Skip to content

Lung squamous cell carcinoma — screening and early detection

TL;DR — Low-dose CT (LDCT) screening reduces lung-cancer mortality in high-risk smokers: NLST reported a 20.0% relative reduction (95% CI 6.8–26.7), and NELSON reported a 10-year mortality rate ratio of 0.76 (0.61–0.94) (Aberle 2011, PMID 21714641; de Koning 2020, PMID 31995683). These are lung-cancer—not LUSC-specific—effects, but LUSC is strongly smoking-linked and therefore concentrated in eligible populations. Its central-airway biology creates a blind spot: flat mucosal lesions may be CT-occult, and a bronchoscopy referral cohort found 115 CT-occult central LUSCs among 10,851 lung cancers (Zhou 2024, PMID 38443800). Population bronchoscopy has no demonstrated mortality benefit; even six-monthly autofluorescence bronchoscopy failed to improve outcomes over standard surveillance in people with low-grade airway lesions (Guisier 2022, PMID 35236723). The central screening problem is selection: simple age/pack-year rules miss high-risk people and can perpetuate sex/race disparities, whereas risk models improve discrimination but select different populations and require local calibration (Katki 2018, PMID 29800127; Pinsky 2021, PMID 33545164).

What randomized screening has established

Trial/evidence set Population and comparison Lung-cancer mortality Harms/implementation anchor Source
NLST 53,454 high-risk US participants; 3 annual LDCT vs chest radiography 20.0% relative reduction (95% CI 6.8–26.7) 24.2% positive screens; 96.4% of positives false positive under original definition Aberle 2011, PMID 21714641
NELSON 13,195 men + 2,594 women; 4 volume-CT rounds vs no screening Men: 10-year rate ratio 0.76 (0.61–0.94) 9.2% had ≥1 additional CT; 2.1% referral rate de Koning 2020, PMID 31995683
USPSTF evidence review 7 RCTs, 86,486 participants NLST IRR 0.85; NNS 323/6.5 y; NELSON IRR 0.75; NNS 130/10 y 17 invasive procedures/1,000 screened from NLST false positives; <1 major complication Jonas 2021, PMID 33687468
Cochrane review 8 mortality RCTs, 91,122 participants RR 0.79 (0.72–0.87); moderate-certainty Invasive-test RR 2.60 (2.41–2.80); estimated overdiagnosis 18% (0–36%) Bonney 2022, PMID 35921047
NLST extended follow-up Median 12.3-y mortality follow-up RR 0.92 (0.85–1.00); dilution-adjusted RR 0.89 (0.80–0.997) NNS 303; no persistent excess incidence NLST Research Team 2019, PMID 31260833

The original NLST false-positive proportion reflects a low size threshold and should not be transplanted directly into modern structured nodule management. NELSON’s volume/doubling-time algorithm produced a much lower referral rate while preserving mortality benefit (de Koning 2020, PMID 31995683).

The Cochrane estimate of a 21% lung-cancer mortality reduction is a useful synthesis, but protocols differed in age, smoking exposure, rounds, interval, comparator, and nodule algorithm. The review graded mortality evidence moderate rather than high and found a wide overdiagnosis interval (Bonney 2022, PMID 35921047).

Why the evidence applies to LUSC—but incompletely

LUSC’s dose-response to smoking is steeper than adenocarcinoma’s: men smoking more than 30 cigarettes/day had a pooled OR of 103.5 (95% CI 74.8–143.2) versus never-smokers (Pesch 2012, PMID 22052329). Screening trials enriched for precisely this exposure history, so LUSC contributes materially to detected cancers.

However, neither NLST nor NELSON was powered to demonstrate a histology-specific mortality reduction. A screening program can reduce overall lung-cancer mortality while having unequal sensitivity across tumor locations, growth patterns, and histologies (Aberle 2011, PMID 21714641; de Koning 2020, PMID 31995683).

