Introduction: Currently, anticancer therapy is mainly based on histology and on giving the same treatment to presumed homogeneous patients. The switch from histology-driven therapy to molecular clinical oncology is correlated with a better understanding of the 'molecular taxonomy' of each tumor that can provide us with targets for specific drugs. Cancer therapy is moving irreversibly towards personalized therapy that benefits selected patients. Once the potential therapeutic targets are identified, the availability of predictive biomarkers is the key element and their prospective evaluation should be a parallel component of the clinical evaluation of a new drug.
Areas covered: The state of the art in clinical results of personalized therapy. The authors discuss the finding that, in patients with advanced disease, a limited number of targeted agents improve overall survival, whilst the majority only have an effect on response rate and/or time to tumor progression, with efficacy limited in time due to acquired resistance.
Expert opinion: The mechanisms leading to resistance are related to tumor cell heterogeneity and in part explained by the cancer stem cell model and genetic instability. The steps toward the optimization of tailored therapy need validated predictive biomarkers, pharmacogenetics analysis and a close collaboration between bench and bedside.