Skip to content

Lung adenocarcinoma — Screening and early detection

TL;DR — Low-dose CT (LDCT) is the only lung-cancer screening modality with randomized mortality evidence: NLST reduced lung-cancer mortality 20% versus chest radiography, and NELSON reduced mortality with a volume-based schedule versus no screening (NLST 2011, PMID 21714641; de Koning 2020, PMID 31995683). The 2021 USPSTF recommends annual LDCT at age 50–80 for people with ≥20 pack-years who currently smoke or quit within 15 years, after shared decision-making and only where diagnostic and treatment capacity exist (USPSTF 2021, PMID 33687470). Benefit is counterweighted by false positives, incidental findings, radiation, invasive work-up, and overdiagnosis; modern volumetric and Lung-RADS pathways reduce but do not eliminate these harms (Bonney 2022, PMID 35921047). Adenocarcinoma is disproportionately represented among screen-detected peripheral nodules, while never-smoker adenocarcinoma remains largely outside evidence-based eligibility. Blood-based early detection is investigational: case-control discrimination is not proof of mortality benefit or safe replacement for CT (Shen 2022, PMID 36514063).

Mortality evidence

Trial Population Intervention / comparator Principal result Key limitation
NLST 53,454 US participants, age 55–74, ≥30 pack-years, current or quit ≤15 y Three annual LDCT vs chest radiography Lung-cancer mortality RR reduction 20.0% (95% CI 6.8–26.7); all-cause mortality reduction 6.7% (1.2–13.6) (NLST 2011, PMID 21714641) Expert centres; radiography comparator; high false-positive count
NELSON 13,195 men plus 2,594 women, age 50–74, smoking-risk criteria Volume-based LDCT at baseline, 1, 3, 5.5 y vs no screening At 10 y, male lung-cancer mortality rate ratio 0.76 (95% CI 0.61–0.94) (de Koning 2020, PMID 31995683) Female subgroup underpowered; European implementation context

NLST’s design compared LDCT with chest radiography, not no screening, and used a low size threshold that labelled many examinations positive (NLST design 2011, PMID 21045183). At the initial round, 27.3% of LDCT examinations were positive, but only 3.8% of positive screens contained lung cancer; most positive results therefore required resolution without cancer diagnosis (NLST 2013, PMID 23697514).

NELSON used nodule volume and volume-doubling time to classify indeterminate nodules, reducing immediate positive calls (de Koning 2020, PMID 31995683). New nodules during incidence rounds cannot be managed exactly like baseline nodules: persisting new solid nodules had appreciable malignancy probability at smaller size, supporting separate thresholds (Walter 2019, PMID 30591535).

Systematic reviews pooling randomized trials confirm a lung-cancer mortality reduction with LDCT, but effect magnitude varies with eligibility, comparator, adherence, nodule protocol, and follow-up (Hunger 2021, PMID 34198856; Bonney 2022, PMID 35921047). Chest radiography and sputum cytology have not shown comparable mortality benefit (Usman Ali 2016, PMID 27130532).

Eligibility is a policy choice

The 2021 USPSTF broadened US eligibility from the 2013 threshold to age 50–80, ≥20 pack-years, current smoking or quit within 15 years, with annual screening stopped after 15 quit-years or when health limits curative treatment (USPSTF 2021, PMID 33687470).

Selection strategy Strength Failure mode
Age + pack-years + quit-years Simple, auditable, aligned with trial entry Excludes high-risk people below thresholds and nearly all never-smokers (Senthil 2023, PMID 37806735)
Multivariable risk model Integrates age, duration, intensity, COPD and other predictors Calibration, threshold, equity, and workflow complexity (Toumazis 2023, PMID 36745885)
Deep-learning/radiograph risk May exploit pre-existing imaging Retrospective selection and transportability; not mortality-tested (Lee 2022, PMID 35699582)
Never-smoker risk enrichment Could address family history and Asian high-incidence settings No randomized mortality trial; overdiagnosis risk (Chang 2024, PMID 38042167)

Economic modelling suggests risk-model selection can be more cost-effective than categorical USPSTF criteria at some thresholds, but model conclusions inherit assumptions about adherence, treatment, life expectancy, and willingness-to-pay (Toumazis 2023, PMID 36745885). A prospective German comparison of PLCOm2012 and NELSON criteria is testing effectiveness directly rather than relying only on modelling (Vogel-Claussen 2025, PMID 41232542).

