In the current study, we performed a complete analysis of gene amplification, copy-number variations (CNV), transcriptional profiling and protein expression of all four HER family receptors, in a series of patients with metastatic breast cancer (MBC), treated with trastuzumab-based regimens. In addition, our analysis included the evaluation of several other factors, such as PTEN and mTOR protein expression by immunohistochemistry (IHC) and PIK3CA mutations.
Formalin-fixed paraffin-embedded tumor tissue samples were collected from 229 patients, considered to be HER2-positive when assessed at the local laboratories. Central review of HER2 status by fluorescence in situ hybridization and/or IHC revealed that of the 229 patients, only 139 (61%) were truly HER2-positive.
In the multivariate analysis, HER3 gene amplification (using the 75th percentile as a cut-off point) (HR = 0.50, 95% CI 0.27-0.93, p = 0.027), HER3 mRNA expression (75th percentile as a cut-off point) (HR = 0.47, 95% CI 0.22-1.01, p = 0.052) and PTEN protein expression (HR = 0.60, 95% CI 0.37-0.98, p = 0.042) retained their favorable prognostic significance for OS in the HER2-positive subgroup. In HER2-negative patients, none of the significant factors that were identified in the univariate analysis retained prognostic ability for OS. Moreover, wild-type PIK3CA was found to be an independent favorable prognostic factor for TTP in the HER2-positive patients (HR = 0.53, 95% CI 0.27-1.02, p = 0.059). In addition, EGFR CNVs (HR = 4.89, 95% CI 1.23-19.36, p = 0.024), HER3 gene amplification (using the median value as a cut-off point) (HR = 0.41, 95% CI 0.16-1.06, p = 0.067), HER3 protein expression (HR = 0.15, 95% CI 0.04-0.52, p = 0.003) and mTOR protein expression (HR = 0.33, 95% CI 0.13-0.84, p = 0.02) independently affected TTP in the HER2-negative subgroup.
The present study suggests that EGFR is a negative, whereas HER3 is a positive prognostic factor in patients with MBC. Since the above associations were not restricted to HER2-positive patients only, a definitive predictive value for trastuzumab treatment is not documented. Given the retrospective nature of the current analysis, our findings should be considered as hypothesis generating.
Clinical trial identification
Legal entity responsible for the study
Hellenic Cooperative Oncology Group
Roche Hellas S.A.
G. Fountzilas: Advisory Board: Roche Hellas S.A. All other authors have declared no conflicts of interest.