Background: Controversy exists concerning the natural history of ductal carcinoma in situ (DCIS) of the breast, including its pathologic expression and treatment. This controversy has been fostered largely by the retrospective nature and limited sample sizes of extant studies.
Method: Resolution of some of these issues was attempted by analyzing the pathologic features of 573 examples of DCIS obtained from a larger cohort of 790 women with DCIS enrolled in Protocol B-17 of the National Surgical Adjuvant Breast Project. This prospective randomized clinical trial was performed to assess the efficacy of local breast irradiation to reduce the incidence of second ipsilateral breast tumors (IBT) after lumpectomy.
Results: Tumor and patient characteristics, including significantly less IBT for those treated by lumpectomy and irradiation than lumpectomy alone, were almost identical for the subset comprising this analysis and the total B-17 cohort reported previously. The presence of moderate/marked comedo necrosis, which was evaluated as an independent parameter rather than as a specific histologic type of DCIS and uncertain/involved lumpectomy margins were the only statistically significant independent predictors of IBT for patients treated by lumpectomy as well as irradiation. The latter markedly reduced the annual hazard rates for the IBT associated with these indicators.
Conclusions: Although not an endpoint of this study, the authors' findings suggest that the beneficial effect of irradiation in reducing IBT after lumpectomy for DCIS occurs with small (< 1.0 cm.) and larger lesions. Moderate/marked comedo necrosis and uncertain/involved lumpectomy margins represent independent predictors of IBT.