BackgroundBreast cancer is the prevailing malignancy among women, exhibiting a discernible escalation in incidence within our nation; hormone receptor‐positive (HR+) human epidermal growth factor receptor 2‐negative (HER2−) breast cancer is the most common subtype. In this study, we aimed to search for a non‐invasive, specific, blood‐based biomarker for the early detection of luminal A breast cancer through proteomic studies.MethodsTo explore new potential plasma biomarkers, we applied data‐independent acquisition (DIA), a technique combining liquid chromatography and tandem mass spectrometry, to quantify breast cancer‐associated plasma protein abundance from a small number of plasma samples in 10 patients with luminal A breast cancer, 10 patients with benign breast tumors, and 10 healthy controls.ResultsThe proteomes of 30 participants in all cohorts were analyzed using the DIA method, and a total of 517 proteins and 3584 peptides were quantified. We found that there were significant differences in plasma protein expression profiles between breast cancer patients and non‐breast cancer patients, and breast cancer was mainly related to lipid metabolism pathways. Finally, the optimal protein combinations for the diagnosis of breast cancer were PON3, IGLV3‐10, and IGHV3‐73 through multi‐model analysis, which had a high prediction accuracy for breast cancer (AUC = 0.92), and the model could also distinguish breast cancer from HC (AUC = 0.92) and breast cancer from benign breast tumor (AUC = 0.91).ConclusionsThe study revealed proteomic signatures of patients with luminal A breast cancer, identified multiple differential proteins, and identified three plasma proteins as potential diagnostic biomarkers for breast cancer. It provides a reference for the screening of biomarkers for breast cancer.