Regional Nodal Irradiation for Low-Risk, Node-Positive Breast Cancer

Abstract

The benefit of regional nodal irradiation (RNI) for node-positive, low-risk breast cancer is controversial. The ongoing TAILOR RT trial is enrolling breast cancer patients with 1-3 involved macroscopic nodes and a low risk Oncotype DX Recurrence Score (< 18) to determine the breast cancer recurrence-free interval (BCRFI) with and without RNI. The objective of this study was to determine if RNI was associated with improved BCRFI in a population of patients similar to those enrolling on TAILOR RT.We interrogated a population-based database, which included all women treated for breast cancer in the province. Inclusion criteria were: age 40-79, pT1-2 pN1 (macroscopically node-positive) breast cancer, and diagnosis between 2005 and 2014. To reproduce the inclusion criteria of TAILOR RT, patients could have had breast-conserving surgery (BCS) or mastectomy & axillary lymph node dissection (ALND) with 1-3 positive nodes, BCS & sentinel lymph node biopsy (SLNB) with 1-2 positive nodes or mastectomy & SLNB with 1 positive node. To select a cohort of patients likely to have Recurrence Score < 18, we only included Luminal A breast cancers. Luminal A was approximated based on the tumor being: ER positive (Allred 6-8/8), PR positive (Allred 6-8/8), HER2-negative, and Grade 1-2. All patients were started on hormonal treatment. The primary endpoint of BCRFI, which was the time to any breast cancer recurrence or breast cancer-related death, was analyzed using multivariate competing risks analysis. Secondary outcomes of locoregional recurrence and distant metastasis were analyzed with multivariate competing risks models, and overall survival was analyzed with multivariate cox analysis.We identified 1,169 eligible women. Median follow-up was 9.2 years. There were 885 patients who received RNI and 284 who did not undergo RNI. The RNI group was younger (median 62 versus 58 years), had a higher rate of nodal involvement, and were more likely to have received chemotherapy (all P < 0.05). The 10-year estimate of BCRFI was 90% in the no-RNI group versus 90% in the RNI group (P = 0.5). On multivariable analysis, RNI was not a significant predictor of BCRFI (HR = 1.0, P = 0.9). At 10-years, locoregional recurrence was 3.3% in the no-RNI group and 1.7% in the RNI group (P = 0.2). Distant metastasis at 10-years was 7.0% for the no-RNI group versus 8.8% for the RNI group (P = 0.5). Finally, overall survival at 10-years was 85% for the no-RNI group and 85% for the RNI group (P = 0.2). All secondary outcomes were analyzed in multivariate models, and, in all cases, RNI receipt was not associated with improved outcomes (all P > 0.05).RNI was not associated with improved outcomes in node-positive, low-risk breast cancers. This work underscores the importance of continued accrual onto the ongoing non-inferiority trial of RNI, TAILOR RT.