Abstract 5576: Isolation and detection of breast cancer cells in blood samples using antibody cocktails selected by large-scale genome-wide screening

Abstract

Abstract Screening and prevention is critical to reducing breast cancer deaths, and mammography is not an ideal tool. Mammography has not reduced invasive breast cancer incidence, it has high numbers of false positives and it detects many non-invasive cancers that are not life threatening. Metastatic breast cancer is the second leading cause of cancer death for women. The development of a new blood test for breast cancer screening could offer significant improvements with the potential to be more accurate, less costly, less intimidating, not subject to age differences, and able to discriminate between cancers subtypes. Detection of breast cancer cells (BCC) in blood samples offers a valuable opportunity for screening. However, BCC in blood are rare and must be identified against a background of millions of other blood cells. Although many technologies have been developed for the isolation and detection of circulating tumor cells using EpCAM and cytokeratins, these tests do not provide levels of sensitivity and specificity required for a cancer screening tool. To address this critical unmet need, we developed a panel of cell surface markers whose expression patterns specifically distinguish breast cancer cells in blood samples. These genes were selected based on a discovery platform tailored to detect early stage breast cancers and include 20 novel putative cell surface molecules that are candidates for breast cancer cell detection and isolation. We prioritized these molecules by specificity of expression in tumor cells including: 1) Ratio >25:1 average breast tumor biopsy expression versus average normal blood expression, 2) High expression relative (>3 fold) in >20% of tumor biopsy samples, 3) High expression (>3 fold) in >40% of breast cancer cell lines, and 4) Expression significantly associated with 10-year outcomes (Kaplan Meier analysis, p<0.05). We combined the top 10 ranking genes in each of these categories, resulting in a list of 20 candidates that represent high priority targets. Antibodies for all 20 genes are commercially available. Expression in micro dissected biopsies was markedly higher than non-breast cancer blood for most. Also, most were expressed at lower levels than EpCAM in normal breast tissue and had background level expression in isolated normal blood cells. The innovation of this strategy is twofold. First, it uses a panel of molecules instead of an individual molecule to detect and isolate breast cancer cells and, second, it relies on molecules that are clinically relevant to breast cancers, rather than relying on generic epithelial cell markers. Current studies include the validation of antibody cocktails with specificity and sensitivity that outperforms the current EpCAM enrichment method and provides a proof of feasibility for screening of breast cancer. Citation Format: Lucas Delmonico, Edward C. Goodwin, Marcia Vasconcellos Fournier. Isolation and detection of breast cancer cells in blood samples using antibody cocktails selected by large-scale genome-wide screening [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5576.