Thrombospondin 1 protein expression relates to good prognostic indices in ductal carcinoma in situ of the breast.

Abstract

AIM: Angiogenesis plays an important role in tumour growth and has been shown to occur around both in situ and invasive tumours. The degree of angiogenesis within tumours depends on the balance of pro-angiogenic and anti-angiogenic factors. One such anti-angiogenic factor is thrombospondin 1 (TSP-1). This study investigates the pattern of expression of TSP-1 in ductal carcinoma in situ (DCIS) of the breast and its relation to the surrounding microvessel pattern and density. MATERIALS/METHODS: The expression of TSP-1 was studied in formalin fixed, paraffin wax embedded sections from 58 cases of pure DCIS, using a monoclonal antibody against TSP-1 and the avidin-biotin-diaminobenzidine immunoperoxidase detection system. Vessels were stained with a monoclonal antibody to the endothelial cell marker CD31. Stromal microvessel density was assessed by counting "hot spots" within 500 micro m of the basement membrane of involved ducts using a 25 point Chalkey graticule. RESULTS: TSP-1 staining of the basement membrane around duct spaces with DCIS was seen in 69% of cases. In addition, staining of the stroma between involved duct spaces was seen in 31% of cases, with a fibrillary pattern identical to that seen in invasive breast carcinomas. In 12% of cases no staining for TSP-1 was seen. Two patterns of vascularity were identified. A cuff of vessels immediately adjacent to the basement membrane of ducts with DCIS was seen in 71% of cases. The presence of stromal TSP-1 was significantly associated with DCIS showing no/little necrosis (p = 0.01) and no/little periductal inflammation (p = 0.04). There was a trend between the presence of stromal TSP-1 and tumour cell negativity for p53 (p = 0.087). The stromal microvessel Chalkey point count ranged between 3.33 and 16. An increased stromal microvessel count was associated with high histological grade (p = 0.02), extensive necrosis (p = 0.047), and pronounced periductal inflammation (p = 0.049). There was no association between the presence of stromal TSP-1 and stromal microvessel density. CONCLUSIONS: TSP-1 is expressed in the stroma around DCIS and in the immediately adjacent basement membrane. Expression of stromal TSP-1 is lost in DCIS with more aggressive histological features. The absence of a relation with microvessel density suggests that other angiogenic factors may play an important role in DCIS.