Written and maintained by CASRAI Editorial Board
Last updated
General information about a published research trial, not medical advice. Decisions about lung cancer screening belong to individuals and their clinicians.
The National Lung Screening Trial (NLST) was a US randomized trial that compared annual low-dose helical computed tomography (CT) with standard chest X-ray in heavy smokers. This page is for researchers, students and research administrators who need the design at a glance, the headline results in both relative and absolute terms, and the route to the trial’s data through the National Cancer Institute (NCI). Facts below come from the official NCI pages, checked 2026-11-07.
What the NLST is and who ran it
NCI describes the NLST as a randomized trial of two screening tests for lung cancer: low-dose helical CT and standard chest X-ray. It was conducted by two groups, ACRIN (the American College of Radiology Imaging Network) and the Lung Screening Study group, and launched in 2002. Initial findings were announced in November 2010 and the primary results were published in the New England Journal of Medicine on June 29, 2011. The trial is registered on ClinicalTrials.gov as NCT00028808. It is a randomized design: participants were randomly assigned to one screening method or the other.
Eligibility
As the NCI NLST page states, the trial enrolled 53,454 current or former heavy smokers who met all of the following:
- age 55 to 74;
- a smoking history of at least 30 pack-years;
- no signs, symptoms or history of lung cancer at entry.
These are the trial’s own entry criteria. They are not a statement of who should be screened today; current screening recommendations are issued separately by bodies such as the US Preventive Services Task Force.
What the trial required: design at a glance
| Element | As described by NCI |
|---|---|
| Participants | 53,454 heavy current or former smokers |
| Comparison | Low-dose helical CT versus standard chest X-ray, by random assignment |
| Screening schedule | Three annual screens |
| Follow-up | About 7 years |
| Registry record | ClinicalTrials.gov NCT00028808 |
Results: relative and absolute figures side by side
NCI reports that participants screened with low-dose helical CT had 15 to 20 percent fewer deaths from lung cancer than those screened with chest X-ray. The same page gives the absolute rates, which put that relative figure in scale:
| Measure | CT arm | Chest X-ray arm |
|---|---|---|
| Lung cancer deaths per 1,000 participants | 17.6 | 20.7 |
| Positive screens, averaged over the three rounds | 24.2% | 6.9% |
The difference between 20.7 and 17.6 is about 3 fewer lung cancer deaths per 1,000 people screened. The higher share of positive CT screens matters for analyses of follow-up testing, not only mortality.
How researchers access NLST data
NCI’s Cancer Data Access System (CDAS) distributes NLST data. The CDAS NLST page sets out these steps:
- Choose a project type. CDAS lists three: Data-Only; Images (images plus data); and EEMS (biospecimens, through the PLCO Etiologic and Early Marker Studies Program).
- Submit a project proposal through CDAS.
- NCI reviews the proposal.
- Sign the agreements if approved: a Data Transfer and Use Agreement and, for biospecimens, a Material Transfer Agreement.
The full dataset described by CDAS covers demographics, screening results, diagnostic procedures, lung cancer outcomes and mortality, including CT and pathology images. Data dictionaries are on the NLST Datasets page. For how such records are classed, see clinical trial data.
Routes that do not go through CDAS
- Some CT and pathology images, with a limited set of clinical variables, are available through NCI Imaging Data Commons and The Cancer Imaging Archive, without a CDAS proposal.
- X-ray images, ACRIN-only data and ACRIN biospecimens are requested from ECOG-ACRIN using its Contact Us form with the subject “Data Sharing”; CDAS states a response within 48 working hours.
Typical phases
Using only the phases NCI describes: launch in 2002; enrollment and three annual screening rounds; follow-up of about 7 years; initial findings (November 2010); peer-reviewed primary results (June 2011); then secondary use of the data by outside investigators through CDAS and the image archives. For a new data request, treat proposal preparation, NCI review and agreement signing as separate stages.
Common mistakes
- Quoting only the relative reduction. “15 to 20 percent” is accurate but should sit beside 17.6 versus 20.7 deaths per 1,000.
- Treating the trial’s eligibility as current screening guidance. The criteria describe who enrolled in the trial, not today’s recommendations.
- Assuming all NLST data comes from CDAS. X-ray images and ACRIN-only data and biospecimens go through ECOG-ACRIN, and some images are open via Imaging Data Commons and The Cancer Imaging Archive.
- Overlooking the positive-screen rate. 24.2% of CT screens were positive versus 6.9% for X-ray, averaged over three rounds.
- Skipping the agreement. Proposal approval is followed by signing the Data Transfer and Use Agreement.
Related programs and data sources
| Resource | What the CDAS page says it provides |
|---|---|
| NCI Cancer Data Access System (CDAS) | Full NLST dataset with images, by proposal, review and Data Transfer and Use Agreement |
| NCI Imaging Data Commons | Some CT and pathology images with limited clinical variables, no CDAS proposal |
| The Cancer Imaging Archive | Some CT and pathology images with limited clinical variables, no CDAS proposal |
| ECOG-ACRIN (Data Sharing form) | X-ray images, ACRIN-only data, ACRIN biospecimens |
For background, see CASRAI’s guides to medical imaging and epidemiology, and the dictionary entries for the clinical trial protocol and informed consent form.
FAQ
What was the National Lung Screening Trial?
A randomized trial of 53,454 heavy smokers aged 55 to 74 comparing annual low-dose helical CT with chest X-ray for three rounds, launched in 2002.
What did it find?
15 to 20 percent lower lung cancer mortality with CT: 17.6 versus 20.7 deaths per 1,000, about 3 fewer per 1,000 screened.
Who could enroll?
Current or former heavy smokers aged 55 to 74 with at least 30 pack-years and no signs, symptoms or history of lung cancer.
How do I get NLST data?
Submit a project proposal in NCI CDAS, pass NCI review and sign the Data Transfer and Use Agreement; some images are available without CDAS from Imaging Data Commons and The Cancer Imaging Archive.
Sources
- NCI, National Lung Screening Trial: https://www.cancer.gov/types/lung/research/nlst, checked 2026-11-07.
- NCI Cancer Data Access System, NLST: https://cdas.cancer.gov/nlst/, checked 2026-11-07.
- ClinicalTrials.gov registry identifier: https://clinicaltrials.gov/study/NCT00028808, checked 2026-11-07.
