Lung cancer subtype risk stratification: Validation in real world cohorts

About This Grant

SUMMARY Lung cancer screening (LCS) using low dose computed tomography (CT) can reduce mortality in individuals with high-risk smoking histories. However, there are critical limitations to current LCS approaches, which rely solely on interpretation of imaging findings, including: a) The mortality benefit is largely driven by patients with adenocarcinoma (AD) lung cancer, with limited benefit for squamous cell carcinoma (SCC) or small cell lung cancer (SCLC). b) CT screening frequently results in “indeterminate nodules” for which clinical management to determine malignancy is based on AD growth trajectories and often relies on repeat imaging. c) Current LCS protocols and resulting guidelines were created based on evidence from trial cohorts that do not reflect the most at-risk populations. The recent United States Preventive Services Task Force recommendations for LCS state, “Research to identify biomarkers that can accurately identify persons at high risk is needed to improve detection and minimize false-positive results.” Our biomarker data show lung cancer histological subtypes display distinct risk factors consistent with their different pathology, etiology and outcomes, leading our multi-disciplinary team to employ a novel lung subtype-specific approach to address the shortcomings of current LCS. Our methods include both detection of specific blood autoantibody levels and quantitative imaging features to assess the distinct risk of AD, SCC and SCLC lung cancer and will be used in concert with existing guidelines to recommend follow-up actions and lead to earlier diagnosis. While tissue analysis will still be used in diagnosis and treatment of lung cancer, our approach for screening will overcome critical limitations of current guidelines to better classify AD indeterminate nodules and increase SCLC and SCC detection sensitivity, thus identifying patients who benefit from immediate action. Our specific aims are to #1: Validate the performance of our subtype specific risk prediction models in existing LCS sample sets. #2: Evaluate the performance and real-world utility of our risk prediction models prospectively in patients undergoing LCS across multiple sites and populations.