Recently there has been a resurgence in interest in the use of HER2 protein overexpression or gene amplification to refine prognostic estimates of breast cancer patient outcomes and to predict which therapies might be most appropriate for individual breast cancer patients. To bring HER2 testing into clinical practice, two hurdles must be cleared. First, HER2 tests must be developed that accurately reflect HER2 status and that can be reliably performed. Second, HER2 test results must be correlated with patient outcome and response to therapy. Much progress has been made in both of these areas, but the full implementation and utility of this work have yet to be realized.