Determining Tumor Heterogeneity and Response to Treatment Ex Vivo Using 3D Mammosphere Cultures Derived from Primary Breast Cancers and Metastases
Hannah L Gilmore, Kristy Miskimen, Analisa Difeo, Lyndsay N Harris. Seidman Cancer Center, Cleveland, OH
Background: One of the major limitations to using commercially available breast cancer (BC) cell lines for research is that they are not accurate representations of the tumor heterogeneity that exists in patients with newly diagnosed disease. In addition, the response of commercial cell lines to current BC therapies does not give an accurate portrayal of patient response in vivo. The studies described here focus on growing fresh BC tissue in 3D cultures called mammospheres. It is believed that 3D cultures better reflect the in vivo growth properties of BC and retain tumor heterogeneity. The goal is to use this model to predict responses to chemotherapeutic agents ex vivo.
Design: Discarded fresh tissue samples from 5 patients with primary or metastatic BC were obtained. This included ER+/HER2-, ER+/HER2+ and ER-/HER2- disease. After digestion, a homogenous single cell suspension was plated in Matrigel, a laminin-rich basement membrane substance that recapitulates the complex extracellular environment, and on Ultra Low Attachment (ULA) plates which are designed to inhibit cell attachment and select for tumor initiating cells. Growth was assessed at varying timepoints by brightfield, immunofluorescence and confocal microscopy.
Results: Mammospheres grew from all tumors types on both the Matrigel and ULA plates. Growth and expansion of the mammosphere structures were seen over time by brightfield microscopy in culture. The presence of epithelial cells was confirmed using immunofluorescence and confocal microscopy. Figure 1 shows (A) representative brightfield images of two mammospheres and (B) shows representative confocal images.
Conclusions: Mammospheres may be grown from fresh primary and metastatic BC samples. Future directions include treating the 3D mammosphere cultures with relevant chemotherapeutic agents such as trastuzumab, bevacizumab, or paclitaxel. Genomic changes that occur with treatment will be assessed using next-generation sequencing, and signatures will be developed to predict response to therapy. Ultimately, these optimized conditions for growing BC tumor cells ex vivo will be used on limited tissue samples from patients as part of clinical trials to guide therapeutic decision making and predict response to therapy.
Tuesday, March 5, 2013 1:00 PM
Poster Session IV # 32, Tuesday Afternoon