Visscher, Daniel W.

Abstract Purpose: The evidence on associations of mammographic breast density with breast cancer risk by combination of tumor aggressiveness features is limited. We examined associations of breast density phenotypes with risk of aggressive breast tumor features by menopausal status, and current postmenopausal hormone therapy. Methods: This study included 2,635 invasive breast cancer cases and 4,059 controls from participants of four nested case-control studies within four established cohorts: the Mayo Mammography Health Study, the Nurses Health Study, Nurses Health Study II, and San Francisco Mammography Registry. Percent breast density, absolute dense and non-dense areas were assessed from digitized film-screen mammograms using a computer-assisted threshold technique and standardized across studies. We used polytomous logistic regression to quantify the associations of breast density measures with risk of breast tumor aggressiveness (defined as presence of 2 or more of the following tumor characteristics: size ≥2cm, grade 2 or 3, or positive nodes), stratified by menopausal status and current hormone therapy (i.e., premenopausal, postmenopausal/estrogen therapy, postmenopausal/combined therapy, and postmenopausal/no hormones). We also evaluated differences in the strength of associations across categories. In a secondary analysis, we examined these associations while excluding cases with mammogram date within 2 years of diagnosis. Results: Positive associations of percent density and dense area and inverse associations of non-dense area with breast cancer risk were stronger in aggressive vs. non-aggressive tumors (OR=2.62, 95%CI 2.08-3.31 vs. OR=1.94, 95%CI 1.62-2.33 for percent density≥51% vs. 11-25%, p-heterogeneity=0.001; OR=1.89, 95%CI 1.54-2.31 vs. OR=1.65, 95%CI 1.41-1.93 for dense area 4th vs. 2nd quartile, p-heterogeneity=0.015; OR=0.56, 95%CI 0.44-0.72 vs. OR=0.71, 95%CI 0.59-0.86 for non-dense area 4th vs 2nd quartile, p-heterogeneity=0.007, respectively). These patterns were similar across all menopausal and hormone therapy groups (P-interactions=0.62, 0.76, and 0.23, for percent density, dense area and non-dense area, respectively). Excluding cases diagnosed within 2 years of mammography resulted in similar findings. Conclusion: Mammographic density phenotypes were more strongly associated with aggressive cancer (having two or more of the following: size ≥2cm, grade 2 or 3, or positive nodes) vs. non-aggressive types of breast cancer across categories of menopause and hormone therapy types. Citation Format: Lusine Yaghjyan, Rulla Tamimi, Kimberly Bertrand, Christopher G. Scott, Matthew R. Jensen, Shane Pankratz, Kathleen Brandt, Daniel Visscher, Aaron Norman, Fergus Cough, John Shepherd, Bo Fan, Yunn-Yi Chen, Lin Ma, Andrew H. Beck, Steven R. Cummings, Karla Kerlikowske, Celine Vachon. Interaction of mammographic breast density with menopausal status and postmenopausal hormone use in relation to the risk of aggressive breast cancer subtypes. [abstract]. In: Proceedings of the AACR Special Conference: Improving Cancer Risk Prediction for Prevention and Early Detection; Nov 16-19, 2016; Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2017;26(5 Suppl):Abstract nr B27.

