Lung squamous cell carcinoma — histology and diagnosis¶
TL;DR — LUSC is diagnosed morphologically by keratinization and/or intercellular bridges; poorly differentiated tumors require immunohistochemical evidence of squamous differentiation, usually diffuse p40, interpreted with an adenocarcinoma marker such as TTF-1 (Travis 2015, PMID 26291008). Across 85 diagnostic studies (17,893 patients), p40 had sensitivity 0.92 (95% CI 0.89–0.95), specificity 0.94 (0.93–0.96), and a diagnostic odds ratio of 377—more specific than p63 (Chen 2022, PMID 35126682). The task is not merely taxonomic: pemetrexed is less effective in squamous tumors, bevacizumab trials excluded or harmed patients with squamous histology, and molecular-testing pathways differ, so classification changes treatment efficacy, safety, and tissue allocation (Scagliotti 2011, PMID 21119545; Sandler 2006, PMID 17167137). Small samples require a minimal panel because excessive stains can exhaust tissue needed for PD-L1 and genomic testing; p40 plus TTF-1 reduced NSCLC-NOS from 46.7% to 14.4% in one prospective biopsy series (Walia 2017, PMID 29168459). A lung lesion with squamous phenotype is not automatically a primary LUSC: metastasis from head-and-neck, cervix, skin, esophagus, or urothelium and mimics such as mesothelioma must be resolved with clinical-radiologic context and focused ancillary testing (Schulte 2020, PMID 32350806; Gruver 2012, PMID 23106579).
Diagnostic definition¶
The 2021 WHO approach remains morphology first, then immunohistochemistry (IHC), then molecular methods where they answer a specific differential or therapeutic question. The edition explicitly separates requirements for resection specimens from those for small biopsies/cytology and lists essential and desirable criteria (Nicholson 2022, PMID 34808341).
| Context | Defensible diagnosis | Required evidence | Qualification |
|---|---|---|---|
| Resection with keratin pearls and/or intercellular bridges | Squamous cell carcinoma, keratinizing | Morphology usually sufficient | Report variant and other clinically relevant features; Travis 2015, PMID 26291008 |
| Resection without obvious keratinization | Squamous cell carcinoma, nonkeratinizing | Squamous IHC required | Exclude adenocarcinoma and neuroendocrine carcinoma; Travis 2015, PMID 26291008 |
| Basaloid morphology | Squamous cell carcinoma, basaloid | Morphology plus squamous phenotype | Distinguish from small-cell, large-cell neuroendocrine, and NUT carcinoma; Travis 2015, PMID 26291008 |
| Small biopsy with clear squamous morphology | NSCLC, squamous cell carcinoma | Morphology; limited confirmatory stain if needed | Avoid unnecessary panels; Travis 2013, PMID 23401443 |
| Small biopsy lacking differentiation, p40 positive/TTF-1 negative | NSCLC, favor squamous cell carcinoma | Pattern and intensity must fit | “Favor” reflects sampling limits; Walia 2017, PMID 29168459 |
| No morphology or lineage-defining IHC | NSCLC-NOS | Diagnosis of exclusion | Preserve tissue and pursue repeat/alternate sampling if treatment depends on subtype; Loo 2010, PMID 20195168 |
Keratinizing, nonkeratinizing, and basaloid categories replaced older subclassifications because the latter had limited reproducibility or clinical utility. Nonkeratinizing tumors specifically require IHC confirmation of squamous differentiation (Travis 2015, PMID 26291008).
Morphologic anchors¶
| Feature | Supports LUSC | Pitfall |
|---|---|---|
| Keratin pearls | Concentric eosinophilic keratin within tumor nests | Keratinization can occur in metastases and adenosquamous carcinoma |
| Dyskeratosis | Individual-cell keratinization | May be focal and absent from a small biopsy |
| Intercellular bridges | Desmosomal connections between polygonal cells | Difficult to see in crushed or poorly fixed specimens |
| Sharp cell borders and dense cytoplasm | Squamous maturation | Not specific without keratinization/bridges |
| Endobronchial dysplasia/CIS adjacent to tumor | Supports pulmonary primary and field origin | Absence does not argue against LUSC; sampling is limited |
| Comedo-type necrosis/cavitation | Common phenotype | Also occurs in other carcinomas and infection |
Morphology alone performs better in well-differentiated resection material than in crushed bronchial biopsies, cytology cell blocks, or necrotic specimens. In a systematic review, pathologist agreement across NSCLC subtypes ranged from 67.1% to 89.6% (κ 0.42–0.84); for the clinically relevant squamous/nonsquamous split it ranged from 77.0% to 94.2% (κ 0.48–0.88) (Paech 2011, PMID 21107286).
