Consensus-Phenotype Integration of Transcriptomic and Metabolomic Data Implies a Role for Metabolism in the Chemosensitivity of Tumour Cells

Using transcriptomic and metabolomic measurements from the NCI60 cell line panel, together with a novel approach to integration of molecular profile data, we show that the biochemical pathways associated with tumour cell chemosensitivity to platinum-based drugs are highly coincident, i.e. they describe a consensus phenotype. Direct integration of metabolome and transcriptome data at the point of pathway analysis improved the detection of consensus pathways by 76%, and revealed associations between platinum sensitivity and several metabolic pathways that were not visible from transcriptome analysis alone. These pathways included the TCA cycle and pyruvate metabolism, lipoprotein uptake and nucleotide synthesis by both salvage and de novo pathways. Extending the approach across a wide panel of chemotherapeutics, we confirmed the specificity of the metabolic pathway associations to platinum sensitivity. We conclude that metabolic phenotyping could play a role in predicting response to platinum chemotherapy and that consensus-phenotype integration of molecular profiling data is a powerful and versatile tool for both biomarker discovery and for exploring the complex relationships between biological pathways and drug response. Author Summary: Resistance to chemotherapy drugs in cancer sufferers is very common. Using a panel of 59 cell lines obtained from different types of cancer we study the links between the genes and metabolites measured in these cells and the resistance the cells show to common cancer drugs containing platinum. In order to combine the information given by the genes and metabolites we introduce a new pathway-based approach, which allows us to explore synergy between the different types of data. We then extend the procedure to look at a wider panel of drugs and show that the pathways we found were associated with platinum are not just the pathways which are frequently selected for a large number of drugs. Given the increasing use of multiple sets of measurements (genes, metabolites, proteins etc.) in biological studies, we demonstrate a powerful, yet straightforward method for dealing with the resulting large datasets and integrating their knowledge. We believe that this work could contribute to developing a personalised medicine approach to treating tumours, where the genetic and metabolic changes in the tumour are measured and then used for prediction of the optimal treatment regime.