Breast cancer recurrence is the primary cause of mortality in breast cancer, and although advances have been made in the treatment of primary breast cancer, recurrent breast cancer remains uncurable. Targeted therapy has had a major impact on survival across multiple cancer types, including breast cancer, however patients who respond to targeted therapy ultimately relapse. Residual disease, the tumor cells that survive initial therapy, represent an attractive therapeutic target, however little is known about the biology of these cells. We use mouse models of breast cancer to investigate the phenotype of residual disease that survives targeted therapy, and to explore approaches to inhibit tumor recurrence. Residual disease exhibits cellular dormancy in models driven by distinct oncogenic pathways. Gene expression profiling reveals that residual tumor cells are enriched for a phenotype associated with normal and neoplastic stem-like cells, but are not enriched for tumor initiative cells. Interventions that inhibit inflammatory signaling inhibit tumor recurrence, however increasing inflammation promotes tumor recurrence. Inflammatory macrophages may be leukocytes that promote inflammation and drive recurrence in these models. Together, our findings present a more comprehensive picture of residual tumor cells surviving targeted therapy, and suggest possible therapeutic strategies for targeting residual disease that gives rise to cancer recurrence.