2611

Background: Use of combined estrogen+progestin (E+P) hormone therapy increases breast cancer risk. Studies evaluating this risk by histologic type suggest that use of E+P is more strongly related to risk of invasive lobular carcinoma (ILC) compared to invasive ductal carcinoma (IDC), however the molecular basis for this difference is largely unknown.

Methods: We conducted a population-based case-control study in western Washington to assess the relationship between E+P use and risk of ILC and IDC. A total of 328 ILC cases, 518 IDC cases, and 469 controls 55-74 years of age were enrolled and completed a detailed in person questionnaire. In addition, tumor tissue specimens from all consenting cases were requested, and thus far pathology review and tumor marker testing has been completed on 229 ILC and 394 IDC cases.

Results: Consistent with previous studies we observed that current use of E+P for 5 years or longer was associated with a 2.5-fold [95% confidence interval (CI): 1.6-4.1] increased risk of ILC, but not with risk of IDC [odds ratio (OR): 1.3; 95% CI: 0.9-2.0]. ILC tumors were more likely to be estrogen receptor positive, e-cadherin negative, and to have low Ki-67, p53, and p21 expression compared to IDC (p<0.0001 for each comparison). Use of E+P was associated with elevated risks of ILC regardless of tumor stage, size, or grade, while E+P use only increased risks of IDC tumors that were early stage, ≤2.0cm, or low grade (ORs ranging from 1.7 to 2.2). When stratified by tumor marker expression, E+P was only associated with risk of ILC tumors that were ER+ (OR=3.3; 95% CI: 1.9-5.8), PR+ (OR=3.7; 95% CI: 2.0-6.7), had low Ki-67 expression (≤15%) (OR=4.4; 95% CI: 2.3-8.2), or had low p21 expression (≤10%) (OR=4.1; 95% CI: 2.2-7.6). In contrast, E+P use was only associated with an increased risk of IDC tumors that had low Ki-67 expression (OR=1.8; 95% CI: 1.1-2.9), but not with IDC tumors expressing other tumor markers.

Discussion: This is the first population-based study involving a centralized pathology review of tumor tissue specimens that has evaluated the relationship between E+P use and risk of ILC and IDC. We confirmed that E+P use is more strongly related to risk of ILC than to risk of IDC. The role of E+P use in the etiology of ILC is further supported by the persistence of this association regardless of tumor stage, size, or grade. Also, our findings that E+P use is only associated with increased risks of ILC with particular tumor marker profiles provides new insight into the molecular basis of this relationship.

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