Author of

• 25938

  +

  P2.01 - Advanced NSCLC (Not CME Accredited Session) (ID 950)

  • Event: WCLC 2018
  • Type: Poster Viewing in the Exhibit Hall
  • Track:
  • Presentations: 1
  • Moderators:
  • Coordinates: 9/25/2018, 16:45 - 18:00, Exhibit Hall

  • 26815

    +

    P2.01-19 - Radiomics Features of Contrast Enhanced CT as Prognostic Factors in Resectable Adenocarcinoma of Lung (ID 12915)

    16:45 - 18:00  |  Author(s): Se Hoon Choi
    • Abstract
    • Slides

    Background
    

    To identify radiomics features as prognostic factors in patients with adenocarcinoma of lung and assess its incremental value to the traditional staging system and clinical-pathologic risk factors.

    a9ded1e5ce5d75814730bb4caaf49419 Method
    

    Total 1085 patients who underwent surgery for lung adenocarcinoma were enrolled in this study (from March 2010 to December 2014, training cohort: n = 749; from January 2015 to February 2016, temporal validation cohort: n = 336). A subset of 80/94 reproducible radiomics features including shape, first order statistics and texture features were identified reproducible and selected for analysis. A radiomics signature to predict (overall survival, OS and recurrence free survival, RFS) was generated by using the least absolute shrinkage and selection operator, or LASSO in training cohort. Association between the radiomics signature and prognosis was explored. Prognostic models incorporating radiomics signature alone and combined clinical-pathologic risk factors including staging system, age, sex, smoking status and adenocarcinoma subtype were tested in the temporal validation cohort.

    4c3880bb027f159e801041b1021e88e8 Result
    

    Results: The radiomics signatures (constructed from 5 features identified from LASSO) were significantly associated with OS and RFS. Compared with traditional staging, the radiomics signature resulted in better performance for the estimation of OS (C-index for radiomic signature vs TNM staging = 0.726 vs 0.689 in training set, 0.798 vs 0.766 in validation set) and RFS (0.760 vs 0.722 in training set, 0.773 vs 0.751 in validation set) in both training and validation cohorts. The combined model of radiomic signature and clinical-pathologic risk factors showed a significant improvement of predictive performance over the TNM staging system in both training and validation cohorts (OS, C-index = 0.774 in training set and 0.857 in validation set; RFS, C-index = 0.796 in training set and 0.837 in validation set; for all, p < 0.05).

    8eea62084ca7e541d918e823422bd82e Conclusion
    

    Contrast enhanced CT-based radiomics provided improved prognostic prediction in resectable lung adenocarcinoma, which might enable a step forward precision medicine and affect personalized postoperative treatment strategies.

    6f8b794f3246b0c1e1780bb4d4d5dc53

    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.