Background: The efficacy of endocrine therapy in breast cancer patients expressing low levels (1-9%) of estrogen receptor (ER) is controversial, and there is limited clinical information about these patients. Methods: Patients diagnosed with stage I to III primary invasive breast cancer with low ER expression between January 2008 and December 2016 from six breast centers in China were recruited for this study. Results: Among the total of 17,216 patients, 457 (2.7%) showed low ER expression. Of the 457 patients, 56.9%, 61.6%, and 71.6% of tumors were human epidermal growth factor 2 (HER2) positive, progesterone receptor ( PR) negative, and Ki-67 status >20%, respectively. Of these 457 patients, 112 patients received neoadjuvant chemotherapy, while 276 received adjuvant chemotherapy, and 170 patients received endocrine therapy. The total pathological complete response (pCR) rate was 26.8%. The rate of patients who received endocrine therapy was center and time-dependent. ER expression, PR expression, Ki-67 status, and lymphovascular invasion (LVI) were associated with endocrine therapy decision-making using univariate logistic regression analysis. However, the adjustment for other covariates revealed that only the ER was significantly associated with the rate of patients receiving endocrine therapy. Compared to patients <5% ER-positive tumors, patients with 5-9% ER-positive tumors were significantly more likely to receive endocrine therapy (OR, 2.882; 95% CI: 1.928- 4.308; P<0.001). The median follow-up time was 30 months, and the five-year recurrence-free survival (RFS) rate was 85%. Younger age and positive lymph nodes were associated with worse RFS. The survival rate was not significantly different between patients who received and those patients who did not receive endocrine therapy. Conclusions: The 2010 ACSO/CAP recommendations increased in the proportion of patients with low ER expression receiving endocrine therapy. However, these patients do not appear to receive help from endocrine therapy.