Virtual Library
Start Your Search
Rebecca Landy
Author of
-
+
MA05 - Lung Cancer Screening (ID 174)
- Event: WCLC 2020
- Type: Mini Oral
- Track: Screening and Early Detection
- Presentations: 1
- Moderators:
- Coordinates: 1/30/2021, 11:45 - 12:45, Scientific Program Auditorium
-
+
MA05.10 - Performance of Draft 2020 USPSTF Lung-Cancer Screening Guidelines and Potential for use of Risk Models to Reduce Racial/Ethnic Disparities (ID 3564)
11:45 - 12:45 | Presenting Author(s): Rebecca Landy
- Abstract
- Presentation
Introduction
For the same age and smoking history as whites, minorities have substantially different lung-cancer risk. The 2013 US Preventive Services Task Force lung cancer screening guidelines recommended screening ever-smokers aged 55-80, with ≥30 pack-years and ≤15 quit-years. The draft 2020 USPSTF lung cancer screening guidelines expanded this to those aged 50-80, with ≥20 pack-years and ≤15 quit-years, and claim to “partially ameliorate racial disparities in screening eligibility” compared to 2013 guidelines. Incorporating individualized prediction-models into USPSTF guidelines may reduce racial/ethnic disparities in lung-cancer screening eligibility. We examine whether draft 2020 USPSTF lung cancer screening eligibility criteria ameliorate disparities in screening eligibility compared to 2013 guidelines, and whether including ever-smokers whose benefit (calculated by an individualized prediction model) exceeded a threshold would reduce racial/ethnic disparities.
Methods
We empirically modeled the performance of NLST-like screening (3 annual CT-screens, 5-years follow-up) among three cohorts of ever-smokers aged 50-80 using the US-representative 2015 National Health Interview Survey: (i) those eligible by USPSTF-2013 (ages 55-80, ≥30 pack-years, ≤15 quit-years), (ii) USPSTF-2020 guidelines (ages 50-80, ≥20 pack-years, ≤15 quit-years), and (iii) augmenting USPSTF-2020 guidelines to also include individuals with ≥12 days of life-gained according to the Life-Years From Screening-CT (LYFS-CT) model (USPSTF2020+LYFS-CT). Among each race/ethnicity, we calculated the number eligible for screening, proportion of preventable lung-cancer deaths prevented (LCD sensitivity), proportion of gainable life-years gained (LYG sensitivity) and screening effectiveness (the number needed to screen to prevent one lung-cancer death), as well as the relative disparities in lung cancer deaths prevented and life-years gained (absolute differences in percentages of model-estimated gainable life-years from NLST-like screening by eligible whites vs minorities).
Results
8.0 million ever-smokers were eligible under USPSTF-2013, of whom 13% were minorities. These guidelines performed best for whites (20% eligible, preventing 55% of preventable lung-cancer deaths, gaining 48% of gainable life-years). The disparities in lung cancer death sensitivity were 15% for African-Americans and Asian-Americans, and 24% for Hispanic-Americans, with similar results for life-year gained sensitivity. Under USPSTF-2020 guidelines, 14.5 million were eligible. 48% of those who became eligible were aged 50-54, and 16% were minorities. Disparities in lung cancer death sensitivity under USPSTF-2020 were similar to those under USPSTF-2013 (absolute differences: 13% for African-Americans, 19% for Asian-Americans, 27% for Hispanic-Americans). 3.5 million additional high-benefit people were added by LYFS-CT, of whom 29% were minorities, including 22% African-Americans, which equated to a 65% increase. Disparities for African-Americans were nearly eliminated (0% for lung cancer death sensitivity, 1% for life-year gained sensitivity). Disparities for Hispanic-Americans were only reduced slightly (to 23%), and disparities for Asian-Americans remained unchanged at 19%, though screening efficiency increased (USPSTF-2020: Hispanic-Americans: NNS/LCD=501, Asian-Americans: NNS/LCD=550; USPSTF-2020+LYFS-CT: Hispanic-Americans: NNS/LCD=442, Asian-Americans: 505).
Conclusion
Draft 2020 USPSTF guidelines may inadvertently increase racial/ethnic disparities versus the 2013 guidelines. Adding high-benefit individuals regardless of race/ethnicity, identified using LYFS-CT, could improve equity.
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.
-
+
P42 - Screening and Early Detection - Risk Modelling and Artificial Intelligence (ID 177)
- Event: WCLC 2020
- Type: Posters
- Track: Screening and Early Detection
- Presentations: 1
- Moderators:
- Coordinates: 1/28/2021, 00:00 - 00:00, ePoster Hall
-
+
P42.07 - Comparative Performance of Lung Cancer Risk Models to Define Lung Screening Eligibility in the United Kingdom (ID 905)
00:00 - 00:00 | Author(s): Rebecca Landy
- Abstract
Introduction
The National Health Service England (NHS) classifies individuals as eligible for lung cancer screening using two prediction models, PLCOm2012 and Liverpool Lung Project-v2 (LLPv2). However, no study has compared the performance of lung cancer risk models in the United Kingdom.
Methods
We analysed current and former smokers aged 40-80 in the UK Biobank (N=217,199), EPIC-UK (N=30,982), and Generations Study (N=25,849). We quantified model calibration (ratio of expected to observed cases, E/O) and discrimination (AUC).
Results
Risk discrimination in UK Biobank was best for the Lung Cancer Death Risk Assessment Tool (LCDRAT, AUC=0.82, 95%CI=0.81-0.84), followed by the LCRAT (AUC=0.81, 95%CI=0.79-0.82) and the Bach model (AUC=0.80, 95%CI=0.79-0.81) (Figure). Results were similar in EPIC-UK and the Generations Study. All models overestimated risk in all cohorts, with E/O in UK Biobank ranging from 1.30 for PLCOm2012 (95%CI=1.23-1.36) to 2.16 for LLPv2 (95%CI=2.05-2.28). Overestimation increased with area-level socioeconomic status. In the combined cohorts, USPSTF criteria classified 50.6% of future cases as screening-eligible. The LCDRAT and LCRAT identified 60.9%, followed by PLCOm2012 (58.3%), Bach (58.1%), and LLPv2 (53.6%).
Conclusion
Discrimination of lung cancer risk models in UK cohorts was highest for LCDRAT and LCRAT, and lowest for LLPv2. Our results highlight the importance of context-specific validation for prediction tools.
