A multivariate analysis of low-dose CT lung cancer screening in a Michigan community hospital network.

Abstract

10576 Background: Lung cancer is the leading cause of cancer deaths in the United States. The USPSTF recommended screening for lung cancer with low-dose chest computer tomography (LDCT) in 2013. This was based on the National Lung Screening Trial (NLST) which showed a 20% relative reduction in mortality from LDCT compared with chest radiography in high-risk individuals. The major NLST eligibility criteria were age 55-74, a 30 + pack year smoking history and current smoking status or having quit in the last 15 years. The impact of other risk factors on screening results besides inclusion criteria is unknown. We sought to identify other factors for the risk of lung cancer detection in screened individuals. Methods: This was a retrospective cohort study employing a Michigan based McLaren Hospital Network Lung Cancer Screening Program (LCSP) database from 2015-2022. Participant eligibility in screening was the USPSTF criteria except for an age range of 50-80. Analyses were stratified by BMI, smoking pack-years, insurance status, smoking status, sex, other CT findings, pre- vs. during COVID pandemic (before and after January 2020). Multivariate logistic regression analysis was used for odds ratios (OR) with 95% confidence intervals (CI) as estimates of the relative risk of an elevated Lung-RADS category ≥3 (high risk of detecting lung cancer). Results: A total of 2638 subjects, 49.91% females, 50.09% males, mean BMI 29.50 ± 7.17 (SD), pack-years 44.52 ±18.29, Medicaid 6.66%, Medicare 32.01%, Private Insurance 57.62%, Self-pay /unknown 3.71%. Current smoker 65.49%, Former smoker 34.51%, other CT findings 51.99%, pre-COVID 46.08%, during COVID pandemic 53.92%. Lung-RADS category 0 (0.04%), 1 (35.33%), 2, (50.61%), 3 (6.56%), 4a (4.85%), 4b (1.55%), 4x (1.06%). A statistically significant association with a Lung-RADS category ≥3 was found with lower BMI, OR=0.980 (95% CI, 0.969-0.991), p=0.0005, and female gender, OR=1.232 (95% CI, 1.039-1.462), p=0.0166 (Table). Conclusions: In our study cohort a lower-than-average BMI and also female gender were statistically correlated with an elevated Lung-RADS category ≥3. These findings might be incorporated into a risk stratification model, however further larger studies are needed to validate our findings. [Table: see text]