The development of rationally designed agents targeting specific biological pathways in tumor development has been heralded as a major paradigm shift in the approach to the treatment of cancer. The application of these agents has lead to promising pre-clinical findings in a variety of human cancer pre-clinical models including lung cancer. Results from initial clinical trials employing targeted agents have shown that in contrast to what was expected, only selected patients may benefit. Determining which patients will benefit remains a major challenge. Findings from the 'Iressa' Dose Evaluation in Lung Cancer and INTACT trials for non-small cell lung cancer (NSCLC) require investigators to return to the laboratory to design relevant pre-clinical studies that improve our ability to predict response to targeted agents in the clinic.