associated with ER, but not with PR or COX-2 expression in carcinoma cells. Measurements of aromatase in WS were not comparable to results from TMAs. Expression of COX-2 and aromatase did not predict response to endocrine therapy. Aromatase in combination with high PR expression may select letrozole treated patients with a longer TTP.

Conclusion: TMAs are not suitable for IHC analysis of in situ aromatase expression and we did not find COX-2 expression in carcinoma cells to be a surrogate marker for aromatase. In situ aromatase expression in tumor cells is associated with ER expression and may thus point towards good prognosis. Aromatase expression in cancer cells is not predictive of response to endocrine therapy, indicating that in situ estrogen synthesis may not be the major source of intratumoral estrogen. However, aromatase expression in combination with high PR expression may select letrozole treated patients with longer TTP.

Trial registration: Sub-study of trial P025 for advanced breast cancer.

Figure 1

Representative pictures of immunohistochemical aromatase staining of primary breast tumor tissue with the antibody #677. (A) WS with high level of aromatase expression. (B) TMA with high level of aromatase expression. (C) WS with low level of aromatase expression. (D) TMA with low level of aromatase expression. (E) WS from a tumor showing heterogeneity in carcinoma cell staining intensity scored as high expression (score value 9). (F) A TMA illustrating discordant low intensity score (score value 1).

Figure 2

Time to progression (TTP) according to high, low and negative expression of COX-2 and Aromatase in primary tumor tissue for patients treated with either letrozole (black) or tamoxifen (grey). (A) TTP according to COX-2 expression: Letrozole treatment, COX-2 high, n = 6; COX-2 low, n = 22; COX-2 negative, n = 13. Tamoxifen treatment, COX-2 high, n = 7; COX-2 low, n = 14; COX-2 negative, n = 11. (B) TTP according to Aromatase expression: Letrozole treatment, Aromatase high, n = 4; Aromatase low, n = 27; Aromatase negative, n = 10. Tamoxifen treatment, Aromatase high, n = 4; Aromatase low, n = 25; Aromatase negative, n = 4.

Figure 3

Time to progression (TTP) according to high, low and negative PR expression in aromatase positive primary tumours from patients treated with letrozole or tamoxifen. (A) Letrozole treatment, Aromatase positive, PR high, n = 16; Aromatase positive, PR low, n = 8; Aromatase positive, PR neg., n = 5. (B) Tamoxifen treatment, Aromatase positive, PR high, n = 9; Aromatase positive, PR low, n = 11; Aromatase positive, PR neg., n = 9.