Deciphering lung adenocarcinoma evolution: Integrative single-cell genomics identifies the prognostic lung progression associated signature

Pengpeng Zhang; Zijun Yang; Zuo Liu; Ge Zhang; Lianmin Zhang; Zhenfa Zhang; Jun Fan

doi:10.1111/jcmm.18408

Article · 2024 Paper GEO V2

Deciphering lung adenocarcinoma evolution: Integrative single-cell genomics identifies the prognostic lung progression associated signature

Journal of Cellular and Molecular Medicine

Full Authors

Pengpeng Zhang
Zijun Yang
Zuo Liu
Ge Zhang
Lianmin Zhang
Zhenfa Zhang
Jun Fan

DOI / publisher Google Scholar search Cite

Research Question

Can cell-state patterns associated with adenocarcinoma in situ, minimally invasive adenocarcinoma and invasive lung adenocarcinoma inform a prognostic expression signature, and which experimental observations support prioritized genes?

Author Evidence Summary

The researchers integrated two public single-cell LUAD series, TCGA transcriptomic and survival data, five independent GEO survival series, exploratory immunotherapy-response datasets and cultured-cell knockdown assays. The prognostic evidence is observational; pseudotime, immune scores and therapy-response computations require separate interpretation from experiments.

Cite this paper

Plain citation

Pengpeng Zhang, Zijun Yang, Zuo Liu, Ge Zhang, Lianmin Zhang, Zhenfa Zhang, Jun Fan. Deciphering lung adenocarcinoma evolution: Integrative single-cell genomics identifies the prognostic lung progression associated signature. Journal of Cellular and Molecular Medicine. 2024. https://doi.org/10.1111/jcmm.18408

Author-controlled citation; verify against the publisher version of record.

BibTeX RIS / EndNote CSL JSON DOI

Research & review context

Research fields: Lung cancer molecular pathology
Computational oncology
Review topics: Cell states in early lung adenocarcinoma
Transcriptomic survival risk signatures
Scientific role: Retrospective LUAD multi-dataset signature discovery with an ancillary PSMB1 cell-line perturbation study.
Evidence position: Observational transcriptomic/prognostic associations and in-vitro functional readouts; no prospective LPAS-guided therapy evidence.
Related concepts: AIS–MIA–IAC comparison
LPAS nine-gene signature
PSMB1 siRNA
Possible citation contexts: Reviews evaluating cross-sectional single-cell LUAD states and prognostic transcriptomic signatures.
Discussions distinguishing cell-line perturbation from clinical proof of a molecular therapeutic target.

Possible literature-review contexts, not evidence of existing citations. Read the findings and Evidence Scope before citing the original article.

Questions & evidence · Evidence limits · Original article & citation downloads

Evidence Scale

Deciphering lung adenocarcinoma evolution: Integrative single-cell genomics identifies the prognostic lung progression associated signature

Integrative single-cell and bulk-transcriptomic analyses nominated a Cluster 6-related lung progression-associated signature (LPAS) associated with survival and inferred immune states in lung adenocarcinoma. PSMB1 silencing reduced proliferative and migratory phenotypes in cultured LUAD cells; these findings do not establish patient treatment efficacy.

Single-cell specimen scale: GSE189357 and GSE150938: 4 AIS, 8 MIA and 9 IAC specimens (21 reported across pathology groups)
Independent survival-series list: GSE13213 (119), GSE29016 (39), GSE30219 (86), GSE31210 (227) and GSE68465 (443) as reported analyzed series
Model discovery: TCGA-LUAD; cohort n not inferred from unspecified filtering
Local stage-associated RNA sequencing: 14 LUAD surgical specimens (reported)
LPAS signature: 9 genes

Counting note

The 21 scRNA-seq specimens, 14 local RNA-seq specimens, TCGA cases, five survival cohorts and mixed-cancer immunotherapy datasets represent distinct analytic layers; neither cross-dataset independence nor a synthetic unique-patient total is inferred.

Key Findings

KF1 Two single-cell series identified heterogeneous epithelial states across early LUAD pathological categories.