Abstract Introduction: Atypical hyperplasia (also known as proliferative disease with atypia), while categorized as a benign finding, has been consistently associated with an increased risk of subsequent breast cancer that persists for decades after initial diagnosis. There are various other lesions that are classified as benign breast disease (BBD) that are routinely identified by pathologists that may also increase risk and be of use to inform breast cancer risk models. We sought to identify which BBD lesions were associated with increased risk of breast cancer in an African American (AA) cohort. Methods: Benign breast biopsies from 3,895 AA women diagnosed with BBD between 1997 and 2010 in metropolitan Detroit were reviewed for 12 benign features including columnar alterations, ductal ectasia, ductal hyperplasia, fibrosis, apocrine metaplasia, lobular hyperplasia, calcifications, cysts, intraductal papilloma, radial scar, sclerosing adenosis, and fibroadenomas. Women were followed for subsequent breast cancer using the Metropolitan Detroit Cancer Surveillance System, part of the Surveillance, Epidemiology, and End Results (SEER) cancer registry. Associations between BBD features and subsequent breast cancer were examined using chi-square tests. Features that had a significant chi-square test result were also combined into scores describing the overall number of BBD features identified, termed busy breast score. Multivariable log-binomial regression with backward selection based on BIC criteria was performed to assess risk ratio and 95% confidence intervals of breast cancer on 12 candidate features. Multivariable log-binomial regression was performed for busy breast score as well. All regression models were adjusted for age at biopsy and overall impression for presence of proliferative disease with or without atypia using non-proliferative disease as the reference. Models were checked for multicollinearity using a variable inflation factor (VIF). Results: Of the 3,895 AA women in the BBD cohort, 210 developed a subsequent breast cancer. The median age at biopsy among those without a subsequent cancer (controls) was 47 years of age, while cases were 53 years of age (p-value<0.001). In univariate analyses, ductal hyperplasia (p-value=0.001), lobular hyperplasia (p-value<0.001), calcifications (p-value<0.004), cysts (p-value=0.003), intra-ductal papilloma (p-value=0.005), sclerosing adenosis (p-value=0.002), columnar alterations (p-value<0.001) and radial scar (p-value=0.001) were associated with subsequent breast cancer. Fibroadenomas were inversely associated with subsequent breast cancer (p-value<0.001). In multivariable model after backwards selection, presence of columnar alterations was associated with increased risk of breast cancer, (RR=1.57, 95% CI: 1.16, 2.13). Combining the 9 BBD features associated with subsequent breast cancer in univariate analysis, those with a busy breast score of 5+ was associated with 2.5 times increased risk compared to those with none of these features (RR=2.52, 95% CI: 1.41, 4.49). As a continuous variable for busy breast score, each additional feature conferred a 20% increase in risk of breast cancer (RR=1.20, 95% CI: 1.10, 1.31). Conclusions: Columnar alterations confer increased risk of breast cancer beyond the risks associated with atypical hyperplasia, as does the presence of multiple types of BBD lesions in a single biopsy. These estimates may help improve the current risk assessment models for African American women and highlight the need for additional research regarding the utility of closer surveillance and potentially chemoprevention for reduction of breast cancer in these women. Citation Format: Michele L. Cote, Wei Chen, Julie J. Ruterbusch, Eman Abdulfatah, Visakha Pardeshi, Asra N. Shaik, Marcel T. Ghanim, MHD Fayez Daaboul, Daniel W. Visscher, Sudeshna Bandyopadhyay, Rouba Ali-Fehmi. Benign breast disease and subsequent breast cancer risk: The Detroit cohort. [abstract]. In: Proceedings of the AACR Special Conference: Improving Cancer Risk Prediction for Prevention and Early Detection; Nov 16-19, 2016; Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2017;26(5 Suppl):Abstract nr A25.

Abstract Introduction: The majority of the 1.6 million breast biopsies performed annually in the United States are benign; however, several breast cancer risk models consider a benign biopsy a factor that increases risk of subsequent breast cancer. Fibroadenomas (FA), benign tumors of epithelial and stromal tissue, occur frequently in premenopausal women and more frequently in African American (AA) women than European American (EA) women. A small increased risk of subsequent breast cancer due to FA has been reported in some studies of EA women. We sought to investigate whether the risk of this lesion differs for AA women. Methods: Benign breast biopsies from 3895 AA women diagnosed between 1997 and 2010 in metropolitan Detroit were reviewed for 12 benign features including FA, ductal ectasia, fibrosis, apocrine metaplasia, ductal hyperplasia, lobular hyperplasia, calcifications, cysts, intraductal papilloma, radial scar, sclerosing adenosis, and columnar alterations. These features were also used to categorize FA into simple and complex FA, where complex FA occurs when FA is accompanied by at least one of the following features: cysts, calcifications, apocrine metaplasia or intraductal papilloma. Women were followed for subsequent breast cancer using the Detroit Surveillance, Epidemiology, and End Results (SEER) cancer registry. Associations between FA and other benign lesions were examined using chi-square tests. Risk of breast cancer was estimated by relative risk ratios and 95% confidence intervals calculated using logistic regression. All models were adjusted for age at biopsy and additionally adjusted for presence of proliferative disease with or without atypia. Results: Of the 3895 AA women, 46.5% presented with FA on biopsy. FA occurred more frequently in biopsies of younger women (p-value < 0.001) and were less likely to include the other benign lesions assessed (all p-value < 0.001). Of the 210 identified breast cancers that developed in this cohort, 72 cancers developed in women with FA on biopsy. FA were associated with a reduction in risk of breast cancer after adjusting for age at biopsy (RR=0.64, 95% CI: 0.48, 0.86) compared to women without FA on biopsy. Reduction in risk was somewhat attenuated when analysis was adjusted for proliferative disease with or without atypia (RR = 0.67, 95%CI: 0.48, 0.936). Similar associations were seen with simple FA (RR =0.61, 95% CI: 0.35, 1.01) and complex FA (RR =0.70, 95% 0.35, 1.01) after adjustment for age at biopsy and proliferative disease with or without atypia. Conclusions: FA are negatively associated with other benign breast disease features. Risk of breast cancer may be reduced in women with FA compared to women with other types of benign lesions. These findings have important implications for modeling breast cancer risk particularly among AA women for whom FAs are common. Citation Format: Asra N. Shaik, Julie J. Ruterbusch, Eman Abdulfatah, Marcel T. Ghanim, MHD Fayez Daaboul, Visakha Pardeshi, Rouba Ali-Fehmi, Daniel W. Visscher, Sudeshna Bandyopadhyay, Michele L. Cote. Fibroadenomas on benign breast biopsy and subsequent breast cancer risk in an African American cohort. [abstract]. In: Proceedings of the AACR Special Conference: Improving Cancer Risk Prediction for Prevention and Early Detection; Nov 16-19, 2016; Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2017;26(5 Suppl):Abstract nr A16.

