Performance of three-biomarker immunohistochemistry for intrinsic breast cancer subtyping in the AMBER consortium

Academic Article


  • Background: Classification of breast cancer into intrinsic subtypes has clinical and epidemiologic importance. To examine accuracy of IHC-based methods for identifying intrinsic subtypes, a three-biomarker IHC panel was compared with the clinical record and RNA-based intrinsic (PAM50) subtypes. Methods: Automated scoring of estrogen receptor (ER), progesterone receptor (PR), and HER2 was performed on IHCstained tissue microarrays comprising 1,920 cases from the African American Breast Cancer Epidemiology and Risk (AMBER) consortium. Multiple cores (1-6/case) were collapsed to classify cases, and automated scoring was compared with the clinical record and to RNA-based subtyping. Results: Automated analysis of the three-biomarker IHC panel produced high agreement with the clinical record (93% for ER and HER2, and 88% for PR). Cases with low tumor cellularity and smaller core size had reduced agreement with the clinical record. IHC-based definitions had high agreement with the clinical record regardless of hormone receptor positivity threshold (1% vs. 10%), but a 10% threshold produced highest agreement with RNAbased intrinsic subtypes. Using a 10% threshold, IHC-based definitions identified the basal-like intrinsic subtype with high sensitivity (86%), although sensitivity was lower for luminal A, luminal B, and HER2-enriched subtypes (76%, 40%, and 37%, respectively). Conclusion: Three-biomarker IHC-based subtyping has reasonable accuracy for distinguishing basal-like from nonbasallike, although additional biomarkers are required for accurate classification of luminal A, luminal B, and HER2-enriched cancers. Impact: Epidemiologic studies relying on three-biomarker IHC status for subtype classification should use caution when distinguishing luminal A from luminal B and when interpreting findings for HER2-enriched cancers. Cancer Epidemiol Biomarkers Prev; 25(3); 470-8.
  • Authors

    Digital Object Identifier (doi)

    Author List

  • Allott EH; Cohen SM; Geradts J; Sun X; Khoury T; Bshara W; Zirpoli GR; Miller CR; Hwang H; Thorne LB
  • Start Page

  • 470
  • End Page

  • 478
  • Volume

  • 25
  • Issue

  • 3