Part Ⅰ Development And Validation Of Clinical Diagnostic Models For The Probability Of Malignancy In Solitary Pulmonary Nodules Part Ⅱ Application Of Non-small Cell Lung Cancer Biomarker To Monitor Recurrence Of Preoperative And Postoperative NSCLC Pa

Part I Development and Validation of Clinical diagnostic Models for the Probability of Malignancy in Solitary Pulmonary NodulesAims With the increasing level of imaging detection technology, more and more solitary pulmonary nodules were found, which has a large part of benign pulmonary lesions. In order to reduce the unnecessary invasive and open chest operation to benign pulmonary lesions, and perform treatment in time for early stage lung cancer. It is critical to develop a method for differentiating between malignant and benign solitary pulmonary nodule. The purpose of this study is, to establish an effective model for differentiating between malignant and benign solitary pulmonary nodule based on large samples.Methods3,358patients with a solitary pulmonary nodule between January2005and March2013were enrolled. All patients have received surgery for pulmonary nodules resection and have achieved pathologic diagnosis. Clinical characters, preoperative biomarkers results and CT scans findings were collected from the patients. All patients were randomly separated into training set (n=1,679) and test set (n=1,679), we used training set to build a diagnostic model for the malignancy probability of SNP, and applied test set to validate the model we built and results of published diagnostic models.Result In the training set, single factor analysis showed that there were8indicators could be independent risk factors for malignant solitary pulmonary nodule. By these indicators, multivariate logistic regression analysis was used to establish a diagnosis model for SNP. The model was made of old age, smoking history, higher tumor markers detection results, lobulation sign and spiculation sign, and then ROC curve was drawn used the predicative result of the model. The AUC of ROC curve for the model was0.935. In the test set, which the area under the curve was0.917, the effectiveness of this model was much batter than other models which had reported before.Conclusion By the large sample, the diagnosis model was built, which made of old age, smoking history, higher tumor markers detection results, lobulation sign and spiculation sign. This model provides a new method for differentiating between malignant and benign solitary pulmonary nodule. Part II Application of non-small cell lung cancer biomarker to monitor recurrence of preoperative and postoperative NSCLC patientsAims Recurrence and metastasis of postoperative patients with non-small cell lung cancer is the major cause of death. It is very important to monitor the postoperative recurrence and obtain early diagnosis of recurrence so that it can be treated early. The purpose of this study is to dynamically collect the blood sample from non-small cell lung cancer patient in postoperative follow-up and find biomarkers which can be used to monitor the postoperative recurrence for non-small cell lung cancer.Methods A total of185non-small cell lung cancer patients’ serum samples was collected in our hospital from2012November to2014February. The time points of collecting the serum samples are preoperative and postoperative follow-up for patients. The levels of CEA、Cyfra21-1、CA125. SCC、NSE and PAK4, which were found as an biomarker of NSCLC, at each time point in patient serum were assayed. The relationship of the dynamically changes of these biomarkers and recurrence or not were analyzed.Results Serum samples of different time points in postoperative stage were obtained from146patients. After surgical operation, the serum concentration level of CEA, Cyfra21-1and PAK4became lower than preoperative serum (P<0.05), for squamous cell lung cancer patients, serum concentration level of SCC also became lower than preoperative serum (P<0.05). When relapse happened, the serum’s concentration level of CEA, Cyfra21-1and PAK4became higher than non-relapse serum (P<0.05), for squamous cell lung cancer patients, the serum’s concentration level of SCC became higher than non-relapse serum (P<0.05) as well. For8patients, whose blood samples of multi-point in follow up were obtained, the relapsed serum’s level of CEA, Cyfra21-1, PAK4and SCC became higher than non-relapse serum. Moreover, before and near the time of relapse, the biomarkers became higher as well. In21patients, pre-relapse serum’s level of Cyfra21-1, PAK4and SCC (14patients with squamous cell lung cancer) have already achieved the level of relapsed; after relapse happened, serum’s level of CEA continue to rise (P<0.05). For one same patients (61patients have not relapsed finally and13patients have relapsed finally), there was no statistical difference (P>0.05) for the different time points serum results of CEA, Cyfra21-1, PAK4and SCC when relapse not happen. Furthermore, for the diagnosis of relapse, the combination of CEA, Cyfra21-1and PAK4showed larger AUC (0.806) than the combination without PAK4(0.785).Conclusions Among the biomarkers applied in clinic, CEA and Cyfra21-1could reflect the tumor state in non-small cell lung cancer patient’s body, moreover, as a novel biomarker, PAK4can also perform this function perfectly. The combination of PAK4plus CEA and Cyfra21-1showed much better performance in the diagnosis for relapse of postoperative patients. Detection of these three biomarkers can play an important role in monitoring for postoperative period of the patients.