Primary angiosarcoma of the breast: clinicopathologic analysis of 49 cases, suggesting that grade is not prognostic.

Mammary angiosarcoma is a rare neoplasm, accounting for about 0.05% of all primary malignancies of the breast. It is currently believed that histologic grading of mammary angiosarcomas plays an important role in prognostication. Forty-nine cases of primary angiosarcoma of the breast were retrieved from our files. Clinical details and follow-up information were obtained from referring pathologists and clinicians, and by chart review. All statistics were performed using Fisher exact test and only P<0.05 was considered significant. Recurrence-free survival and overall survival curves were established using Statistica software version 5.5 (StatSoft Inc). All patients were female with ages ranging from 15 to 74 years (mean 41.5, median 40). Peak incidence was between the ages of 30 and 50 years. All tumors examined were located within breast parenchyma with or without minor cutaneous involvement. The right side was more commonly affected than the left side (66% vs. 29.5%). Tumor was bilateral at presentation in 2 cases (4.5%). Tumor size varied from 0.7 to 25 cm (mean 6.7, median 5). Most patients presented with a palpable, painless mass. Two patients had a history of prior radiation treatment for breast carcinoma. Histologically, primary tumors were graded using Rosen's 3-tier system: 17 tumors (35.4%) as low grade, 17 (35.4%) as intermediate grade, and 14 (29.2%) as high grade. Forty-six patients were treated surgically, 11 underwent chemotherapy, and 12 patients received radiotherapy. Follow-up was available in 41 patients (83.7%, median duration 29 mo). Ten patients (24.4%) showed evidence of local recurrence within 11 to 60 months (median 36) after diagnosis. Twenty-four patients (58.5%) thus far have developed metastases, which were most commonly to lung, liver, skin, and bone. Time interval between diagnosis and metastasis ranged from 2 to 144 months (median 34). Eighteen patients (44%) so far have died of disease and 1 died of presumably disseminated breast carcinoma. Five patients (12.2%) are alive with disease and 15 patients (36.6%) are alive with no evidence of disease. Statistical analysis evaluating correlation between tumor grade and size, and rate of local recurrence, metastasis, and death owing to disease showed no significant difference among tumors of different grades. The median recurrence-free and overall survival rates for the entire cohort were 2.8 and 5.7 years, respectively. In conclusion, mammary angiosarcoma is a rare disease that affects relatively younger patients. This tumor seems to have an overall similar clinical course as other types of angiosarcoma arising in skin or soft tissue; it carries a moderate risk of local recurrence, and a high risk of metastasis and death. In this large series, there is no correlation between histologic grade and patient outcome, more in line with angiosarcomas at other sites.