Virtual Library

Start Your Search

C. Straus



Author of

  • +

    P1.04 - Poster Session/ Biology, Pathology, and Molecular Testing (ID 233)

    • Event: WCLC 2015
    • Type: Poster
    • Track: Biology, Pathology, and Molecular Testing
    • Presentations: 1
    • +

      P1.04-012 - Using Computed Tomography Scans to Assess the Histology of Malignant Pleural Mesothelioma (ID 3003)

      09:30 - 09:30  |  Author(s): C. Straus

      • Abstract
      • Slides

      Background:
      The purpose of this study is to assess the histology of malignant pleural mesothelioma using computed tomography (CT) based imaging.

      Methods:
      28 patients with malignant pleural mesothelioma were used (histologies: 17 epithelioid, 11 biphasic). A CT scan was acquired for each patient prior to surgical resection of the tumor. A radiologist identified and outlined the tumor boundary on each CT section that demonstrated tumor. These outlines were analyzed to determine the total volume of disease present, the mean volume of disease per outlined section, and the distribution of Hounsfield Unit (HU) values throughout the outlined tumor. These parameters were used to differentiate tumors of epithelioid and biphasic histologies. For each parameter, cutoffs were determined to maximize the extraction of biphasic cases from the entire cohort, while minimizing the extraction of epithelioid cases.

      Results:
      Discernable differences were extracted from the images of the two different histologies of the disease. Figure 1 shows the mean HU value, the standard deviation and skew in the distribution in the HU values, and the volume of tumor represented on each CT section demonstrating disease. With regard to HU distribution, the biphasic cases generally had a higher mean HU value. For example, 73% of the biphasic cases had a mean value greater than 30, compared to only 29% of the epithelioid cases. Biphasic cases also tend to have a more negative skew in their HU distribution; 73% of biphasic cases had a skew value less than -1, compared to 35% of epithelioid cases. It was also seen that biphasic cases also tended to have a higher volume of tumor present throughout their disease presenting CT sections. There were promising results from extracting the biphasic cases by using optimized cutoffs from gathered data. The criteria used were as follows: Cases that exhibited more than 9 mL of tumor per outlined CT section, or exhibited a mean HU value greater than 10 as well as a skew in HU values less than -1 were extracted from the cohort and identified as biphasic. Of the cases that match these criteria, 10 were actually biphasic while 6 were actually epithelioid. These results are 91% specific, missing only one biphasic case, and 65% specific, correctly excluding 11 of the 17 epithelioid cases. Figure 1 Figure 1 – Comparison of Epithlioid and Biphasic cell types.



      Conclusion:
      This study demonstrates that CT-based imaging may be a useful tool for the assessment of tumor histology through image analysis.

      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.