Abstract P3-06-38: Optical metabolic imaging predicts therapeutic response in breast cancer tumors and organoids

Abstract

Abstract The standard of care for breast cancer patients includes treatment with chemotherapy, antiestrogens, and targeted inhibitors. There are more than 100 drugs approved to treat breast cancers with more in the pipeline of discovery and approval; however, few biomarkers have been identified for individualized, patient-specific treatment planning. Currently, drug regimens are chosen based on the presence or absence of specific receptors, including progesterone receptor (PR), estrogen receptor (ER), and human epidermal growth factor receptor 2 (HER2) and patient response is assessed by changes in tumor size weeks after treatment. Unfortunately, many patients do not initially respond to therapy or develop a resistance, and face a greater risk of recurrence and death. Herein we present a novel, light-based technology that can accurately predict individual tumor response to anti-cancer drugs based on drug-induced changes in metabolism. Cellular metabolism is a powerful indicator response to treatment, because the oncogenic drivers targeted by therapeutic agents often regulate cellular metabolism. In this study, we demonstrate the sensitivity of optical metabolic imaging to predict therapeutic response in xenografts in vivo, and in primary tumor derived organoids. Optical metabolic imaging utilizes mulitphoton fluorescence intensity and fluorescence lifetime imaging to probe the concentrations and protein-binding dynamics of cellular NADH and FAD, two coenzymes of metabolism. A composite endpoint, the OMI index, provides a robust, dynamic readout of cellular metabolism. First, we probed metabolic changes within ER+/HER2+ tumors treated with trastuzumab, an anti-HER2 antibody. After 48h of a single injection of trastuzumab (HER2 antibody), the OMI index of responsive tumors (ER+/HER2+) decreased (p<0.001), but remained unchanged in resistant tumors. To optimize clinical utility of this technology, we developed protocols to generate organoids from fresh biopsy samples. This allows high-throughput screening to directly test individual tumor response to a panel of drugs. We validated this approach in trastuzumab responsive and resistant tumors: the OMI index decreased in responsive ER+/HER2+ organoids treated with trastuzumab, paclitaxel, and XL147 (PI3K inhibitor) (p<0.05). Combination therapies resulted in the lowest OMI index values (p<0.001). Measurements on patient-derived organoids resulted in similar findings: reductions in ER+ tumors treated with tamoxifen and paclitaxel (p<0.05), and the greatest reductions in OMI index observed with combination treatments. The high-resolution images of OMI allow segmentation of the cells within each organoid for cellular-level analysis of heterogeneity. We identified heterogeneity profiles of resistant cell populations within the xenograft and human organoids and were able to track the emergence of resistant sub-populations of cells over the first 72 hours of drug treatment. Heterogeneity analysis is important for clinical utility of this technology to ensure the optimal drug regimen is selected. With these results, optical metabolic imaging shows potential for development into a high-throughput screen to test the efficacy of a panel of drugs to direct clinical therapy selection and expedite pre-clinical studies. Citation Format: Alex J Walsh, Rebecca S Cook, Melinda E Sanders, Carlos L Arteaga, Melissa C Skala. Optical metabolic imaging predicts therapeutic response in breast cancer tumors and organoids [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 P3-06-38.