Background: We sought to explore the impact of locoregional recurrence (LRR) on survival in breast cancer (BC) patients in British Columbia treated in the modern era. Methods: A retrospective cohort study design identified patients diagnosed with stage I-III BC from 04/2005-12/2013 treated with surgery and who had a subsequent LRR. Exclusions were death or distant metastasis within 120 days of LRR, bilateral previous/synchronous BC, and other invasive cancers. After LRR, overall survival (OS) and factors associated with OS, including clinical subtype and adjuvant therapy (AdTx), were examined. We defined clinical subtypes as: Luminal (Lum) A-estrogen receptor (ER) and progesterone receptor (PR) positive, HER2 negative, and grade 1 or 2; Lum B-as Lum A but grade 3, or as Lum A but only one of ER or PR positive; triple negative (TNBC)-ER and PR and HER2 negative; and HER2 positive (with any ER, PR). In the absence of earlier LRR, we defined adequate AdTx as: (a) TNBC: >=50% of planned chemotherapy (Chx), (b) HER2 positive: >=50% of planned Chx and >=8 cycles of anti-HER2 therapy, (c) Lum A, B: >=4 years of endocrine therapy and (d) after partial mastectomy or positive final margins: >=50% of radiation therapy dosage. Results: The final cohort had 492 patients with a median follow-up of 7.2 years from LRR and 11.8 years from diagnosis. LRR was local in 69.3% (n=341) and regional +/- local in 30.7% (n=151). Compared with local only, regional recurrences were associated with higher T and N stage, grade, and Lum status (p<=0.01). Biomarkers were re-evaluated at LRR in 82% and changed from initial diagnosis in 32% of those tested: ER expression 3.8% gain, 6.1% loss; PR expression 9.1% gain, 15.1% loss; HER2 overexpression 3.7% gain and 4.8% loss. Over half of patients (n=255, 52%) did not receive adequate AdTx, either by choice or recommendation. A similar proportion with local vs. regional recurrence had inadequate AdTx. Time to death from 1st LRR did not vary significantly between local vs. regional recurrences (median 2.7 years). OS after LRR was lowest in TNBC (median 3.1 years, 24.2% 10-year OS) and longest in Lum A (median not reached, 64.7% 10-year OS) (Table). Conclusions: Our data provide rates of OS after LRR in the era of modern adjuvant therapy. OS after LRR varied by clinical subtype, with TNBC faring the worst, and Lum A the best. Over half had not received adequate AdTx. Despite similar treatment options, OS after LRR was significantly longer for Lum A than B subtypes, underscoring the need for therapy tailored to biology. OS was low in all other subtypes, emphasizing the importance of avoiding LRR.