Correlation of clinical and pathologic features with outcome in patients with ductal carcinoma in situ of the breast treated with breast-conserving surgery and radiotherapy.

PURPOSE: Although breast-conserving surgery followed by radiotherapy (RT) has become a standard treatment option for patients with ductal carcinoma in situ of the breast, risk factors for ipsilateral breast tumor recurrence (IBTR) in these patients remain an active area of investigation. The purpose of this study was to evaluate the impact of clinical and pathologic features on long-term outcome in a cohort of DCIS patients treated with breast-conserving surgery plus RT.

METHODS AND MATERIALS: Between 1973 and 1998, 230 patients with DCIS were treated with breast-conserving surgery plus RT at our institution. All patients were treated by local excision followed by RT to the breast to a total median tumor bed dose of 64 Gy. Adjuvant hormonal therapy was used in only 20 patients (9%). All available clinical, pathologic, and outcome data, including ipsilateral and contralateral events, were entered into a computerized database. The clinical and pathologic variables evaluated included detection method, mammographic appearance, age, family history, histologic subtype, presence of necrosis, nuclear grade, final margin status, and use of adjuvant hormonal therapy.

RESULTS: As of December 15, 2000, with a median follow-up of 8.2 years, 17 patients had developed a recurrence in the ipsilateral breast, resulting in a 5- and 10-year IBTR rate of 5% and 13%, respectively. Contralateral breast cancer developed in 8 patients, resulting in a 10-year contralateral recurrence rate of 5%. Patient age, family history, histologic subtype, margin status, and tumor grade were not significantly associated with recurrence on univariate analysis. A significantly higher rate of local relapse was observed in patients with the presence of necrosis. The 10-year relapse rate was 22% in 88 patients with necrosis compared with 7% in 142 patients without necrosis (p <0.01). In multivariate analysis, the presence of necrosis remained a significant predictor of local relapse. No breast relapses occurred among the 8 patients with positive margins, and three relapses developed among 21 patients with close margins. The rate of IBTR in those with close/positive margins did not differ from the rate in those with negative or unknown margins. It is also notable that none of the 20 patients treated with adjuvant tamoxifen had developed IBTR or a contralateral event to date, although the follow-up on this group was still too short to reach significance.

CONCLUSION: In this cohort of uniformly treated patients with a relatively long follow-up, the presence of necrosis was a significant predictor of local relapse. However, positive or close margin status was not a significant predictor of local relapse. Although none of the patients receiving tamoxifen had a recurrence in the ipsilateral or contralateral breast, longer follow-up is required to assess the effect of tamoxifen on these end points.