Basal-like breast cancer, as defined by gene expression profiling, is associated with aggressive phenotype and poor clinical outcome. Recent immunohistochemical validation suggested that basal-like subtype could be characterized by staining for cytokeratin (CK) 5/6 and epidermal growth factor receptor (EGFR) in triple negative breast cancers (TNBCs). Most of studies evaluated surrogate immunopannel of biomarkers to define basal-like breast cancer subtypes only in the TNBCs, although not all basal-like breast cancers are triple negative breast cancers (TNBCs). The significance of basal marker expression in other than triple negative breast cancer remains to be evaluated. To define prognostic impact of basal marker expression in HER2 positive breast cancer, we investigated cytokeratin (CK) 5/6 and epidermal growth factor receptor (EGFR) expression in patients with HER2 positive early breast cancer.
Patients and Methods:
Biomarker evaluation was performed using five immunohistochemical surrogate panel of estrogen receptor (ER), progesterone receptor (PR), HER2, CK 5/6 and EGFR in HER2 positive early breast cancers. Amplification of HER2 was confirmed by fluorescent in situ hybridization. HER2-positive breast cancer was classified by expression of basal markers (either EGFR or CK5/6) as “basal HER2- positive” (patients with HER2- positive disease who express basal markers) and “non-basal HER2” (patients with HER2-positive disease who did not express basal markers). We compared the prognostic significance of the basal marker expression between two groups.
HER2 overexpression was found in 24.8% of early breast cancers with available tissue specimens from the primary tumor (236 of 952 cases). Basal marker co-expression was identified in 12.7% of HER2 positive early breast cancers. (30 of 236 patients) Basal HER2 positive breast cancer was significantly associated with age greater than 50 years (P=0.012), absence of ER (P < 0.001) and PR (P=0.004). The basal marker co-expression in patients with HER2 amplified early breast cancers demonstrated poorer overall survival (basal positive vs. basal negative, 85.6 months [95% confidence interval (CI), 70.8- 100.3 months] vs. 122.3 [C.I. 133.7-139.9], P=0.001) and disease free survival (DFS) (44.6 months [95% CI, 14.7-74.8] vs 110.7 months [95% CI, 96.4-123.9]; P=. 008) respectively. In lymph node positive group, basal marker expression retained its statistical significance at the multivariate level (P=0.047) HER2 positive breast cancer with lymph node involvement with basal marker expression showed substantially poorer overall survival with 2.1-fold (95% CI, 1.0-4.2) risk for death.
Considerable number of HER2 positive breast cancer co-expressed basal markers. Our data demonstrated that simultaneous basal marker expression in HER2 positive early breast cancer is associated with poor clinical outcome. The molecular significance of basal marker expression in HER2 positive breast cancer needs to be further investigated.
Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P4-09-03.