Seldi-tof-ms Screening Of Differentially Expressed Proteins In The Lungs Of Benign And Malignant Disease Identification And Clinical Decision-making

PART ONE:Establish the reliability of SELDI-TOF-MS techniqueObjective: To Establish the reliability of SELDI-TOF-MS technique.Material and methods: Data of one plasma sample in 6 randomize chip location was analyzed, to evaluated the reliability and stability of the technology. The coefficient variation (CV) of protein M/Z value and protein intensity value were calculated. .Results: The coefficient variation (CV) of protein M/Z value was 4.88E-04(< 1‰). The coefficient variation (CV) of protein intensity was 0.12(<0.2).Conclusion: SELDI-TOF-MS was an stable and reliable proteomic technique.PART TWO:Detection Plasma Proteomic Patterns Of Lung Cancer With Different Pathological FeaturesObjective: To detect plasma proteomic patterns in lung cancer, screen the protein related with disease diagnosis,stage and other clinicopathological characters.Material and methods: Surface enhanced laser desorption/ionization time of flight mass spectrometry technique and weak cation exchanger (WCX2) was used in detecting the plasma proteomic pattern of 108 lung cancer patients,26 benign lung disease. Biomarker Wizard Software was used to analyze the data, the protein expression related with disease diagnosis and staging was analyzed.Results: 1. The difference of plasma proteomic patterns between lung cancer and benign lung disease cases was analyzed, 11 proteins with different M/Z values were found to be with statistical significance (P<0.05). Proteins with M/Z values 1450.24,1471.95,6794.65,8289.41,8381.62,14374.77 were upregulated in lung cancer cases; Proteins with M/Z values of 7617.22,11457.74,15590.85,15813.76 and 22938.34 were downregulated in lung cancer cases.2. The difference of plasma proteomic patterns betweenⅠstage lung cancer and benign lung disease cases was analyzed, 43 proteins with different M/Z values were found to be with statistical significance (P<0.05). Proteins with M/Z values 1411.4,1447.63,1471.95,1479.1,1496.96,1520.47,1548,1564.67,1641.54,2803.03,2841.1,4768.71,6802.1,7005.24,8181.59,8289.41,9573,9754.75,12054.43,14184.82,14374.77,16454.35,16678.98,22938.34,24283.03,24602.05,29125.61 were upregulated in lung cancer cases, Proteins with M/Z values 1217.46,2696.98,4874.59,6311.76,6520.82,7617.22,7983.73,8524.95,8788.99,9045.58,9176.61,11457.74,13483.64,15589.4,15813.76,27561.76 were downregulated in lung cancer cases.3. The difference of plasma proteomic patterns between different stage was analyzed, 40 proteins with different M/Z values were found to be with statistical significance (P<0.05). Proteins with M/Z values 1203.76,1223.51,2695.20,4873.45,5181.75,6315.51,6350.75,6516.06,7627.34,8386.03,8530.45,8776.55,9045.58,9165.15,11456.55,13481.16,14840.41,15604.89,15836.05,22961.58 and 27548.47 were upregulated in late stage cases. Proteins with M/Z values 1411.40,1450.24,1471.95,1479.10,2803.03,2841.10,6441.20,6576.90,8289.41,8918.22,,9573.00,9772.02,14184.82,14374.77,16454.35,16678.98,24283.03,24602.05,29125.61 were downregulated in late stage cases.4. The difference of plasma proteomic patterns between different cell differentiation group were analyzed, 28 proteins with different M/Z values 1203.76,1223.51,1411.4,1450.24,1471.95,1479.1,2695.2,2803.03,2841.1,4874.59,5181.75,6315.51,6509.97,6780.06,7627.34,8289.41,8776.55,9573,9772.02,11456.55,13495.44,14374.77,15604.89,15836.05,22961.58,24283.03,24602.05,29125.61 were found to be with statistical significance (P<0.05).5. The difference of plasma proteomic patterns between lymph node metastasis and without lymph node metastasis group were analyzed, 40 proteins with different M/Z values were found to be with statistical significance (P<0.05). Proteins with M/Z values 1203.76,1223.51,2695.2,4873.45,5181.75,6315.51,6350.75,6441.2,6516.06,6576.9,7627.34,8386.03,8530.45,8776.55,9165.15,11456.55,13481.16,14840.41,15604.89,15836.05,22961.58,27548.47 were upregulated in lymph node metastasis group. Proteins with M/Z values 1411.4,1450.24,1471.95,1479.1,2803.03,2841.1,6780.06,8289.41,8918.22,9573,9772.02,14184.82,14374.77,16454.35,16678.98,24283.03,24602.05,29125.61 were downregulated in lymph node metastasis group. Furthermore, The difference of plasma proteomic patterns between N1 and N2 metastasis group were analyzed, 8 proteins with different M/Z values were found to be with statistical significance (P<0.05). Proteins with M/Z values 1203.76,1223.32,4515.97,9022.82,9175.79,9233.91were upregulated in N2 metastasis group. Proteins with M/Z 11445.96,23551.51 were downregulated in lymph node metastasis group. At last, we found only one protein with M/Z value 8409.17 was to be with statistical significance (P<0.05) between single node metastasis group and group with more than 2 node metastasis;we found two proteins with M/Z value 8409.17 and 8606.14 were be with statistical significance (P<0.05) between group with≤3 node metastasis and group with more than >3 node metastasis.6. The difference of plasma proteomic patterns between different T stage group were analyzed, 32 proteins with different