Objective

Recapitulation of the tumor microenvironment is widely acknowledged as being key to the development of more predictive 3D in-vitro efficacy models. To facilitate this, we demonstrate simple, robust workflows for creating, monitoring and automatically quantifying features, such as morphology, growth and death of two tumor multi-spheroids co-culture models using real time live cell analysis.

96-well plate-based assays were established using a panel of breast tumor multi spheroids cultured on a base of extracellular matrix (ECM, Matrigel®) in co-culture with either stromal cells or immune cells. SKBr-3 cells, when cultured alone failed to produce spheroids, however inclusion of normal human dermal fibroblasts (NHDFs) had a striking effect on spheroid formation. NHDFs co-cultured with MDA-MB-231 cells reveal a time-dependent morphological effect, transitioning from a stellate appearance at early time-points, before reverting to a more clustered round appearance beyond 6 days. Using a novel brightfield image analysis algorithm, the growth and size (area) of spheroids could be determined over time non-invasively without the use of fluorescent labels. The inhibition of growth caused by a range of standard of care and cytotoxic agents (Lapatinib, ZK164015, camptotecin) aligned with the known expression profile of receptors in MCF-7, MDA-MB-231, SKBr-3 and BT-474 targeted by these agents.  When co-cultured with immune cells (PBMCs), multi-spheroid viability could be assessed (over 10 days) by measurement of a fluorescent protein (RFP) loss.  In this model, Herceptin induced a concentration-dependent, antibody-dependent cell-mediated cytotoxicity (ADCC) of Her-2 expressing tumor cells.

These data highlight the value of live-cell analysis to study the temporal impact of stromal cells on tumor spheroid morphology and pharmacological profiles.  In addition, co-culturing with immune cells enables the study of ADCC.  Spheroid assays, incorporating ECM and additional cell types have the potential to provide more translational models for the study of the tumor microenvironment.