RESUMO
Objective To use the liquid protein combined with MALDI-TOF-MS for screening the serum differential peptides markers in lung adenocarcinoma patients and to establish the lung adenocarcinoma diag-nosed prediction model for founding the potential markers for the diagnosis of lung adenocarcinoma.Methods 37 patients with lung adenocarcinoma and 33 healthy subjects and benign lung disease which were made up in control group were collected,in the two groups the age and the sex were matched.The two groups were ran-domly divided into training group(30 cases of lung adenocarcinoma,26 cases of control)and test group(7 ca-ses of lung adenocarcinoma,7 cases of control)according to 3:1.T he differential diagnosis of lung adenocarci-noma and control group was performed by liquid chip-time-of-flight mass spectrometry and software ClinPro-Tools 3.0 to establish a prediction model of lung adenocarcinoma.The diagnostic model was validated by using serum samples from the test group to assess the diagnostic efficacy of the model.Results Nine peptide peaks with significant differences(P<0.05)were obtained by ClinProTools 3.0 software analysis.The up-regulated peaks in lung adenocarcinoma(m/z)were 8 976.5,4 469.05,4 966.78,8 925.5,4 531.05,and the down-reg-ulated m/z were 3 304.44,8 594.76,3 266.82,3 195.52.According to the genetic algorithm(GA),the lung ad-enocarcinoma diagnosis and prediction model was established.The overall recognition ability of the model was 94.49%.The model was evaluated by the test group.The results showed that the sensitivity of the model was 100.0% and the specificity was 85.7%.Conclusion Among lung adenocarcinoma patients,serum benign lung disease and healthy,there are differences in the serum peptide.T he use of differential peptide peaks to estab-lish lung adenocarcinoma diagnostic prediction model for the early diagnosis of lung adenocarcinoma provides a new method.