Evaluating %LAA-950 Thresholds for Optimizing Lung Cancer Screening Frequency in Emphysema Patients: Limitations and Considerations

Le résumé fourni par la source

Abstract Rationale Emphysema, quantified by the percentage of low attenuation area at -950 Hounsfield units (%LAA-950) on CT imaging, has a well-documented association with an elevated risk of lung cancer. Identifying patients at a low risk of lung cancer within two years could allow for less frequent screening. This study aimed to determine an optimal %LAA-950 threshold that could reliably identify patients with minimal lung cancer risk over two years, potentially supporting biennial screening intervals. Methods This retrospective, multi-center cohort study included data from 4,642 patients at Lahey Hospital and Medical Center (LHMC), with a validation cohort of 1,254 patients from Mount Auburn Hospital (MAH). Baseline lung cancer screening CT scans conducted at LHMC between 2012 and 2017 were followed through 2019 to assess lung cancer development within two years. We utilized 4D CT imaging to generate %LAA -950 scores. We analyzed %LAA-950 thresholds, focusing on identifying an optimal cut point below which patients exhibited a low lung cancer incidence. Cox Logistic regression and Chi-square tests were conducted to compare lung cancer outcomes based on %LAA-950 thresholds. Results During the study period, lung cancer was diagnosed in 190 patients (4.1%) at LHMC and in 60 patients (4.8%) at MAH. At LHMC, a %LAA-950 threshold of 0.07% was associated with a lower risk of lung cancer, with a hazard ratio of 0.56 (CI: 0.34-0.95, p=0.03), indicating a potentially protective effect. In contrast, patients with a %LAA-950 threshold of 5% showed increased lung cancer risk, with a hazard ratio of 1.45 (CI: 1.03-2.04, p=0.03). Further analysis revealed that among 577 patients at LHMC with %LAA-950 below 0.07%, 15 (2.6%) still developed lung cancer within two years, which challenges the assumption that low %LAA-950 alone can reliably identify patients for reduced screening frequency. Patients with %LAA-950 ≥0.07% had a significantly higher lung cancer incidence, with 191 cases (4.7%), supported by a statistically significant chi-square test (p=0.022). Additionally, among patients with Lung-RADS negative (1 or 2) scans and %LAA-950 below 0.07%, 8 cases (1.8%) developed lung cancer within two years, suggesting that %LAA-950 alone may not be sufficient to determine screening frequency. Conclusion Our findings indicate that some patients with low %LAA-950 scores still develop lung cancer within two years. Therefore, %LAA-950 thresholds alone may not be adequate to justify biennial screening. These results suggest that combining %LAA-950 thresholds with other risk factors, such as Lung-RADS assessment, may be necessary to optimize screening intervals safely.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.