Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Randomized Controlled Trial
. 2025 Aug 28;16(1):8060.
doi: 10.1038/s41467-025-63471-6.

A comparative analysis of heterogeneity in lung cancer screening effectiveness in two randomised controlled trials

Max Welz  1   2 , Carlijn M van der Aalst  3 , Andreas Alfons  4 , Andrea A Naghi  4   5 , Marjolein A Heuvelmans  6   7   8 , Harry J M Groen  9 , Pim A de Jong  10 , Joachim Aerts  11 , Matthijs Oudkerk  7 , Harry J de Koning #  3 , Kevin Ten Haaf #  12 ; NELSON trial consortium
Collaborators, Affiliations
Randomized Controlled Trial

A comparative analysis of heterogeneity in lung cancer screening effectiveness in two randomised controlled trials

Max Welz et al. Nat Commun. .

Abstract

Clinical trials demonstrate that screening can reduce lung cancer mortality by over 20%. However, lung cancer screening effectiveness (reduction in lung cancer specific mortality) may vary by personal risk-factors. Here we evaluate heterogeneity in lung cancer screening effectiveness through traditional sub-group analyses, predictive modelling approaches and machine-learning in individual-level data from the Dutch-Belgian lung cancer screening trial (NELSON; 14,808 participants, 12,429 men, 2377 women, 2 persons with an unknown sex) and the National Lung Screening Trial (NLST; 53,405 participants, 31,501 men, 21,904 women). We find that screening effectiveness varies by pack-years (screening effectiveness ranges across trials: lowest groups = 26.8-50.9%, highest groups = 5.5-9.5%), smoking status (screening effectiveness ranges across trials: former smokers = 37.8-39.1%, current smokers = 16.1-22.7%) and sex (screening effectiveness ranges across trials: women = 24.6-25.3%; men = 8.3-24.9%). Furthermore, screening effectiveness varies by histology (screening effectiveness ranges across trials: adenocarcinoma = 17.8-23.0%, other lung cancers = 24.5-35.5%, small-cell carcinoma = 9.7%-11.3%). Screening is ineffective for squamous-cell carcinoma in NLST (screening effectiveness = 27.9% (95% confidence interval: 69.8% increase to 4.5% decrease) mortality increase) but effective in NELSON (screening effectiveness = 52.2% (95% confidence interval: 25.7-69.1% decrease) mortality reduction). We find that variations in screening effectiveness across pack-years, smoking status, and sex are primarily explained by a greater prevalence of histologies with favourable screening effectiveness in these groups. Our study shows that heterogeneity in lung screening effectiveness is primarily driven by histology and that relaxing smoking-related screening eligibility criteria may enhance screening effectiveness.

