| Summary: | Purpose: Human epidermal growth factor receptor 2 (HER2) positivity is an important prognostic and predictive indicator in breast cancer. HER2 status is determined by immunohistochemistry and fluorescent in situ hybridization (FISH), which are potentially inaccurate techniques as a result of several technical factors, polysomy of chromosome 17, and amplification or overexpression of CEP17 (centromeric probe for chromosome 17) and/or HER2. In South Africa, HER2-positive tumors are excluded from a MammaPrint (MP; Agendia BV, Amsterdam, Netherlands) pretest algorithm. Clinical HER2 status has been reported to correlate poorly with molecular subtype. The aim of this study was to investigate the correlation of clinical HER2 status with BluePrint (BP) molecular subtyping. Methods: Clinico-pathologic and genomic information was extracted from a prospectively collected central MP database containing records of 256 estrogen receptor–positive and/or progesterone receptor–positive tumors. Twenty-one tumors considered HER2 positive on immunohistochemistry or FISH were identified for this study. Results: The median age of patients was 56 years (range, 34 to 77 years), with a median tumor size of 16 mm (3 to 27 mm). Four (19%) tumors were confirmed HER2-enriched subtype, six (29%) were luminal A, and 11 (52%) were luminal B. The positive predictive values of HER2/CEP17 ratio ≥ 2 and HER2 copy number ≥ 6 were only 29% and 40%, respectively. The differences in means for HER2/CEP17 ratio were significant between BP HER2-enriched versus luminal (P = .0249; 95% CI, 0.12 to 1.21) and MP high-risk versus low-risk tumors (P = .0002; 95% CI, 0.40 to 1.06). Conclusion: Of the 21 tumors considered clinically HER2 positive, only four were HER2-enriched subtype with BP, indicating an overestimation of HER2 positivity. FISH testing has a poor positive predictive value.
|
|---|