An H&E digital-slide study found κ 0.25 for 44 detailed WHO diagnoses, 0.48 for ten major categories, and 0.55 for the squamous/nonsquamous dichotomy. Better differentiation, slide quality, confidence, and pulmonary-pathology expertise improved agreement (Grilley-Olson 2013, PMID 22583114).
Minimal immunohistochemical panel¶
| Marker | Typical interpretation | Meta-analytic performance for pulmonary LUSC | Main caveat |
|---|---|---|---|
| p40 (ΔNp63) | Nuclear, preferably diffuse/strong | Sensitivity 0.92 (0.89–0.95); specificity 0.94 (0.93–0.96) | Also labels squamous tumors from other organs; Chen 2022, PMID 35126682 |
| p63 | Nuclear squamous/basal marker | Sensitivity 0.92 (0.90–0.94); specificity 0.83 (0.80–0.86) | More adenocarcinoma cross-reactivity than p40; Chen 2022, PMID 35126682 |
| CK5/6 | Cytoplasmic/membranous high-molecular-weight keratin | Sensitivity 0.90 (0.87–0.93); specificity 0.91 (0.89–0.93) | Mesothelioma and other tumors can be positive; Chen 2022, PMID 35126682 |
| DSC3 | Membranous desmosomal marker | Sensitivity 0.81 (0.73–0.88); specificity 0.95 (0.85–0.98) | Less sensitive; useful as a focused rule-in stain; Chen 2022, PMID 35126682 |
| TTF-1 | Nuclear pneumocytic/adenocarcinoma marker | Not a squamous marker | Clone-dependent focal staining and rare biphenotypic tumors require context; Inamura 2018, PMID 29538329 |
| Napsin A | Granular cytoplasmic adenocarcinoma marker | Not a squamous marker | Histiocytes and renal/thyroid tumors can confound; Tran 2016, PMID 26447895 |
p40 is preferred to p63 because their sensitivities are similar while p40 is more specific. In the 85-study meta-analysis, diagnostic odds ratios were 377 for p40, 120 for CK5/6, 94 for DSC3, and 70 for p63 (Chen 2022, PMID 35126682).
A practical first-line panel for an undifferentiated NSCLC is one squamous marker (p40) and one glandular marker (TTF-1), with mucin or Napsin A added only when needed. In 263 prospective biopsies, morphology classified 53.2%; p40/TTF-1 reduced NSCLC-NOS from 46.7% to 14.4% and produced 90% concordance with available resections (Walia 2017, PMID 29168459).
An earlier bronchial-biopsy study used mucin, p63, and TTF-1 to subtype 73% of initially unclassifiable cases with 86% accuracy, reducing unclassified cases to 7% when prospectively applied (Loo 2010, PMID 20195168). Exact rates differ because cohorts, markers, thresholds, and reference standards differ.
Interpreting discordant markers¶
| Pattern | Working interpretation | Next step |
|---|---|---|
| p40 diffuse; TTF-1 negative | Squamous differentiation | Confirm primary site clinically; do not infer lung origin from p40 alone |
| TTF-1 diffuse; p40 negative | Adenocarcinoma favored | Add Napsin A/mucin only if morphology or clone behavior is discordant |
| p40 and TTF-1 both negative | NSCLC-NOS or alternate lineage | Review morphology; consider neuroendocrine, metastatic, sarcomatoid, SMARCA4-deficient, or NUT entities |
| p40 and TTF-1 in separate populations | Possible adenosquamous carcinoma | A small biopsy cannot quantify whole-tumor components reliably |
| p40 and TTF-1 in same cells | Rare biphenotypic carcinoma or technical issue | Repeat/orthogonal stains and molecular workup; do not force a routine binary label |
| Focal weak p40 only | Insufficient alone | Correlate with keratinization, CK5/6/DSC3, and exclude p63-family cross-reactivity |
Antibody clone matters. Comparative studies found variable performance among TTF-1, Napsin A, and p40 clones, so a result is inseparable from laboratory validation and cutoff (Tran 2016, PMID 26447895). Agreement among CK5/6, p40, and p63 is high but not perfect; a panel should resolve a differential, not create one through indiscriminate staining (Kriegsmann 2019, PMID 30798982).