LUSC feature Screening implication Evidence boundary
Heavy-smoking concentration High representation in pack-year-selected cohorts Eligibility is an exposure proxy, not a histology test; Pesch 2012, PMID 22052329
Central endobronchial origin Flat or subtly thickened mucosal disease may be CT-occult Referral-cohort evidence, not population sensitivity; Zhou 2024, PMID 38443800
Cavitation/necrosis May make a lesion conspicuous but also mimic infection Small retrospective prognosis study; Onn 2005, PMID 16183941
Faster symptomatic presentation through obstruction/bleeding Can create interval cancers despite screening Histology-stratified interval-cancer data are sparse
Airway field cancerization New lesions can arise away from an observed dysplastic focus Bronchoscopic high-risk cohorts; van Boerdonk 2015, PMID 26275031

The central-airway blind spot

In a retrospective cohort assembled through a cancer-center bronchoscopy service, 130 of 10,851 lung cancers (1.20%) were CT-occult: 115 central LUSCs and 15 squamous precursors. All occurred in middle-aged or older men with heavy-smoking histories (Zhou 2024, PMID 38443800). Selection into bronchoscopy and retrospective image review prevent this from estimating LDCT sensitivity in an invited screening population.

Sputum cytology can enrich for central airway lesions. In high-risk cohorts, cytologic atypia followed by autofluorescence bronchoscopy identified early central cancers, but these designs do not demonstrate mortality benefit and are vulnerable to verification bias (Lam 2009, PMID 19010567; Kennedy 2005, PMID 16022912).

Autofluorescence bronchoscopy increases detection of metaplasia/dysplasia compared with white light but also finds lesions that regress. In SELEPREBB, 364 people with low-grade lesions were randomized to standard surveillance or AFB every six months; invasive cancer/persistent CIS did not differ (OR 0.63, 95% CI 0.20–1.96) (Guisier 2022, PMID 35236723).

Therefore:

  • LDCT remains the mortality-tested population screening modality (Bonney 2022, PMID 35921047).
  • Bronchoscopy is a diagnostic response to symptoms, airway abnormalities, or selected high-risk findings—not a proven universal screening add-on (Guisier 2022, PMID 35236723).
  • A negative CT does not end evaluation of persistent hemoptysis, focal wheeze, lobar collapse, or suspicious sputum cytology (Zhou 2024, PMID 38443800).

Eligibility: rules versus risk

The 2021 USPSTF recommendation lowered the starting age and smoking threshold relative to its predecessor, expanding eligibility, but fixed cutoffs remain a lossy representation of risk (Jonas 2021, PMID 33687468; Reese 2021, PMID 33433600).

Selection approach Variables Strength Limitation
Age + pack-years + quit-time rule Three simple variables Transparent, operationally easy Sharp thresholds; pack-years compress duration/intensity and omit COPD, family history, deprivation
PLCOm2012 Demographic, smoking, clinical variables Well-calibrated in two US validation cohorts; AUC 0.75–0.79 among top models Calibration can fail in deprived or non-US populations; Katki 2018, PMID 29800127; Lebrett 2020, PMID 32631933
LLP models Smoking plus occupational/family/clinical risk Incorporates asbestos and pneumonia history Threshold-dependent misses; community validation required; Lebrett 2020, PMID 32631933
LCRAT/LCDRAT Cancer incidence or death risk Can include competing mortality in selection logic More complex implementation; Katki 2018, PMID 29800127

At a uniform five-year risk threshold of 2.0%, nine models selected between 7.6 million and 26 million US ever-smokers. Four better-performing models selected 7.6–10.9 million and agreed on 73% of individuals, but no consensus threshold existed (Katki 2018, PMID 29800127).

In a deprived Manchester community program, PLCOm2012 ≥1.51% selected 1,429 participants and detected 62 of 71 cancers predicted over six years; NLST criteria would have missed 11 of 62 observed cancers. The local cancer detection rate among NLST-eligible participants was about 4.3%, versus 1.7% after two rounds in NLST, suggesting that PLCOm2012 underestimated risk in this setting (Lebrett 2020, PMID 32631933).

Equity of eligibility

Finding Estimate Source
Eligibility-to-incidence ratio in non-Hispanic Black men 30%–50% lower than non-Hispanic White men under evaluated guidelines Pinsky 2021, PMID 33545164
Revised vs prior eligibility, women 25.9% → 36.4% among ever-smokers in surveyed states Reese 2021, PMID 33433600
Revised vs prior eligibility, Black respondents 16.3% → 28.8% Reese 2021, PMID 33433600
Revised vs prior eligibility, Hispanic respondents 10.5% → 18.7% Reese 2021, PMID 33433600

Expanded criteria reduce but do not remove inequity. In the revised-rule analysis, adjusted eligibility odds remained lower for women (OR 0.88, 95% CI 0.79–0.99), Black respondents (0.43, 0.33–0.56), and Hispanic respondents (0.70, 0.62–0.80) versus reference groups (Reese 2021, PMID 33433600).

