Long-term cancer risk associated with lung nodules observed on low-dose screening CT scans

Abstract

ObjectiveNon-calcified nodules (NCNs) associated with false positive low-dose CT (LDCT) lung cancer screens have been attributed to various causes. Some, however, may represent lung cancer precursors. An association of NCNs with long-term lung cancer risk would provide indirect evidence of some NCNs being cancer precursors. MethodsLDCT arm participants in the National Lung Screening Trial (NLST) received LDCT screens at baseline and years 1-2. The relationship between NCNs found on LDCT screens and subsequent lung cancer diagnosis over different time periods was examined at the person and lobe level. For the latter, a lobe had a cancer outcome only if the cancer was located in the lobe. Separate analyses were performed on baseline and post-baseline LDCT findings; for the latter, those with baseline NCNs were excluded and only new (non-pre-existing) NCNs examined. Raw and adjusted rate-ratios (RRs) were computed for presence of NCNs and subsequent lung cancer risk; adjusted RRs controlled for demographic and smoking factors. Results26,309 participants received the baseline LDCT screen. Over median 11.3 years follow-up, 1675 lung cancers were diagnosed. Adjusted RRs for time periods 0–4, 4–8 and 8−12 years following the baseline screen were 5.1 (95 % CI:4.4–5.9), 1.5 (95 % CI:1.3–1.9) and 1.5 (95 % CI:1.2-1.8) at the person-level and 14.7 (95 % CI:12.6–17.2), 2.6 (95 % CI: 2.0–3.4) and 2.2 (95 % CI:1.6–2.9) at the lobe-level. 18,585 participants were included in the post-baseline analysis. Adjusted RRs for periods 0–4, 4–8 and 8−11 years were 5.6 (95 % CI: 4.5–7.0), 1.9 (95 % CI: 1.3–2.7) and 1.6 (95 % CI: 0.9–2.9) at the person-level and 19.6 (95 % CI:14.9–25.3), 2.5 (95 % CI:1.3–4.7) and 3.3 (95 % CI:1.4–7.6) at the lobe-level. Raw RRs were similar. ConclusionNCNs are associated with excess long-term lung cancer risk, suggesting that some may be lung cancer precursors.