1043 Background: In estrogen receptor (ER) positive MBC, mutations (e.g. ESR1), identified from tumor biopsies or cfDNA, confer resistance. The concordance between mutations observed in tumor and cfDNA and the implications for response to Z-endoxifen, a potent anti-estrogen, are unknown. Methods: We previously conducted a phase I trial of Z-endoxifen in endocrine refractory, ER positive MBC. Seven dose levels were considered ranging from 20 to 160 mg/day followed by expansion cohorts (EC) of 40, 80, and 100 mg/day. Pretreatment blood samples (all pts) and fresh tumor biopsies (EC) were collected prospectively. Tumor and cfDNA were evaluated by targeted NGS. Results: 41 pts (38 evaluable) were enrolled. Prior endocrine therapy included aromatase inhibitors (37/38, 97%), fulvestrant (22/38, 58%) and tamoxifen (26/38, 68%). Substantial endoxifen exposure without DLTs at doses above 80 mg/day led to a halt in dose escalation and opening of the EC. Overall clinical benefit (stable > 6 months [7 pts.] or partial response by RECIST criteria [3 pts.]) rate was 26.3% (95%CI: 13.4-43.1%). cfDNA was obtained from 36 pts and mutations were identified in 13 (36%) including ESR1 [ Y537N or D538G] (5), PIK3CA [ H1047R or E542K] (8), TP53 [ K132R, R248Q, R267Q, or H179Y] (4), AKT ( Q79K) (1), and KRAS ( G12D) (1). In 5 pts with cf ESR1 mutations, 4 had additional cfDNA mutations including PIK3CA (3), and TP53 (1). PFS was shorter in pts with cfDNA mutations relative to those without (median: 61 vs.132 days; log-rank p = 0.021). Discordance was observed between tumor and cfDNA mutations where 3/7 PIK3CA tumor mutations were detected by cfDNA, 1/2 TP53 tumor mutations were detected by cfDNA, and 0/1 AKT tumor mutations were detected by cfDNA. Conversely, 2 pts had cfDNA mutations (either ESR1, TP53 or AKT) undetected in tumor. Conclusions: The absence of cfDNA mutations in patients with endocrine resistant, MBC treated with Z-Endoxifen was associated with significantly longer PFS. Given the discordance between tumor and cfDNA sequence data, future studies must determine which approach maximizes prognosis and prediction of benefit for estrogen-targeted therapy. Clinical trial information: NCT01327781.

533 Background: Current guidelines define TNBC as complete absence of estrogen (ER) and progesterone receptor (PR), without HER2 amplification. However, the prognostic impact of clinical and histopathological factors, sTILs, and AdjCT in TNBC meeting these strict criteria is unknown. Methods: From a cohort of 9985 women who underwent upfront surgery for M0 breast cancer (BC) at Mayo Clinic Rochester from 1985-2012, 1159 pts with ER negative or low (≤10%) BC were identified for central ER/PR/HER2 staining and HER2 FISH (IHC2+ only) to select those with TNBC by modern definitions. Cox proportional hazards models were used to assess the impact of clinicopathological variables on invasive disease-free (IDFS) and overall survival (OS). Results: Tumors from 605 pts (median age 56.3 yrs) met criteria for TNBC (ER < 1%, PR < 1% and HER2 0, 1 or 2+ and FISH negative). 51% were T1, 65% N0, 88% grade 3, and 75% had Ki67 > 15%. Histologically, 39% were anaplastic, 26% invasive ductal, 16% medullary, 8% metaplastic, 6% apocrine and 5% others. Median sTILs was 20% (0-90%). 55% pts received AdjCT [21% anthracycline (A), 19% A + taxane, and 15% other]. Median follow-up for IDFS and OS were 7.4 and 10.6 yrs, respectively. Multivariate analyses demonstrated that higher N stage (p < 0.01), lower sTILs (p = 0.01) and no AdjCT (p < 0.01) were independently associated with worse IDFS and OS. Histology (medullary subtype) was associated with better IDFS in univariate (HR 0.56, 95% CI, 0.35-0.89) but not in multivariate analyses, once sTILs were accounted for. Among systemically untreated pts (n = 182), higher N (p < 0.01) and lower sTILs (p = 0.04) were associated with worse IDFS. For systemically untreated T1N0 pts (n = 111), the 5-yr IDFS was 70% (95% CI, 61-81) [T1a: 83% (95% CI, 63-100), T1b: 68% (95% CI, 52-88) and T1c: 67% (95% CI, 55-83)], compared to 78% (95% CI, 68-84) for T1N0 pts treated with AdjCT. Conclusions: In TNBC pts, N stage, sTILs and receipt of AdjCT were independently prognostic for IDFS and OS. sTILs remained prognostic for IDFS in systemically untreated TNBC. In N0 TNBC, the risk of recurrence or death was substantial in the absence of chemotherapy, even for those with T1 tumors.