Evaluation of new monoclonal antibodies in detection of estrogen receptor, progesterone receptor, and Her2 protein expression in breast carcinoma cell block sections using conventional microscopy and quantitative image analysis

Accurate assessment of estrogen receptor (ER), progesterone receptor (PR), and Her2 status of breast carcinomas is critical for predicting response to systemic therapies. Recently, developed rabbit monoclonal antibodies (RMab) are reported to have higher sensitivity than murine monoclonal antibodies (Mab). This study compares RMabs against FDA‐approved Mab (FMab) in breast carcinoma cell block sections using visual and image quantification. Cell blocks from 52 breast cancers were studied. Immunohistochemistry using RMab (ER, PR, and Her2) was compared with FMabs (ER, PR, Dako) and HercepTest (HerFDA). Fluorescent in situ hybridization (FISH) was used as a reference standard for Her2. Slides were later scanned and reanalyzed with an automated cellular imaging system (ACIS III, Dako). Frequency of ER (38.5% vs. 36.5% for visual; 55.8% vs. 57.7% for image) and PR (28.8% vs. 36.5% for visual; 50% vs. 51.9% for image), and concordance (overall agreement is 71.2% and 75% for visual and image ER; and 84.6% and 59.6% for visual and image PR) were similar for both FMab and RMab, respectively. Overall agreement (53.8% vs. 77.1% for visual and image detection, respectively, using HerFDA and RMab) is poor to moderate for Her2. Visual Her2 (RMab) has the highest concordance (94.1%), and visual HerFDA has the lowest concordance (35.3%) with FISH. ER and PR analysis (FMab vs. RMab) are almost comparable using both detection methods with good overall agreement. For Her2 overexpression, RMab proved to be superior to HerFDA and showed excellent agreement with FISH results with both quantitative detection methods. Diagn. Cytopathol. 2009. © 2009 Wiley‐Liss, Inc.