Abstract: Objective To analyze the characteristic of serum proteins in non-small cell lung cancer ( NSCLC) patients, establish serum markers pattern for the diagnosis of NSCLC. Methods Surface enhanced laser desorption ionization time of flight mass spectormetry( SELDI-TOF-MS ) technology was used to analyze serum samples including 100 cases of NSCLC patients and100 cases of healthy controls. Biomarker Pattern Software (BPS) was used to detect the protein peaks significantly different between NSCLC patients and healthy controls. And then established a diagnostic pattern.This pattern was further valuated by a blind test. Results Sixteen significantly different protein peaks were found in serum samples between NSCLC patients and healthy controls. Eight up-regulated protein peaks and eight down-regulated protein peaks （P<0.001）were identified in serum samples of patients of NSCLC comparing with healthy controls.Using BPS, a diagnostic pattern was established with 100% sensitivity and 100% specificity . A blind test generated a sensitivity of 96.15％ and specificity of 95.83％ respectively. Three up-regulated protein peaks（P<0.05）were identified in serum samples of patients of NSCLC with smoking history comparing with no smoking history. Two up-regulated protein peaks（P<0.01）were identified in serum samples of patients of squamous carcinoma comparing with adenocarcinoma .No significantly different protein peak was found in serum samples of NSCLC patients with different clinical stages. Conclusion SELDI-TOF-MS technology can identify the significantly different protein peaks and establish a diagnostic pattern with high sensitivity and specificity.It will provide a highly accurate approach for the diagnosis of NSCLC.

Key words: SELDI, carcinoma, non-small cell lung cancer, proteomics