HER2 Testing and Correlation With Efficacy of Trastuzumab Therapy

As a result of the availability and clinical efficacy of trastuzumab (Herceptin), clinicians are now faced with a dilemma regarding the accurate identification of patients with HER2 overexpression. In the October 2002 issue of ONCOLOGY, Monica Fornier and colleagues critically reviewed the methodologies currently available for HER2 testing in their article, "HER2 Testing and Correlation With Efficacy of Trastuzumab Therapy."[1] The most widely used technique for detection of HER2 overexpression at the protein level is immunohistochemistry (IHC). However, the problems with this technique are, as correctly stated by the authors, that multiple primary antibodies—each of which binds to different epitopes and has differing sensitivities and specificities—are currently in clinical use. Moreover, different cutoff values are used to distinguish overexpressing form non-overexpressing tumors.[2] Detection of HER2/neu alteration is therefore not well standardized, and although IHC detection methods are widely commercially available, some of these assays may be suboptimal for this purpose. Other IHC Limitations Another concern is that the degree of IHC staining intensity for HER2 is subjective and qualitative. As a consequence, it may be difficult to know whether a sample scored as HER2-positive in one laboratory will be confirmed as HER2-positive by another laboratory. Furthermore, formalin fixation of tumor samples and storage in paraffin can result in epitope degradation so that sensitivity is lost over time in archival clinical material.[3-5] To overcome this problem, some assays use a technique called antigen retrieval, which consists of boiling the tumor sample using heat or microwave radiation, in order to rehabilitate epitopes that have been lost during tissue processing and storage. Many believe that the loss of antigenicity in paraffin-embedded specimens can be corrected by application of heat-induced antigen retrieval to the paraffin-embedded tissue sections. However, this approach has not been validated, before introduction of these reagents and methods, by systematic evaluation of clinical breast cancer specimens with known amplification/overexpression levels. Detection Rate Variability As summarized in the review by Fornier et al, reported detection rates for IHC differ among clinical laboratories. This discrepancy has generated considerable confusion among clinicians, pathologists, and patients regarding the appropriate methods for HER2 testing. To address this issue, we recently used each of the four US Food and Drug Administration (FDA)-approved assays for detecting HER2 alterations, to determine which is the most accurate. We tested these assays on breast cancer specimens in which we had previously determined amplification and overexpression levels using solid matrix blotting techniques as a "gold standard." In this comparative study, the sensitivity of the fluorescence in situ hybridization (FISH) assays for HER2 gene amplification was high—95.4% for the Vysis PathVision assay (Vysis, Downers Grove, Ill) and 95.4% for the Ventana INFORM assay (Ventana, Tucson). These sensitivity levels were followed by those for IHC assays performed with the Ventana CB11 monoclonal antibody (72.1%) and the DAKO HercepTest (69.8%, DAKO, Carpinteria, Calif).[5] The HercepTest was performed according to the approved protocol, as described by the manufacturer, using heat treatment for antigen retrieval. The protocol provided by Ventana Medical Systems and a Ventana automated slide stainer were used for the CB11 IHC assay. Both FISH assays were performed according to their respective FDA-approved protocols. The specificity of HER2 testing methods can also vary considerably among laboratory studies, as