R.F. Abi Raad

Influence of margin status and tumor bed boost dose on local recurrence rate in breast-conserving therapy: does a higher radiation dose to the tumor bed overcome the effect of close or positive margin status in breast-conserving therapy?

Purpose/Objective: Two randomized trials of patients treated by breast conserving therapy have demonstrated a significant increase in local control, for patients with negative margins, when a tumor bed boost was used in addition to whole breast irradiation. However, in patients with positive or close margins, the adequacy of tumor bed boost for local control and the optimal boost dose remain controversial. Published reports have suggested that tumor bed boost, even at higher doses, is not sufficient to overcome the negative effect of close or focally positive margins. The purpose of this study is to assess the impact of boost dose on local recurrence rate according to margin status. Materials/Methods: Six hundred thirty-five patients with clinical Stage I or II breast cancer were treated by breast-conserving surgery and radiation between 1992 and 1998 at the Massachusetts General Hospital. The median follow up was 54 months (range 6 ‐120). As an institutional policy, all pathologic specimens were re-reviewed upon referral and prior to de finitive treatment. Final margin status was determined based on the distance of invasive or in situ carcinoma from inked surgical margin: negative (no residual tumor or at least 2 mm from margin), close (less than 2 mm from margin), focally positive, or grossly positive. Patients received 50 Gy to the whole breast using 2 Gy per fraction, followed by a 10 Gy en face electron boost in 5 fractions for patients with negative margins and a 15‐18 Gy (15Gy/5 fractions in 28, 18 Gy/9 fractions in 80) high dose boost for focally-close, focally-positive or grossly-positive margins. Results: Sixty-five percent of the patients were T1, 35% T2, 72% N0, and 28% N1-2. Margin status was as follows: negative 71% (n450), close 18.4% (n117), and focally positive 11% (n68). There were no patients with grossly positive margin. The 5-year Kaplan-Meier local recurrence rates (LR) in the different margin groups were: 1.4% for negative margin, 3.2% for close margin, and 5.6% for focally close margin (p0.1169). When close (2mm) or focally positive margins were grouped together, the 5-year local failure rate was 4.1% (compared to LR of 1.4% in negative margin, p0.0115). Overall, the local recurrence rate was not statistically different in the standard versus high boost dose (p0.1745). Furthermore, in each of the margin groups, there was no statistically signi ficant difference between standard tumor bed boost and high tumor bed boost dose. The 8-year LR was identical to the 5-year LR. 5-year overall survival rates were 92.1% for negative margin, 89.9% for close margin and 79.4% for focally positive margin (p0.0087). Eight-year overall survival rates were 82.4%, 87.8%, and 64% (p0.0019), respectively. On univariate and multivariate analysis, only margin status was signi ficant in predicting local recurrence. Lymph node status, tumor size, age, and boost dose were not predictive of local recurrence (p0.12, p0.67, p0.98, and p0.34). Conclusions: This study shows increased local recurrence rate with close or focally positive final surgical margin in breast conserving therapy, which was not significantly altered by increasing the boost dose. Patients with negative surgical margin prior to radiation had low local recurrence rate of 1.4%, which was durable at 8 years. Patients with close or positive margin, however, had nearly 3x higher local recurrence rate, observed in the first 5 years of follow-up and persistent at 8 years. A risk of recurrence around 5% may still be regarded as acceptable in breast conserving therapy, but improving the margins surgically can further improve local control. Patients with close or focally positive margins should therefore undergo re-excision to render them margin-negative prior to radiation.