Background: Post-mastectomy radiation therapy (PMRT) is part of standard regimen in the management of breast cancer. It has been shown to decrease the incidence of local regional recurrence (LRR) and improve breast cancer survival. Yet, its role in patients with 1-3 positive lymph nodes (N1) remains unclear. The lack of biological markers to evaluate the efficacy of PMRT for N1 patients leads to excessive treatment of low-risk patients and ignore of those at high risk. A clinical-genomic model RecurIndex® (RI-LR), based on a genomic database from mRNA expression as well as clinical factors (lymph node involvement, ER status,age at diagnosis, and lymphovascular invasion), is mainly developed for predicting the LRR in early stage breast cancer patients. Method: The preliminary result enrolled 107 breast cancer patients with N1 involvement in a retrospective study atthe fourth hospital of Hebei Medical University. Kaplan Meier method is used to calculate the survival rates in terms of LRR-free interval (LRFI). The log rank test and cox regression model are applied for the survival difference between two independent groups, and for investigating prognostic factors related to high-risk patients. The primary endpoint is LRFI. Results: With a median follow-up of 84 [IQR 64-84] months, RI-LR partitioned these N1 patients into 70% high-risk group and 30% low-risk group. LRR rate was 14.7% in the high-risk group whereas only 3.1% in the low-risk group. The survival curve clearly showed a partitioned trend of two groups (5-year LRFI 87.5%vs 100%, p = 0.085). Patients identified in the RI-LR high-risk group showed significantly higher LRFI if they had PMRT compared to those without PMRT (5-year LRFI 94% vs 70%, p =0.043), suggesting that higher radiosensitive patients could be identified by RI-LR. Multivariate analysis revealed that high-risk patients treated with PMRT has a 71 % reduction of LRR compared to those without PMRT (HR 0.29, 95%CI: 0.08-1.00, p = 0.051). Conclusions:The present study provided robust evidence that RI-LR could partition N1 breast cancer patients into good and poor prognosis of LRR. RI-LR is capable of identifying high-risk patients who would benefit from PMRT. This observation warrants to validate in a larger cohort