|| Checking for direct PDF access through Ovid
Human epidermal growth factor receptor 2 (ERBB2/HER2) is amplified and overexpressed in 20% to 25% of breast carcinomas, correlates with poor outcome, and is an indication for treatment with trastuzumab. Accurate assessment of ERBB2 status is crucial for proper prognosis and to offer appropriate treatment for patients. ERBB2 status is generally determined by immunohistochemistry or fluorescence in situ hybridization (FISH), and sporadically by quantitative real-time polymerase chain reaction (PCR). We developed a new algorithm, termed quantitative PCR algorithm (QPA) score, and compared its performance with the gold standard FISH assay. The QPA is a computation of the relative number of copies of the ERBB2 gene with respect to a nonstandard, short-arm centromeric sequence on chromosome 17, and referenced to a single-copy gene, RPP30. This provides a more reliable determination of ERBB2 amplification, reducing the false polysomy 17 error. A total of 69 breast carcinoma samples were tested for quantitative real-time PCR and FISH, and the degree of concordance was analyzed. Sixty-two cases were in agreement between the 2 methods, and the contingency study assigned a κ value of 0.729 for their correlation. A receiver operating characteristic analysis was used to determine the optimal cut-off point for ERBB2 amplification, which was estimated at a QPA=1.53 (sensitivity=0.863; specificity=0.944). Our data conclude that the QPA is able to determine ERBB2 gene status with high accuracy, while also overcoming the limitations of conventional techniques and providing better cost-effectiveness.