The fixed 15-year quit threshold creates discontinuity despite risk changing continuously with age and time since cessation. Risk-adapted starting and stopping ages are being modelled, but have not displaced trial-anchored categorical recommendations (Frick 2025, PMID 41632141).

Nodule management

Screening nodules and incidental nodules arise in different populations and use different systems. Lung-RADS is designed for an organized screening programme; Fleischner guidance addresses incidentally detected nodules outside screening. Applying one system in the other context changes pretest probability and follow-up intensity (Nam 2022, PMID 35803268).

Nodule feature Why it matters Typical management logic
Size / volume Strong baseline malignancy predictor Below threshold: interval CT; above threshold: short-interval CT, PET/CT, or tissue evaluation
Growth / volume-doubling time Separates stable scars from active lesions NELSON used volumetry to resolve indeterminate nodules (de Koning 2020, PMID 31995683)
Solid vs part-solid vs ground glass Correlates with invasive component and adenocarcinoma spectrum Persistent subsolid nodules often require longer surveillance
New nodule Incident nodules can be malignant at smaller size Apply incident-round thresholds (Walter 2019, PMID 30591535)
Patient operability and preference Determines whether detection can produce benefit Screening is inappropriate when curative treatment would not be pursued (USPSTF 2021, PMID 33687470)

The management objective is not “biopsy every nodule”; it is to preserve mortality benefit while minimizing procedures for benign disease. Volumetry, structured reporting, multidisciplinary review, and tracking systems are programme components, not optional add-ons (Adams 2023, PMID 36563698).

Harms and trade-offs

Harm Evidence signal Mitigation
False positive Initial NLST LDCT positivity 27.3%; cancer in 3.8% of positive screens (NLST 2013, PMID 23697514) Higher thresholds, volumetry, serial imaging, structured algorithms
Overdiagnosis Excess indolent cancers persist in trial follow-up; magnitude depends on duration and model (Bonney 2022, PMID 35921047) Long follow-up; avoid immediate intervention for low-risk subsolid nodules
Radiation Repeated CT plus diagnostic imaging adds cumulative dose (Milleron 2017, PMID 27956083) Low-dose protocols; avoid unnecessary scans
Invasive work-up of benign disease Biopsy/surgery can cause pneumothorax, bleeding, or resection Multidisciplinary probability-based evaluation
Incidental findings Cardiovascular, thyroid, adrenal and other abnormalities trigger cascades Defined reporting and referral pathways
Anxiety / preference conflict Indeterminate results create prolonged uncertainty Shared decisions and explicit result communication

Overdiagnosis is a population concept: a pathologically real cancer that would not have caused symptoms or death during the person’s lifetime. It cannot be identified with certainty in an individual. Short follow-up inflates estimates because screening advances diagnosis time; very long follow-up can still be confounded by subsequent screening in controls (Hunger 2021, PMID 34198856).

Implementation gap

Mortality efficacy requires invitation, eligibility assessment, shared decision-making, scan completion, nodule tracking, diagnostic resolution, and repeat annual attendance. Failure at any step converts an efficacious test into an ineffective programme (Adams 2023, PMID 36563698).

US uptake remains low and unequal. Reviews describe lower eligibility or participation among women, racial/ethnic minorities, rural populations, and people with low income or limited access; the 2021 criteria improve representation but do not remove structural barriers (Senthil 2023, PMID 37806735; Reid 2025, PMID 42038935). Puerto Rico data illustrate how eligibility, use, and being up-to-date must be reported separately (Castañeda-Avila 2025, PMID 40612588).

Community-health-worker education can be implemented in underserved settings, but pilot feasibility is not the same as completed annual screening or mortality impact (Williams 2021, PMID 32594413). Qualitative synthesis across underserved communities identifies shared determinants spanning knowledge, clinician recommendation, transport, cost, distrust, and competing priorities (Pandey 2025, PMID 41338709).

