Abstract P1-01-06: Quantitative breast DCE-MRI risk biomarkers and breast tumor immunohistochemistry phenotypes: A preliminary assessment

Abstract

Abstract Purpose Fibroglandular tissue (FGT) and background parenchymal enhancement (BPE) estimated from breast dynamic contrast enhanced MRI (DCE-MRI) have been correlated with breast cancer risk. We performed a preliminary study assessing in a small cohort of breast cancer patients the relationship between FGT and BPE both quantified from cancer-unaffected breasts and breast tumor’s immunohistochemistry phenotypes, including testing of estrogen (ER), progesterone (PR), and human epidermal growth factor receptor 2 (HER-2/neu) receptors. Materials and Methods We retrospectively identified 51 breast cancer patients who were diagnosed with unilateral breast cancer from 2009-2011. Six out of the 51 were excluded for analysis due to missing data in ER, PR, or Her-2 testing in their pathology reports. The 45 cancer patients (mean 47.3±7.6 YO; 24 IDC, 4 ILC, 16 mixed of IDC and DCIS, and 1 papillary carcinoma) consist of 25 premenopausal and 20 postmenopausal women. Breast DCE-MRI scans acquired at time-of-diagnosis were analyzed using fully automated computer algorithms. From the contralateral cancer-free breast, four quantitative variables were computed for each patient: the absolute volumes of FGT and BPE (i.e., |FGT| and |BPE|) and their relative amount over the whole-breast volume (i.e., FGT% and BPE%). BPE were assessed over the FGT region of the breast, and separately from each of the three sequential DCE post-contrast-subtracted sequences (i.e., SUB1, SUB2, and SUB3). Multivariable logistic regressions were performed to assess the quantitative FGT/BPE variables for distinguishing/predicting the status (+/-) of ER, PR, and HER-2 testing. Area under the curve (AUC) of the receiver operating characteristic (ROC) was used to evaluate the regression models’ discriminative performance. Odds ratios (ORs) were reported with 95% confidence intervals (CIs). Results In this cohort, the most common phenotypes were ER+/PR+ (14/45 each HER2+ and HER2-) and triple negative (9/45). Premenopausal status was associated with tumors that were HER-2+ (AUC=0.66; p=0.04), ER+ (AUC=0.73; p=0.02), and PR+ (ACU=0.66; p=0.04). FGT% (AUC=0.74; p=0.017; OR=5.3 [95% CI: 1.3, 21.1]) and BPE%-SUB3 (AUC=0.69; p=0.02; OR = 1.4 [95%CI: 1.1, 1.9]) were discriminative of HER-2 status; after adjusting for menopausal status, FGT% remained the association (AUC=0.77; p=0.04; OR=4.4 [95% CI: 1.1, 18.5]) but BPE%-SUB3 did not (p=0.06). BPE%-SUB1 was discriminative of PR status (AUC=0.70; p=0.03; OR=2.3 [95%CI: 1.1, 4.8]) but did not contribute to predict when adjusting for menopausal status (AUC=0.75; p=0.06). All FGT and BPE variables were not associated (p>0.05) with ER status. Conclusions In this preliminary study, FGT% and BPE% were associated with HER-2 and PR status, and none of FGT and BPE was associated with ER status, of breast tumor. Clinical Relevance Quantitative MRI variables of FGT and BPE that are predictive of breast cancer risk are related to breast tumor immunohistochemistry phenotypes. While FGT and BPE cannot replace the testing of these phenotypes, they may contribute to assess the risk of molecular subtypes of breast cancer, with potential implications for improving breast cancer screening strategies. Citation Format: Shandong Wu, Margarita L Zuley, Brenda F Kurland, Rachel C Jankowitz, Jules Sumkin, David Gur. Quantitative breast DCE-MRI risk biomarkers and breast tumor immunohistochemistry phenotypes: A preliminary assessment [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-01-06.