Real-world outcomes of a diverse safety-net lung cancer screening program.

Abstract

155 Background: Lung cancer screening (LCS) with low-dose CT (LDCT) reduces lung cancer mortality, yet populations at highest risk often face barriers to screening and remain underrepresented in screening studies. Characterizing participants and outcomes in real-world safety-net screening programs is important to determine whether LCS is effectively reaching vulnerable populations and achieving meaningful clinical outcomes. We evaluated characteristics and outcomes of participants enrolled in a large urban safety-net LCS program. Methods: We analyzed 3,872 individuals enrolled in the safety-net LCS program at the University of Illinois Health system. Demographic, socioeconomic, clinical, and screening characteristics were evaluated, including age, race/ethnicity, household income, insurance status, smoking history, and Lung-RADS category. Participants were categorized as no cancer (Group 0, n=3,749), incident lung cancer diagnosed during follow-up (Group 1, n=58), or lung cancer present at baseline screening (Group 2, n=65). Annual screening adherence was defined as completion of a repeat LDCT 11–15 months after the prior screening examination. Descriptive analyses characterized the screening population and evaluated adherence, cancer detection, and stage at diagnosis. Results: Among 3,872 screened individuals, 63.3% identified as non-Hispanic Black, 13.3% as Hispanic, and 14.6% as non-Hispanic White. Mean age was 62 years. Nearly half of participants (47%) reported annual household incomes below $15,000. Most participants were publicly insured, including 40.1% covered by Medicaid and 40.4% by Medicare. Current smokers comprised 63.8% of the cohort; 74% reported menthol cigarette use. Overall, 123 individuals were diagnosed with lung cancer (3.2% detection rate); 65 cancers were identified at baseline screening and 58 during follow-up. Abnormal screening findings (Lung-RADS ≥3) were present in 21.4% of participants and were more common among individuals diagnosed with lung cancer. Annual screening adherence was 85%. Among patients diagnosed with lung cancer, those diagnosed during longitudinal follow-up were more likely to have Stage I-II disease than those diagnosed on initial screening (77% vs 63%). Stage IV disease accounted for 15% of baseline-detected cancers and <1% of cancers diagnosed during longitudinal follow-up. Conclusions: This safety-net LCS program successfully reached a diverse, high-risk population characterized by socioeconomic vulnerability and a high prevalence of smoking. The program achieved a 3.2% cancer detection rate while maintaining strong longitudinal screening engagement. The predominance of early-stage cancers among individuals diagnosed during follow-up highlights the potential value of ongoing screening in high-risk populations. Ongoing analyses will evaluate outcomes and communication strategies to support equitable lung cancer screening.