Radiation Therapy Risk Factors for Development of Lymphedema in Patients Treated With Regional Lymph Node Irradiation for Breast Cancer

Purpose: We previously evaluated the risk of breast cancer-related lymphedema (LE) with the addition of regional lymph node irradiation (RLNR) and found an increased risk when RLNR is used. Here we analyze the association of technical radiation therapy (RT) factors in RLNR patients with the risk of LE development. Methods and Materials: From 2005 to 2012, we prospectively screened 1476 women for LE who underwent surgery for breast cancer. Among 1507 breasts treated, 172 received RLNR and had complete technical data for analysis. RLNR was delivered as supraclavicular (SC) irradiation (69% [118 of 172 patients]) or SC plus posterior axillary boost (PAB) (31% [54 of 172]). Bilateral arm volume measurements were performed pre- and postoperatively. Patients' RT plans were analyzed for SC field lateral border (relative to the humeral head), total dose to SC, RT fraction size, beam energy, and type of tangent (normal vs wide). Cox proportional hazards models were used to analyze associated risk factors for LE. Results: Median postoperative follow-up was 29.3 months (range: 4.9-74.1 months). The 2-year cumulative incidence of LE was 22% (95% confidence interval [CI]: 15%-32%) for SC and 20% (95% CI: 11%-37%) for SC plus PAB (SC+PAB). None of the analyzed variables was significantly associated with LE risk (extent of humeral head: P=.74 for <1/3 vs >2/3, P=.41 for 1/3 to 2/3 vs >2/3; P=.40 for fraction size of 1.8 Gy vs 2.0 Gy; P=.57 for beam energy 6 MV vs 10 MV; P=.74 for tangent type wide vs regular; P=.66 for SC vs SC+PAB). Only pretreatment body mass index (hazard ratio [HR]: 1.09; 95% CI: 1.04-1.15, P=.0007) and the use of axillary lymph node dissection (HR: 7.08, 95% CI: 0.98-51.40, P=.05) were associated with risk of subsequent LE development. Conclusions: Of the RT parameters tested, none was associated with an increased risk of LE development. This study underscores the need for future work investigating alternative RLNR risk factors for LE