Histopathologic Characteristics Predicting HER-2/neu Amplification in Breast Cancer


Address correspondence and reprint requests to: Helena Chang, MD, PhD, Director, Revlon/UCLA Breast Center, 200 UCLA Medical Plaza, Suite B265-1, Los Angeles, CA 90095-7028, USA, or e-mail: hchang@mednet.ucla.edu.


Abstract:  The HER-2/neu gene is a proto-oncogene that is amplified in 10–30% of breast cancers. New drugs for targeted therapy, such as Herceptin, are effective for patients with HER-2/neu-positive tumors, making it necessary to have a noncostly and accurate method to assess HER-2/neu status. We studied the correlation of findings made by fluorescent in situ hybridization (FISH) and immunohistochemistry (IHC) staining and the possibility of combining IHC and other clinicopathologic characteristics of breast tumors to predict FISH-determined HER-2/neu status. The clinicopathologic characteristics analyzed were the size of the tumor, p53, lymph-vascular invasion, estrogen/progesterone receptors (ER/PR), tumor grade, axillary lymph node status, and patient age. A total of 199 cases of invasive breast cancer studied at the UCLA Pathology Laboratory during 2003 were included in this study. Tumors with IHC 0, 1+, 2+, and 3+ scores were found to be FISH positive in 3.5%, 6.4%, 25.7%, and 81.5% of the respective groups. Our study showed a strong association between the FISH-negative and IHC scored 0 and 1+ tumors, suggesting that the FISH test may not be necessary in these cases (p < 0.0001). Although the concordance between IHC 3+ and FISH positive is high, 18% of the patients with overexpression of HER-2/neu fail to show gene amplification by FISH. HER-2/neu positivity was found to be proportionally associated with increasing grade in infiltrating ductal carcinoma (p < 0.0001). p53-positive tumors are more likely to be HER-2/neu amplified (p = 0.0003). Tumors that are negative for ER/PR are also associated with HER-2/neu positivity by FISH (31.15%, p = 0.0016). FISH-determined HER-2/neu status is not associated with histologic type, tumor size, nodal status, lymph-vascular invasion, or patient age.