Originally published as Ann Surg Oncol Early Release 10.1245/ASO.2004.07.001 on February 9, 2004
Annals of Surgical Oncology 11:316-321 (2004)
© 2004 Society of Surgical Oncology
Avoidance of Adjuvant Radiotherapy in Selected Patients With Invasive Breast Cancer
Susan H. Lee, MD,
Maureen A. Chung, MD, PhD,
David Chelmow, MD and
Blake Cady, MD
From The Breast Health Center, Program in Women’s Oncology, Women & Infants’ Hospital of Rhode Island, Providence, Rhode Island (SHL, MAC, BC); Department of Surgery, Brown University School of Medicine, Providence, Rhode Island (MAC, BC); and Department of Obstetrics and Gynecology, Division of General Obstetrics and Gynecology, Tufts-New England Medical Center, Boston, Massachusetts (DC).
Correspondence: Address correspondence and reprint requests to: Blake Cady, MD, University Surgical Associates-APC 120, Rhode Island Hospital, 593 Eddy Street, Providence, RI 02903; Fax: 401-868-2313; E-mail: bcady{at}usasurg.org
 |
ABSTRACT
|
|---|
Background: Current standard therapy for invasive breast carcinoma is mastectomy or breast conservation with adjuvant radiation. Data from randomized trials suggest no advantage for radiotherapy after lumpectomy in highly selected patients. Selective radiotherapy would make contemporary breast cancer therapy more rational with decreased morbidity and expense.
Methods: A total of 163 patients were treated by breast conservation without adjuvant radiation between 1978 and June 2003. They declined radiation after discussion or had medical contraindications. The great majority were postmenopausal and had lower-grade T1 tumors with resection margins 
1 cm and no nodal metastases. The goal was to identify patients with favorable prognostic features for omission of postsurgical irradiation without impaired local recurrence or survival.
Results: Twenty patients (12%) had local recurrences; 17 (10%) were invasive, and 3 (2%) were ductal carcinoma-in-situ. An ideal patient subgroup >50 years of age with grade 1 or 2 cancers 
1.5 cm in diameter and with surgical margins 1 cm was empirically defined. Of 80 such patients, 5 (6%) had local recurrence; 3 (3.5%) were invasive, and 2 (2.5%) were ductal carcinoma-in-situ.
Conclusions: A defined ideal subset of older breast cancer patients with smaller, lower-grade cancers and adequate excision margins can be treated with lumpectomy without irradiation and with minimal local recurrence.
Key Words: Breast cancer • Radiation • Recurrence • Lumpectomy
|
INTRODUCTION
|
|---|
Adjuvant radiation after breast-conservation therapy for cancer is regarded as standard therapy today in the United States. Ten randomized, controlled trials1–10 have compared breast-conservation surgery (BCS) and adjuvant radiotherapy with BCS alone; radiotherapy decreased the risk of local recurrence by 67%, but overall survival was the same. Subset analysis in these studies2,3,6,9 suggested that radiotherapy contributed little to local control in patients >55 years of age with small cancers of favorable histology (non-high grade), negative nodes, and adequate resection margins. With increased mammographic screening, breast cancers are detected at smaller sizes with lower grades, fewer nodal metastases, and improved survival. Because more prognostically favorable tumors are thus discovered, it is important to identify a subgroup of patients who, after breast conservation, may be able to avoid radiation and the attendant time, travel, expense, and morbidity that it entails.
|
METHODS
|
|---|
A total of 163 patients treated with BCS and no adjuvant radiotherapy in the practice of 1 author (B.C.) at the New England Deaconess Medical Center (Boston, MA) from 1978 to 1997 and the staff at the Breast Health Center of Women & Infants’ Hospital (Providence, RI) from 1997 through June 2003 were followed up. Patients declined radiotherapy after detailed discussions of standard practices and options, which were encouraged. Some patients declined radiation despite recommendations or had medical conditions that precluded radiation treatment, e.g., connective tissue disease, advanced age, or infirmity. Patients with inadequate resection margins initially were recommended for re-excision to ensure a tumor-free margin 1 cm if they were to forgo irradiation, but some patients declined additional surgery. The patients were followed up by regular mammography and clinical examinations. Chart review and tumor registry data as of June 2003 were used to determine current status in all patients.
