Author of

• 25948

  +

  P2.11 - Screening and Early Detection (Not CME Accredited Session) (ID 960)

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

  • 27136

    +

    P2.11-14 - Malignancy Associated Change and The LuCED® Test for Detection of Early Stage Lung Cancer (ID 14044)

    16:45 - 18:00  |  Author(s): Chris Presley
    • Abstract
    • Slides

    Background
    

    Early detection remains the most reliable and effective strategy for curing lung cancer. Many approaches, however, are limited by poor sensitivity or specificity that increase health care costs and potentially risk patient health through unneeded procedures.

    The association between cell morphology and cancer has been established in the pathology literature. However, through the so-called field effect, cancer can introduce subtle morphological changes into non-cancerous cells that are proximal to the tumor site.

    The Cell-CT™ platform and LuCED® test represent a promising new method for detecting lung cancer with high (92%) sensitivity and (95%) specificity based on cytologically abnormal cells. In this research we investigate use of the Cell-CT to detect malignancy associated changes in normal cells near the cancer with a view towards enhancing LuCED test performance.

    a9ded1e5ce5d75814730bb4caaf49419 Method
    

    We present a study of 3D morphological alterations in non-cancer cells obtained from sputum of healthy subjects and biopsy confirmed lung cancer patients. Three major cell types were analyzed from 235 patients: bronchial epithelial columnar, squamous intermediate, and mature macrophages. We used the Cell-CT™ platform to measure over 700 different structural biomarkers for each cell. The measurements were used to define prominent clusters of cells through a hierarchical process that were then used under supervised learning, with case status as ground truth, to create classifiers that optimally separated cells from cancer vs. normal cases.

    4c3880bb027f159e801041b1021e88e8 Result
    

    The table gives classifier development and performance characteristics:

    Cell Type	Number of cells from cancer patients	Number of cells from normal patients	Cluster-based Supervised Learning – aROC
    Squamous Intermediate	2316	684	0.94
    Macrophages	4960	5040	0.94
    Columnar cells	3227	3234	0.92

    8eea62084ca7e541d918e823422bd82e Conclusion
    

    Our results indicate that the Cell-CT can discriminate cell features that are too subtle to distinguish by a human. The study suggests that detection of cells with Malignancy Associated Changes may be used to further enhance the LuCED test’s performance beyond published levels.

    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.