To determine the incidence of, and risk factors for, regional nodal failure (RNF) and to evaluate the effectiveness of, and indications for, regional nodal irradiation (RNI) in patients with Stage I-II breast cancer treated with breast-conserving therapy.

A total of 1500 cases of Stage I-II breast cancer were treated with breast-conserving therapy between February 1980 and December 2000. All patients underwent excisional biopsy, and 925 (62%) underwent re-excision. Level I-II axillary lymph node dissection was done in 94% of patients. The lymph nodes were pathologically involved in 335 patients (22%); 255 with 1-3 nodes and 80 with >/=4 nodes involved. All patients received whole breast irradiation to a median dose of 45 Gy, and 97% received a tumor bed boost to a median dose of 61 Gy. Treatment included the breast only in 1309 patients (87%), and the breast and regional lymphatics in 191 (13%).

With a median follow-up of 8.1 years, 35 patients had failure within the regional nodes: 12 patients (6%) who received RNI and 23 patients (2%) who did not. The 5- and 10-year rate for any RNF was 1.9% and 2.8%, respectively. The 5 and 10-year rates of axillary failure and supraclavicular failure were 0.6% and 1.0% and 0.9% and 1.6%, respectively. In patients with >/=4 positive lymph nodes, RNI reduced the 10-year rate of any RNF from 11% to 2% (p = 0.024), the rate of axillary failure from 5% to 0% (p = 0.019), and the rate of supraclavicular failure from 11% to 2% (p = 0.114). RNI did not affect the rate of axillary failure or supraclavicular failure in patients with 1-3 positive nodes. In node-negative patients, the rate of RNF was significantly greater if <6 nodes were removed at the time of axillary dissection. Multiple clinical, pathologic, and treatment-related factors were analyzed for association with RNF. On univariate analysis, RNF was associated with the number of nodes excised, number of positive nodes, percentage of positive nodes, size of nodal metastasis, presence of angiolymphatic invasion, estrogen receptor status, age, systemic chemotherapy, and RNI. Three subsets of patients had unusually high rates of RNF, those with >/=67% nodes positive (16%), nodal metastasis >/=2.0 cm (44%), or age 10 nodes excised, respectively (p = 0.05).

Failure within the regional lymph nodes as an isolated site of first relapse is uncommon in patients with Stage I-II breast cancer treated with breast-conserving therapy. RNI can significantly reduce the rate of RNF (axillary failure) in patients with >/=4 positive lymph nodes. The maximal size of the lymph node metastasis was found to be the only significant independent predictor of RNF, with nodal metastases >/=2.0 cm associated with extremely high regional failure rates. Despite this, young age and the extent of axillary dissection (particularly as related to the number of positive nodes) also appear to be important and should be considered when evaluating patients for RNI. Inadequate axillary dissection was not only associated with increased regional failure, but also reduced survival.