Background: Optimal approach to early-stage breast cancer (BC) management in women ≥70y is unclear. Studies have demonstrated improved local control but no survival advantage with use of radiation therapy (RT) after breast-conservation therapy (BCT). The impact on survival of other management decisions, including breast procedure, axillary staging (AS), systemic treatment, and patient comorbidities is unknown. We evaluate the effect of each treatment decision on overall survival (OS), disease-free survival (DFS), and breast-cancer specific survival (BCSS). Methods: Patients age ≥70y with cT1N0/ER+/HER2- tumors undergoing surgery from 2011-2013 were identified from a prospectively maintained, single-institution database. Clinicopathologic features were evaluated. 10-year estimated mortality from comorbidity was calculated using a Suemoto index (SI). Univariate and multivariable associations were assessed using Cox modeling. Kaplan-Meier method was used to estimate OS and DFS. Results: 338 patients were identified: 312 (92%) underwent BCT; 26 (8%) had mastectomy. Median age was 75.5y (70-101); median tumor size was 1.0cm (0.1-1.9); median SI was 42.5(22-99). With a median follow-up of 60 mos, 25 patients died—1 from BC, for an OS rate of 92.8% and BCSS of 99.7%. 9 patients developed a locoregional recurrence and 1 developed distant recurrence, representing a 5-year DFS rate of 97.8%. There were no differences in DFS by age groups (≤75, 76-80, > 80; p = 0.1). On univariate analysis, DFS did not vary by age, breast procedure, or receipt of chemotherapy. It was improved in those who had AS and a lower SI (p < 0.01). OS was better in those patients who had AS, had recurrence score performed and had radiation therapy, reflecting lower SI. On multivariable analysis, lack of AS (HR 10.5, p < 0.01) and higher SI (HR 1.03, p < 0.01) were the only variables associated with worse OS. Lack of AS became non-significant in those > 80y. No treatment decision led to improved BCSS. When comparing patients with local or systemic recurrence vs those without recurrence, there was no difference in OS (p = 0.4) or BCSS (p = NS). Conclusions: Decisions in early-stage BC in the elderly are complex due to competing morbidities. While DFS and OS may vary based on treatment variables, no specific care component affected BCSS. Treatment discussions in elderly BC patients should emphasize the risks/benefits of each care component, taking into consideration the patient’s comorbidities and allowing for the patient’s understanding of outcomes and quality of life.