The patients were monitored for local and regional nodal recurrences and distant metastases. Local recurrence was defined as the clinical reappearance of the original cancer in the primary surgical field or its immediate surrounding tissue. Regional nodal recurrence was the appearance of clinical metastases in the regional lymph nodes through the persistence and growth of previously established nodal metastatic malignant cells. Metastases were defined as the presentation of disease at a distance from the primary breast cancer in distant organs.
Individual patient and tumor characteristics (age, grade of tumor, size of tumor, margin distance, extensive intraductal component presence, estrogen receptor or progesterone receptor status, and tamoxifen therapy) were tested for significance as univariate predictors for local and total recurrence by using the 
2 method and by calculating the relative risk (RR) with 95% confidence intervals (CI). Multivariate logistic regression was used to determine the significance of the four selected ideal characteristics as predictors for local and total recurrence. Characteristics from subset analyses in randomized trials2,3,6,9 and the Van Nuys prognostic index for ductal carcinoma-in-situ (DCIS)11 were considered in selecting these characteristics. Further analyses of these features were performed to determine whether they were associated with lower local recurrence rates in our population. Univariate outcome comparisons for the ideal versus the nonideal group were performed with Kaplan-Meier estimates for means and Cox regression for RR and CI. The program for statistical analysis was SPSS 11.5 (SPSS Inc., Chicago, IL). The institutional review boards of Women & Infants’ Hospital and the New England Deaconess Hospital approved the study.
|
RESULTS
|
|---|
The age range of the patients was 41 to 96 years, with a median of 64.5 years. Most patients (90%) were older than 50 years, and most were likely postmenopausal. Cancers had usually been detected by screening mammography (79%), were of smaller size (T1a, 29%; T1b, 49%; T1c, 19%), were of lower histological grade (grades 1 or 2, 81%), and had no clinical evidence of axillary nodal metastases. When surgical axillary evaluation was performed, 95% of nodes were histologically negative. The resection margins were adequate (1 cm) in 83% of patients.
The 163 patients included 138 with no evidence of recurrence and 25 (15%) with local recurrence or metastases (Table 1). Recurrences were local in 20 (12%) patients, regional in 4 (3%) patients, or distant metastases in 11 (7%) patients; 8 patients had more than 1 site of recurrence. Because adjuvant radiotherapy is administered only for local control, patients with local recurrence were the major interest of this study. A total of 143 patients had no local recurrence, whereas 20 (12%) had local recurrence: 17 (10%) with invasive disease and 3 (2%) with DCIS (Table 1).
All patients were monitored for local recurrence and regional and distant metastases. The follow-up period was 0 to 209 months, with a median of 56 months. The patient with 0 months of follow-up had her surgery without plans for radiotherapy but then was lost to follow-up. Only nine patients had less than a minimum follow-up of 12 months. Five of these patients were in the ideal group, of which one patient died within 4 months of diagnosis of congestive heart failure and of which two were diagnosed in 2002. Three were in the nonideal group, of which one patient was diagnosed in 2002; all had no evidence of disease (NED) at the last known follow-up. One was in the unknown group; however, she died of causes unrelated to this breast cancer at 7 months.
The univariate characteristics found to be statistically significant as individual predictors for local recurrence were grade 3 vs. 1 or 2 (RR, 5.02; 95% CI, 2.26–11.15) and size >1.5 vs. 1.5 cm (RR, 3.02; 95% CI, 1.08–8.43). The same characteristics were found to be statistically significant as individual predictors for total recurrence (grade 3 vs. 1 or 2: RR, 4.52; 95% CI, 2.19–9.32; size >1.5 vs. 1.5 cm: RR, 4.28; 95% CI, 2.10–8.72). However, for total recurrence, age 50 vs. >50 years (RR, 2.3; 95% CI, 1.00–5.28) was close to significant. Margin distance, an extensive intraductal component presence, estrogen receptor or progesterone receptor status, and tamoxifen therapy were not found to be statistically significant in local or total recurrence.