Small samples and tissue stewardship¶
Most advanced NSCLC is diagnosed from small biopsy or cytology rather than resection. The sample must support four tasks: establish malignancy, assign lineage, stage when possible, and preserve material for predictive biomarkers (Travis 2013, PMID 23401443; Nicholson 2022, PMID 34808341).
| Specimen | Strength | Limitation | Evidence |
|---|---|---|---|
| Endobronchial forceps biopsy | Architecture and visible central lesion | Crush, necrosis, superficial sampling | Small-biopsy classification framework; Travis 2013, PMID 23401443 |
| EBUS-TBNA cell block | Diagnosis and nodal staging in one procedure | Variable tumor fraction/cell-block processing | 774-patient study: subtype in 77%; NSCLC-NOS reduced by IHC; Navani 2012, PMID 22505743 |
| Transthoracic core | Good peripheral-tumor yield | Pneumothorax/hemorrhage; limited spatial sampling | Specimen choice follows lesion location and patient risk |
| Cytology smear/cell block | Minimally invasive; can support IHC/molecular testing | Architecture limited; fixation varies | IHC can subtype cytology when validated; Ao 2014, PMID 24746197 |
| Resection | Whole-tumor architecture, mixed components | Available only in operable disease | Required for confident adenosquamous quantification; Li 2018, PMID 30147334 |
In the 774-patient EBUS-TBNA study, EGFR testing succeeded in 107 of 119 requested cases (90%); diagnostic accuracy for NSCLC was 91% (95% CI 89–93). IHC halved the adjusted odds of an NSCLC-NOS label (OR 0.50, 95% CI 0.28–0.82) (Navani 2012, PMID 22505743). Modern broad genomic panels impose additional material demands not measured by that 2009–2011 cohort.
Adenosquamous carcinoma and sampling¶
Adenosquamous carcinoma contains both glandular and squamous components and is diagnosed most reliably in a resection, where each can be assessed across the tumor. A small biopsy showing only p40-positive cells may sample the squamous component of an adenosquamous carcinoma and misrepresent its driver probability (Li 2018, PMID 30147334).
Microdissection of both components in 16 adenosquamous tumors found most driver mutations in the shared trunk, supporting monoclonal origin; EGFR and PI3K-pathway alterations occurred in 30% and 25%, respectively (Vassella 2015, PMID 26068980). This is a small European cohort, but it demonstrates why mixed morphology should trigger molecular attention rather than be treated as ordinary LUSC.
Primary lung LUSC versus metastasis¶
No single squamous marker establishes organ of origin.
| Competing primary | Useful evidence | Critical pitfall |
|---|---|---|
| HPV-associated oropharynx | Compare prior tumor; high-risk HPV DNA or E6/E7 mRNA | p16 alone is insufficient: 8/25 primary lung LUSCs expressed p16 but all tested HPV-negative; Schulte 2020, PMID 32350806 |
| Urothelial carcinoma | CK7, CK20, GATA3, uroplakin III vs CK14/DSG3 pattern | GATA3 labeled 23% of primary pulmonary LUSC in one study; use a panel; Gruver 2012, PMID 23106579 |
| Mesothelioma | Mesothelial markers plus epithelial marker such as claudin-4 | Claudin-4 specificity for NSCLC vs epithelioid mesothelioma was 100%, but sensitivity across common NSCLC was 82%; Naso 2020, PMID 32333920 |
| Cutaneous/cervical/esophageal SCC | Clinical history, imaging distribution, viral/molecular comparison when informative | Morphology and p40/CK5/6 overlap extensively |
| Thymic squamous carcinoma | Mediastinal epicenter, CD5/CD117 context | No stain is perfectly specific; anatomy is decisive |
The p16 study directly demonstrates the danger of using a surrogate outside its validated context: 32% of primary lung LUSCs showed some p16 expression, while none had transcriptionally active HPV by E6/E7 RNA in situ hybridization (Schulte 2020, PMID 32350806).
Predictive biomarker handoff¶
Histologic diagnosis should lead to, not replace, predictive testing. PD-L1 IHC is technically distinct from lineage IHC and has assay-, platform-, and scoring-specific requirements; multicenter testing shows reproducibility is achievable but not perfect (Vigliar 2019, PMID 31111063). Molecular testing should not be categorically withheld when clinical or pathologic features suggest an actionable driver; the testing framework is covered in biomarkers (Lindeman 2018, PMID 29398453).
