Risk Stratification Model for Prediction of Locoregional Recurrence in Patients with Pathologic T1-2N0 Breast Cancer after Mastectomy

J. Wang,S. Wang,Yi Tang,Jing Hao,Guang-Yi Sun,Jie Jin,Y. Liu,Ying Song,W. Wang,Hui Fang,Hu Ren,Zihua Yu,Y. Li
DOI: https://doi.org/10.1016/j.ijrobp.2019.06.697
2019-01-01
Abstract:To identify the risk factors of locoregional recurrence (LR) in patients with pathologic T1-2N0 breast cancer in a real-world and distinguish individuals warrant postmastectomy radiotherapy (PMRT) Patients who had pathologic T1-2N0 breast cancer from 1999 to 2014 were retrospectively reviewed. Locoregional recurrence was defined as reappearance in the ipsilateral chest wall, axilla, infra/supraclavicular fossa, or internal mammary chain. The association of locoregional recurrence with pretreatment factors was evaluated using multi-variable proportional hazards regression. A total of 4841 patients were eligible patients in this study, and all underwent mastectomy plus axillary node dissection without PMRT. 186 LR developed within a median follow-up period for survivors of 56.4 months (range, 1–222 months). Multivariate analysis revealed age < 50yeras old (p<0.001, hazard ratio [HR] =1.795), T2 stage (p=0.008, HR=1.510), negative expressions of estrogen receptor (ER) and progesterone receptor (PR) (p<0.001, HR=1.921), and tumor location in inner quadrant (p<0.001, HR=1.960) were independent predictor of LR. There were 768 patients (15.9%) with no risk factors, 3328 patients (68.7%) with 1-2 risk factors, and 745 patients (15.4%) with 3-4 risk factors. Patients with 0, 1-2, and 3-4 risk factors had 5- and 10-year LR failure of 2.3% and 5.6%, 3.6% and 5.9%, 7.1% and 8.2%, respectively (p < 0.001). 5-year metastasis-free survival and overall survival (OS) were significantly lower for patients with 3-4 factors (89.2% and 93.9%) compared with patients with 1-2 (92.7% and 96.4%) or 0 (97.4% and 99.1%) risk factors. Risk model based on age, T stage, ER/PR and tumor location can stratify patients with pT1-2 N0 breast cancer into subgroups with different LR and outcomes. Patients with 3-4 risk factors may benefit from PMRT.
What problem does this paper attempt to address?