When multivariate analysis of the four ideal characteristics was performed, grade 3 vs. 1 or 2 (RR, 7.92; 95% CI, 2.62–23.91) was the only statistically significant characteristic. Age (50 vs. >50 years; RR, 3.54; 95% CI, .78–16.00) and size (>1.5 vs. 1.5 cm) were not statistically significant with multivariate modeling.
Sixteen patients had lymph node dissection, and 44 patients had sentinel lymph node biopsies. Only 3 of these 60 patients had evidence of axillary metastases: 2 had macrometastases, and 1 had a micrometastasis. They did not have recurrences, and all were NED at their last follow-up, a median of 70 months (range, 27–138 months) after surgery. One hundred patients did not have surgical nodal analysis; they all had a negative axillary clinical examination, and three patients had no accurate data.
Univariate outcome comparisons of the type of recurrence comparing nonideal versus ideal groups were all statistically significant: local recurrence (RR, 3.57; 95% CI, 1.37–9.09), regional recurrence (P = .041; 2 was performed because few patients had regional recurrence), distant metastases (RR, 12.5; 95% CI, 1.75–100), and total recurrence (RR, 4.76; 95% CI, 1.25–1.81). The differences in time to recurrence, disease-free survival (DFS), and total survival were not statistically significant.
An ideal group of patients was defined empirically by analyses that revealed lower local recurrence rates for selected features (Table 2). The favorable factors were age older than 50 years, cancers 1.5 cm in maximum diameter, tumor grade 1 or 2, and resection margins of at least 1 cm. Of the 80 patients so defined, local recurrences were diagnosed in 5 (6.0%): 3 invasive (3.5%) and 2 DCIS (2.5%). Of these five patients, four were alive with NED after a median DFS of 74 months (range, 15–86 months) since local recurrence and a median total follow-up of 104 months (range, 70–174 months). The fifth patient died of disease (DOD), with a DFS of 52 months to the local recurrence and a total survival of 138 months. This patient was noted to have an invasive local recurrence, and the tumor registry defined her as DOD 86 months after the local recurrence. Distant metastases were assumed; however, the clinical circumstances and survival duration call into question whether the breast cancer was the cause of death.
Sixty-seven patients did not qualify for the ideal group because they did not possess all four of the ideal characteristics. These patients had 20 (30%) total recurrences. Nine (13%) had only local recurrences, of which eight were invasive (two DOD and six NED) and one was DCIS (NED at 29 months). Once local recurrence only was determined, patients were treated with local excision only (declined adjuvant radiotherapy), local excision and radiotherapy, or mastectomy. The exact breakdown data were unavailable. Six patients had local recurrence accompanied by regional or distant metastases (five DOD and one alive with disease). Two patients had regional and distant metastases (both DOD). Three patients had distant metastasis without local recurrence or regional nodal metastases (two DOD and one NED). In the patient classified as NED, the records indicated an abdominal exploration with small-bowel metastases, but she had no other evidence of disease. Because of the DFS of 144 months to the diagnosis of abdominal metastatic disease and no other evidence of disease, this most likely was a second primary cancer of gastrointestinal or ovarian origin. However, because of the arbitrary classification of the tumor registry, she was classified as distant metastases only and remains NED.
