Skip to content

Lung adenocarcinoma — Molecular landscape

TL;DR — Lung adenocarcinoma is a collection of genomically defined diseases rather than one molecular entity. TCGA found a mean 8.9 somatic mutations/Mb in 230 resected tumors and recurrent alteration of RTK–RAS–RAF, PI3K–mTOR, p53, cell-cycle, and chromatin pathways (Cancer Genome Atlas 2014, PMID 25079552). Driver frequencies vary sharply with smoking, ancestry, sex, stage, and assay: EGFR and fusions enrich in never-smokers and East Asian cohorts, KRAS and high tobacco-signature burden in smokers, while TP53, STK11, and KEAP1 co-alterations modify phenotype and treatment response. Multiregion TRACERx and whole-genome studies show that truncal drivers coexist with subclonal evolution, chromosomal instability, and treatment-selected resistance, making a single biopsy an incomplete snapshot (Frankell 2023, PMID 37046096). Frequency tables therefore require a named population and assay; no percentage below is universal.

Genomic architecture

Pathway Recurrent events Clinical meaning
RTK–RAS–RAF EGFR, KRAS, ALK/ROS1/RET/NTRK fusions, BRAF, MET, ERBB2 Primary targetable partition
p53 TP53 mutation/deletion Genomic instability and adverse co-context in several drivers
Cell cycle CDKN2A loss, RB1 alteration, CCND amplification Proliferation; RB1/TP53 loss relates to lineage transformation
PI3K–mTOR PIK3CA, PTEN, AKT/mTOR events Usually co-drivers; few established monotherapy standards
Oxidative stress KEAP1/NFE2L2 Aggressive biology and immune/chemotherapy resistance association
Chromatin/RNA SETD2, ARID1A, SMARCA4, MGA Heterogeneity; emerging synthetic-lethal targets

TCGA’s 230 resected adenocarcinomas identified 18 significantly mutated genes and RIT1 activation, but early-stage surgical sampling and historical ancestry composition constrain frequency estimates (Cancer Genome Atlas 2014, PMID 25079552). Morphologic subgroups also differ molecularly, showing that histology and genotype are correlated but non-redundant (Ci 2020, PMID 32953482).

Actionable driver map

Driver Approximate range in adenocarcinoma Enrichment Therapeutic class
EGFR sensitizing ~10–15% Western; ~40–50% many East Asian advanced cohorts Never-smoking, East Asian ancestry, female sex EGFR TKI
KRAS G12C ~13% in Western adenocarcinoma Tobacco exposure, European ancestry Covalent KRAS G12C inhibitor
ALK fusion ~3–7% Younger age, never/light smoking ALK TKI
ROS1 fusion ~1–2% Never/light smoking ROS1 TKI
BRAF V600E ~1–2% No single demographic rule BRAF + MEK inhibition
MET exon 14 ~3–4% Older age; pleomorphic histology enrichment MET TKI
RET fusion ~1–2% Never/light smoking Selective RET TKI
ERBB2 mutation ~2–4% Never/light smoking enrichment HER2-directed ADC
NTRK fusion <1% Histology-agnostic rare event TRK inhibitor

These ranges synthesize testing cohorts and are deliberately approximate; clinical testing should not be omitted because a patient lacks an enriched phenotype. PIONEER prospectively found EGFR mutations in 51.4% of 1,450 evaluable advanced Asian adenocarcinomas, with variation by country, sex, ethnicity, and smoking (Shi 2014, PMID 24419411). A Brazilian series linked EGFR frequency to Asian genetic ancestry and found an independent prognostic role for KRAS, illustrating within-country admixture effects (Leal 2019, PMID 30824880).

Hispanic/Latino sequencing likewise shows heterogeneity that self-identified ethnicity alone cannot capture (Gimbrone 2017, PMID 28911955). Analyses across Asian and non-Asian datasets report sex-biased molecular features that interact with smoking and ancestry; pooled “male versus female” percentages can therefore be misleading (Li 2023, PMID 36964522).

Smoking and never-smoker biology

Dimension Tobacco-associated pattern Never-smoker pattern
Mutation burden Higher, tobacco-signature substitutions Lower average burden, but heterogeneous
Common drivers KRAS, TP53; actionable events still occur EGFR and kinase fusions enriched
Geography Smoking history remains dominant Strong geographic variation even after never-smoking restriction
Immune context Higher neoantigen load on average Driver-positive tumors often immunologically “cold”

Sherlock-Lung analyzed 871 treatment-naive never-smokers from 28 locations: KRAS mutations were 3.8 times more common in North American/European than East Asian never-smoker adenocarcinomas, while EGFR and TP53 were more prevalent in East Asia (Díaz-Gay 2025, PMID 40604281). The study also identified geographically varying mutational signatures, arguing against a single causal exposure.