M/Z values were found to be with statistical significance (P<0.05). Proteins with M/Z values 1203.76,1223.51,2695.20,4874.59,5181.75,6315.51,6350.75,6509.97,7627.34,8386.03,8530.45,11456.55,13495.44,14840.41,22961.58,27548.47 were upregulated in T3+T4 group. Proteins with M/Z values 1411.40,1450.24,1471.95,1479.10,2803.03,2841.10,8918.22,9573.00,9772.02,14184.82,14374.77,16454.35,16678.98,24283.03,24602.05,29125.61 were downregulated in T3+T4 group.7. The difference of plasma proteomic patterns between tumor≤2cm and >2cm group were analyzed, 12 proteins with different M/Z values of 1203.76,1450.24,1471.95,1479.10,2803.03,4874.59,5181.75,8289.41,16454.35,16678.98,22961.58,27548.47were found to be with statistical significance (P<0.05); The difference of plasma proteomic patterns between tumor≤3cm and >3cm group were analyzed, 9 proteins with different M/Z values of 1203.76,1411.40,1450.24,1471.95,2803.03,5181.75,6780.06,11456.55,15604.89,15836.05 were found to be with statistical significance (P<0.05); The difference of plasma proteomic patterns between tumor≤5cm and >5cm group were analyzed, 11 proteins with different M/Z values of 1411.40,1450.24,1471.95,2803.03,5181.75,6576.90,6780.06,8918.22,14840.41,15604.89,15836.05 were found to be with statistical significance (P<0.05); There was no difference between tumor≤7cm and >7cm group.8. The difference of plasma proteomic patterns between different age group was analyzed. There was no difference found to be with statistical significance between age≤50 and age >50 years old group. One protein with M/Z value of 24283.03 was found to be with statistical significance (P<0.05) between age≤60 and age >60 years old group. 6 protein with M/Z values of 2841.10,6576.90,6690.84,6887.11,8289.41,15604.89 were found to be with statistical significance (P<0.05) between age≤70 and age >70 years old group.9. There was no difference of plasma proteomic patterns between different gender group (P>0.05).Conclusion: Plasma proteomic patterns were detected to be with great difference between lung cancer cases and benign lung disease cases. These screened proteins may be of great importance in tumor generating and developing. Tumor diameter,lymph node metastasis,tumor stage,cell differentiation and age were important factors influencing plasma proteomic patterns. Furthermore, whether there were mediastinal lymph node metastasis had greater importance than metastasis lymph node number in influencing plasma proteomic patterns. There were greater plasma proteomic patterns differences between≤70 and >70 years old group than other group. There were many plasma protein differences in different tumor diameter groups (≤2cm or >2cm,≤3cm or >3cm,≤5cm or >5cm).PART THREE:Establish The Clinical Model Related With Lung CancerObjective: To establish the diagnostic pattern model for lung cancer diagnosis and staging in clinical work.Material and methods: Biomarker Pattern Software was use in establishing the diagnostic pattern of detecting early lung cancer,differentiating lymph node metastasis and differentiating the stage of disease.Results: 1.Diagnostic model of early stage lung caner detection 1) Plasma protein obtained from 54Ⅰstage lung cancer and 26 benign lung diseases were used to generate decision tree by Biomarker Pattern Software. Proteins with M/Z value of 1217.45and 2696.98 were automatically selected to constructed the model. Its sensitivity reached 96.296%, specificity reached 100%.2) 35 cases of pulmonary nodule≤2cm were used as blind assessment, an accuracy of 82.9%(29/35),sensitivity of 81.3%(13/16),specificity of 84.2%(16/19),positive predictive value of 81.3%(13/16),negative predictive value of 84.2%(16/19)were obtained.2. Diagnostic model of lymph node metastasis1) Plasma protein obtained from 108 lung cancer cases were used to generate decision tree by Biomarker Pattern Software. Proteins with M/Z value of 1479.10,6576.89,6780.05,8776.55 were automatically selected to constructed the model. Its sensitivity reached 97.83%, specificity reached 91.94%.2) Another 35 cases of lung cancer were used as blind assessment, an accuracy of 74.3%(26/35),sensitivity of 72.2%(13/18), specificity of 76.5%(13/17),positive predictive value of 76.5%(13/17),negative predictive value of 72.2%(13/18)were obtained.3. Diagnostic model to distinguish early and later stage of lung cancer1) Plasma protein obtained from 54Ⅰstage and 28Ⅲstage lung cancer were used to generate decision tree by Biomarker Pattern Software. Proteins with M/Z value of 1224.05,1447.63,1479.10 were automatically selected to constructed the model. Its sensitivity reached 94.44%, specificity reached 100%.2) 32 cases ofⅠstage andⅢstage among 35 cases were used to used as blind assessment. An accuracy of 84.4%(27/32),sensitivity of 80.0%(12/15), specificity of 88.2%(15/17),positive predictive value of 85.7%(12/14),negative predictive value of 83.3%(15/18)were obtained.Conclusion: The constructed diagnostic models were helpful to early lung cancer detection. The constructed diagnostic model of lymph node metastasis, and the model of distinguish early and late lung cancer were of great importance in clinical work.