Pack-years can encode structural inequity because intensity, duration, menthol use, occupational exposure, and access to documentation vary across groups. Risk-model selection may improve efficiency but can reproduce biased source data unless calibration and access are monitored (Smeltzer 2023, PMID 36208717; Potter 2024, PMID 38537159).

Nodule management and false positives

The purpose of structured systems is to distinguish a screen-positive finding requiring surveillance from one requiring immediate invasive diagnosis.

Metric Result Context/source
NLST positive screens 24.2% over three rounds Original threshold; Aberle 2011, PMID 21714641
NLST positives that were false positive 96.4% “False positive” includes benign findings resolved noninvasively; Aberle 2011, PMID 21714641
Lung-RADS retrospective positive rate 27.6% → 10.6% 2,180 baseline screens; McKee 2015, PMID 25176499
Lung-RADS positive predictive value 6.9% → 17.3% 1,603 with follow-up; McKee 2015, PMID 25176499
Community program invasive procedure 5.1% of 2,003 participants Included malignant, false-positive, and incidental indications; Manyak 2023, PMID 36781101
Invasive procedure for false-positive lung nodule 0.4% of participants Multidisciplinary program adhering to algorithm; Manyak 2023, PMID 36781101
Complication in participants without malignancy 0.15% Same single-center program; Manyak 2023, PMID 36781101

The contrast between NLST’s 96.4% false-positive proportion and a modern program’s 0.4% invasive-procedure rate for false-positive nodules illustrates why “false positive” should not be equated with surgery or major harm (Aberle 2011, PMID 21714641; Manyak 2023, PMID 36781101).

Overdiagnosis, radiation, and incidental findings

Overdiagnosis is a cancer detected by screening that would not have become clinically consequential in the person’s lifetime. It cannot be identified with certainty in an individual; it is estimated from long-term excess incidence. Estimates in the USPSTF review ranged from 0% to 67%, while the Cochrane pooled estimate was 18% with a 95% CI spanning 0%–36% (Jonas 2021, PMID 33687468; Bonney 2022, PMID 35921047).

Extended NLST follow-up found nearly identical cumulative incidence (1,701 LDCT vs 1,681 radiography; RR 1.01, 95% CI 0.95–1.09), reducing concern that the original excess represented large persistent overdiagnosis (NLST Research Team 2019, PMID 31260833). LUSC may be less prone to indolent overdiagnosis than ground-glass/lepidic adenocarcinoma, but a histology-specific estimate is not established.

Incidental findings affected 4.4%–40.7% across studies reviewed for USPSTF. Some create cascades; others identify actionable cardiovascular or extrapulmonary disease (Jonas 2021, PMID 33687468). Radiation-induced cancer is modeled rather than directly observable at screening-program scale and accumulates with repeated scans.

Comorbidity and net benefit

Screening requires enough life expectancy and treatment fitness to convert early detection into avoided death. A post-hoc PLCO/NLST comorbidity index found mortality benefit in intermediate quintiles but none in the highest (cause-specific HR 0.99, 95% CI 0.79–1.23); that highest quintile contained more squamous cancers and untreated lung cancer (Gendarme 2025, PMID 39798695).

This creates a LUSC-specific tension: heavy smokers can have both the highest LUSC risk and the least physiological reserve for surgery or radiotherapy. Excluding them solely for comorbidity may deepen inequity; screening without pathways to pulmonary rehabilitation, smoking cessation, stereotactic radiotherapy, or limited resection may fail to deliver benefit.

Screening as a cessation opportunity

In an 818-person randomized trial embedded in screening, eight counseling calls plus eight weeks of nicotine patches produced 3-month bioverified abstinence of 9.1% versus 3.9% with three calls/two weeks of patches (OR 2.70, 95% CI 1.44–5.08). Differences were not significant at six or twelve months (Taylor 2022, PMID 35818122).