PubMed Disclaimer

Conflict of interest statement

Competing interests: C.M.v.d.A. declares roles in the WHO-IARC European Code Against Cancer working group, B3care user committee and Expert Mission Cancer Screening in Georgia outside of the submitted work. H.J.G. declares consulting fees from Eli Lilly outside of the submitted work. P.A.d.J. research support from Philips Healthcare to their institute outside of the submitted work. J.A. declares speakers fees from Eli Lilly, MSD and BIOCAD, patents from Pamgene and Amphara, data safety monitoring/advisory board participation for Eli-Lilly, Amphara, BIOCAD and MSD, boardmembership of the IASLC and stock ownership in Amphera outside of the submitted work. H.J.d.K. declares consulting fees from Bayer and honoraria from TEVA/Menarini/Astra Zeneca outside of the submitted work. K.t.H. declares grants from NIH, Horizon 2020, University of Zurich, Cancer Research UK, Cancer Australia and the Australian Ministry of Health, speakers fees from Johnson&Johnson and Centre Hospitalier Universitaire Vaudois paid to their institute and travel support from the Rescue Lung Society outside of the submitted work. The remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Relative screening effectiveness for overall LCM by trial and diferent risk-factors aggregated across the considered methods.
Based on N = 400 lung cancer deaths in NELSON and N = 977 lung cancer deaths in NLST. The screening effectiveness estimates and 95% confidence intervals were based on the medians of the point estimates of the rate-ratios, effect models and causal forests. Similarly, the lower and upper bounds of the 95% confidence intervals were based on the medians of the lower and upper bounds across the considered methods. The estimates for the individual methods can be found in Supplementary Figs. S6–S9. Source data are provided as a Source Data file. LC Lung Cancer, LCM Lung Cancer Mortality, NLST National Lung Screening Trial, Dutch-Belgian Lung Cancer Screening Trial (NELSON Nederlands–Leuvens Longkanker Screenings Onderzoek).
Fig. 2
Fig. 2. Screening effectiveness for histology-specific mortality by methodology and trial.
Based on N = 178 Adenocarcinoma deaths, N = 94 Squamous-cell carcinoma deaths, N = 43 Other lung cancer deaths and N = 84 Small-cell carcinoma deaths in NELSON and N = 393 Adenocarcinoma deaths, N = 184 Squamous-cell carcinoma deaths, N = 176 Other lung cancer deaths and N = 209 Small-cell carcinoma deaths in NLST. The screening effectiveness estimates were based on the point estimates of each of the specific methods, by trial. The error bars reflect the 95% confidence intervals of the estimates. The adenocarcinoma-specific NLST estimate for the risk-model approach which uses LLPv3 risk in its first stage includes an interaction-effect between first-stage risk and screening effectiveness. The figure represents the estimate for the screening effectiveness parameter without the interaction effect. Source data are provided as a Source Data file. RM Risk-modelling, ADN Adenocarcinoma, SQM Squamous-cell carcinoma, OTH Other lung cancers, SCLC Small-cell carcinoma, NLST National Lung Screening Trial, Dutch-Belgian Lung Cancer Screening Trial (NELSON Nederlands–Leuvens Longkanker Screenings Onderzoek).
Fig. 3
Fig. 3. Histology distribution by risk-factor in the CT-arms of NELSON and NLST.
Source data are provided as a Source Data file. ADN Adenocarcinoma, SQM Squamous-cell carcinoma, OTH Other lung cancers, SCLC Small-cell carcinoma, NLST National Lung Screening Trial, Dutch-Belgian Lung Cancer Screening Trial (NELSON Nederlands–Leuvens Longkanker Screenings Onderzoek).
Fig. 4
Fig. 4. Histology-specific lung cancer mortality rate differences in NELSON and NLST.
Based on N = 178 Adenocarcinoma deaths, N = 94 Squamous-cell carcinoma deaths, N = 43 Other lung cancer deaths and N = 84 Small-cell carcinoma deaths in NELSON and N = 393 Adenocarcinoma deaths, N = 184 Squamous-cell carcinoma deaths, N = 176 Other lung cancer deaths and N = 209 Small-cell carcinoma deaths in NLST. The lung cancer mortality rate differences represent the difference in lung cancer mortality rate per 1000 person-years in the CT arm compared to the control arm by histology in each trial. The error bars reflect the 95% confidence intervals of the mortality rate differences.The percentages below the mortality rate differences represent the relative differences in mortality rates compared to the control-arm, along with their 95% confidence intervals. Source data are provided as a Source Data file. ADN Adenocarcinoma, SQM Squamous-cell carcinoma, OTH Other lung cancers, SCLC Small-cell carcinoma, NLST National Lung Screening Trial, Dutch-Belgian Lung Cancer Screening Trial (NELSON Nederlands–Leuvens Longkanker Screenings Onderzoek).
Fig. 5
Fig. 5. Differences in lung cancer incidence rates between the screen and control arms of NELSON and NLST by histology and stage.
Based on N = 372 Adenocarcinomas, N = 174 Squamous-cell carcinomas, N = 93 Other lung cancers and N = 103 Small-cell carcinoma in NELSON and N = 985 Adenocarcinomas, N = 461 Squamous-cell carcinomas, N = 298 Other lung cancers and N = 287 Small-cell carcinomas in NLST. The lung cancer incidence rate differences represent the difference in lung cancer incidence rate per 1000 person-years in the CT arm compared to the control arm by stage and histology in each trial. The error bars reflect the 95% confidence intervals of the incidence rate differences. Source data are provided as a Source Data file. ADN Adenocarcinoma, SQM Squamous-cell carcinoma, OTH Other lung cancers, SCLC Small-cell carcinoma, NLST National Lung Screening Trial, Dutch-Belgian Lung Cancer Screening Trial (NELSON Nederlands–Leuvens Longkanker Screenings Onderzoek).
See this image and copyright information in PMC

References

    1. de Koning, H. J. et al. Reduced lung-cancer mortality with volume CT screening in a randomized trial. N. Engl. J. Med.382, 503–513 (2020). - PubMed
    1. The National Lung Screening Trial Research Team. Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. N. Engl. J. Med.365, 395–409 (2011). - PMC - PubMed
    1. Pinsky, P. F., Church, T. R., Izmirlian, G. & Kramer, B. S. The National Lung Screening Trial: Results stratified by demographics, smoking history, and lung cancer histology. Cancer119, 3976–3983 (2013). - PMC - PubMed
    1. Becker, N. et al. Lung cancer mortality reduction by LDCT screening-Results from the randomized German LUSI trial. Int. J. Cancer146, 1503–1513 (2020). - PubMed
    1. Mathilde, M. W. W. et al. Results of the randomized Danish lung cancer screening trial with focus on high-risk profiling. Am. J. Respir. Crit. Care Med.193, 542–551 (2016). - PubMed

Publication types

LinkOut - more resources