Novel therapies, particularly those that target vascular endothelial growth factor (VEGF) and epidermal growth factor receptor (EGFR), have improved the treatment of advanced non-small-cell lung cancer (NSCLC). The search continues for biomarkers that can predict which patients are most likely to benefit from these therapies. At the same time, new research is helping to define how clinical and histopathologic features, such as ethnicity and histologic subtype, influence treatment decisions. Numerous studies have assessed potential biomarkers that may predict outcomes following treatment with anti-EGFR therapies, including mutations in EGFR and KRAS genes, EGFR gene copy number by fluorescence in situ hybridization, and EGFR protein expression by immunohistochemistry (IHC). Although mounting evidence supports the role of EGFR mutations in selecting therapy in clinical practice, in other cases, the emerging data have painted a complex and often contradictory picture, making it difficult to foresee how this information could be used in clinical practice. The discrepancies in published reports may be because of differences in patient populations or variations in methodology for assessing biomarkers; they also may reflect our incomplete understanding of the role of the EGFR pathway in NSCLC. Further assessment of these and other novel biomarkers is warranted to help define which patients with advanced NSCLC may derive the most benefit from targeted therapies.