Many drugs have been discontinued during phase II/III breast cancer clinical trials due to lack of clinical efficacy, indicating shortcomings in predictive value of preclinical data. Nutrient availability in the tumour cell microenvironment and the dimensionality of in vitro tumour cells likely impact on drug responsiveness. Global proteomics experiments were conducted to assess the impact of nutrient availability and dimensionality of culture. Protein set enrichment analyses identified “pathways in cancer”, “focal adhesion” and “ECM receptor in interaction” related to cell culture dimensionality in MDA-MB-231 cells. In MCF-7 cells, 4 pathways were influenced by medium composition, and 2 pathways were influenced by cell culture dimensionality (2D vs. 3D). These pathways were also identified using KEGG analyses. Eight drugs were selected for investigation according to the differential expression of their putative or known target proteins. The influence of medium composition on drug effectiveness was explored using the “Melbourne Medium” (MM), developed to have nutritionally physiological levels of metabolites as compared with conventional (hyper-nutritional) cell culture medium (CM). The influence of dimensionality on drug effectiveness was also explored, using an innovative 3D viability assessment combining automated confocal microscopy and image analysis. Dimensionality of culture appeared to have a greater influence on the proteome and drug effects than variation in nutrient levels. The number of differentially expressed proteins in the different media was greater in 2D than 3D. We conclude that the risk of qualifying inactive compounds in preclinical assessment may be mitigated using additional models incorporating physiological media and 3-dimensionality.
Read full abstract