Background: Until recently, patients (pts) with stage IV breast cancer at diagnosis (BC) were generally treated with exclusive systemic therapy. There is recent retrospective evidence of the potential benefits of local surgery as part of the treatment plan. We attempted to evaluate, in a large retrospective series, local control (LC) rates achieved by loco-regional radiation therapy (LRRT) alone or in combination with surgery. Methods: We retrospectively reviewed the prospectively registered data of all pts treated at the Institut Gustave Roussy since 1990, for BC with evidence of distant metastases at diagnosis and who have received a loco-regional treatment. Survival curves were determined by the Kaplan-Meier method and Cox models were used for adjusted comparisons. Results: Between 1990 and 2003, 239 BC pts met the inclusion criteria. Median follow-up 6.5 years, median age: 55 years; ER-positive 68%, visceral metastases 51%. Primary chemotherapy was given in 86% of pts. Pts were categorized for loco-regional treatment as having received LRRT only (group 1; n=147) or breast and axillary surgery ± LRRT (group 2; n=92). Hormone therapy was given in 57% of ER positive BC pts until disease progression. LC rates at 3-years FU, metastases-progression-free survival (MPFS) and overall survival (OS) rates are shown in Table 1. Conclusions: Our results suggest that loco-regional treatment is an important part of the treatment of BC pts with metastases at diagnosis. Radiation therapy alone or in association with surgery provides an excellent local control. Better MPFS and OS rates in group 2 may be related to more favorable prognostic factors at diagnosis influencing the treatment choice.