Screening and smoking cessation

LDCT does not substitute for tobacco treatment. Screening encounters create repeated opportunities for cessation, and the combined mortality effect is expected to exceed either approach alone (Moldovanu 2021, PMID 33718048). Randomized screening participation can itself alter smoking behaviour, complicating comparisons of CT-result effects (Pedersen 2016, PMID 27195275).

A systematic review found heterogeneous cessation interventions embedded in LDCT programmes and no single proven delivery model; pharmacotherapy plus behavioural support remains the evidence-based tobacco-treatment frame, while screening programmes must measure abstinence rather than merely referrals (Iaccarino 2019, PMID 30753860).

Never-smoker screening

TALENT prospectively screened risk-enriched Taiwanese never-smokers or very light former smokers and found many early-stage adenocarcinomas, particularly among people with family history (Chang 2024, PMID 38042167). Because there was no unscreened randomized control, the study establishes detection yield—not lung-cancer mortality reduction, all-cause benefit, or acceptable overdiagnosis.

Never-smoker screening is therefore a research question, not a direct extension of NLST. Baseline incidence, driver spectrum, competing mortality, nodule prevalence, and treatment thresholds differ across countries; a strategy useful in Taiwan may have a lower positive predictive value elsewhere.

Blood-based and AI early detection

Liquid-biopsy studies evaluate ctDNA mutations, methylation, fragmentation, circulating cells, microRNAs, and proteins. A meta-analysis found diagnostic and prognostic signals in early-stage NSCLC, while emphasizing heterogeneity and the need for clinical-utility evaluation (Shen 2022, PMID 36514063).

Technology Reported promise Required evidence before screening use
ctDNA methylation Case-control discrimination; inexpensive targeted assays possible (Wang 2023, PMID 37330511) Prospective asymptomatic sensitivity by stage, false-positive work-up, mortality endpoint
Methylation + fragmentomics Multimodal signal and tissue localization (Nguyen 2023, PMID 37819044) External calibration at screening prevalence
Broad liquid biopsy Multiple analytes accessible without tissue (Casagrande 2023, PMID 36768828) Demonstrated incremental benefit over LDCT, not just AUC
Imaging AI Risk/diagnostic prediction from existing images (Lee 2022, PMID 35699582) Multi-centre prospective validation and fairness monitoring

Early-stage tumors shed little ctDNA, producing the central sensitivity problem. Conversely, clonal haematopoiesis and benign conditions can create false positives. A high case-control area under the curve can coexist with poor positive predictive value in a low-prevalence screening population.

Open questions

Minimum outcome set for new screening strategies

Any never-smoker, blood-first, or AI-assisted strategy should report invitation, uptake, stage distribution, false-positive procedures, interval cancers, radiation, overdiagnosis estimate, lung-cancer mortality, all-cause mortality, and equity by prespecified groups. Detection rate or AUC alone is not clinical utility.

  • Can risk-model eligibility improve deaths prevented per scan without worsening racial, sex, or socioeconomic inequity (Toumazis 2023, PMID 36745885)?
  • What randomized design can establish benefit versus overdiagnosis for never-smoker LDCT (Chang 2024, PMID 38042167)?
  • Which nodule-volume and growth thresholds retain mortality benefit while minimizing invasive benign work-up (de Koning 2020, PMID 31995683)?
  • What programme interventions improve completion of the entire screening cascade, not only ordering or first-scan uptake (Reid 2025, PMID 42038935)?
  • Can a blood or AI test add stage-I sensitivity to LDCT at an acceptable false-positive rate and ultimately reduce mortality (Shen 2022, PMID 36514063)?