There were 16 patients older than 50 years of age with cancers 1.5 cm who could not be clearly defined as ideal or nonideal because the data for tumor grade (5 patients), resection margin (9 patients), or both (2 patients) were not available. They were placed in an unknown group and omitted from statistical comparison of the completely defined ideal group (Table 3). When defined, grade and margin data were consistent with the ideal group (i.e., grade of 1 or 2 and margins 1 cm). Given that 82% of our patients with a known grade had grade 1 or 2 tumors, that 82% had resection margins >1 cm, that the known factors were ideal, and that they were all alive with NED, it could be assumed that these patients were also in our ideal subset. With this assumption on a separate analysis, the expanded ideal group consisted of 96 patients, and 5 (5%) patients had a local recurrence (Table 4).
|
DISCUSSION
|
|---|
Ten randomized prospective controlled trials1–10 have compared BCS and adjuvant radiotherapy with BCS alone. The trials differed in selection of patient age, tumor size, nodal status, extent of surgery, type of adjuvant radiotherapy, and use of adjuvant systemic hormonal therapy and chemotherapy. In a compilation of all these studies, the median breast cancer recurrence rate was 21% (range, 5%–39%) with BCS alone versus 7% (range, 2%–18%) with BCS and adjuvant radiotherapy. The median follow-up was 66 months (range, 24–240 months). In our ideal group of patients and our expanded ideal group of patients, the local recurrence was 6% after a median follow-up of 61 months (range, 1–147 months) and was 5% with a median follow-up of 59 months (range, 1–174 months); these breast cancer recurrence rates are similar to those of trial participants treated by BCS and radiation.
In 4 of 10 studies, patients were stratified to identify patient or tumor characteristics associated with low recurrence rates without adjuvant radiation. In the trial of Clark et al.,2 after a follow-up of 7.6 years, women older than 50 years of age with tumors <2 cm had a 22% local recurrence rate; women older than 50 years of age with grade 1 or 2 tumors had a 24% local recurrence rate; and women with tumors <1 cm had a 28.5% local recurrence rate. Women in the irradiated groups had an 11% local recurrence rate. In the study of Fisher et al.,3 the benefit of radiotherapy was found to be independent of the nodal status of the patient. After a 20-year follow-up, local recurrence in node-negative patients treated with BCS alone and BCS and radiotherapy was 36% and 17%, respectively. In node-positive patients, it was 44% and 9%, respectively, for BCS alone versus BCS and radiotherapy. In the Swedish trial,6 after 10 years of follow-up, women older than 55 years of age with a noncomedo or nonlobular carcinoma had a local recurrence rate of 11% without radiotherapy versus 6% with radiotherapy. In the Milan III trial,9 patients older than 65 years of age in both the irradiated and nonirradiated groups had a 5% local recurrence rate after a 10-year follow-up. Again, our local recurrence rates of 6% in the ideal group and 5% in the expanded ideal group are similar to the recurrence rate after radiotherapy of the low-risk patients as defined by these studies. This may be attributed to a selection of four factors associated with low rates of recurrence.
Univariate analysis demonstrated grade 3 and size >1.5 cm to be significant predictors for local and all recurrences. A surgical margin <1 cm was not found to be significant, but this could be attributed to only 25 (15%) of all patients having nonideal margins; thus, the number of subjects may too small to calculate statistical significance. It is interesting to note that in the trials analyzing subset populations, those with the higher local recurrence rates had smaller tumor-free margins: Clark et al.2 had margins of .5 to 1 cm, and Fisher et al.3 had no minimal margins (tumor not transected), in contrast to 2 cm in the Swedish trial6 and 2 to 3 cm in the Milan III trial.9 Our results demonstrate that patient and tumor characteristics can be used to select a group of patients who may derive little or no additional benefit from adjuvant radiotherapy after BCS for cancer.
|
SUMMARY
|
|---|
An ideal group of 80 breast cancer patients was empirically defined for avoidance of adjuvant radiation after BCS on the basis of older age (>50 years), smaller size (1.5 cm), lower grade (grade 1 or 2), and adequate surgical margins (1 cm). The local recurrence rate was only 6% after a median DFS of 64 months (range, 15–86 months). Median follow-up was 61 months (range, 1–174 months); only one patient was DOD. When patients from the unknown group (grade or margin data not available) were included in the ideal group, the expanded ideal group of 96 patients had a local recurrence rate of 5% and a median DFS of 64 months (range, 15–86 months). The median follow-up was 59 months (range, 1–174 months), and no additional patient was DOD. In patients who had one or more nonideal characteristics, the local recurrence rate was 22%.
