Background: Estrogen positive/HER2 negative (ER+/HER2-) early (e) breast carcinoma (BC) is an heterogeneous entity, on the prognostic and predictive plan. Several prognostic multigene tests have been developed to identify patients in whom chemotherapy could be safely avoided. Proteomic is a complementary approach as variations in mRNA expression only account for ≈40% of the tumor-encoded protein range and it is taking into account, as other OMICs the tumoral microenvironment Sequential Windowed Acquisition of All Theoretical fragment (SWATH-MS) proteomic approach let an accurate and reproducible label-free quantification of large proteome. To our knowledge, no study has been conducted in a large cohort of luminal BC by SWATH-MS. The aim of this study was to establish a proteomic cartography of ER+/HER2- eBC and identify prognostic biomarkers. Methods: Frozen primary tumors were collected from 157 ER+/HER2- eBC treated in the ICO cancer center between 2006 and 2009. Patients were included if they fulfilled the following criteria: 1) ductal carcinoma; 2) unilateral; 3) first occurrence; 4) have received adjuvant chemotherapy. Clinicopathologic characteristics as outcomes were collected. Each sample was analyzed using SWATH-MS acquisition method as previously described (Aebersold et al, 2012). Peak extraction of the SWATH data was performed using either the Spectronaut software (ver 8.0, Biognosys, Switzerland). Peptide identification results were filtered with a q-value of < 1%. We performed clustering analysis (fuzzy clustering method) based on the 15% of most variant proteins. Functional annotation of clusters based on GO biological process terms enrichment (GOEA) was performed by means of ToppGene and GORILLA web tools. Results: The median of follow-up was 8.34 years. Respectively 32, 4 and 7 patients presented a metastatic, locoregional and controlateral recurrence. 684 among 4555 proteins represented the 15% of most variants proteins.Two ER+/HER2- eBC clusters were identified (C1 [23%] and C2 [77%]) by means of fuzzy clustering and GOEA. Two significant clinicopathological differences were observed between the two subgroups: more unifocal tumors in C1 (P = 0.0415) and mostly a clear better outcome in term of Disease Free Survival (DFS), Distant DFS (DDFS) and Overall survival (OS) in patients belonging to C2 (cf table 1). Functional annotation found that C1 was characterized by mRNA processing and protein synthesis (GO:0006396: RNA processing; GO:0008380: RNA splicing; GO:0016071: mRNA metabolic process; GO:0022613: ribonucleoprotein complex biogenesis), and C2 by a high immune response (GO:0002757: immune response-activating signal transduction; GO:0050778: positive regulation of immune response; GO:0002253: activation of immune response; GO:0050776: regulation of immune response). Differential protein expression according to the C1-C2 clusters will be presented at the meeting. Conclusion: Proteomic cartography by SWATH-MS can clearly distinguish two ER+/HER2- eBC subgroups with clear different prognosis with a better outcome for C2 patients compared to C1 patients. High immune response observed in C2 could underlie this difference with results that must be confirmed on external cohort. Nevertheless this approach could be considered as a complementary approach, helpful for clinical decision for administration of adjuvant treatment.