Virtual Library

Start Your Search

P. Jain



Author of

  • +

    MA 17 - Locally Advanced NSCLC (ID 671)

    • Event: WCLC 2017
    • Type: Mini Oral
    • Track: Locally Advanced NSCLC
    • Presentations: 1
    • +

      MA 17.11 - Prediction of Response to Trimodality Therapy Using CT-Derived Radiomic Features in Stage III Non-Small Cell Lung Cancer (NSCLC) (ID 10336)

      16:55 - 17:00  |  Author(s): P. Jain

      • Abstract
      • Presentation
      • Slides

      Background:
      There are no clinically validated biomarkers to identify patients with locally advanced NSCLC who benefit from trimodality therapy (TMT) (i.e. neoadjuvant chemoradiation (NAT) followed by surgery). In this study, we evaluate radiomic (i.e. computer extracted imaging) features of tumor phenotype as potential predictors of pathological response.

      Method:
      123 patients with stage III NSCLC who received TMT were selected for this study. Of these, 33 patients including those with distant metastasis at presentation and those without baseline pre-NAT CT scans were excluded. Lung tumors were retrospectively contoured on 3D SLICER software by an expert reader. A total of 1542 radiomic features (textural and shape) were extracted from intra and peritumoral region using the MATLAB® 2016a platform (Mathworks, Natick, MA). A random forest (RF) machine classifier was trained with the most predictive features identified on the training set (n=45) and then validated on an independent test set (n=45). The primary endpoint of our study was pathological response defined as the percentage of the residual viable tumor.

      Result:
      90 patients with NSCLC were included for analysis with a median age of 64 years (38−88), and 54.4 % men. Tumor histology was predominantly adenocarcinoma (71.1%), stage IIIA (94.4%), with positive N2 nodes (91.1%). Pathological response was achieved in 36 (40%) patients; labeled responders (R) and the rest 54 (60%) were labeled non-responders (NR). No statistically significant difference was found in clinical characteristics. We identified five radiomic features (intratumoral and peritumoral textural patterns) predictive of pathological response (Area under Receiver Operating Characteristic (ROC) Curve = 0.7806, RF classifier). Figure 1



      Conclusion:
      Texture features extracted from within and around the lung tumor on CT images were predictive of pathological response to NAT. Additional validation of these quantitative image-based biomarkers is warranted for accurate early identification of responders who could be potentially spared surgery.

      Only 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, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      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.