A 1,024-tumor whole-genome landscape linked divergent evolutionary trajectories to tobacco exposure, ancestry, sex, endogenous processes, and LINE-1 retrotransposition (Zhang 2026, PMID 41372401). This scale improves subgroup resolution but does not convert associations into individual exposure attribution.

Co-mutations as biological context

Co-event Common driver context Observed association Current status
TP53 EGFR, ALK, KRAS More chromosomal instability, mixed response, shorter control in several cohorts Prognostic; not a validated drug-selection rule
STK11 KRAS Immune-excluded phenotype and poorer checkpoint outcomes Negative-risk marker; prospective selection unresolved
KEAP1/NFE2L2 KRAS and smoking-associated disease Oxidative-stress programme, aggressive course No established targeted standard
RB1 + TP53 EGFR Small-cell transformation risk Supports re-biopsy at atypical progression
SMARCA4 Diverse Aggressive phenotype and altered immune context Emerging classification/therapeutic relevance

In metastatic EGFR/TP53 co-mutant adenocarcinoma, TRACERx-linked work associated chromosomal instability and genome doubling with mixed intra-patient TKI responses (Hobor 2024, PMID 38871738). This supplies a plausible mechanism for heterogeneous response but does not prove that TP53 status alone should choose the initial TKI.

Clonality and evolution

TRACERx sampled 1,644 tumor regions from 421 NSCLC patients and found significant subclonal selection affecting classical genes including TP53 and KRAS; intratumor heterogeneity and copy-number instability tracked relapse and outcome (Frankell 2023, PMID 37046096). A single core biopsy can therefore miss subclonal resistance, rare high-grade morphology, or spatially restricted immune states.

Evolutionary stage Dominant process Sampling implication
Initiation Truncal driver and early copy-number events Primary driver often detectable across sites
Diversification Subclonal mutation, genome doubling, chromosomal instability One region underestimates heterogeneity
Metastatic spread Clonal bottleneck and organ selection Metastasis may differ from primary
Therapy Selection of resistant pre-existing or acquired clones Re-biopsy/plasma at progression can change treatment

ORACLE was developed around clonally expressed genes to reduce spatial sampling bias and then prospectively validated as a survival-associated expression biomarker; clinical utility still requires proof that using it improves decisions (Biswas 2025, PMID 39789179). AI histology systems similarly aim to integrate whole-slide heterogeneity, but external validation and explainability remain necessary (Pan 2024, PMID 38200244).

Tumor microenvironment

Single-cell studies show adenocarcinoma-specific immune ecosystems rather than a uniform “inflamed/non-inflamed” binary. Paired early-tumor, normal-lung, and blood profiling mapped tumor-associated myeloid and lymphoid changes (Lavin 2017, PMID 28475900). Multiregion sequencing of 186,916 cells from five early adenocarcinomas demonstrated spatial evolution of malignant, epithelial, stromal, and immune states (Sinjab 2021, PMID 33972311).

Cross-histology single-cell analysis of 72,475 cells identified distinct immune landscapes in adenocarcinoma versus squamous carcinoma (Wang 2022, PMID 36008393). A TP53-focused multiomic atlas further linked TP53 mutation to loss of alveolar identity and multicellular tissue remodelling (Zhao 2025, PMID 41057692). These atlases generate mechanisms and candidate biomarkers; sample sizes and analytic pipelines remain barriers to clinical thresholds.

Genotype–histology relationships

Morphology Molecular enrichment Caveat
Lepidic/nonmucinous EGFR enrichment in many cohorts Not specific; all drivers require testing
Invasive mucinous KRAS common; NRG1/ERBB2 fusions in subsets RNA testing important when DNA panel is negative
Solid/high grade TP53, smoking signatures, higher burden Heterogeneous and not a surrogate for PD-L1
Micropapillary Aggressive behaviour and nodal spread No unique actionable driver

Open questions

What would change practice

The practical threshold is not whether a feature is statistically associated with outcome; it is whether a reproducible assay changes treatment allocation and improves outcomes. Co-mutations, lineage states, and microenvironment classes therefore remain research biomarkers until tested prospectively against a defined alternative.

Candidate layer Required next evidence
Co-mutation panel Prospective treatment-by-biomarker interaction
Clonal expression score Locked assay, cutoff, external calibration, decision trial
Single-cell state Reducible bulk/spatial surrogate with reproducible sampling
Mutational signature Exposure attribution validated across populations
Plasma subclonality Proof that acting on low-frequency clones improves outcome

Longitudinal sampling should retain lesion site, treatment exposure, and time because evolution is the signal rather than a nuisance.

