The Impact of Boost Dose and Margins on the Local Recurrence Rate in Breast Conserving Therapy: Results From the EORTC Boost-No Boost Trial
Reviewer: Christine Hill, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 29, 2007
Presenter: H. Jones Presenter's Affiliation: Holzer Cancer Center for Cancer Care, Gallipolis, OH Type of Session: Plenary
Breast conserving therapy (BCT), which consists of lumpectomy followed by adjuvant radiation treatment (RT), is currently a standard option for patients with early-stage breast cancer.
Although BCT allows breast preservation and provides similar long-term overall survival benefit that is similar to mastectomy, it is associated with significant risk of ipsilateral breast tumor recurrence (IBTR), ranging from 0.5% to 2.5% per year.
In a recent overview of the randomized trials, the Early Breast Cancer Trialists group found IBTR to be correlated with overall survival, such that reduction in IBTR would be expected to reduce risk of death from breast cancer [EBCTG, Lancet, 2005].
Current radiation techniques after lumpectomy generally consist of 45-50 Gray (Gy) delivered to the whole breast, followed by a boost dose delivered to the tumor bed. The boost is delivered most commonly with an en face electron field, although photon or brachytherapy boosts are used less often.
This prospective, randomized trial was carried out to evaluate the necessity of boost delivery following whole-breast RT in preventing IBTR.
Materials and Methods
All patients enrolled on this trial underwent lumpectomy followed by whole breast RT of 50 Gy.
5318 patients had a microscopically complete excision. These patients were randomized to receive either no boost or a 16 Gy boost. 251 patients had a microscopically incomplete excision, and were randomized to either a 10 Gy boost or a 26 Gy boost.
A subset analysis was performed on 1724 patients whose pathologic specimens were reviewed centrally at the Netherlands Cancer Institute.
Final margin status (FMS) was evaluated in relation to age, tumor size, receptor status, histology, and use of adjuvant systemic therapy.
Margin status was defined as follows:
Negative: Tumor ≥ 2mm from specimen edge, or no tumor present on re-excision pathology.
Close: Tumor present < 2mm from specimen edge.
Postive: Specimen edge involved with tumor (level of involvement was not scored during this study).
Median follow-up data for the 5318 patients with microscopically complete excision was 10.8 years.
IBTR rate was 12% in patients who did not receive a boost versus 7% in patients who did (p < 0.0001).
The hazard ratio of IBTR was 0.59 (0.46 – 0.76) in favor of the boost.
This difference was observed in all age categories; however, the absolute risk reduction was largest in patients ≤ 40 years (23.9% versus 13.5%, p = 0.0014).
The rate of severe fibrosis was 1.6% in patients who did not receive a boost versus 4.4% for those who did.
At 10 years, cumulative incidence of IBTR in the group of patients who had incomplete tumor resections was 17.5% versus 10.8% for patients receiving a 10 Gy boost versus those receiving a 26 Gy boost (p > 0.1).
Of the subset of patients with invasive breast cancer whose pathology underwent centralized review, 78% had negative margins, 12% had close margins, 7% had margins positive for invasive disease, and 3% remained unclassified.
Of the subset of patients with ductal carcinoma in situ (DCIS) whose pathology was reviewed centrally, 57% had negative margins, 26% had close margins, and 15% had margins positive for DCIS.
Overall, this subset analysis demonstrated factors associated with increased risk of IBTR to be age < 50, presence of DCIS at the specimen margin and high grade tumor histology (either invasive or DCIS). Invasive tumor at specimen edge was not a prognostic factor for IBTR. Patients with these risk factors were found to benefit most from the addition of a boost to whole breast RT:
For patients with DCIS, 10 year IBTR rates were found to be 8% for patients with negative margins, 10% for patients with close margins, and 14% for those with positive margins (p = 0.02).
For those DCIS patients with positive margins, the IBTR rate was decreased from 15% to 6% with the addition of a boost (p < 0.001). For those with negative margins, the IBTR rate did not differ significantly with the addition of a boost (p = 0.77).
For patients noted to have high grade DCIS, the IBTR rate was decreased from 17% to 5% with the addition of a boost.
For patient with invasive breast cancer, 10 year IBTR rates were 6% for those with negative margins, 8% for those with close margins, and 11% for those with positive margins (p = 0.24).
For those patients with invasive breast cancer and positive resection margins, the IBTR rate was decreased from 13% to 4% with the addition of a boost (p < 0.01). Again, for those patients with negative margins, no significant difference was observed in IBTR rate with boost versus no boost (p = 0.87).
For patients with high grade invasive cancer, the IBTR rate was decreased from 19% to 7% with the addition of a boost.
On a multivariate analysis, age > 50 years and receipt of boost therapy were the factors that reduced the risk of IBTR the most.
The authors conclude that age < 50 years and high grade invasive tumor histology with or without DCIS are more significant prognostic factors for IBTR than is final margin status.
Additionally, the addition of a boost to whole breast RT reduces the effect of involved margins and decreases the risk of IBTR in high risk patients.
Current BCT techniques involve the use of whole breast RT after lumpectomy, followed by a boost, usually delivered with an en face electron field to the tumor bed.
This study supports the addition of a boost to reduce risk of IBTR.
In light of the recent correlation between IBTR and overall survival demonstrated by the Early Breast Cancer Trialists Group, the use of a boost after EBRT would be expected to improve overall survival, although this was not examined as part of this study.
Current practices often involve multiple re-excision surgeries in attempt to achieve negative surgical margins. Interestingly, in this study, the presence of margins with involvement of invasive tumor did not significantly impact the risk of IBTR. The presence of DCIS at the tumor margin did impact the risk of IBTR, however.
Additionally, addition of a boost to patients with margins involved with either DCIS or invasive cancer significantly reduced the risk of IBTR. This was not the case for patients with negative margins, for whom addition of the boost did not significantly impact risk of IBTR.
These data imply that re-excision surgery may be avoidable in patients with microscopically involved margins, as the addition of the boost may offset any increased risk of IBTR imposed by disease present at the margin.
As a caveat, however, the amount of tumor present at the margin was not scored in this analysis. The presence of gross disease at the margin would be expected to be of considerably different clinical importance than would microscopic disease.
Although patients in this trial were treated with several different boost doses, this data was not presented as part of this abstract. This data will certainly be of interest in the future, as high risk patients (those of young age or with high grade tumor histology, as identified in this analysis) may in fact benefit from higher boost dosing.
Partially funded by an unrestricted educational grant from Bristol-Myers Squibb.