Background
Randomized, open-label phase 3 (total N=4004; 46 centers, 13 countries; enrolled 1996-2004). Stage I-III breast cancer with involved axillary nodes or a centrally/medially located primary. This report: unplanned subset analysis of 20-year outcomes in the node-negative (pN0) subgroup (n=1778; ~44% of each arm), evaluating internal mammary + medial supraclavicular (IM-MS) nodal irradiation.
Results
Interventions and follow up: Arm A: IM-MS irradiation 50 Gy in 25 fractions (added to standard locoregional treatment)
Arm B: No IM-MS irradiation
Primary endpoint: OS (ITT)
mFollow up: 22.2 yr (IQR 20.1-24.5)
Results: OS at 20 yr: 69.0% vs 68.4% (HR 0.98; 95% CI 0.83-1.15; P=.77)
Breast cancer mortality at 20 yr: 10.0% vs 14.2% (HR 0.70; 95% CI 0.53-0.92; P=.010)
Non-breast-cancer/unknown-cause mortality: 20.9% vs 17.4% (HR 1.24; 95% CI 1.00-1.53; P=.048)
Arm B: No IM-MS irradiation
Primary endpoint: OS (ITT)
mFollow up: 22.2 yr (IQR 20.1-24.5)
Results: OS at 20 yr: 69.0% vs 68.4% (HR 0.98; 95% CI 0.83-1.15; P=.77)
Breast cancer mortality at 20 yr: 10.0% vs 14.2% (HR 0.70; 95% CI 0.53-0.92; P=.010)
Non-breast-cancer/unknown-cause mortality: 20.9% vs 17.4% (HR 1.24; 95% CI 1.00-1.53; P=.048)
Adverse events
Late cardiac fibrosis (left-sided disease): 3.4% vs 2.8%
Lung fibrosis (any grade): 6.4% vs 2.1%
Lung fibrosis (any grade): 6.4% vs 2.1%
Conclusions
In node-negative patients with central/medial tumors, IM-MS irradiation significantly reduced 20-year breast cancer mortality (HR 0.70), but the gain was offset by higher non-breast-cancer mortality after 15 years, yielding no OS benefit. Longest-follow-up randomized breast RT dataset; emphasizes very long-term follow-up and modern organ-sparing techniques.
Key Limitations
Unplanned subset analysis, hypothesis-generating only; 1996-2004 era 2D/3D techniques delivered higher heart/lung doses than modern IMRT/VMAT with deep-inspiration breath-hold; systemic therapy standards outdated (pre-modern taxane/anti-HER2/endocrine optimization); cause-of-death attribution over 20+ years is imprecise.
Clinical Context
The 15-year trial analysis similarly showed reduced breast cancer mortality without OS gain. Guidelines (NCCN/ESTRO) support individualized comprehensive nodal irradiation, primarily in node-positive or high-risk disease. These data argue for careful selection and cardiac/lung-sparing delivery when considering internal mammary irradiation in node-negative medial/central tumors.