"The dawn of the age of personalized cancer treatment provides the promise to identify the right drug for the individual that will greatly improve the patient’s outcome. It is well known that cancer drugs work only in small subset of patients. For many of these agents, there are putative markers of response in the literature but very few are been used in clinical practice. More importantly, new anticancer agents are targeted to specific cancer genes and are expected to be effective only in tumors in which these genes are mutated or otherwise abnormal. Consequently, the success of personalized treatment of cancer patients depends on matching the most effective therapeutic regimen with the characteristics of the individual patient, balancing benefit against risk of adverse events. The primary challenge in achieving this goal is the heterogeneity of the disease, recognizing that breast, lung, colon and the majority of cancers are not single diseases but rather an array of disorders with distinct molecular mechanisms.High-throughput technologies such as next-generation sequencing, gene expression microarrays have the capacity to dissect this heterogeneity and now doing an unbiased interrogation of the human genome, which allows strategies to search for novel, previously unsuspected, biomarkers of drug response and afford opportunities to match therapies with the characteristics of the individual patient’s tumor.Here we propose to develop an integrative bioinformatics approach as part of CNIO’s personalized cancer treatment platform that will predict treatment response and select new biomarkers based on the integration of genomics data and drug response analyzed in patients with pancreatic cancer and personalized xenografted mice. A cross-disciplinary integrative effort that will convert the information contained in multidimensional data sets into useful biomarkers that can classify patient tumors by prognosis and response to therapeutic modalities."