EARLY DETECTION OF LUNG CANCER IN AN ASYMPTOMATIC HIGH-RISK POPULATION BY LOW-DOSE COMPUTED TOMOGRAPHY SCAN AND MOLECULAR MARKERS: VALIDATION OF A PERSONALIZED RISK MODEL TO OPTIMIZE THE SCANNING INTERVAL

Objectives: Low-dose helical chest computed tomography (LDCT) screening may benefit individuals at an increased risk for lung cancer, but uncertainty exists about the potential harm from screening. In this study we present a per- sonalized risk model to select the target population and optimize the scanning interval. Furthermore we intend to validate the model of microRNA signature for the early detection of non-small-cell lung cancer (NSCLC). Methods: This is a multicentre prospective study (started July 2012) designed to recruit 10 000 high-risk individuals in six Italian institutions. It includes current and former (≤15 years) smokers, >55 years of age, with a 30 pack- years or more smoking history, who are offered LDCT, blood sample and smoking cessation counselling. A personalized risk model was developed. The discriminatory ability of the models was assessed in the validation sets by examining the areas under the receiver operating characteristic curves. People with annual lung cancer risk ≥0.6% receive the miRNA test (TaqMan® Low Density Array Human MicroRNA Panel) and annual LDCT. The <0.6% risk groupreceivesLDCTeverytwoyears.Theaccuracy,sensitivityandspecificity of microRNA and LDCT will be assessed. Results: Preliminary results indicate that a 34 microRNA test can identify early stage NSCLC with 80% accuracy and that the risk model is a sensitive predictor of NSCLC. Conclusions: Validation of the personalized risk model (0.6% cut-off) will be helpful for identifying lower risk persons in whom the screening interval can be safely increased. If the non-invasive miRNA signature is validated in our in- dependent prospective cohort, it may become an important first-line screen- ing tool.