The result supports integrated treatment while showing that a one-time “teachable moment” is insufficient for durable cessation. Screening messages must avoid implying that annual CT cancels tobacco risk.

Early-detection research beyond anatomy

Candidate Rationale Current limitation
Airway gene-expression signatures Measure field injury in normal-appearing bronchus Requires bronchoscopy or validated nasal/buccal surrogate; Beane 2019, PMID 31015447
Sputum cytology/molecular assays Direct sampling of central exfoliating tumors Sensitivity and reproducibility inadequate for stand-alone screening; Lam 2009, PMID 19010567
Blood ctDNA/methylation Potentially histology-agnostic Stage-I sensitivity and false-positive management remain limiting
Radiomics/AI Integrates nodule texture and growth External validation, calibration drift, and opaque thresholds
Combined risk + biomarker Could reduce imaging burden or prioritize high-risk fields Must prove incremental mortality benefit, not only AUC

Open questions

  • What is LDCT sensitivity for central LUSC and CIS in an invited screening cohort, not a bronchoscopy referral population (Zhou 2024, PMID 38443800)?
  • Can a sputum, nasal, or blood marker identify CT-occult progressive airway lesions while avoiding the overdiagnosis seen with indiscriminate bronchoscopy (Guisier 2022, PMID 35236723; Beane 2019, PMID 31015447)?
  • Which eligibility model maximizes LUSC deaths prevented after explicitly accounting for COPD, cardiovascular death, operability, and treatment access (Katki 2018, PMID 29800127; Gendarme 2025, PMID 39798695)?
  • Do structured nodule algorithms maintain sensitivity for cavitary and endobronchial squamous lesions while reducing false positives (McKee 2015, PMID 25176499; Onn 2005, PMID 16183941)?
  • What intervention sustains cessation beyond three months in the screening setting (Taylor 2022, PMID 35818122)?

