The 2006 Worldwide Overview of the Effects of Local Treatments for Early Breast Cancer On Long-term Outcome? "Meta-analysis of the Randomized Trials of Radiotherapy After Mastectomy With Axillary Clearance"

Reviewer: John P. Plastaras, MD, PhD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 7, 2006

Presenter: Paul McGale, PhD
Presenter's Affiliation: Clinical Trial Service Unit, United Kingdom
Type of Session: Plenary


  • Randomized trials have shown that post-mastectomy radiation therapy ( PMRT ) confers both a local control and survival benefit in certain patient populations.
  • It is generally accepted that patients with 4 or more positive axillary nodes (N4+) benefit from PMRT, whereas patients with no positive axillary nodes (N0) do NOT benefit from PMRT .
  • It is unclear whether patients with 1-3 positive axillary nodes (N1-3) benefit from PMRT .
  • The EBCTCG meta-analysis draws on an enormous database of randomized trials that may help answer these questions using primary patient data, without publication bias.

Materials and Methods

The Oxford overview includes over 300,000 early breast cancer patients pooled from over 400 randomized trials performed at over 150 trial centers.


  • PMRT reduced local recurrences in all groups, with the biggest benefit seen in patients with the highest local recurrence rates. There were fewer local recurrences after 5 years.
    • 5-yr local recurrence:
      • N0: No RT = 5.8% vs. PMRT = 2.4% (2.8% gain)
      • N1-3: No RT = 24.7% vs. PMRT = 5.3% (15.7% gain)
      • N4: No RT = 40.6% vs. PMRT = 12.9% (22.3% gain)
  • Breast cancer mortality was improved by PMRT only in node-positive patients. For every 4 local recurrences prevented, 1 breast cancer death is prevented.
    • 15-yr breast cancer mortality:
      • N0: No RT = 26.6% vs. PMRT = 26.0% (non-significant difference)
      • N1-3: No RT = 50.9% vs. PMRT = 43.3% (7.6% gain), p=0.002
      • N4: No RT = 76.4% vs. PMRT = 69.5% (6.9% gain), p=0.0008
  • Overall survival was improved by PMRT in node-positive patients with efficacy comparable to systemic therapy. In N0 patients, PMRT decreased survival.
    • 15-yr overall mortality rates:
      • N0: No RT = 41.3% vs. PMRT = 37.4% (3.9% loss), p=0.0005
      • N1-3: No RT = 56.1% vs. PMRT = 50.9% (5.3% gain), p=0.05
      • N4: No RT = 79% vs. PMRT = 72.8% (6.2% gain)

Author's Conclusions

  • In node-positive breast cancer patients (N1-3 and N4+), PMRT significantly reduced local recurrence, breast cancer mortality, and overall mortality.
  • In node negative patients (N0), PMRT reduced local recurrence, but decreased overall survival.
  • Survival after radiotherapy is a balance of risk and benefits. With improvements in radiotherapy technique, the risks may be reduced.

Clinical/Scientific Implications

  • This study, which combined both published and unpublished randomized trials, demonstrated that node positive women benefit from PMRT in terms of both local control and survival.
  • The advantage of the EBCTCG meta-analysis lies in its substantial size, length of follow-up, and lack of publication bias. However, it is limited by the lack of data on prognostic/predictive factors and heterogeneity of treatment technique.
  • Other pathologic factors, such as lymphovascular invasion seen on nodal evaluation, may help to better inform the ultimate decision to use PMRT .