Single-cell cross-sectional profiles; inferCNV and trajectory scores estimate rather than observe clonal evolution.

Evidence for KF1
Measure	Value
Pathological specimens	4 AIS, 8 MIA and 9 IAC in GSE189357/GSE150938
Computational annotation	12 broad cell types; 8 re-clustered inferred tumor epithelial groups

Source locator: PMC11149493: Methods §2.1 Data set source (jcmm18408-sec-0003); Results §3.1 (jcmm18408-sec-0024); Figure 2A–H.

Permalink to KF1

KF2 Cluster 6-associated transcriptomic features were enriched during the reported AIS-to-IAC comparison.

Proportional and pseudotime associations do not establish observed within-patient malignant transformation.

Evidence for KF2
Measure	Value
State of interest	Epithelial Cluster 6
Associated programs	E2F targets and G2M checkpoint signatures
Inference methods	inferCNV, pseudotime and SCENIC

Source locator: PMC11149493: Results §3.2 Biological characteristic analysis of tumour clusters (jcmm18408-sec-0025); Figure 3A–H.

Permalink to KF2

KF3 LPAS was constructed from nine genes and associated with overall survival in TCGA and five GEO cohorts.

Retrospective survival discrimination and cohort reapplication, not independent prospective treatment guidance.

Evidence for KF3
Measure	Value
Selection	41 univariately associated module genes, followed by 9-gene LPAS
TCGA time-dependent ROC AUC	1-year 0.713; 5-year 0.675
Independent series	GSE13213, GSE29016, GSE30219, GSE31210, GSE68465
Model composition	PSMB1, ENY2, NPTN, RPS16, HYAL3, KLK8, CAMTA1, MAGEH1, AKR1A1

Source locator: PMC11149493: Methods §2.8 (jcmm18408-sec-0010); Results §§3.3–3.4 (jcmm18408-sec-0026, jcmm18408-sec-0027); Figures 4B–G and 5A–F.

Permalink to KF3

KF4 LPAS scores were associated with computationally inferred immune infiltration and treatment-related expression patterns.

Immune deconvolution, TIDE and SubMap are analytical scores or expression matching; they are not verified patient-level benefit from immunotherapy.

Evidence for KF4
Measure	Value
Low-score association	Higher inferred immune/stromal signatures in TCGA
Drug results	CTRP/PRISM compound-screen associations are exploratory
Therapy-response analyses	SubMap/TIDE-based association; mixed external malignancy cohorts

Source locator: PMC11149493: Results §§3.5, 3.7, 3.9 (jcmm18408-sec-0028, jcmm18408-sec-0030, jcmm18408-sec-0032); Figures 6A–C, 8D–G, 10A–L.

Permalink to KF4

KF5 PSMB1 knockdown changed LUAD cell-line growth and motility endpoints.

Culture experiments provide selected cell-functional evidence but not an in-vivo antitumor effect or human therapeutic validation.

Evidence for KF5
Measure	Value
Experimental systems	A549 and H1299 human LUAD cell lines
Intervention	PSMB1-targeting siRNA
Reported assays	CCK-8, colony formation, wound healing and Transwell

Source locator: PMC11149493: Methods §§2.15–2.19 (jcmm18408-sec-0017 through jcmm18408-sec-0021); Results §3.10 (jcmm18408-sec-0033); Figure 11A–D.

Permalink to KF5

Study Design & Model Development

Study profile

Study Design: Secondary integration of human LUAD single-cell and bulk expression cohorts with retrospective survival modeling, computational immune/drug-response inference and in-vitro PSMB1 perturbation.
Evidence Type: Human transcriptomic association, retrospective model evaluation and LUAD cell-line functional assays.
Population: Early pathological LUAD tissue samples across AIS, MIA and IAC; independent retrospective GEO/TCGA LUAD survival cohorts and additional tumor-treatment series from heterogeneous cancers.
Primary Endpoint: Association of the LPAS expression score with overall survival in retrospective LUAD series.
Secondary Endpoint: Single-cell state contrasts across pathological stages; immune/stromal signature associations; computational immunotherapy/drug-response hypotheses; PSMB1 knockdown cell phenotypes.
External dataset evaluation: Yes

