Twenty-Year Follow-up of a Randomized Trial Comparing Total Mastectomy, Lumpectomy, and Lumpectomy Plus Irradiation for the Treatment of Invasive Breast Cancer

Reviewer: Roberto J. Santiago, M.D.
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 24, 2002

Authors: Bernard Fisher, M.D., Stewart Anderson, Ph.D., John Bryant, Ph.D., et al.
Source: N Engl J Med 2002, 347:1233-1241


  • The National Surgical Adjuvant Breast and Bowel Project (NSABP) protocol B-04 showed that over 25 years of follow-up there was no significant difference in survival between women treated with a radical mastectomy and those treated with less extensive surgery.
  • These results prompted a second randomized trial, B-06, to determine whether lumpectomy with or without radiation therapy was as effective as total mastectomy for the treatment of stage I or II breast tumors that were 4 cm or less in diameter.


  • Out of the 2163 women randomized between 1976 and 1984, the present analysis was limited to 1851 women for whom follow-up data were available and nodal status was known.
  • Axillary nodes (level I and II in the lumpectomy-containing arms) were removed regardless of the treatment assignment.
  • Tumor-free specimen margins were required in the lumpectomy-containing arms.
  • Radiotherapy consisted of 50 Gy to the whole breast (no boost), but not the axilla.
  • Women with positive axillary nodes received adjuvant melphalan and fluorouracil.
  • Lumpectomy-specimen margins were found to contain tumor in some patients.
    • These women underwent total mastectomy but were analyzed according to their original treatment assignment.
  • The end points (calculated from the date of surgery) were disease-free survival (DFS), distant-disease-free survival (DDFS), and overall survival (OS).
  • A diagnosis of a second cancer was considered an event for DFS.
  • A first recurrence in the chest wall or in the operative scar, but not in the ipsilateral breast, were classified as a local recurrence.
  • Recurrences in the supraclavicular nodes were classified as regional occurrences.
  • For analysis of DDFS, events included all second cancers, including tumors in the contralateral breast.
  • Kaplan-Meier and cumulative-incidence estimates of the outcome were obtained.


  • The distribution of the women among the treatment groups according to age, tumor size, and nodal status was similar
    • About 60% were = 50 years of age
    • Slightly more than 50% had tumors 2.0 cm in diameter
    • 62% had negative nodes, 26% had 1-3 positive nodes, and 12% had = 4
    • Estrogen-receptor status was determined for ~75% of tumors in each arm, 64% of which were positive
    • Tumor was found in the margins of:
      • 64 of the 634 women assigned to lumpectomy
      • 61 of the 628 assigned to lumpectomy and irradiation
  • The rate of local recurrence at 20 years in women with tumor-free margins was decreased from 39.2% in the lumpectomy alone group to 14.3% in the irradiated group (P<0.001)
  • The benefit of radiation therapy was independent of the nodal status
  • Following lumpectomy alone most recurrences occurred within the first 5 years (73.2%) but in the irradiated group most occurred after the first 5 years (59.5%)
  • The DFS for the whole group was 36.8%
  • There were no differences in DFS, DDFS and OS among the 3 treatment arms (p=0.26, p=0.34, p=0.57, respectively)
  • There was a nearly significant increase in DFS for the lumpectomy plus irradiation group, as compared with the lumpectomy alone group (OR, 0.87; P=0.07)
  • The most frequent first events were distant recurrences (in 24.5%)
  • Lumpectomy followed by irradiation, as compared with lumpectomy alone, was associated with a marginally significant decrease in deaths due to breast cancer (OR, 0.82; P=0.04) partially offset by an increase in deaths from other causes (OR, 1.23; P=0.21)

Author's Conclusions

  • After 20 years of follow-up, no significant difference in overall survival among women who underwent mastectomy and those who underwent lumpectomy with or without postoperative breast irradiation was found.
  • The risk of death due to breast cancer was slightly reduced after lumpectomy and irradiation as compared with lumpectomy alone.
  • The incidence of a recurrence in the ipsilateral breast is significantly reduced by lumpectomy and breast irradiation, as compared with lumpectomy alone.
  • A substantial proportion of events occur after five years of follow-up.
  • No increase in the risk of cancer in the contralateral breast was observed with the use of postoperative breast irradiation.


  • Breast-conserving surgery plus whole breast radiotherapy (BCT) yields equivalent long-term overall survival to mastectomy and high rates of local control even for stage I and II breast carcinomas with poor prognostic features
  • The study confirm the findings of the meta-analyses reported by the Early Breast Cancer Trialists' Collaborative GrouP</li>
  • Most women with early stage breast cancer are candidates for BCT provided that the margins of lumpectomy specimens are free of tumor, and therefore, every woman should be informed about BCT as an alternative and of the suitability of the procedure in her particular case
  • Continued follow-up long after BCT is indicated
  • Local failures that occur many years after BCT often represent new primary tumors
  • Systemic therapy is now frequently administered after lumpectomy, regardless of nodal status, to reduce the risk of distant metastases as well as reduce the rate of ipsilateral breast tumor recurrence
  • The incidence of local recurrence among women with negative nodes who receive modern systemic therapy in addition to radiation therapy is about 5% at 10 years
  • Although controversy exists on whether lumpectomy or quadrantectomy is preferable, it would be inappropriate to choose between them on the basis of a comparison of this NSABP trial and the study by Veronesi et al. because of significant differences in the patient populations in the two trials