**3.2 Benefits and pitfalls for targeting kinases in cancers**

The development of small-molecule kinase inhibitors for the treatment of cancer has continued to be of intense interest. Notably, many inhibitors have received FDA approval with approximately another 150 are in preclinical and clinical phase trials. Despite these important advances, many factors have confounded the clinical efficacy of these kinase-targeted drugs including the challenges of tumor heterogeneity and microenvironment as well as the emergence of mutations that confer drug resistance. Another major challenge of kinase inhibition is that of the development of adverse side effects. Some classic examples of this include dermatologic complications and cardiotoxicity associated with inhibition of EGFR and vascular endothelial growth factor receptor (VEGFR), respectively. Furthermore, there is an urgent need to develop relevant models of resistance in response to kinase inhibitors in efforts to overcome this resistance via potential synergistic combinatorial therapies.

Another critical issue facing clinical trial design with kinase-targeted agents is that of determining the types of tumors that are most likely to respond to specific kinase inhibitors and thus identify the subsets of patients who will likely benefit from these treatments. To combat this issue, many studies have been dedicated to identifying certain "kinase dependencies" in cancer cells that would make them more susceptible to inhibition. These so-called dependencies are primarily based on the existence of constitutively activate kinases achieved by gene mutation, amplification, or fusion. Among the potential approaches to identifying signatures of kinase dependency are proteomic profiling, next-generation sequencing and various applications utilizing phospho-specific antibodies against numerous specific kinase substrates. Additional mechanisms of kinase dependency include impairment of the function of phosphatases, the negative regulators of phosphorylation as is the case with mutations in the phosphatase and tensin homolog (PTEN) tumor suppressor gene. The consequence of *PTEN* loss is signal propagation through downstream kinases such as Akt. Moreover, growing evidence from isogenic human and mouse models also suggests that this type of indirect avenue of kinase dependency may be analogous to direct, activating mutations in the kinases themselves.

Finally, there are also cases in which the beneficial effect of a kinase inhibitor is counteracted by an additional genetic lesion in a compensatory signaling pathway. Therefore, studies to identify such secondary events are urgently needed. Taken together, these evidences underscore the critical need to optimize the use of kinase inhibitors against cancers by continued detailed molecular characterization

of tumor tissues. It will be critical to develop new compounds that circumvent acquired resistance to the first-generation kinase inhibitors for patients with refractory disease.
