Virtual Library

Start Your Search

Hormuzd Katki



Author of

  • +

    MA20 - Implementation of Lung Cancer Screening (ID 923)

    • Event: WCLC 2018
    • Type: Mini Oral Abstract Session
    • Track: Screening and Early Detection
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/25/2018, 15:15 - 16:45, Room 206 F
    • +

      MA20.03 - Surprising Implications of Proposed Risk-Thresholds for Selecting US Ever-Smokers into CT Lung-Cancer Screening (ID 13834)

      15:25 - 15:30  |  Presenting Author(s): Hormuzd Katki

      • Abstract
      • Presentation
      • Slides

      Background

      Many studies show that using risk-models to select ever-smokers for screening may be more effective and efficient than current US Preventive Service Task Force (USPSTF) guidelines. Current National Comprehensive Cancer Network (NCCN) guidelines permit screening ever-smokers with at least 1.3% 6-year lung-cancer risk. Here we re-evaluate assertions, originally based on pre-2015 or non-representative data, that currently proposed risk-thresholds would screen no more ever-smokers than current USPSTF guidelines and at higher effectiveness and efficiency.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      Using the 2015 US National Health Interview Survey (NHIS), we estimate the number of ever-smokers eligible for lung-cancer screening according to 3 lung-cancer risk-thresholds: 1.3% (NCCN) and 1.51% 6-year risks by the PLCOM2012 risk-model and 1.9% 5-year risk by the Lung Cancer Risk Assessment Tool (LCRAT). The NCCN and 1.51% thresholds were based on data from the Prostate Lung Colorectal and Ovarian (PLCO) cancer screening trial, a non-representative sample of US ever-smokers 1993-2001. The 1.9% threshold was based on pre-2015 but representative data (NHIS 2010-2012). Using previously published methodology to estimate 5-year outcomes following 3 annual CT lung-screens, we calculate screening effectiveness (the number needed to screen (NNS) to prevent 1 lung-cancer death) and efficiency (the number of false-positive CT screens per prevented lung-cancer death).

      4c3880bb027f159e801041b1021e88e8 Result

      8.0M US ever-smokers were eligible for lung-screening by USPSTF guidelines in 2015. Surprisingly, millions more were eligible according to risk-thresholds: 12.6M, 11.3M, and 9.2M ever-smokers were eligible at the 1.3% (NCCN), 1.51%, and 1.9% thresholds, respectively. We estimated effectiveness by USPSTF guidelines as NNS=194, which worsened to 222 and 207 for the 1.3% (NCCN) and 1.51% thresholds respectively, but improved to 172 for the 1.9% threshold. We estimated screening efficiency by USPSTF guidelines as 133 false-positives per prevented death, which worsened to 150 and 141 for the 1.3% (NCCN) and 1.51% thresholds respectively, but improved to 122 for the 1.9% threshold. The PLCOm2012 risk threshold that would select 8.0M 2015 ever-smokers is substantially higher than NCCN guidelines (2.28% vs. 1.3% 6-year risk).

      8eea62084ca7e541d918e823422bd82e Conclusion

      Compared to current USPSTF guidelines, the 1.3% (NCCN) and 1.51% risk thresholds (6-year risks by PLCOm2012) would screen many millions more US ever-smokers and likely at lower effectiveness and efficiency. Although the 1.9% threshold (5-year risk by LCRAT) also chose more ever-smokers than USPSTF guidelines, it may screen them more effectively and efficiently. Our findings demonstrate that risk-thresholds developed using older or non-representative data should be re-evaluated using current and representative data.

      6f8b794f3246b0c1e1780bb4d4d5dc53

      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.