One hundred forty-five consecutive women who underwent mastectomy with or without radiation therapy for DCIS from 1998 to 2005 were included in this retrospective analysis. Only patients with pure DCIS were eligible; patients with microinvasion were excluded. The primary endpoint was local recurrence, defined as recurrence on the chest wall; regional and distant recurrences were secondary endpoints. Outcomes were analyzed according to margin status (positive, close (¡Ü2 mm), or negative), location of the closest margin (superficial, deep, or both), nuclear grade, necrosis, receptor status, type of mastectomy, and receipt of hormonal therapy.
The primary cohort consisted of 142 patients who did not receive postmastectomy radiation therapy (PMRT). For those patients, the median follow-up time was 7.6 years (range, 0.6-13.0 years). Twenty-one patients (15 % ) had a positive margin, and 23 patients (16 % ) had a close (¡Ü2 mm) margin. The deep margin was close in 14 patients and positive in 6 patients. The superficial margin was close in 13 patients and positive in 19 patients. One patient experienced an isolated invasive chest wall recurrence, and 1 patient had simultaneous chest wall, regional nodal, and distant metastases. The crude rates of chest wall recurrence were 2/142 (1.4 % ) for all patients, 1/21 (4.8 % ) for those with positive margins, 1/23 (4.3 % ) for those with close margins, and 0/98 for patients with negative margins. PMRT was given as part of the initial treatment to 3 patients, 1 of whom had an isolated chest wall recurrence.
Mastectomy for pure DCIS resulted in a low rate of local or distant recurrences. Even with positive or close mastectomy margins, the rates of chest wall recurrences were so low that PMRT is likely not warranted.