Classification and diagnostic prediction of colorectal cancer using protein profiling of serum and bioinformatics / 浙江大学学报·医学版

ABSTRACT

<p><b>OBJECTIVE</b>To develop a bioinformatic tool and to use it to identify proteomic patterns in serum, distinguishing colorectal cancer from colorectal adenoma and healthy individuals.</p><p><b>METHODS</b>182 serum samples including 55 colorectal cancer patients, 35 colorectal adenoma and 92 healthy individuals were subjected to analysis by surface-enhanced laser desorption/ionization (SELDI) mass spectrometry.</p><p><b>RESULT</b>The diagnostic pattern combined of 4 candidate biomarkers (M/Z 5911, 8922, 8 944, and 8817) could separate colorectal patients from healthy control with a specificity of 93.3%, sensitivity of 90.9%, and Youden index value of 0.84242. The diagnostic pattern combined of 7 candidate biomarkers (M/Z 17247, 18420 ,5911, 9294, 4654, 21694, and 21742) could separate colorectal cancer patients from colorectal adenoma patients with a specificity of 83.2%, sensitivity of 89.3%, and Youden index value of 0.72484.</p><p><b>CONCLUSION</b>Combination of SELDI with bioinformatics tool can identify some new biomarkers from the sera of colorectal cancer patients, which has a high sensitivity and specificity to distinguish colorectal cancer patients from healthy control.</p>