Internal mammary chain and medial supraclavicular lymph node irradiation in stage I-III breast cancer (EORTC trial 22922/10925): an unplanned subset analysis of 20-year outcomes in patients with node-negative breast cancer
Mené sur 4 004 patientes atteintes d'un adénocarcinome mammaire unilatéral de stade I à III avec atteinte ganglionnaire axillaire ou présentant une tumeur primitive située en position centrale ou médiale (durée médiane de suivi : 22,2 ans ; âge médian : 55 ans), cet essai randomisé international évalue l'efficacité, du point de vue de la survie globale, d'une irradiation des ganglions mammaires internes et sus-claviculaires médians
Background: Results from several studies, including our 15-year analysis, showed improved disease-free survival and distant metastasis-free survival, reduced breast cancer-related mortality, and variable effects on overall survival with the addition of partial or comprehensive regional lymph node irradiation after surgery in patients with breast cancer. Here, we present the scheduled 20-year analysis of the EORTC trial 22922/10925, evaluating the role of internal mammary and medial-supraclavicular (IM-MS) nodal irradiation in patients with stage I–III breast cancer. Patients without confirmed axillary nodal involvement (pathological N stage 0 [pN0]) were eligible for the trial when the primary tumours were centrally or medially located.
Methods: EORTC 22922/10925 was a randomised, open-label, phase 3 trial done across 46 radiation oncology departments from 13 countries. Eligible participants were women up to 75 years of age with unilateral, histologically confirmed, stage I–III breast adenocarcinoma with involved axillary nodes or a central or medially located primary tumour. Patients were randomly assigned (1:1) centrally using minimisation to receive IM-MS irradiation at 50 Gy in 25 fractions (IM-MS irradiation group) or no IM-MS irradiation (control group). Stratification was done for institution, menopausal status, site of the primary tumour within the breast, type of breast and axillary surgery, and pathological T and N stage. Patients and investigators were not masked to treatment allocation. Only patients without clinically and histopathologically confirmed nodal involvement (pN0) were included in this unplanned subgroup analysis. The primary outcome was overall survival analysed according to the intention-to-treat principle. Safety analyses were conducted in the subset of eligible patients treated per protocol. This trial is registered with ClinicalTrials.gov (on Nov 1, 1999; NCT00002851) and closed after its final analysis.
Findings: Between Aug 5, 1996, and Jan 13, 2004, 4004 patients were randomly assigned. 890 (44·5%) of 2002 patients in the IM-MS irradiation group and 888 (44·4%) of 2002 patients in the control group had pN0 disease. The median age of participants was 55 years (IQR 48–63). At a median follow-up of 22·2 years (IQR 20·1–24·5), 276 (31·1%) of 888 patients in the IM-MS irradiation group and 277 (31·1%) of 890 patients in the control group died because of any cause; 87 (31·5%) of 276 and 122 (44·0%) of 277 patients died due to breast cancer, respectively. At 20 years, the overall survival rate was 69·0% (95% CI 65·4–72·3) in the IM-MS irradiation group and 68·4% (64·8–71·6%) in the control group (hazard ratio [HR] 0·98 [95% CI 0·83–1·15], p=0·77). The breast cancer mortality rate was 10·0% (8·0–12·3) in the IM-MS irradiation group and 14·2% (11·9–16·8) in the control group (HR 0·70 [95% CI 0·53–0·92], p=0·010), but cumulative mortality of unknown cause or not breast cancer cause was 20·9% (95% CI 17·9–24·1) in the IM-MS irradiation group and 17·4% (14·7–20·3) in the control group (HR 1·24 [95% CI 1·00–1·53]; p=0·048). In patients with left-sided breast cancer, late cardiac fibrosis occurred in 15 (3·4%) of 445 patients in the IM-MS irradiation group and 12 (2·8%) of 436 patients in the control group. Lung fibrosis at any grade occurred in 55 (6·4%) of 855 patients in the IM-MS irradiation group and 18 (2·1%) of 877 patients in the control group.
Interpretation: Breast cancer mortality at 20 years was significantly lower in the IM-MS irradiation group, whereas non-breast cancer mortality was numerically higher in the IM-MS irradiation group after 15 years, resulting in no long-term benefit of IM-MS irradiation on overall survival. Our results emphasise the importance of very long-term follow-up and advanced irradiation techniques to reduce the dose to organs of interest.
The Lancet Oncology , article en libre accès, 2026