Objective To establish the diagnostic model for lung cancer by protein fingerprint techniques and to further explore the serum protein biomarker for early diagnosis and surgical effect assessment of lung cancer . Methods Serum samples from 32 healthy controls , 38 lung cancer patients , and 12 lung benign tumor patients were analyzed by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS) combining with magnetic bead technique to establish proteomic spectra .The data were categorized and analyzed with Biomarker Patterns Software (BPS) to develop a preliminary diagnostic model of serum protein fingerprint of lung cancer.Efficiency of this diagnostic model was tested .Combining with the diagnostic model , the proteomic spectra before and after surgery were compared to identify the proteins suitable as indicators of surgical effect . Results Within the mass-to-charge ratio ( m/z) range of 1000-50 000, 215 protein peaks were detected and marked in the enrolled patients and healthy controls .Of these protein peaks , 6 peaks were identified as showing statistically significant difference in expression ( m/z 1115.37, 1929.70, 3217.57, 3246.34, 3318.57, 11 508.90, P<0.05) .The primary sensitivity for diagnosing lung cancer was 92.11%(35/38) and its corresponding specific-ity was 90.91%(40/44).The cross validation suggested that the sensitivity and specificity were both 86.67%(13/15).The protein peaks with m/z being 1115.37, 1929.70, 3246.34, and 11508.90 were significantly in-creased in lung cancer patients ( P<0.05) , and significantly reduced after surgery compared with before surgery in these patients (P<0.01), suggesting potential value of these 4 protein peaks in diagnosis and treatment effect as-sessment of lung cancer .Conclusion The serum protein fingerprint diagnostic model for lung cancer based on SELDI-TOF-MS technique yields fairly high sensitivity and specificity , which may provide innovative thoughts for identification of biomarker for early diagnosis and treatment effect assessment of lung cancer .