The realization of the promise of personalized molecular medicine will require the efficient development and implementation of novel targeted therapeutics. The goal will be to deliver the right drug to the right patient at the right time at the right dose. This effort will require an integration of information from the DNA, RNA and protein level into predictors of which patients are likely to respond to particular therapies. The overall likelihood of response to particular drugs represents the interaction between predictors of sensitivity with predictors of resistance. Efficient clinical trials testing these precepts will require the development and implementation of novel trial designs. It is likely that we will need to increase the size of phase I and II trials to allow the identification and validation of molecular markers at the same time as the initial evaluation the toxicity and efficacy of targeted therapeutics. This will come with the advantage of being able to deliver targeted therapeutics to enrol a much smaller population of patients selected for the likelihood to respond in phase III trials accelerating the approval of effective targeted therapeutics.

The phosphatidylinositol 3'kinase (PI3K) pathway is aberrant at multiple levels across a wide variety of tumours making it the most common activating aberration in cancer. This has led to the development and now early clinical testing of drugs targeting multiple components of the pathway. The efficient utilization of these drugs will require the ability to accurately determine mutation and activation status in tumours as well as determining the interaction between the PI3K pathway and other pathways in driving tumour pathophysiology. Using a novel accurate and sensitive mass spectroscopy based sequencing approach, we have evaluated mutations in the PI3K pathway across more than 500 breast cancer samples. We have also implemented a high throughput functional proteomics approach designated reverse phase protein arrays to characterize the level and activity of multiple signalling pathways. We demonstrate than an integrated analysis of mutation, proteins levels and protein activity is able to predict lack of response to trastuzumab in patients and to novel drugs targeting the PI3K pathway in vitro. This demonstrates that the response to targeted therapeutics is due to an interaction of markers of sensitivity and markers of resistance and provides important approaches for patient selection.