Gene expression studies have identified a basal phenotype of breast cancer; these are hormone receptor and HER2-negative cancers with poor prognosis. High levels of cyclin E and Skp2, and low levels of p27 have previously been individually associated with both basal-like breast cancer and a poor outcome after diagnosis. The goal of this study was to first confirm the prognostic value of these biomolecular markers using a breast cancer tissue microarray. Second, we also test the hypothesis that the combined phenotype of high cyclin E, low p27, and high Skp2 would be a strong predictor of outcome and would be closely associated with the basal phenotype of breast cancer. Our cohort consisted of 438 cases of breast cancer and the median follow-up was 15.4 years. The tissue microarray was constructed from archival tumor blocks and we used commercially available antibodies for biomarker immunostaining. Cyclin E was positive in 46% of cases, p27 was negative in 62%, and Skp2 was positive in 35%. We found cyclin E and Skp2 to be prognostic for breast cancer-specific survival in univariate analyses, but p27 was not prognostic. The strongest predictor of outcome was the combination of cyclin E positive and Skp2 positive (difference in survival of 19% at 10 years, P = .0009). This combination was present in 78 (27%) of 288 cases for which data on both biomarkers were available. This combination was also highly associated with young age at diagnosis, grade 3 tumors, ER-negative status, HER2-negative status, and the basal biomarkers epidermal growth factor receptor and cytokeratin 5/6. However, in a multivariate model including standard clinicopathologic variables, this combination was not found to have independent prognostic significance. In conclusion, the combination of high cyclin E and Skp2 expression predicts for poor prognosis in breast cancer in univariate analysis only, it is associated with high risk features, and it is associated with the basal phenotype.