Image cytometric validation of breast carcinoma markers (ER, HER2 and MIB-1) using tissue microarrays: rabbit monoclonal vs. FDA-approved antibodies.

Abstract

Objective: To use the ACIS III (Dako, Carpinteria, California, U.S.A.) with tissue microarrays (TMAs) to compare rabbit monoclonal antibodies (RMab) for ER, HER2, and MIB-1 with FDA-approved monoclonal (FMab) and polyclonal (FPab) antibodies.

Study design: TMAs of 43 breast cancers were used. Immunohistochemistry was performed using RMab (LabVision, Fremont, California, U.S.A.): ER (SP1; 1/100), HER2 (SP3; 1/100), and MIB-1 (SP6; 1/200). FMPab (Dako) used: ER (1D5; 1/50), HercepTest kit and MIB-1 (MIB-1; 1/160). The stained TMAs were quantitated visually and by image cytometry (ACIS III).

Results: The overall agreement between RMab and FMab for ER using visual (98.45%) and image analysis (97.56%) was excellent, with a kappa level of 0.89 and 0.94, respectively. For HER2, the overall agreement between RMab and FPab was fair for visual (67.44%) and substantial (87.50%) for image analysis, with a kappa level of 0.32 and 0.72, respectively. For MIB-1, there was fair (64.29%) to poor (43.33%) agreement between MIB-1 RMab and FMab using visual and image analysis, with a kappa level of 0.47 and 0.16, respectively.

Conclusion: RMabs for ER (SP1) and HER2 (SP3) are almost comparable to their counterpart, FDA antibodies; however, MIB-1 RMab (SP6) shows poor concordance with FMab in TMA. Image analysis shows a better concordance than visual quantitation assessment specifically for ER and HER2.