Background. For patients with node-positive breast cancer whose axilla is clinically downstaged after neoadjuvant chemotherapy (NAC), targeted axillary dissection (TAD) has been adopted at several institutions. This study compared axillary nodal recurrence between TAD and sentinel lymph node biopsy (SLNB) alone. Methods. Consecutive patients with stage II or III biopsy-proven node-positive breast cancer treated with NAC from August 2018 to June 2022 were identified. Patients who became clinically node-negative after NAC and had tumor-free SLNB were evaluated. The patients were divided into two groups: the TAD and SLNB-alone groups. Results. Of the 377 patients treated with NAC, 143 (38 %) had stage II or III node-positive breast cancer, 105 (73 %) were converted to ycN0, and 44 (42 %) had tumor-free SLNB and avoided an axillary lymph node dissection (ALND). Of the 44 patients, 25 (57 %) had TAD, and 19 (43 %) had SLNB alone. The TAD and SLNB-alone groups were clinically similar. The median tumor size was 2.7 cm (range, 1.9-3.4 cm). The SLNB-alone approach was less likely to retrieve the biopsy-proven clipped node (clipped node retained: overall [n = 5/37], TAD [n = 1], SLNB alone [n = 4]; p = 0.03). Adjuvant radiotherapy (RT) was administered to 40 patients (91 %) and regional nodal RT to 32 patients (73 %). During a median follow-up period of 28 months, no axillary nodal recurrences were found in either group. Conclusions. For the patients with stage II or III node-positive breast cancer who became cN0 after NAC, with tumor-free sentinel nodes, axillary nodal recurrence rates were low after both TAD and SLNB alone despite rates of higher non-retrieval of the clipped node in the SLNB-alone group. These findings suggest that either method affords excellent staging and regional control.