Why classification is a safety decision¶
| Decision | Consequence of calling a tumor squamous | Evidence |
|---|---|---|
| Pemetrexed | Avoid as the histology interaction favors nonsquamous disease | Scagliotti 2011, PMID 21119545 |
| Bevacizumab | Conventional regimens exclude squamous histology because of pulmonary-hemorrhage concern | Sandler 2006, PMID 17167137 |
| First-line chemotherapy partner | Taxane or gemcitabine backbones are used rather than pemetrexed | Histology became a treatment modifier; Scagliotti 2011, PMID 21119545 |
| Genomic testing | Lower average driver yield, not zero yield | Clinical features, small biopsy, and mixed histology can justify broad testing; Lindeman 2018, PMID 29398453 |
| Clinical-trial eligibility | Many studies are histology-specific | Misclassification can exclude or expose patients incorrectly |
The safety logic does not justify overconfidence. When a tiny sample supports only “NSCLC, favor squamous,” the report should preserve that uncertainty and recommend integration with imaging, smoking history, and molecular results (Travis 2013, PMID 23401443).
Open questions¶
- Can a standardized two-marker algorithm reproduce expert classification across laboratories while preserving enough tissue for PD-L1, DNA, and RNA testing (Walia 2017, PMID 29168459; Navani 2012, PMID 22505743)?
- What fraction of “pure” LUSC on small biopsy is adenosquamous carcinoma or metastasis at resection, and how often does that change treatment (Li 2018, PMID 30147334; Vassella 2015, PMID 26068980)?
- Can digital pathology improve the modest H&E-only κ without embedding site-specific training bias (Grilley-Olson 2013, PMID 22583114)?
- How should rare TTF-1/p40 biphenotypic tumors be classified and treated (Nicholson 2022, PMID 34808341)?
- What is the real-world rate of harm from squamous/nonsquamous misassignment in the chemo-immunotherapy era (Scagliotti 2011, PMID 21119545; Sandler 2006, PMID 17167137)?
Related pages¶
- central-airway biology and presentation — precursor morphology and bronchoscopic sampling.
- molecular landscape — genomic features that complement but do not define morphology.
- chemotherapy and histology constraints — treatment consequences of the diagnosis.
- biomarkers — PD-L1 and molecular-testing workflow after subtype assignment.
- red flags and safety concerns — consequences of misclassification and inadequate samples.
References¶
- Nicholson AG, Tsao MS, Beasley MB, Borczuk AC, Brambilla E, Cooper WA, et al. The 2021 WHO Classification of Lung Tumors: Impact of Advances Since 2015. J Thorac Oncol. 2022;17(3):362-387. PMID 34808341
- Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JHM, Beasley MB, et al. The 2015 World Health Organization Classification of Lung Tumors: Impact of Genetic, Clinical and Radiologic Advances Since the 2004 Classification. J Thorac Oncol. 2015;10(9):1243-1260. PMID 26291008
- Travis WD, Brambilla E, Riely GJ. New pathologic classification of lung cancer: relevance for clinical practice and clinical trials. J Clin Oncol. 2013;31(8):992-1001. PMID 23401443
- Chen H, Katakura S, Horita N, Namkoong H, Kato I, Hara Y, et al. Immunohistochemical markers to diagnose primary squamous cell carcinoma of the lung: a meta-analysis of diagnostic test accuracy. Ther Adv Med Oncol. 2022;14:17588359211065152. PMID 35126682
- Inamura K. Update on Immunohistochemistry for the Diagnosis of Lung Cancer. Cancers (Basel). 2018;10(3). PMID 29538329
- Tran L, Mattsson JS, Nodin B, Jönsson P, Planck M, Jirström K, et al. Various Antibody Clones of Napsin A, Thyroid Transcription Factor 1, and p40 and Comparisons With Cytokeratin 5 and p63 in Histopathologic Diagnostics of Non-Small Cell Lung Carcinoma. Appl Immunohistochem Mol Morphol. 2016;24(9):648-659. PMID 26447895
- Kriegsmann K, Cremer M, Zgorzelski C, Harms A, Muley T, Winter H, et al. Agreement of CK5/6, p40, and p63 immunoreactivity in non-small cell lung cancer. Pathology. 2019;51(3):240-245. PMID 30798982
- Walia R, Jain D, Madan K, Sharma MC, Mathur SR, Mohan A, et al. p40 & thyroid transcription factor-1 immunohistochemistry: A useful panel to characterize non-small cell lung carcinoma-not otherwise specified (NSCLC-NOS) category. Indian J Med Res. 2017;146(1):42-48. PMID 29168459
