Lung cancer has a high incidence worldwide and thus, population-based screening programmes have been proposed as a feasible approach to improve the outcomes of patients with this disease. Now, the publication of 10-year follow-up results of the NELSON trial further informs this debate.

On the basis of a questionnaire sent to ~600,000 individuals of 50–74 years of age, 15,792 respondents with a history of smoking were randomly assigned (1:1) to be part of the screening group (four rounds of low-dose CT screening for lung cancer at baseline, 1, 3 and 5.5 years) or to the control group (no screening). Data on randomization date, sex, date of lung-cancer diagnosis, and date and cause of death were obtained after approximately 5, 7 and 10–11 years of follow up. Among male participants, 6,583 were assigned to the screening group and 6,612 to the control group. The average screening uptake was 90.0%, with 22,600 CT scans performed (9.2% of which needed to be repeated). A positive result was obtained in 467 CT scans (2.1%), subsequently resulting in 203 screening-detected lung cancers.

At 10 years of follow up, the cumulative incidence of lung cancer was 5.58 and 4.91 cases per 1,000 person-years (344 and 304 lung cancers) in the screening and control group, respectively (rate ratio 1.14, 95% CI 0.97–1.33) and thus, 59.0% of cancers detected in the screening group were screening-detected cancers. These cancers were more often diagnosed at stage IA–B (58.6%) than non-screening-detected cancers in the screening group (14.2%) and cancers detected in the control group (13.5%). The frequency of diagnosis of stage IV lung cancers was 9.4%, 51.8% and 45.7%, respectively. Cumulative mortality from lung cancer was 2.50 and 3.30 deaths per 1,000 person-years (156 and 206 deaths) in the screening and control group, respectively (rate ratio 0.76, 95% CI 0.61–0.94). All-cause mortality was 13.93 and 13.76 deaths per 1000 person-years, respectively.

Overall, these results suggest that population-based CT screening can increase the likelihood of detecting lung cancers at an early stage. Therefore, together with optimal treatment, this approach could improve the chance of long-term survival.