The incidence and clinical relevance of tumor cells contaminating the stem cell products of patients with advanced breast cancer treated with high-dose chemotherapy is uncertain because prior studies used small sample sizes and lacked standardization of the immunocytochemistry (ICC) detection method used. We evaluated blood stem cell and bone marrow samples obtained from 535 women with metastatic breast cancer who received high-dose chemotherapy and unmanipulated mobilized blood stem cell support. Of the patients tested, 20.6% and 26.3% had blood stem cell and bone marrow contamination, respectively. Blood stem cell contamination was significantly more frequent in patients with marrow involvement than in patients without marrow involvement (35% versus 18.4%, respectively; P = .009). In fact, according to multivariate analysis results, marrow involvement was the only significant predictor for blood stem cell product contamination. Patients without marrow involvement who had fewer apheresis procedures were also observed to have a significantly lower incidence rate of blood stem cell contamination than patients who had more procedures (P < or = .008), and patients who received combined chemotherapy and cytokine mobilization therapy had less contamination than patients who received cytokine alone (P = .0001). Combined mobilization therapy appears to be associated with a lower incidence of contamination as a result of fewer apheresis procedures rather than through an antitumor effect of chemotherapy (P < or = .001). Patients with ICC-negative blood stem cell products had significantly longer progression-free survival (PFS) and overall survival (OS) than did patients with ICC-positive blood stem cell products (median PFS, 401 versus 291 days, respectively, P = .007; median OS, 1060 versus 697 days, P = .009) . However, multivariate analysis did not reveal any significant independent predictors of survival outcomes. Thus, further study is needed to determine if contaminating tumor cells in the stem cell products of breast cancer patients ever directly impact survival outcomes or are only indicative of residual in vivo disease in high-dose chemotherapy recipients.