- Loo PS, Thomas SC, Nicolson MC, Fyfe MN, Kerr KM. Subtyping of undifferentiated non-small cell carcinomas in bronchial biopsy specimens. J Thorac Oncol. 2010;5(4):442-7. PMID 20195168
- Ao MH, Zhang H, Sakowski L, Sharma R, Illei PB, Gabrielson E, et al. The utility of a novel triple marker (combination of TTF1, napsin A, and p40) in the subclassification of non-small cell lung cancer. Hum Pathol. 2014;45(5):926-34. PMID 24746197
- Paech DC, Weston AR, Pavlakis N, Gill A, Rajan N, Barraclough H, et al. A systematic review of the interobserver variability for histology in the differentiation between squamous and nonsquamous non-small cell lung cancer. J Thorac Oncol. 2011;6(1):55-63. PMID 21107286
- Grilley-Olson JE, Hayes DN, Moore DT, Leslie KO, Wilkerson MD, Qaqish BF, et al. Validation of interobserver agreement in lung cancer assessment: hematoxylin-eosin diagnostic reproducibility for non-small cell lung cancer: the 2004 World Health Organization classification and therapeutically relevant subsets. Arch Pathol Lab Med. 2013;137(1):32-40. PMID 22583114
- Zakowski MF, Rekhtman N, Auger M, Booth CN, Crothers B, Ghofrani M, et al. Morphologic Accuracy in Differentiating Primary Lung Adenocarcinoma From Squamous Cell Carcinoma in Cytology Specimens. Arch Pathol Lab Med. 2016;140(10):1116-20. PMID 27552093
- Navani N, Brown JM, Nankivell M, Woolhouse I, Harrison RN, Jeebun V, et al. Suitability of endobronchial ultrasound-guided transbronchial needle aspiration specimens for subtyping and genotyping of non-small cell lung cancer: a multicenter study of 774 patients. Am J Respir Crit Care Med. 2012;185(12):1316-22. PMID 22505743
- Li C, Lu H. Adenosquamous carcinoma of the lung. Onco Targets Ther. 2018;11:4829-4835. PMID 30147334
- Vassella E, Langsch S, Dettmer MS, Schlup C, Neuenschwander M, Frattini M, et al. Molecular profiling of lung adenosquamous carcinoma: hybrid or genuine type? Oncotarget. 2015;6(27):23905-16. PMID 26068980
- Schulte JJ, Steinmetz J, Furtado LV, Husain AN, Lingen MW, Cipriani NA. Metastatic HPV-Associated Oropharyngeal Versus Primary Pulmonary Squamous Cell Carcinoma: is p16 Immunostain Useful? Head Neck Pathol. 2020;14(4):966-973. PMID 32350806
- Gruver AM, Amin MB, Luthringer DJ, Westfall D, Arora K, Farver CF, et al. Selective immunohistochemical markers to distinguish between metastatic high-grade urothelial carcinoma and primary poorly differentiated invasive squamous cell carcinoma of the lung. Arch Pathol Lab Med. 2012;136(11):1339-46. PMID 23106579
- Naso JR, Churg A. Claudin-4 shows superior specificity for mesothelioma vs non-small-cell lung carcinoma compared with MOC-31 and Ber-EP4. Hum Pathol. 2020;100:10-14. PMID 32333920
- Klebe S, Swalling A, Jonavicius L, Henderson DW. An immunohistochemical comparison of two TTF-1 monoclonal antibodies in atypical squamous lesions and sarcomatoid carcinoma of the lung, and pleural malignant mesothelioma. J Clin Pathol. 2016;69(2):136-41. PMID 26281863
- Miettinen M, Sarlomo-Rikala M. Expression of calretinin, thrombomodulin, keratin 5, and mesothelin in lung carcinomas of different types: an immunohistochemical analysis of 596 tumors in comparison with epithelioid mesotheliomas of the pleura. Am J Surg Pathol. 2003;27(2):150-8. PMID 12548160
- Vigliar E, Malapelle U, Bono F, Fusco N, Cortinovis D, Valtorta E, et al. The Reproducibility of the Immunohistochemical PD-L1 Testing in Non-Small-Cell Lung Cancer: A Multicentric Italian Experience. Biomed Res Int. 2019;2019:6832909. PMID 31111063
- Lindeman NI, Cagle PT, Aisner DL, Arcila ME, Beasley MB, Bernicker EH, et al. Updated Molecular Testing Guideline for the Selection of Lung Cancer Patients for Treatment With Targeted Tyrosine Kinase Inhibitors: Guideline From the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology. J Mol Diagn. 2018;20(2):129-159. PMID 29398453
- Scagliotti G, Brodowicz T, Shepherd FA, Zielinski C, Vansteenkiste J, Manegold C, et al. Treatment-by-histology interaction analyses in three phase III trials show superiority of pemetrexed in nonsquamous non-small cell lung cancer. J Thorac Oncol. 2011;6(1):64-70. PMID 21119545
- Sandler A, Gray R, Perry MC, Brahmer J, Schiller JH, Dowlati A, et al. Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med. 2006;355(24):2542-50. PMID 17167137