Consideration should be given for BCS without adjuvant breast radiotherapy in such selected ideal patients after a thorough discussion, including a description of standard alternatives and risks and benefits of radiotherapy. If treated with surgery alone, these patients would still be candidates for radiotherapy in the future if local recurrence occurred. Patients not in the ideal group should be considered for adjuvant radiotherapy after BCS.
|
ACKNOWLEDGMENTS
|
|---|
The acknowledgments are available online in the fulltext version at www.annalssurgicaloncology.org. They are not available in the PDF version.
Supported by a grant from the Susan G. Komen Breast Cancer Foundation.
|
FOOTNOTES
|
|---|
When 163 patients were treated by breast-conservation surgery without radiotherapy, a subset of older patients with smaller, lower-grade cancers and adequate surgical margins was defined that had a 6% local recurrence rate. Lumpectomy without adjuvant radiotherapy is safe for highly selected patients.
Received for publication July 1, 2003.
Accepted for publication November 18, 2003.
|
REFERENCES
|
|---|
- Blamey RW. The British Association of Surgical Oncology Trial (BASO II) of the treatment of small differentiated node negative tumors (abstract). Breast Cancer Res Treat 1999; 47: 50.
- Clark RM, Whelan T, Levine M, et al. Randomized clinical trial of breast irradiation following lumpectomy and axillary dissection for node-negative breast cancer: an update. J Natl Cancer Inst 1996; 88: 1659–64.[Abstract/Free Full Text]
- Fisher B, Anderson S, Bryant J, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med 2002; 347: 1233–41.[Abstract/Free Full Text]
- Forrest AP, Stewart HJ, Everington D, et al. Randomized controlled trial of conservation therapy for breast cancer: 6-year analysis of the Scottish trial. Scottish Cancer Trials Breast Group. Lancet 1996; 348: 708–13.[CrossRef][Medline]
- Holli K, Saaristo R, Isola J, et al. Lumpectomy with or without postoperative radiotherapy for breast cancer with favourable prognostic features: results of a randomized trial. Br J Cancer 2001; 84: 164–9.[CrossRef][Medline]
- Liljegren G, Holmberg L, Bergh J, et al. 10-Year results after sector resection with or without postoperative radiotherapy for stage I breast cancer: a randomized trial. J Clin Oncol 1999; 17: 2326–33.[Abstract/Free Full Text]
- Renton SC, Gazet JC, Ford HT, et al. The important resection margin in conservation surgery for breast cancer. Eur J Surg Oncol 1996; 22: 17–22.[CrossRef][Medline]
- Spooner D, Morrison JM, Oates GD, et al. The role of radiotherapy in early breast cancer (stage I). A West Midlands Breast Group prospective randomized collaborative study (BR 3002). Breast 1995; 4: 231.
- Veronesi U, Marubini E, Mariani L, et al. Radiotherapy after breast conserving surgery in small breast carcinoma: long-term results of a randomized trial. Ann Oncol 2001; 12: 997–1003.[Abstract/Free Full Text]
- Wolmark N, Digman J, Margolese R, et al. The role of radiotherapy and tamoxifen in the management of node negative invasive breast cancer-1.0 cm treated with lumpectomy: preliminary results of NSABP protocol B-21 (abstract). Proc Am Soc Clin Oncol 2000; 19: 70a.
- Silverstein MJ. The University of Southern California/Van Nuys prognostic index. In: Silverstein MJ, ed. Ductal Carcinoma In Situ of the Breast. Philadelphia: Lippincott Williams & Wilkins, 2002: 464–5.
This article has been cited by other articles:
|
|
|
|
 
B. Cady
Simplification of Breast Cancer Surgery
Ann. Surg. Oncol.,
January 1, 2005;
12(1):
6 - 8.
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