  • Can co-mutation states prospectively select immune, targeted, or combination therapy rather than merely stratify prognosis?
  • How much spatial sampling is enough to capture clinically relevant heterogeneity (Frankell 2023, PMID 37046096)?
  • Which never-smoker mutational signatures correspond to preventable exposures (Díaz-Gay 2025, PMID 40604281)?
  • Can clonal expression or plasma monitoring overcome single-biopsy bias and improve survival (Biswas 2025, PMID 39789179)?
  • Which lineage and microenvironment states mediate persistence after targeted therapy?

References

  1. Cancer Genome Atlas Research Network, et al. Comprehensive molecular profiling of lung adenocarcinoma. Nature. 2014. PMID 25079552
  2. Frankell AM, et al. The evolution of lung cancer and impact of subclonal selection in TRACERx. Nature. 2023. PMID 37046096
  3. Díaz-Gay M, et al. The mutagenic forces shaping the genomes of lung cancer in never smokers. Nature. 2025. PMID 40604281
  4. Zhang T, et al. Uncovering the role of LINE-1 in the evolution of lung adenocarcinoma. Nature. 2026. PMID 41372401
  5. Shi Y, et al. A prospective, molecular epidemiology study of EGFR mutations in Asian patients with advanced non-small-cell lung cancer of adenocarcinoma histology (PIONEER). J Thorac Oncol. 2014. PMID 24419411
  6. Leal LF, et al. Mutational profile of Brazilian lung adenocarcinoma unveils association of EGFR mutations with high Asian ancestry and independent prognostic role of KRAS mutations. Sci Rep. 2019. PMID 30824880
  7. Gimbrone NT, et al. Somatic Mutations and Ancestry Markers in Hispanic Lung Cancer Patients. J Thorac Oncol. 2017. PMID 28911955
  8. Li X, et al. Sex-biased molecular differences in lung adenocarcinoma are ethnic and smoking specific. BMC Pulm Med. 2023. PMID 36964522
  9. Ci B, et al. Molecular differences across invasive lung adenocarcinoma morphological subgroups. Transl Lung Cancer Res. 2020. PMID 32953482
  10. Hobor S, et al. Mixed responses to targeted therapy driven by chromosomal instability through p53 dysfunction and genome doubling. Nat Commun. 2024. PMID 38871738
  11. Biswas D, et al. Prospective validation of ORACLE, a clonal expression biomarker associated with survival of patients with lung adenocarcinoma. Nat Cancer. 2025. PMID 39789179
  12. Pan X, et al. The artificial intelligence-based model ANORAK improves histopathological grading of lung adenocarcinoma. Nat Cancer. 2024. PMID 38200244
  13. Lavin Y, et al. Innate Immune Landscape in Early Lung Adenocarcinoma by Paired Single-Cell Analyses. Cell. 2017. PMID 28475900
  14. Sinjab A, et al. Resolving the Spatial and Cellular Architecture of Lung Adenocarcinoma by Multiregion Single-Cell Sequencing. Cancer Discov. 2021. PMID 33972311
  15. Wang C, et al. The heterogeneous immune landscape between lung adenocarcinoma and squamous carcinoma revealed by single-cell RNA sequencing. Signal Transduct Target Ther. 2022. PMID 36008393
  16. Zhao W, et al. A cellular and spatial atlas of TP53-associated tissue remodeling defines a multicellular tumor ecosystem in lung adenocarcinoma. Nat Cancer. 2025. PMID 41057692
  17. Travis WD, et al. International association for the study of lung cancer/american thoracic society/european respiratory society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol. 2011. PMID 21252716
  18. Chang WC, et al. Pulmonary invasive mucinous adenocarcinoma. Histopathology. 2024. PMID 37867404
  19. Howlader N, et al. The Effect of Advances in Lung-Cancer Treatment on Population Mortality. N Engl J Med. 2020. PMID 32786189
  20. Byun J, et al. Genome-wide association study of familial lung cancer. Carcinogenesis. 2018. PMID 29924316
  21. Hill W, et al. Lung adenocarcinoma promotion by air pollutants. Nature. 2023. PMID 37020004
  22. Masago K, et al. Genomic Landscape of Resected Invasive Mucinous Adenocarcinoma of the Lung. Clin Lung Cancer. 2025. PMID 40634197
  23. Nicholson AG, et al. The 2021 WHO Classification of Lung Tumors: Impact of Advances Since 2015. J Thorac Oncol. 2022. PMID 34808341
  24. Rokutan-Kurata M, et al. Validation Study of the International Association for the Study of Lung Cancer Histologic Grading System of Invasive Lung Adenocarcinoma. J Thorac Oncol. 2021. PMID 33905897
  25. Rivera GA, et al. Lung Cancer in Never Smokers. Adv Exp Med Biol. 2016. PMID 26667338