ABSTRACT
BACKGROUND: Currently no standardized blood test exists for breast cancer screening or staging purposes. The goals of this study were to use proteomic mass spectrometry approaches for profiling, fractionation, and identification of serum proteins from breast cancer patients for discovery of new biomarkers of stage and nodal status. STUDY DESIGN: Samples from 150 patients were collected preoperatively for patients undergoing breast biopsy. Serum was processed using weak cation exchange (WCX) fractionation and analyzed with matrix-assisted laser desorption ionization time of flight mass spectrometry. Spectra were processed and group profiles, peak statistics, and cross-validation scores were determined using a k-nearest neighbor genetic algorithm. Pools of subgroups based on stage, race, and obesity were processed with WCX fractionation followed by trypsin digestion. Differentially expressed proteins and peptides were identified by tandem mass spectrometry. RESULTS: Matrix-assisted laser desorption ionization time of flight proteomic profiling using WCX capture of serum proteins resulted in correct cancer stage classifications ranging from 72% to 84%. Nodal status was classified correctly with 88% cross-validation scores. Levels of endogenous low mass peptide fragments derived from kininogen, fibrinogen, plasminogen, and inter-alpha-trypsin inhibitor heavy chain 4 protein were increased in cancer stage III and stage IV samples. Adding trypsin digestions with WCX capture indicated increased levels of alpha-2-HS-glycoprotein, prothrombin, and serum amyloid A in stage IV samples. Obesity, but not race, was a factor in the relative levels of detected proteins/peptides. CONCLUSIONS: WCX fractionation alone or with trypsin digestion of serum suggest it can be possible to use a panel of proteins to predict breast cancer stage and nodal status. Additional study is required on the role of inflammatory molecules in breast cancer development.
