The prognostic and predictive values of ECD-HER-2.

The search for a simple, sensitive test to reliably determine prognosis and predict response to therapy in patients with cancer is an important area of translational research. In this issue of Clinical Cancer Research, Hayes et al. (Clin. Cancer Res., 7: 601-604, 2001) report the results of an ancillary Cancer and Leukemia Group B protocol designed to determine whether the circulating extracellular domain of HER-2/neu (ECD-HER-2) was indicative of prognosis or predictive of response to therapy in women with metastatic breast cancer. Results were drawn from a sample of 242 patients of whom 89 had elevated values of the protein. These women had been enrolled in a variety of Cancer and Leukemia Group B protocols evaluating either the efficacy of dose in the use of megestrol acetate as second-line hormonal treatment or in patients enrolled into several chemotherapeutic protocols, many containing doxorubicin. They report that patients with pretreatment elevation of ECD-HER-2 had a worse prognosis than those who did not, but that there was no convincing correlation of elevated ECD-HER-2 with response to either endocrine or chemotherapy. Although the small number of patients and the retrospective study design allows one to draw only tentative conclusions, this report raises several important issues for the conduct of translational research and points to several new hypotheses for future testing.