The 2 largest and most recent randomized trials investigating the benefit from adjuvant radiotherapy to the chest wall and regional lymph nodes in N1 breast cancer after mastectomy are the DBCG 82 b and c trials by the Danish Breast Cancer Cooperative Group (DBCG) [10
]. In the DBCG 82 b and c trials, 3,083 pre- and postmenopausal high-risk breast cancer patients were randomized to +/– adjuvant PMRT in addition to their systemic therapy. In the DBCG 82 b protocol, 1,708 pre- and perimenopausal patients were included, and in the DBCG 82 c protocol, 1,375 postmenopausal patients below 70 years of age were included. The protocols were open for inclusion in the period of 1982–1990, and the high-risk criteria at that time were node-positive disease and/or a T3 or T4 tumor and/or skin or deep fascia invasion.
Based on the surgical guidelines used during the period of inclusion in these protocols, the median number of removed lymph nodes was 7. Therefore, in the following we discuss results from a selected subgroup of 1,152 node-positive patients where 8 or more lymph nodes were removed during surgery. The surgical treatment was macroradical mastectomy and axillary dissection with the aim to remove the nodes of level I and partly level II and all macroscopically enlarged nodes. The systemic therapy in the premenopausal patients consisted of 8–9 cycles of CMF (cyclophosphamide, methotrexate, 5-fluorouracil), and for the postmenopausal patients tamoxifen 30 mg daily for 48 weeks (further details in [10
]). Patients randomized to radiotherapy were treated with 48–50 Gy in 22–25 fractions, 5 weeks. The radiotherapy fields included the thorax wall, the internal mammary nodes, and the periclavicular and axillary nodes. Further details are given in [14
]. After a median follow-up of 18 years (range 15–22 years), a substantial and equal survival benefit from locoregional PMRT was documented in patients with 1–3 and 4+ positive lymph nodes in the axilla [12
]. During the first 15 years post treatment, 179 patients (16%) were diagnosed with LRR and 762 patients (66%) died. The frequency of LRR in the radiotherapy group was 4% and in the none-radiotherapy group 26%; the corresponding 15-year actuarial values were 6 and 37%, respectively, p < 0.001, and relative risk (RR) 0.12 (95% confidence interval (CI) 0.07–0.19). Using the endpoint overall survival, 61% had died in the radiotherapy group and 71% in the none-radiotherapy group; the 15-year actuarial values were 39 and 29%, respectively, p = 0.015, and RR 0.63 (95% CI 0.49–0.81). Morbidity from PMRT was evaluated in these patients, and importantly no excess cardiac mortality was identified after a median follow-up of 12 years [13
]. Results from the RACE study have recently confirmed no differences in cardiac morbidity and mortality between Danish patients irradiated with PMRT for left- and right-sided breast cancer [20
In 1,000 patients included in the DBCG 82 b and c trials, tumor tissue was available for further examination. Thus immunohistochemical studies of classical histopathologic parameters such as estrogen receptors (ER) and progesterone receptors and HER2 status together with T and N status were related to the pattern of recurrence in the irradiated and non-irradiated patients [17
]. This showed that in the patients with ‘good prognosis’ defined as pT1, pN1, grade 1, ER+, and HER2–, there was a 1:1 relationship between LRR and overall survival (in the non-irradiated group, 11% more LRR after 5 years translated into 12% worse overall survival after 15 years). In the subgroup of worst prognosis defined as pT3+, pN2+, and grade 3, there was a difference of 35% more LRR in the non-irradiated group after 5 years but this did no result in any difference in overall survival after 15 years. The conclusion therefore was that the irradiated patients with the best prognosis in both the DBCG 82 b and c trials had significantly better locoregional control and a significant gain in overall survival of the same magnitude.