Boston, Massachusetts
March 8–11, 2020


Conference Dates and Location: 
March 4–7, 2019 | Seattle, Washington
Abstract Number: 



Subhashini A. Sellers1, Andrew Edmonds1, Catalina Ramirez1, Sushma Cribbs2, Igho Ofotokun2, Laurence Huang3, Alison Morris4, Meredith C. McCormack5, Ken M. Kunisaki6, Maria P. Rivera1, M. Brad Drummond1, Adaora Adimora1

1University of North Carolina at Chapel Hill, Chapel Hill, NC, USA,2Emory University, Atlanta, GA, USA,3University of California San Francisco, San Francisco, CA, USA,4University of Pittsburgh, Pittsburgh, PA, USA,5Johns Hopkins University, Baltimore, MD, USA,6Minneapolis VA Health Care System, Minneapolis, MN, USA

Abstract Body: 

Based on the National Lung Screening Trial (NLST), US Preventive Services Task Force (USPSTF) recommends screening with low-dose chest computed tomography scan for adults aged 55-80 with >30 pack-year smoking history who are current smokers or quit within the last 15 years. Persons living with HIV (PLWH) are at increased risk for lung cancer but were excluded from the NLST. This study evaluated the performance characteristics of NLST criteria in confirmed lung cancer cases and matched controls from observational cohorts of men and women with HIV. We also explored alternative thresholds to improve lung cancer detection rates.


We selected all confirmed lung cancers among PLWH who were current/former smokers and ≥40 years at diagnosis in the Women's Interagency HIV Study (WIHS) and the Multicenter AIDS Cohort Study (MACS). Controls, selected from each cohort, were PLWH with no reported lung cancer during all follow-up visits, matched on 5-year age windows. Clinical and demographic characteristics, and proportions meeting NLST screening criteria, were compared. Alternative thresholds included iterative reductions in age, pack-years, and quit date.


We identified 44 WIHS women and 17 MACS men with HIV and incident lung cancer (Table). Lung cancer incidence was 270 and 104 per 100,000 person-years in women and men, respectively (p<0.001). Race and income did not differ between cases and controls. Compared to controls, women with lung cancer had a significantly lower median CD4 count but no significant difference in median viral load. In men, there were no significant differences in these markers of HIV infection between cases and controls. Only 16% of women and 24% of men with lung cancer met USPSTF screening criteria. Optimal age and pack-year screening criteria in women (age 49-75, ≥16 pack-year history) yielded 52% sensitivity and 75% specificity. In men, optimal criteria (age 43-75, >19 pack-year history) yielded sensitivity (82%) and specificity (76%).


Current USPSTF lung cancer screening guidelines performed poorly in PLWH, as <25% of lung cancer cases met criteria. Alternative thresholds of age, smoking history, and quit date can better identify PWLH to screen for lung cancer. Among PLWH, lung cancer risk was higher in women than men. This study demonstrates the need for risk prediction modeling incorporating sex and markers of HIV infection to identify high risk individuals who would benefit from screening despite not meeting current USPSTF criteria.


Session Number: 
Session Title: 
Presenting Author: 
Subhashini Sellers
Presenter Institution: 
University of North Carolina at Chapel Hill