Currently used CA15–3 and CEA have found their clinical application particularly in the follow-up of patients with advanced disease. Novel biomarkers are urgent, especially for improving early diagnosis as well as for discriminating between benign and malignant disease. In the present study, we used a proteomic approach based on surface-enhanced laser desorption/ionization–time of flight–mass spectrometry screening with the aim of identifying differentially expressed 2–30 kDa proteins in plasma of patients with malignant (65 cases) and benign (88 cases) breast lesions with respect to 121 healthy controls. We found that the most promising SELDI peaks were those corresponding to hepcidin-25 and ferritin light chain. We evaluated the capability of these peaks in predicting malignant and benign breast lesions using the area under the receiver operating characteristic curve (AUC). The results showed a good capacity to predict malignant breast lesions for hepcidin-25 AUC: 0.82; 95% confidence interval (CI) 0.75–0.90 and ferritin light chain (AUC: 0.86; 95% CI 0.79–0.92). Conversely, a weak and satisfactory capability to predict benign breast lesion was observed for hepcidin-25 (AUC: 0.63; 95% CI 0.41–0.85) and ferritin light chain (AUC: 0.73; 95% CI 0.49–0.97). A significant association between HER2 status and hepcidin-25 was observed and the distribution of transferrin and ferritin were found significantly different in patients with breast cancer when compared with that of controls. This study provides evidence that hepcidin and ferritin light chain level in plasma may be of clinical usefulness to predict malignant and benign disease with respect to healthy controls.