Protein kinase inhibitors have become important cancer therapeutics, more than 30 have been approved for use in humans and more than 250 molecules are undergoing clinical trials. Polypharmacology is commonly observed for these drugs requiring thorough target deconvolution to understand their mechanism of action. Here, we report on the target spectrum of 242 clinical kinase inhibitors and 320 protein kinases using a quantitative and dose resolved chemical proteomics approach in cancer cell lines. The data revealed many novel targets for existing drugs, provides a view on the currently druggable kinome and further off-targets and suggests novel potential applications in immune or cancer therapy. Integration with phosphoproteomics data can be used to define the core pathways affected by kinase inhibitors providing a rational for combination treatments. We show that Cabozantinib may be re-purposed to treat FLT3-ITD positive AML patients and identify MELK as a novel survival marker and potential actionable target in NSCLC. All data is freely available in proteomicsDB to aid in basic and translational clinical research and kinase drug discovery.