Standardization of HER2 immunohistochemistry in breast cancer by automated quantitative analysis

Arch Pathol Lab Med. 2009 Sep;133(9):1413-9. doi: 10.5858/133.9.1413.


Context: There is critical need for standardization of HER2 immunohistochemistry testing in the clinical laboratory setting. Recently, the American Society of Clinical Oncology and the College of American Pathologists have submitted guidelines recommending that laboratories achieve 95% concordance between assays and observers for HER2 testing.

Objective: As a potential aid to pathologists for achieving these new guidelines, we have conducted an examination using automated quantitative analysis (AQUA analysis) to provide a standardized HER2 immunohistochemistry expression score across instruments (sites), operators, and staining runs.

Design: We analyzed HER2 expression by immunohistochemistry in a cohort (n = 669) of invasive breast cancers in tissue microarray format across different instruments (n = 3), operators (n = 3), and staining runs (n = 3). Using light source, instrument calibration techniques, and a new generation of image analysis software, we produced normalized AQUA scores for each parameter and examined their reproducibility.

Results: The average percent coefficients of variation across instruments, operators, and staining runs were 1.8%, 2.0%, and 5.1%, respectively. For positive/negative classification between parameters, concordance rates ranged from 94.5% to 99.3% for all cases. Differentially classified cases only occurred around the determined cut point, not over the entire distribution.

Conclusions: These data demonstrate that AQUA analysis can provide a standardized HER2 immunohistochemistry test that can meet current guidelines by the American Society of Clinical Oncology/College of American Pathologists. The use of AQUA analysis could allow for standardized and objective HER2 testing in clinical laboratories.

MeSH terms

  • Algorithms
  • Biomarkers, Tumor / metabolism
  • Breast Neoplasms / metabolism*
  • Breast Neoplasms / pathology
  • Carcinoma, Ductal, Breast / metabolism*
  • Carcinoma, Ductal, Breast / pathology
  • Female
  • Fluorescent Antibody Technique, Indirect / methods
  • Fluorescent Antibody Technique, Indirect / standards*
  • Guidelines as Topic
  • Humans
  • Image Processing, Computer-Assisted / methods*
  • Middle Aged
  • Receptor, ErbB-2 / metabolism*
  • Reproducibility of Results
  • Tissue Array Analysis


  • Biomarkers, Tumor
  • ERBB2 protein, human
  • Receptor, ErbB-2