A study of diagnostic model for serum protein finger-print of thyroid cancer patients based on SELDI technique

Objective: To establish a diagnostic model for serum protein finger-print of thyroid cancer patients. Methods: serum samples of 40 patient with thyroid cancer and 30 healthy subjects were tested by surface enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS). Protein finger-print was obtained, and 2 specific differentially expressed proteins were selected. A diagnostic model was established with the help of bioinformatics. Results: By comparing the data of differentially expressed proteins from 2 groups, and 121 protein peaks were obtained within the range 2000Da - 20000Da. Eighteen protein peaks showed significant difference in 2 groups (P <; 0.01), 5 protein peaks with mass to charge (M/Z) value 2050.69, 2940.50, 3942.92, 5345.54 and 6918.33 were selected by the Biomarker Pattern Software to construct decision-tree classification diagnostic model for serum differentially expressed proteins. In the test group of this model, the accuracy was 81.4% (57/70), sensitivity 80.0 (32/40), and specificity 83.3% (25/30). Conclusions: the model for serum protein finger-print is able to achieve the optimal effects distinguishing the thyroid cancer patient and healthy individuals. SELDI technology is a quick and effective tool in the diagnosis of thyroid cancer and screening of specific tumor marker.