References

  1. National Lung Screening Trial Research Team, Aberle DR, Adams AM, Berg CD, Black WC, Clapp JD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365(5):395-409. PMID 21714641
  2. de Koning HJ, van der Aalst CM, de Jong PA, Scholten ET, Nackaerts K, Heuvelmans MA, et al. Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial. N Engl J Med. 2020;382(6):503-513. PMID 31995683
  3. Jonas DE, Reuland DS, Reddy SM, Nagle M, Clark SD, Weber RP, et al. Screening for Lung Cancer With Low-Dose Computed Tomography: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2021;325(10):971-987. PMID 33687468
  4. Bonney A, Malouf R, Marchal C, Manners D, Fong KM, Marshall HM, et al. Impact of low-dose computed tomography (LDCT) screening on lung cancer-related mortality. Cochrane Database Syst Rev. 2022;8(8):CD013829. PMID 35921047
  5. National Lung Screening Trial Research Team. Lung Cancer Incidence and Mortality with Extended Follow-up in the National Lung Screening Trial. J Thorac Oncol. 2019;14(10):1732-1742. PMID 31260833
  6. Pesch B, Kendzia B, Gustavsson P, Jöckel KH, Johnen G, Pohlabeln H, et al. Cigarette smoking and lung cancer--relative risk estimates for the major histological types from a pooled analysis of case-control studies. Int J Cancer. 2012;131(5):1210-9. PMID 22052329
  7. Zhou J, Yu B, Guo P, Wang S. The insufficiency of CT examination in early detection of central lung squamous cell carcinoma and squamous epithelial precancerous lesions. BMC Cancer. 2024;24(1):299. PMID 38443800
  8. Onn A, Choe DH, Herbst RS, Correa AM, Munden RF, Truong MT, et al. Tumor cavitation in stage I non-small cell lung cancer: epidermal growth factor receptor expression and prediction of poor outcome. Radiology. 2005;237(1):342-7. PMID 16183941
  9. van Boerdonk RA, Smesseim I, Heideman DA, Coupé VM, Tio D, Grünberg K, et al. Close Surveillance with Long-Term Follow-up of Subjects with Preinvasive Endobronchial Lesions. Am J Respir Crit Care Med. 2015;192(12):1483-9. PMID 26275031
  10. Lam B, Lam SY, Wong MP, Ooi CG, Fong DY, Lam DC, et al. Sputum cytology examination followed by autofluorescence bronchoscopy: a practical way of identifying early stage lung cancer in central airway. Lung Cancer. 2009;64(3):289-94. PMID 19010567
  11. Kennedy TC, Franklin WA, Prindiville SA, Cook R, Dempsey EC, Keith RL, et al. High prevalence of occult endobronchial malignancy in high risk patients with moderate sputum atypia. Lung Cancer. 2005;49(2):187-91. PMID 16022912
  12. Guisier F, Deslee G, Birembaut P, Escarguel B, Chapel F, Bota S, et al. Endoscopic follow-up of low-grade precancerous bronchial lesions in high-risk patients: long-term results of the SELEPREBB randomised multicentre trial. Eur Respir J. 2022;60(3). PMID 35236723
  13. Katki HA, Kovalchik SA, Petito LC, Cheung LC, Jacobs E, Jemal A, et al. Implications of Nine Risk Prediction Models for Selecting Ever-Smokers for Computed Tomography Lung Cancer Screening. Ann Intern Med. 2018;169(1):10-19. PMID 29800127
  14. Lebrett MB, Balata H, Evison M, Colligan D, Duerden R, Elton P, et al. Analysis of lung cancer risk model (PLCO(M2012) and LLP(v2)) performance in a community-based lung cancer screening programme. Thorax. 2020;75(8):661-668. PMID 32631933
  15. Pinsky PF, Lau YK, Doubeni CA. Potential Disparities by Sex and Race or Ethnicity in Lung Cancer Screening Eligibility Rates. Chest. 2021;160(1):341-350. PMID 33545164
  16. Reese TJ, Schlechter CR, Potter LN, Kawamoto K, Del Fiol G, Lam CY, et al. Evaluation of Revised US Preventive Services Task Force Lung Cancer Screening Guideline Among Women and Racial/Ethnic Minority Populations. JAMA Netw Open. 2021;4(1):e2033769. PMID 33433600
  17. Smeltzer MP, Liao W, Faris NR, Fehnel C, Goss J, Shepherd CJ, et al. Potential Impact of Criteria Modifications on Race and Sex Disparities in Eligibility for Lung Cancer Screening. J Thorac Oncol. 2023;18(2):158-168. PMID 36208717
  18. Potter AL, Xu NN, Senthil P, Srinivasan D, Lee H, Gazelle GS, et al. Pack-Year Smoking History: An Inadequate and Biased Measure to Determine Lung Cancer Screening Eligibility. J Clin Oncol. 2024;42(17):2026-2037. PMID 38537159
  19. McKee BJ, Regis SM, McKee AB, Flacke S, Wald C. Performance of ACR Lung-RADS in a clinical CT lung screening program. J Am Coll Radiol. 2015;12(3):273-6. PMID 25176499
  20. Manyak A, Seaburg L, Bohreer K, Kirtland SH, Hubka M, Gerbino AJ. Invasive Procedures Associated With Lung Cancer Screening in Clinical Practice. Chest. 2023;164(2):544-555. PMID 36781101
  21. Gendarme S, Irajizad E, Long JP, Fahrmann JF, Dennison JB, Ghasemi SM, et al. Impact of Comorbidities on the Mortality Benefits of Lung Cancer Screening: A Post-Hoc Analysis of the PLCO and NLST Trials. J Thorac Oncol. 2025;20(5):565-576. PMID 39798695
  22. Taylor KL, Williams RM, Li T, Luta G, Smith L, Davis KM, et al. A Randomized Trial of Telephone-Based Smoking Cessation Treatment in the Lung Cancer Screening Setting. J Natl Cancer Inst. 2022;114(10):1410-1419. PMID 35818122
  23. Beane JE, Mazzilli SA, Campbell JD, Duclos G, Krysan K, Moy C, et al. Molecular subtyping reveals immune alterations associated with progression of bronchial premalignant lesions. Nat Commun. 2019;10(1):1856. PMID 31015447
  24. Lancaster HL, Heuvelmans MA, Oudkerk M. Low-dose computed tomography lung cancer screening: Clinical evidence and implementation research. J Intern Med. 2022;292(1):68-80. PMID 35253286
  25. Adams SJ, Stone E, Baldwin DR, Vliegenthart R, Lee P, Fintelmann FJ. Lung cancer screening. Lancet. 2023;401(10374):390-408. PMID 36563698