Identification of serum proteins as prognostic and predictive markers of colorectal cancer using surface enhanced laser desorption ionization-time of flight mass spectrometry.

Colorectal cancer (CRC) is the second most common cause of cancer related death. Prognosis is highly dependent on stage at diagnosis making early detection mandatory. This study aimed to identify novel disease specific biomarkers of CRC, validate our previously identified biomarkers of CRC and identify serum biomarkers predicting treatment response and for monitoring. Serum of patients with metastatic CRC was collected, according to a predefined schedule, prior to start of standard first-line chemotherapy with oxaliplatin and capecitabine and serially before each 3 weekly treatment cycle and analyzed for proteomic profile by standardized SELDI-TOF MS. Serum proteomic mass spectrometry data of all subjects were processed using the tbimass R-package and proteomic profiles of CRC patients were compared with those of matched normal control subjects. Furthermore, changes in proteomic profiles during the course of chemotherapy were recorded according to treatment response. In total, 42 patients with advanced CRC were treated and mean follow-up was 13.5 months. The response rate was 50% and the median overall survival 19.5 months (95% CI: 16-23). By comparing CRC patients and healthy controls we identified 13 potential biomarkers of CRC (m/z 2.0-31.9 kDa) whereas two proteins, m/z 14060 and 28100 Da (apolipoprotein A-I), were highly significant (p<0.0001). Comparison of responding and non-responding patients identified 6 proteins potentially predicting response, where of m/z 3330 Da was significant (p=0.007). Serial analysis identified 2 proteins, m/z 2022 and 28100 Da, that changed during chemotherapy in accordance with response. We identified 13 m/z values discriminating between CRC patients and healthy controls, including the previously identified apolipoprotein A-I as a candidate biomarker for CRC and treatment monitoring.