References

  1. National Lung Screening Trial Research Team, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011. PMID 21714641
  2. de Koning HJ, et al. Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial. N Engl J Med. 2020. PMID 31995683
  3. US Preventive Services Task Force, et al. Screening for Lung Cancer: US Preventive Services Task Force Recommendation Statement. JAMA. 2021. PMID 33687470
  4. National Lung Screening Trial Research Team, et al. The National Lung Screening Trial: overview and study design. Radiology. 2011. PMID 21045183
  5. National Lung Screening Trial Research Team, et al. Results of initial low-dose computed tomographic screening for lung cancer. N Engl J Med. 2013. PMID 23697514
  6. Adams SJ, et al. Lung cancer screening. Lancet. 2023. PMID 36563698
  7. Walter JE, et al. Persisting new nodules in incidence rounds of the NELSON CT lung cancer screening study. Thorax. 2019. PMID 30591535
  8. Bonney A, et al. Impact of low-dose computed tomography (LDCT) screening on lung cancer-related mortality. Cochrane Database Syst Rev. 2022. PMID 35921047
  9. Hunger T, et al. Lung Cancer Screening with Low-Dose CT in Smokers: A Systematic Review and Meta-Analysis. Diagnostics (Basel). 2021. PMID 34198856
  10. Usman Ali M, et al. Screening for lung cancer: A systematic review and meta-analysis. Prev Med. 2016. PMID 27130532
  11. Toumazis I, et al. Risk Model-Based Lung Cancer Screening : A Cost-Effectiveness Analysis. Ann Intern Med. 2023. PMID 36745885
  12. Senthil P, et al. Update on Lung Cancer Screening Guideline. Thorac Surg Clin. 2023. PMID 37806735
  13. Lee JH, et al. Deep Learning to Optimize Candidate Selection for Lung Cancer CT Screening: Advancing the 2021 USPSTF Recommendations. Radiology. 2022. PMID 35699582
  14. Vogel-Claussen J, et al. Effectiveness of NELSON versus PLCOm2012 lung cancer screening eligibility criteria in Germany (HANSE): a prospective cohort study. Lancet Oncol. 2025. PMID 41232542
  15. Frick C, et al. Risk-Adapted Lung Cancer Screening Starting Ages for Former Smokers. JAMA Netw Open. 2025. PMID 41632141
  16. Nam JG, et al. Evaluation and Management of Indeterminate Pulmonary Nodules on Chest Computed Tomography in Asymptomatic Subjects: The Principles of Nodule Guidelines. Semin Respir Crit Care Med. 2022. PMID 35803268
  17. Milleron B, et al. [Lung screening]. Rev Pneumol Clin. 2017. PMID 27956083
  18. Reid MJ, et al. Examining implementation strategies to reduce disparities in lung cancer screening uptake by state Medicaid expansion status: a scoping review. Prev Oncol Epidemiol. 2025. PMID 42038935
  19. Castañeda-Avila MA, et al. Lung Cancer Screening Eligibility, Uptake, and Adherence in Puerto Rico, 2022. JTO Clin Res Rep. 2025. PMID 40612588
  20. Williams LB, et al. Using Implementation Science to Disseminate a Lung Cancer Screening Education Intervention Through Community Health Workers. J Community Health. 2021. PMID 32594413
  21. Pandey S, et al. Common Determinants of Lung Cancer Screening Uptake in Three High-Risk and Underserved Communities. J Am Coll Radiol. 2025. PMID 41338709
  22. Moldovanu D, et al. Lung cancer screening and smoking cessation efforts. Transl Lung Cancer Res. 2021. PMID 33718048
  23. Pedersen JH, et al. Smoking cessation and lung cancer screening. Ann Transl Med. 2016. PMID 27195275
  24. Iaccarino JM, et al. Combining smoking cessation interventions with LDCT lung cancer screening: A systematic review. Prev Med. 2019. PMID 30753860
  25. Chang GC, et al. Low-dose CT screening among never-smokers with or without a family history of lung cancer in Taiwan: a prospective cohort study. Lancet Respir Med. 2024. PMID 38042167
  26. Shen H, et al. Potential clinical utility of liquid biopsy in early-stage non-small cell lung cancer. BMC Med. 2022. PMID 36514063
  27. Casagrande GMS, et al. Liquid Biopsy for Lung Cancer: Up-to-Date and Perspectives for Screening Programs. Int J Mol Sci. 2023. PMID 36768828
  28. Nguyen VTC, et al. Multimodal analysis of methylomics and fragmentomics in plasma cell-free DNA for multi-cancer early detection and localization. Elife. 2023. PMID 37819044
  29. Wang Z, et al. Early detection and stratification of lung cancer aided by a cost-effective assay targeting circulating tumor DNA (ctDNA) methylation. Respir Res. 2023. PMID 37330511