The objection of this study is to investigate whether the cavity margin (CM) status has different predictive efficacy for local-regional recurrence (LRR) in patients who have received or have not received neoadjuvant chemotherapy (NAC) before breast-conserving surgery. We identified 61 patients who received NAC before breast-conserving surgery. A nonrandomized unmatched cohort of 295 patients without history of receiving NAC were also included in this study. Clinicopathological features and follow-up data were abstracted and analyzed. Patients in the NAC-treated group had more advanced diseases when compared with patients in the nonNAC-treated group. With a median follow-up of 42 months, the LRR-free survival rate of patients with positive CMs was significantly lower than that of patients with negative CMs in the NAC-treated group. This distinction was not observed in the nonNAC-treated group. Univariate and multivariate analysis revealed that positive CM was the only independent predictive factor for LRR in the NAC-treated group but not in nonNAC-treated patients. CM status had different predictive efficacy for LRR in different settings. Association between CM status and LRR was observed in NAC-treated patients rather than nonNAC-treated patients. More extensive surgical treatment might be needed in NAC-treated patients when their CMs are positive.