Lung cancer is the most common non-AIDS defining cancer (NADC) and leading source of NADC mortality amongst HIV infected (HIV+) individuals. HIV+ persons have a greater burden of lung cancer due to higher smoking rates combined with an independent HIV related increased risk of lung cancer. Unfortunately, most lung cancers are clinically diagnosed at an advanced stage and have 5-year survival rates <15%. Earlier detection strategies to improve lung cancer mortality among HIV+ persons are urgently needed. The National Lung Screening Trial (NLST) recently demonstrated that low-dose screening chest computed tomography (CT) led to a 20% reduction in lung cancer mortality among high-risk, HIV uninfected (HIV-) smokers. As a result, lung cancer screening is now recommended by the National Comprehensive Cancer Network for high-risk smokers. Extrapolating results of the NLST to HIV+ individuals is challenging. The increased risk and burden of lung cancer in this aging population would make HIV+ persons excellent candidates for screening. However, HIV+ persons also experience considerable multimorbidity and can have higher mortality from competing risks. Thus, how to apply results of the NLST to HIV+ persons is unclear, as some patients may not survive long enough to derive benefits of screening. Additionally, HIV+ persons may be more likely to have false positive CT scans due to prior lung disease and immunocompromise. Consequently, the morbidity associated with the work-up of benign nodules, which are quite common, could be significant. Given that an RCT limited to HIV+ patients is very unlikely in the near future, we propose to develop a mathematical model that estimates the potential benefits of screening, identifies the appropriate candidates for screening, and determines the best screening regimen. As the health care burden of screening CT can be substantial, we will first determine the clinical consequences that result from performing screening chest CT scans among HIV+ individuals by analyzing results of CT scans and data from a large cohort study. Then, we will construct a model to estimate the impact of lung cancer screening on mortality among HIV+ persons; we will compare this model and its results to one generated by our collaborators in the NCI-funded Cancer Intervention and Surveillance Modeling Network (CISNET). After base case analyses, we will consider screening in different HIV+ populations, mortality from competing causes, and the optimal regimen and age for screening. The analyses proposed are poised to provide relevant, clinically useful results to inform the care of HIV+ patients. The results will also inform policy makers who are in the process of developing national guidelines for lung cancer screening, and will address early in the process of dissemination of lung cancer screening the role for screening in HIV.
Lung cancer is a leading cause of cancer death among HIV-infected persons. The risks and benefits of lung cancer screening among HIV-infected persons is unknown. We will determine the impact of lung cancer screening on mortality among HIV-infected persons using a mathematical modeling approach.
Showing the most recent 10 out of 16 publications