Data modalities

Two publicly available LUAD scRNA-seq series
TCGA and GEO transcriptomic and clinical survival data
Exploratory external treatment-response expression datasets
Local LUAD RNA sequencing
A549 and H1299 siRNA perturbation assays

Model development

Model Name: Lung progression-associated signature (LPAS)
Model Family: Nine-gene expression-based survival model
Development: TCGA-LUAD retrospective transcriptomic and survival data
External Evaluation: Five GEO LUAD retrospective survival cohorts; horizon-specific AUCs vary.
Clinical Boundary: No prospective LPAS-guided clinical decision validation.

What This Study Adds

Connects an early-LUAD epithelial-state signature to multi-cohort retrospective survival risk stratification.
Adds cell-line perturbation observations for PSMB1 while maintaining distinct levels of molecular association and function.

Evidence Scope

Supports

Cluster 6 and LPAS show reproducible associations in the reported transcriptomic cohorts.
Higher LPAS corresponds to less favorable retrospective LUAD survival in the evaluated cohorts.
PSMB1 silencing reduces selected proliferation and movement phenotypes in A549/H1299 assays.

Does Not Establish

A directly traced longitudinal path from an AIS cell to an IAC cell or a causal origin of Cluster 6.
Independent prospective diagnostic, treatment selection, or outcome improvement from LPAS.
Clinical benefit from PSMB1 inhibition, LPAS-guided immunotherapy or computationally prioritized compounds.
Unique participant counts aggregated across partially overlapping or heterogeneous public datasets.

Limitations

Early stages and bulk outcome datasets are distinct study populations; pseudotime is not lineage tracing.
Source dataset descriptions include heterogeneous immunotherapy studies, and the article uses different counting contexts rather than a single combined n.
Single-cell immune estimates and simulated drug sensitivity are not direct measurements of clinical response.
PSMB1 siRNA provides selected in-vitro evidence, not an in-vivo drug efficacy or clinical trial.

When This Study Is Useful to Cite

CU1: What evidence links early LUAD cell states with stage-associated molecular signatures?
Cross-sectional transcriptomic comparison and computational state inference, not longitudinal lineage tracing.
Evidence: KF1, KF2
CU2: What is the evidence for prognostic LPAS?
Retrospective TCGA development and five GEO series, with no prospective intervention.
Evidence: KF3
CU3: What experimental evidence supports PSMB1 involvement in LUAD?
Two LUAD cell-line knockdown readouts, not demonstrated human benefit.
Evidence: KF5

What This Study Should Not Be Cited to Claim

A directly traced longitudinal path from an AIS cell to an IAC cell or a causal origin of Cluster 6.
Independent prospective diagnostic, treatment selection, or outcome improvement from LPAS.
Clinical benefit from PSMB1 inhibition, LPAS-guided immunotherapy or computationally prioritized compounds.
Unique participant counts aggregated across partially overlapping or heterogeneous public datasets.

