MYC amplification has been reported as a prominent feature of secondary angiosarcomas (SAS). The differential diagnosis between atypical vascular lesion (AVL) and low-grade angiosarcoma (AS) can be occasionally very difficult or even impossible, and MYC amplification status has been pointed as an important diagnostic tool to distinguish cutaneous vascular lesions of the breast. We assessed MYC amplification and protein expression status by fluorescent in situ hybridization (FISH) and immunohistochemistry (IHC), respectively, in 49 patients diagnosed with breast AS, and 30 patients diagnosed with post-radiation AVL of the breast. Clinical and pathological features, and follow-up data were collected, and survival analyses were performed. Among 37 patients with SAS, twenty patients had tumors with high-level MYC amplification and protein overexpression (54 %). None of primary angiosarcomas (PAS) or AVL cases showed MYC amplification or protein expression. Concordance between MYC amplification (FISH) and protein expression (IHC) was 100 % in AVL, PAS, and SAS. Survival analysis of the SAS patients demonstrates that those with MYC amplification had a significantly worse overall survival compared to cases without MYC amplification (P = 0.035). There was a non-significant trend toward a poor disease-free survival between cases with and without MYC amplification (P = 0.155). Our findings show that MYC amplification is a highly specific but poorly sensitive marker for SAS and, therefore, a negative result does not exclude the diagnosis of angiosarcoma. MYC amplification was associated with adverse prognosis, suggesting a prognostic role of MYC amplification status on SAS of the breast.

Angiosarcomas of the breast (ASB) are rare, representing <1% of breast malignancies. They can develop as a primary or secondary malignancy, also called post-radiation angiosarcoma. The aim of the this study is to discuss diagnosis, treatment, and outcome of primary and secondary ASB patients, diagnosed and treated in a single institution, over a 10-year period and to further compare the two conditions. We retrieved 28 consecutive cases of ASB, diagnosed from 1999 to 2009 at the European Institute of Oncology. Clinical and pathologic findings and survival analyses were performed. Of the 28 cases (27 women and 1 man), eight were primary breast angiosarcomas (PBA) and 20 were secondary breast angiosarcomas (SBA). Median follow-up was 23 months (range 1-112 months). Type of treatment (conservative or radical surgery) did not affect survival in both types of angiosarcomas. The clinical course observed was characterized by a high rate of local recurrence rather than distant metastasis or death from disease. There was a correlation between histological grade and prognosis of angiosarcomas with high-grade tumors presenting worse prognosis. SBA had a poorer prognosis compared to PBA. Our data indicate that primary and secondary ASB are distinct clinical and pathological features. SBA showed worse prognosis and was more often diagnosed in the study period compared to PBA. Physicians who care for patients who have been treated with radiation must be aware of its potential to induce angiosarcoma and stay vigilant in its detection.

Atypical vascular lesions (AVL) that occur in the field of prior radiation therapy for breast carcinoma are placed within the differential diagnosis with low grade angiosarcoma and other benign vascular lesions. Although considered a benign entity, the exact biological behavior of AVLs is not fully established because of the small number of cases reported in the literature. We aim to further characterize these lesions clinically and histopathologically, and to study their behavior. We report a series of 30 patients with AVL of the breast occurring after radiation exposure, diagnosed and treated at the European Institute of Oncology, Italy. Immunohistochemical study was performed in all cases, using CD31, D2-40, CD105, and Ki-67 antibodies. Twenty-seven patients were treated with standard doses of conventional adjuvant radiation therapy for the prior breast carcinoma. Three patients were treated with intraoperative radiotherapy with electrons. The post-radiation latency interval from breast carcinoma to AVL was 48.5 months (ranged from 1 to 146 months). Most of the lesions were classified as lymphatic type (78.6 %) based on D2-40 positivity. No extension into subcutaneous tissue or significant atypia was noted in all cases. Despite the fact that the AVL of our series have shown benign behavior in 93.3 %, one patient developed local recurrence of AVL, and two cases progressed to angiosarcoma at the previous AVL site. Further studies should be conducted to better understand the clinical behavior and to propose additional histopathologic diagnostic criteria to distinguish AVL from low grade angiosarcoma and those AVL with increased risk for malignant progression. Concerning current treatments of AVL, we recommend complete excision with free surgical margins and close follow up.

The aim of this study was to identify the prognostic factors associated with the risk of loco-regional recurrence (LRR) of women undergoing mastectomy and complete axillary dissection without radiotherapy. We analyzed data from 650 women operated between 1997 and 2001 in a single institution. Median follow-up was 10 years. Overall survival was 89.8 % at 5 years and 76.6 % at 10 years. The 10-year cumulative incidence of LRRs was 10.0 % (5.0, 10.5, 15.8, and 18.5 % in patients with 0, 1-3, 4-9, and ≥10 positive lymph nodes (LNs), respectively). Sixty-two (9.5 %) LRRs were observed, 5 (0.8 %) of which occurred in the axillary LNs. Supraclavicular LNs recurrences (n = 16, 2.5 %) occurred more frequently in patients with four or more positive LNs, Ki-67 ≥ 20 % or extensive peritumoral vascular invasion (PVI). At multivariable analysis, nodal status was the only prognostic factor for local events, while nodal status, Ki-67 and PVI were significant prognostic factors for recurrences in the regional LNs. Moreover, within each category of positive LNs, high values of Ki-67 and extensive PVI were associated with the highest risk of LRR while low values of Ki-67 and absence of extensive PVI were associated with the lowest risk of LRR. Women with node-negative tumors have the lowest risk of LRR and represent the group of patients that might benefit the least from radiotherapy. PVI and Ki-67 might help tailoring PMRT indications among patients with positive LNs. Finally, the very low incidence of recurrences in the axillary LNs raises questions about the inclusion of the axilla in the radiation field.