Prevalence of High-Risk Nodules in Screened Individuals Selected According to Standard Guideline Criteria Vs a Lung Cancer Risk Prediction Calculator in the Alberta Lung Cancer Screening Program

SESSION TITLE: Top Posters - Electronic session SESSION TYPE: Original Investigation Poster PRESENTED ON: Monday, October 30, 2017 at 12:00 PM - 01:30 PM PURPOSE: To compare the rate of detection of high-risk nodules according to participant eligibility criteria in the Alberta Lung Cancer Screening Program (ALCSP). METHODS: ALCSP is an ongoing study which aims to screen 800 high-risk Albertans for lung cancer with low-dose computed tomography (CT). Eligible participants have a ≥1.5% lung cancer risk over 6 years, using an epidemiologic risk-prediction model (RPM) developed using Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial data, and/or meet the National Lung Screening Trial (NLST) inclusion criteria (ever smokers aged 55-74 years, ≥30 pack-years of smoking and <15 years since quitting). Lung nodules were classed by malignancy risk: the probability that a lung nodule will be diagnosed as cancer within a 2-4 year follow-up period. We assessed the proportion of individuals either meeting or not meeting these inclusion criteria and determined the prevalence of high risk (≥10% malignancy risk) nodules on baseline CT in participants according to eligibility criteria. RESULTS: Baseline questionnaires were received from 1,453 individuals. 500 (34.4 %) met the criteria for the ALCSP and underwent baseline CT screening [114 (23.0%) met the RPM criteria only, 82 (16.4%) met the NLST criteria only, and 304 (60.6%) met both criteria]. Of these 500 individuals, 385 (77%) had at least one detected lung nodule of any size. A trend towards more frequent detection of nodule(s) with a malignancy risk higher than 10 % was noted in the RPM group (RPM: 4.4% (5/114) [95% Confidence Interval (CI): 2-10%], and NLST: 0% (0/82) [95% CI: 0-4%]. P-value=0.076). CONCLUSIONS: Participants meeting RPM but not NLST criteria may harbor more high-risk lung lesions than the converse. No high-risk nodules have been identified in the group exclusively meeting the NLST criteria. Full analysis of the cohort will be performed once all 800 participants have been screened. CLINICAL IMPLICATIONS: Selection of an optimal at-risk population for screening is a key factor in overall program cost-effectiveness. NSLT criteria may not represent the optimal approach to determine lung cancer screening eligibility, and use of RPM criteria may represent a more accurate method in particular to exclude low-risk individuals unlikely to benefit from screening. DISCLOSURE: The following authors have nothing to disclose: Niloofar Taghizadeh, Paul Burrowes, Paul MacEachern, Rommy Koetzler, James Dickinson, Ashley Mae Gillson, Huiming Yang, Martin Tammemagi, Erika Penz, Sachin Pendharkar, Stephen Lam, Andrew Graham, Jessica Culling, Eric Bédard, Alain Tremblay No Product/Research Disclosure Information