Droplet digital polymerase chain reaction (ddPCR), which could perform thousands of PCRs on a nanoliter scale simultaneously, would be an attractive method to massive parallel sequencing for identifying and studying the significance of low-frequency rare mutations. Recent evidence has shown that the key potential mechanisms of the failure of aromatase inhibitors–based therapy involve identifying activating mutations affecting the ligand-binding domain of theESR1gene. Therefore, the detection ofESR1mutations may be useful as a biomarker predicting an effect of the treatment. We aimed to develop a ddPCR-based method for the sensitive detection ofESR1mutations in 325 breast cancer specimens, in which 270 primary and 55 estrogen receptor–positive (ER+) metastatic breast cancer (MBC) specimens. Our ddPCR assay could detect theESR1mutant molecules with low concentration of 0.25 copies/μL. According to the selected cutoff,ESR1mutations occurred in 7 (2.5%) of 270 primary breast cancer specimens and in 11 (20%) of 55 ER+ MBC specimens. Among the 11 MBC specimens, 5 specimens (45.5%) had the most commonESR1mutation, Y537S, 4 specimens (36.3%) each had D538G, Y537N, and Y537C. Interestingly, 2 patients had 2ESR1mutations, Y537N/D538G and Y537S/Y537C, and 2 patients had 3ESR1mutations, Y537S/Y537N/D538G. Biopsy was performed in heterochrony in 8 women twice. In 8 women, 4 women had primary breast cancer and MBC specimens and 4 women had 2 specimens when treatment was failure. Four of these 8 women acquiredESR1mutation, whereas noESR1mutation could be identified at first biopsy. ddPCR technique could be a promising tool for the next-generation sequencing-free precise detection ofESR1mutations in endocrine therapy resistant cases and may assist in determining the treatment strategy.