P1.03-035 Does Screening with Low-Dose Computed Tomography (LDCT) of Asbestos Exposed Subjects Reduce Mortality for Lung Cancer (LC)?

Abstract

Our previous prospective non-randomized ATOM002 study showed that LDCT screening of asbestos exposed subjects can identify LC at an earlier, and potentially more curable, stage than chest radiographs (CXR) (Fasola et al, The Oncologist 2007). The ATOM002 participants were selected from subjects enrolled in a surveillance program for asbestos exposed workers at the Monfalcone Occupational Health Unit in the Friuli Venezia Giulia (FVG) region, Italy. Here, we report a cohort mortality study of asbestos exposed subjects from that surveillance program, comparing outcomes in the ATOM002 participants and contemporary nonparticipants. Within a cohort of 2,433 asbestos exposed subjects, we compared mortality between the ATOM study participants (who had additional baseline and 1 year LDCT) and nonparticipants (n=926 and 1,507, respectively). The follow-up period spanned the years 2002-2011. Cox models were performed to assess survival for all causes, all cancers, LC and malignant pleural mesothelioma. Final models estimating mortality hazard ratios (HR) were adjusted for smoking habits, age, level of asbestos exposure and Charlson-Quan comorbidity index. For external comparison, we estimated the standardized mortality rate ratio (SMR) using FVG regional standard rates. There was a significant 59.3% (95%CI: 3.9-82.8) reduction in adjusted mortality for LC among ATOM002 participants vs. nonparticipants. LC crude mortality was 99.4 per 100,000 person-year in participants (8 LC deaths) compared to 430.4 per 100,000 person-year in nonparticipants (50 LC deaths). Mortality was also reduced for all causes (HR=0.61; 95%CI 0.44-0.84), but not for all cancers (HR=0.97; 95%CI 0.62-1.50) or malignant pleural mesothelioma (HR=0.86; 95%CI 0.31-2.41). Compared with regional mortality rates, a trend towards reduced mortality for LC was found among ATOM002 participants (SMR =0.55; 95%CI 0.24-1.09), in contrast to a statistically significant increase in the nonparticipants (SMR = 2.07; 95%CI 1.53-2.73). In our cohort, participation in the LDCT based screening study was associated with reduced mortality for LC compared to empiric CXR based public health surveillance. To our knowledge, this is the first report suggesting reduction in mortality for LC with LDCT screening in an asbestos exposed population. LDCT screening might therefore be a reasonable approach for surveillance in these populations.