RESUMEN
The aim of this study was to detect the pretreatment serum protein profiles of breast cancer patients by mass spectrometry (MS) to screen candidate tumor biomarkers, which will supply a simple, accurate, and minimally invasive method to predict the axillary lymph node metastasis of breast cancer. We used magnetic bead-based weak cation-exchange chromatography followed by matrix-assisted laser desorption and ionization time-of-flight MS to detect proteins in the sera of 54 cases of axillary node-negative breast cancer, 47 cases of axillary node-positive breast cancer, and 101 healthy controls. The protein profiles were analyzed to screen tumor biomarkers and lymph node metastasis-associated proteins to establish and verify a diagnostic model. Comparison of the protein profiles between the two cancer groups resulted in a total of 111 discriminate m/z peaks that were associated with breast cancer. Furthermore, 40 discriminate m/z peaks were detected between breast cancer patients with and without axillary node metastases. Four protein m/z peaks at 5,643, 4,651, 2,377, and 2,240 were used to construct a diagnosis model, and cross-validation indicated that breast cancer with and without axillary node metastasis was identified with 87.04% sensitivity (47/54), 87.23% specificity (41/47), and 87.13% accuracy (88/101). These proteins could potentially be used as predictive biomarkers to distinguish between breast cancer patients with or without lymph node metastasis.