Receptor tyrosine kinases play a major role in human carcinogenesis. Overexpression of the epidermal growth factor receptor (EGFR) has been associated with poor clinical outcome in several types of cancer. In principle, as with HER2, the EGFR status of a tumor should predict the likelihood of response to EGFR-targeted therapy. However, clinical data have failed to demonstrate a relationship between EGFR expression and response to the EGFR-targeted compounds cetuximab, gefitinib and erlotinib. Recently, patients reported to be EGFR negative have been shown to respond to cetuximab. Possible explanations include methodological failures or most likely heterogeneity and complexity of the mechanisms of EGFR-mediated molecular carcinogenesis. Immunohistochemistry is the most widely used method for measuring EGFR expression; however, its value is limited by lack of methodological standardization. Other approaches to measuring EGFR such as amplification assays are currently being introduced but need further testing before they can enter clinical practice. Mutational analysis seems also fruitful in predicting response to anti-EGFR small molecules. Further work is needed to identify how EGFR contributes to the carcinogenic and metastatic processes.