Evidence Matrix

Row ID	Component	Context	Finding	Evidence Level	Scope	Source Locator	Evidence refs
evidence-b2-doi-10-1111-jcmm-18408-1	KF1: Two single-cell series identified heterogeneous epithelial states across early LUAD pathological categories.	Single-cell cross-sectional profiles; inferCNV and trajectory scores estimate rather than observe clonal evolution.	Two single-cell series identified heterogeneous epithelial states across early LUAD pathological categories. Pathological specimens: 4 AIS, 8 MIA and 9 IAC in GSE189357/GSE150938; Computational annotation: 12 broad cell types; 8 re-clustered inferred tumor epithelial groups	Single-cell cross-sectional profiles; inferCNV and trajectory scores estimate rather than observe clonal evolution.	Single-cell cross-sectional profiles; inferCNV and trajectory scores estimate rather than observe clonal evolution.	PMC11149493: Methods §2.1 Data set source (jcmm18408-sec-0003); Results §3.1 (jcmm18408-sec-0024); Figure 2A–H.	KF1
evidence-b2-doi-10-1111-jcmm-18408-2	KF2: Cluster 6-associated transcriptomic features were enriched during the reported AIS-to-IAC comparison.	Proportional and pseudotime associations do not establish observed within-patient malignant transformation.	Cluster 6-associated transcriptomic features were enriched during the reported AIS-to-IAC comparison. State of interest: Epithelial Cluster 6; Associated programs: E2F targets and G2M checkpoint signatures; Inference methods: inferCNV, pseudotime and SCENIC	Proportional and pseudotime associations do not establish observed within-patient malignant transformation.	Proportional and pseudotime associations do not establish observed within-patient malignant transformation.	PMC11149493: Results §3.2 Biological characteristic analysis of tumour clusters (jcmm18408-sec-0025); Figure 3A–H.	KF2
evidence-b2-doi-10-1111-jcmm-18408-3	KF3: LPAS was constructed from nine genes and associated with overall survival in TCGA and five GEO cohorts.	Retrospective survival discrimination and cohort reapplication, not independent prospective treatment guidance.	LPAS was constructed from nine genes and associated with overall survival in TCGA and five GEO cohorts. Selection: 41 univariately associated module genes, followed by 9-gene LPAS; TCGA time-dependent ROC AUC: 1-year 0.713; 5-year 0.675; Independent series: GSE13213, GSE29016, GSE30219, GSE31210, GSE68465; Model composition: PSMB1, ENY2, NPTN, RPS16, HYAL3, KLK8, CAMTA1, MAGEH1, AKR1A1	Retrospective survival discrimination and cohort reapplication, not independent prospective treatment guidance.	Retrospective survival discrimination and cohort reapplication, not independent prospective treatment guidance.	PMC11149493: Methods §2.8 (jcmm18408-sec-0010); Results §§3.3–3.4 (jcmm18408-sec-0026, jcmm18408-sec-0027); Figures 4B–G and 5A–F.	KF3
evidence-b2-doi-10-1111-jcmm-18408-4	KF4: LPAS scores were associated with computationally inferred immune infiltration and treatment-related expression patterns.	Immune deconvolution, TIDE and SubMap are analytical scores or expression matching; they are not verified patient-level benefit from immunotherapy.	LPAS scores were associated with computationally inferred immune infiltration and treatment-related expression patterns. Low-score association: Higher inferred immune/stromal signatures in TCGA; Drug results: CTRP/PRISM compound-screen associations are exploratory; Therapy-response analyses: SubMap/TIDE-based association; mixed external malignancy cohorts	Immune deconvolution, TIDE and SubMap are analytical scores or expression matching; they are not verified patient-level benefit from immunotherapy.	Immune deconvolution, TIDE and SubMap are analytical scores or expression matching; they are not verified patient-level benefit from immunotherapy.	PMC11149493: Results §§3.5, 3.7, 3.9 (jcmm18408-sec-0028, jcmm18408-sec-0030, jcmm18408-sec-0032); Figures 6A–C, 8D–G, 10A–L.	KF4
evidence-b2-doi-10-1111-jcmm-18408-5	KF5: PSMB1 knockdown changed LUAD cell-line growth and motility endpoints.	Culture experiments provide selected cell-functional evidence but not an in-vivo antitumor effect or human therapeutic validation.	PSMB1 knockdown changed LUAD cell-line growth and motility endpoints. Experimental systems: A549 and H1299 human LUAD cell lines; Intervention: PSMB1-targeting siRNA; Reported assays: CCK-8, colony formation, wound healing and Transwell	Culture experiments provide selected cell-functional evidence but not an in-vivo antitumor effect or human therapeutic validation.	Culture experiments provide selected cell-functional evidence but not an in-vivo antitumor effect or human therapeutic validation.	PMC11149493: Methods §§2.15–2.19 (jcmm18408-sec-0017 through jcmm18408-sec-0021); Results §3.10 (jcmm18408-sec-0033); Figure 11A–D.	KF5

Q&A

Which pathological stages were compared in the single-cell analysis?

Four AIS, eight MIA and nine IAC specimen profiles were described across two public scRNA-seq series. They are specimens at different pathological stages, not serial samples tracking the same person.

