Endocrine therapy (ET) is typically the first-line treatment in hormone receptor-positive, human epidermal growth receptor 2-negative (HR+/HER2-) breast cancer (BC) patients; however, almost half of the patients exhibit ET resistance after initial response. Mechanisms of ET resistance are not fully understood, and require an integrated epigenetic, transcriptomic and genetic approach. We conducted ATAC, RNA, and whole-exome sequencing on primary and metastatic BC (pBC and mBC) from 459 HR+/HER2- patients. pBC samples were derived from tumor resections of patients, which were then treated with adjuvant ET and experienced either a relapse or sustained remission over a five-year period. Samples of mBC were obtained from metastatic lesions of ET-resistant patients. We performed analysis of chromatin accessibility, gene expression, and mutational data aiming to understand relapse mechanisms in pBC and resistance mechanisms in mBC. PAM50 probability scoring of our cohort of HR+/HER2- BC revealed subgroups within pBC and mBC, which resembled PAM50 subtypes of BC. In pBC, the “Basal” and “Her2” subgroups displayed the highest percentage of relapse, compared to the “LumA” and “LumB” subgroups. TP53 alterations were enriched in “Basal” and “Her2” subgroups, while MAP3K1 alterations were more frequent in “LumA” subgroup. In mBC, ESR1 mutations characterized “LumB” tumors, Her2” subgroup exhibited ERBB2 alterations, and “LumA” subgroup was enriched with FOXA1 mutations. In both pBC and mBC, the subgroups were enriched with gene sets pertaining to their associated PAM50 subtypes. “LumB” subgroup was associated with higher estrogen response. “Basal” and “Her2” subgroups showed enrichment of gene sets related to true Basal and Her2 subtypes, and transcription factor motifs of AP-1 and TFAP2. “Basal” and “Her2” subgroups was characterized by poor survival. PAM50 probability scoring was useful to dissect heterogeneity and to unveil insights into molecular mechanisms associated with the risk of relapse and ET resistance within HR+/HER2- BC. These findings may improve prognostic accuracy and patient stratification in HR+/HER2- BC cohorts