Abstract
Background: The selection of cancer drugs for an individual patient is still based mostly on cancer type and stage. Predictive tests are needed to make individualized and more efficient pharmacological cancer treatment possible. Objective: To provide an overview of available, possible future development and principles for development of predictive tests for individualized selection of cancer drugs. Methods: Overview of published data. Results/conclusion: Despite increased knowledge in cancer biology, only limited progress has been made in the development and use of predictive tests. However, rapid progress in this field will be possible using already available and emerging technologies, but requires a paradigm shift in principles for the development and use of cancer drugs. Assessment of drug activity in intact tumor cells and tumor cell gene expression signatures are considered to have greatest potential for the development of versatile predictive tests.