Is Circulating HER-2 a Prognostic Factor for Metastatic Breast Cancer Patients?

Abstract & Commentary

Synopsis: The prognostic and predictive significance of circulating levels of extracellular domain of HER-2 (ECD/HER-2) in patients with metastatic breast cancer was evaluated in 242 metastatic breast cancer patients. Elevated levels of ECD/HER-2 were present in 35-40% of patients and were associated with a poorer prognosis of these patients. However, elevated levels were not associated with resistance to endocrine therapy or with sensitivity to anthracycline-based chemotherapy. Additional study is needed to determine the clinical use of detecting circulating ECD/HER-2.

Source: Hayes DF, et al. Clin Cancer Res. 2001;7: 2703-2711.

Biologic markers have demonstrated use for treatment decision making for patients with metastatic breast cancer. The estrogen receptor (ER) and progesterone receptor (PR) status of breast cancers represent important biologic markers that influence decision making about endocrine-based therapies for these patients. Additional biologic markers could provide valuable tools for treatment decision making. A candidate marker for these patients is the c-erb B-2 gene and its protein products (also designated HER-2 and c-neu). The protein product of the HER-2 proto-oncogene is a 185-kd transmembrane glycoprotein with intracellular tyrosine kinase activity that is overexpressed in 20-40% of human breast cancers.1 In addition, circulating levels of extracellular domain of HER-2 (ECD/HER-2) have been detected in blood and other body fluids of breast cancer patients.2 Several studies have evaluated the potential role of HER-2 as a prognostic factor for breast cancer patients. While some conflicting results exist in the literature, available data suggest that overexpression or amplification of HER-2 is a weak to moderate-negative prognostic factor in the setting of primary breast cancer.1 Several studies have suggested a weak to moderate-negative predictive ability of HER-2 regarding response to adjuvant endocrine therapy or to adjuvant alkylating agent therapy.1 In addition, a moderate positive predictive value has been associated with HER-2 and response to adjuvant anthracycline-based therapy.1 However, the HER-2 status of the tumor is not currently considered to be a sufficiently powerful factor upon which to base treatment decisions about overall adjuvant systemic therapy or adjuvant endocrine therapy.1 Anthracyclines are suggested to be the preferred therapy when adjuvant chemotherapy is recommended.1 For patients with metastatic disease, the HER-2 status is a strong predictive factor for response to trastuzumab.1 Thus, monitoring circulating levels of ECD/HER-2 in patients with metastatic breast cancer have the potential to be a useful biologic marker for these patients.

This study by Hayes and associates is a laboratory monitoring study of 242 breast cancer patients entered into Cancer and Leukemia Group B (CALGB) prospective therapeutic trials for metastatic breast cancer. Eligible patients were entered into 1 of 8 therapeutic protocols involving endocrine therapy (1 of 3 doses of megestrol acetate [103 patients in 1 study]) or chemotherapy (139 patients receiving standard or investigational chemotherapy in 7 studies). Plasma samples were obtained for the ECD/HER-2 assay prior to treatment. Elevated baseline levels of ECD/HER-2 were detected in 89 of the 242 patients (37%). The elevated levels of ECD/HER-2 were significantly associated with PR levels (but not ER levels), number of prior treatments, and visceral metastases. Elevated initial ECD/HER-2 levels were univariately associated with shorter overall survival (OS). However, after adjusting for other clinical variables (number of prior treatments, primary treatment regimen, performance score, and tissue ER content), ECD/HER-2 levels were not independently correlated with OS. The median OS for patients on megestrol acetate with elevated pretreatment ECD/HER-2 was 20.2 months compared with 27.8 months for patients without elevated pretreatment ECD/HER-2 and was significant by univariate analysis (P = 0.007). However, this interaction did not reach a conventional level of significance following multivariate analysis to account for prior treatments, performance score, and tissue ER content (P = 0.063). There was no significant difference for overall survival, time to treatment progression, or response rates to chemotherapy between patients with elevated and without elevated pretreatment ECD/HER-2 levels. Hayes et al conclude that 40% of metastatic breast cancer patients have elevated circulating levels of ECD/HER-2, and elevated circulating levels of ECD/HER-2 were associated with worse survival. However, elevated levels of ECD/HER-2 were not associated with resistance to endocrine therapy or sensitivity to anthracycline-based therapy.

Comment by Mark R. Albertini, MD

Additional informative biomarkers would represent a welcome advance to help direct therapy for patients with metastatic breast cancer. The protein product of HER-2 is an attractive candidate for this role.3 Tissue measurement of HER-2 is a strong predictive factor for response to trastuzumab, and a potential role for directing adjuvant therapies has been suggested.1 Prospective randomized clinical trials will be needed to clarify the use of this marker for adjuvant treatment decisions. The current study addresses the issue of whether baseline levels of ECD/HER-2 will be useful to direct treatment decisions for patients with metastatic disease. Hayes et al successfully demonstrate the ability to obtain and analyze data for this marker within the context of cooperative group clinical trials. They also demonstrate and confirm detection of elevated levels of ECD/HER-2 in approximately 40% of patients with metastatic breast cancer. The association of a poorer prognosis with elevated baseline levels of ECD/HER-2 suggests a potential prognostic role for this marker. However, no ability to predict response to endocrine or chemotherapy interventions was demonstrated for this marker. Thus, the question remains as to whether circulating HER-2 is more than just a marker of tumor burden?3

Several aspects of this study merit consideration to determine the potential use of ECD/HER-2 to predict response of metastatic breast cancer patients to either hormonal or chemotherapeutic intervention. The only hormonal intervention in this study was megestrol acetate. Thus, responses to other hormonal interventions require further investigation. In addition, the patients receiving second-line megestrol acetate had already progressed on first-line hormonal therapy and could represent a selected group of patients with a poorer prognosis. The analysis of patients receiving anthracycline-based chemotherapy did not demonstrate an interaction between pretreatment ECD/HER-2 levels and sensitivity to chemotherapy. However, the HER-2 positive patients did equally well as the HER-2 negative patients even though the HER-2 status may have identified them as having a worse prognosis. Additional study with patients receiving uniform anthracycline and nonanthracycline therapy will be needed to best evaluate this question. The potential role of ECD/HER-2 to direct trastuzumab therapy merits investigation. In addition, the ability of sequential monitoring of ECD/HER-2 to demonstrate response to therapy or disease recurrence requires clinical testing. The evaluation of ECD/HER-2 as a biologic marker for patients with metastatic breast cancer is clinically feasible and merits additional investigation.


1. Yamauchi H, et al. J Clin Oncol. 2001;19:2334-2356.

2. Hait WN. Clin Cancer Res. 2001;7:2703-2710.

3. Baselga J. Clin Cancer Res. 2001;7:2605-2607.