High-level ERBB2 gene amplification is associated with a particularly short time-to-metastasis, but results in a high rate of complete response once trastuzumab-based therapy is offered in the metastatic setting.

Despite patient selection based on ERBB2 overexpression, not all patients benefit from trastuzumab therapy. We have investigated whether a ERBB2 gene dosage effect might provoke increased biological aggressiveness and altered trastuzumab sensitivity. Absolute ERBB2 copy numbers ("CN") and ERBB2/centromer 17 ratios ("R") were measured by FISH analysis in tumors of 127 patients receiving trastuzumab-based treatment for Her-2/neu overexpressing metastatic breast cancer. CN and R were both significantly associated with shorter time to first metastasis (TTM) (CN: OR: 1.099, 95% CI: 1.042-1.159; R: OR: 1.211, 95% CI: 1.080-1.357) and longer PFS (CN: OR: 0.917, 95% CI: 0.867-0.969; R: OR: 0.840, 95% CI: 0.743-0.949) in a continuous variable Cox's regression model. Tumors with ERBB2/centromer 17 ratios of <2.2 had a significantly shorter TTM (p = 0.002) and significantly longer PFS (p = 0.003) than tumors with low-level (R: 2.2-6) and high-level amplification (R: >6). Interestingly, when ERBB2 copy numbers were analyzed, a significantly shorter TTM (p = 0.001) and longer PFS (p = 0.026) were observed in the group with high-level amplified CN (CN: >13), while no difference was observed between non- and low-level amplified CN. R, but not CN, was an independent predictor of complete (CR; OR: 1.685; 95% CI: 1.122-2.532) and partial (PR; OR: 1.704; 95% CI: 1.136-2.556) response in logistic regression analysis. CR (p = 0.016) rates were significantly higher in the high-level amplification group (R > 6), but no difference existed in response rates between non- and low-level amplified tumors in Chi-square tests. High-level ERBB2 amplification is associated with shorter TTM, but improved response to trastuzumab in metastatic breast cancer.