Virtual Library

 x 
My Library Virtual Library FAQ

Start Your Search

S. Li



Author of

  • +

    P1.03 - Poster Session with Presenters Present (ID 455)

    • Event: WCLC 2016
    • Type: Poster Presenters Present
    • Track: Radiology/Staging/Screening
    • Presentations: 1
    • +

      P1.03-074 - Combined Use of PET/CT and Clinical Features Yields a Higher Diagnostic Rate of Mediastinal Lymph Node Metastasis in Lung Adenocarcinoma (ID 4936)

      14:30 - 14:30  |  Author(s): S. Li

      • Abstract
      • Slides

      Background:
      The aims of this study were to investigate the correlation between [8]F-fluorodeoxyglucose (FDG) uptake of the primary tumor and mediastinal lymph node metastasis (MLNM) in lung adenocarcinoma, and to improve the diagnostic capability of tumor FDG uptake and other risk factors in predicting occult MLNM preoperatively.

      Methods:
      We reviewed 360 consecutive pulmonary adenocarcinoma patients who underwent preoperative PET/CT scan and subsequent surgery. Resected tumors were classified according to the 2011 IASLC/ATS/ERS classification. Univariate and multivariate analysis were conducted to evaluate the associations between clinicopathological variables and MLNM. The receiver operating characteristic (ROC) curve analysis was performed to quantify the predictive value of these factors.

      Results:
      Of all the 360 patients, 54 were pathological N2 diseases. On univariate analysis, CEA level, nodule type, nodal SUVmax, tumor SUVmax, size, location and histologic subtype were associated with MLNM. On multivariate analysis, CEA≥5.0 ng/ml (p < 0.001), solid nodule (p = 0.012), tumor SUVmax ≥ 3.7 (p < 0.027), nodal SUVmax ≥ 2.0 (p < 0.001) and centrally located tumors (p = 0.035) were independent risk factors that associated with MLNM. The area under the ROC curve (AUC) for tumor SUVmax in predicting MLNM was 0.764 and AUC of nodal SUVmax was 0.730. The combined use of five factors yielded a higher AUC of 0.885 for N2 disease. The tumor SUVmax among histologic subtypes differed significantly (p < 0.001).

      Conclusion:
      Primary tumor SUVmax of PET/CT was shown a good predictor for MLNM in patients with lung adenocarcinoma, and the underlying mechanism may attribute to the close association between tumor FDG uptake and IASLC/ATS/ERS adenocarcinoma subtypes. The combined use of tumor SUVmax with factors like nodal SUVmax, solid nodule, centrally located tumor and increased CEA level improved the diagnostic capacity for predicting N2 disease preoperatively.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

  • +

    P1.08 - Poster Session with Presenters Present (ID 460)

    • Event: WCLC 2016
    • Type: Poster Presenters Present
    • Track: Surgery
    • Presentations: 1
    • +

      P1.08-004 - Prediction of Surgical Outcome by Modeling Based on Risk Factors of Morbidity Following Pulmonary Resection for Lung Cancer in the Elderly (ID 5394)

      14:30 - 14:30  |  Author(s): S. Li

      • Abstract
      • Slides

      Background:
      Surgical treatment for elderly patients with lung cancer presents more challenges compared with general population. The aim of the study was to predict surgical outcome following pulmonary resection in the elderly with lung cancer by developing a clinical model.

      Methods:
      Clinical records of 525 patients aged over 70 years who underwent pulmonary resection for lung cancer in a single center were reviewed. Patients were divided into three ordered categories of surgical outcome according to the Clavien–Dindo classification. Using a development cohort of 401 patients, an ordinal logistic regression was performed to develop a prediction model for surgical outcome. The model was internally validated by bootstrap method and externally validated by another cohort of 124 patients. Two previous models were tested as benchmarks of our model.

      Results:
      The model was developed based on five risk factors of morbidity: ASA classification (p<0.001), pulmonary disease (p=0.001), tumor size (p=0.011), tumor location (p=0.015) and surgical approach (p=0.036). C-statistics of the model was similar to bootstrapping one. Hosmer-Lemeshow test showed a good goodness-of-fit. In external validation the performance of our model was superior to the two previous models. Figure 1

      Prediction Performance of the Present and Previous Models
      Models c-statistic (95%CI) Hosmer-Lemeshow test
      The present model 0.75 (0.69-0.80) 0.674
      After bootstrapping 0.75 (0.68-0.80) 0.671
      External validation 0.70 (0.64-0.75) 0.382
      Kates M, et al (2009) 0.63 (0.57-0.69) 0.115
      Poullis M, et al (2013) 0.61 (0.54-0.67) 0.091




      Conclusion:
      Our model displayed an acceptable ability to predict surgical outcome in elderly patients undergoing pulmonary resection for lung cancer.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.