OP26 Does prior pulmonary tuberculosis affect the positivity rate of low-dose computed tomography during lung cancer screening?

Screening for lung cancer (LC) with low-dose computed tomography (LDCT) reduces mortality. However, high false-positives due to residual lung lesions from prior tuberculosis (TB) is a concern in TB-endemic countries. We prospectively screened all high-risk individuals (aged 55-74 years with ≥30 pack-year [PY] history of smoking and patients of chronic obstructive pulmonary disease [COPD] aged 50-74 years with ≥20 PY) using LDCT at a tertiary care center. Evidence of prior TB (history of intake of anti-TB therapy or radiological sequelae) were noted. Evaluation and follow-up were performed as per National Comprehensive Cancer Network guidelines (NCCN v2.2019). Solid or part-solid nodules ≥6 mm, endobronchial abnormality, or mediastinal lymphadenopathy were considered positive. We also conducted exploratory analysis with nodule cut-off of ≥4 mm (according to NLST algorithm). Screen-positive rate (SPR) and LC detection rate (LCDR) were calculated. Our primary objective was to study the SPR after one round of LDCT scan. Our secondary objectives were LCDR, identifying predictors of positive LDCT results including prior TB, and analyzing costs incurred during LDCT screening. 253 individuals (98.4% male, mean [standard deviation] age of 62.3 [6.2] years, 96% had COPD) were included. Evidence of prior TB was observed in 197 (77.9%) individuals. SPR and LCDR were 32% (n=81) and 1.6% (n=4), respectively. Lung nodule, intrathoracic lymphadenopathy, and endobronchial abnormality were detected in 73, 5, and 3 individuals, respectively. Screen-positive individuals had heavier smoking exposure (40.9 [17.6] vs. 35.1 [14.2] PY, P=0.009). On multivariate analysis, only PY of smoking predicted a positive LDCT scan (OR [95% CI], 1.02 [1.01-1.04]; P=0.007). On exploratory analysis, the SPR was 45.1% with NLST cut-off. Both PY (OR 1.02 [1.00 to 1.04]; P=0.03) and prior TB (OR 1.98 [1.04 to 3.75]; P=0.03) were associated with a positive LDCT scan with NLST cut-off. Median cost per individual (subsidized rate applicable in public sector) for LDCT screening was 3516 INR (US$47) which is nearly 28% of the median monthly family income of the participants. LDCT SPR in this study from a TB-endemic region was higher than that reported previously from non-endemic regions (NLST and NELSON). Higher SPR could be attributed to prior TB when using NLST cut-off, but not with NCCN cut-off. However, LCDR was similar.