Identification of the Biomarkers for the Prediction of Efficacy in the First-Line Chemotherapy of Metastatic Colorectal Cancer Patients Using SELDI-TOF-MS and Artificial Neural Networks.
Ying Yuan,Hong Shen,Mo-Dan Li,Chun-Wen Tan,Jie-Kai Yu,Xue-Feng Fang,Shu Zheng
DOI: https://doi.org/10.1200/jco.2012.30.15_suppl.e14026
IF: 45.3
2012-01-01
Journal of Clinical Oncology
Abstract:e14026 Background: For patients with metastatic colorectal cancer, both FOLFOX regimen and FOLFIRI regimen are considered as first-line choices. However, only about half of patients could have an obvious response, and there are no clinically useful markers that could predict the response to these regimens respectively. A major clinical challenge is to identify the subset of patients who could benefit from those regimens. We aimed at identifying serum protein patterns which could predict the efficacy of chemotherapy. Methods: Serum from 70 patients diagnosed as metastatic colorectal cancer before first-line chemotherapy were collected and analyzed for protein patterns using artificial neural networks (ANN) analysis of surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS). Among the 70 cases, 44 patients received FOLFOX chemotherapy, while the other 26 patients received FOLFIRI chemotherapy. After four cycles of the treatment, RECIST criteria were used to define the responders (R) and non-responders (NR). Results: With the method of SELDI-TOF-MS,a potential predicting pattern consisting of 6 biomarkers (9,298m/z, 9,282m/z, 4,606m/z, 9,198m/z, 7,775m/z and 2,266m/z) was identified in the patients receiving FOLFOX chemotherapy. Using this predicting pattern, the responders could be separated from the non-responders with a sensitivity of 92.9%, and a specificity of 81.3%. Another potential predicting pattern that consisted of 7 biomarkers (4,321m/z, 4,305m/z, 2,648m/z, 2,952m/z, 4,121m/z, 4,292m/z and 3,980 m/z) was identified in the patients who have received FOLFIRI chemotherapy. Using this predicting pattern, the responders could be separated from the non-responders with a sensitivity of 92.3% and a specificity of 92.3 %. Conclusions: Two potential patterns for the prediction of efficacy of FOLFOX or FOLFIRI chemotherapy were established in this preliminary study. These protein patterns could be helpful for oncologists to make a decision in selecting suitable first-line regimens for metastatic colorectal cancer patients.