Evidence: KF1

What is the nine-gene LPAS model?

LPAS is a retrospective expression-based LUAD survival score containing nine selected genes, including PSMB1. It was developed from TCGA data and evaluated in five additional GEO survival series.

Evidence: KF3

Was LPAS externally evaluated?

Five additional GEO LUAD survival datasets were evaluated, with differing reported AUCs and follow-up horizons. This is retrospective external dataset evaluation, not a prospective treatment-impact trial.

Evidence: KF3

Does inferred cluster progression prove actual tumor-cell fate transitions?

No. InferCNV, changing cluster frequencies and pseudotime reconstruct possible molecular patterns from cross-sectional data; they do not directly track cells through AIS, MIA and IAC.

Evidence: KF1, KF2

Can low LPAS identify patients who will benefit from immunotherapy?

The article reports in-silico immune scores and SubMap/TIDE comparisons plus treatment-cohort associations, but no prospective LPAS-guided treatment study establishes patient-level clinical benefit.

Evidence: KF4

What did the PSMB1 experiments establish?

PSMB1 siRNA reduced cell growth, colony formation and motility-related outcomes in A549/H1299 culture experiments. It does not establish in-vivo therapeutic efficacy or a validated drug target in patients.

Evidence: KF5

Concepts & Entities

Disease And Stage

Lung adenocarcinomaAdenocarcinoma in situMinimally invasive adenocarcinoma

Data And Methods

Single-cell transcriptomicsinferCNV and pseudotimeRetrospective survival modeling

Molecular And Translation

Cluster 6LPASPSMB1 knockdown

Related Research

Novel post-translational modification learning signature reveals B4GALT2 as an immune exclusion regulator in lung adenocarcinoma
A distinct LUAD prognostic-expression and immune-microenvironment investigation focused on PTMLS and B4GALT2; it does not independently validate LPAS. · DOI: 10.1136/jitc-2024-010787
Advancing predictive markers in lung adenocarcinoma: A machine learning-based immunotherapy prognostic prediction signature
Another transcriptomic LUAD immunotherapy-prognosis modeling study, related by predictive intent but not a common prospective validation cohort. · DOI: 10.1002/tox.24284

Publication & Provenance

DOI: https://doi.org/10.1111/jcmm.18408
Doi Url: https://doi.org/10.1111/jcmm.18408
Publisher: https://onlinelibrary.wiley.com/doi/10.1111/jcmm.18408
Pmc Url: https://pmc.ncbi.nlm.nih.gov/articles/PMC11149493/
PMCID: PMC11149493
License: CC BY 4.0, as shown in article-specific PMC JATS permissions. This evidence page is original factual synthesis.
License Url: https://creativecommons.org/licenses/by/4.0/
Source Of Truth: Version-of-record article, J Cell Mol Med 2024, published online 4 June 2024; PMC11149493 JATS and Figure 2–11 legends in the supplied source pack.
Supporting Material: Article Table S1 was obtained; no synthetic patient aggregation is used.
Identity Note: Publisher/PMC author order is retained by existing publication master; JATS ORCID attribution for one coauthor conflicts with another related paper and is not used to revise structured identities.
Scientific Boundary: No inference of lineage tracing, prospective clinical decision value or immunotherapy benefit from computational association.

This author-maintained evidence summary separates source-located transcriptomic association, retrospective risk modeling, computational therapy hypotheses, and cell-line functional experiments. Cite the publisher article for scientific results.

Study-level evidence table

Study-level author-maintained evidence summary. Not participant-level data and not a ready-to-pool meta-analysis dataset. For formal citation, use the original article DOI.

Download study-level evidence CSV (UTF-8)

Formula safety: cells starting with =, +, -, @ (after leading whitespace), or tab/CR/LF are prefixed with an apostrophe. Treat all cells as text; remove only that safety prefix when recovering the original string.

Verify publication identity

These records identify the same original work; database inclusion is not evidence of scientific validity or clinical benefit.

PubMed 38837585 PMC PMC11149493 OpenAlex record

All Publications Homepage