Treatment for breast sarcoma: a large, single-centre series.

AIMS: To report outcomes in breast sarcoma in the context of a major series from a tertiary referral centre.
METHODS: Retrospective analysis was performed on patients with histologically-proven breast sarcoma treated between 1996 and 2006. Kaplan-Meier survival curves were constructed and differences assessed by Log-Rank and Wilcoxon tests.
RESULTS: 63 patients were identified; 57 underwent treatment with curative intent. 24 patients had undergone previous radiotherapy. 36 patients who underwent primary surgery elsewhere were referred for further treatment, of which 22 had at least one involved margin from primary resection. Surgery performed and margins status varied between patients undergoing primary surgery at this institution (n = 21; WLE = 8, mastectomy = 12, chest wall resection = 1, involved margins = 2 [10%]) or at a referring institution (n = 36; lumpectomy = 25, mastectomy = 11, involved margins = 22 [61%]), although there was no difference in tumour size or previous radiotherapy status. Previous irradiation was associated with poor prognosis. A greater proportion of these patients required primary mastectomy to ensure adequate clearance; the majority of the post-irradiation tumours were angiosarcomas (15/19) and significantly more relapsed locally (P < 0.001). All patient disease-free survival (DFS) rates were 71% at 2 and 42% at 5 years. DFS improved when primary surgery was undertaken at a high-volume sarcoma unit; 2-yr 84%vs75%; 5-yr 58%vs37%. There was a trend towards worse DFS with increasing size and increasing grade of tumour but this did not attain significance.
CONCLUSIONS: Radiation-induced breast sarcoma has worse local recurrence rates compared to primary breast sarcoma. Involved margins were fewer at a specialist unit, which may translate into improved outcome.