Dual PASNet predicts relevant markers and key pathways of LSCC
Xudan Zhou
School of Artificial Intelligent Medicine, Guilin Medical University
Qinglin Yang
Department of Digital Art, Guangxi International Business Vocational College
Yuxin Zhang
School of Artificial Intelligent Medicine, Guilin Medical University
Xiaoli Chen
School of Artificial Intelligent Medicine, Guilin Medical University
Jin Luo
School of Artificial Intelligent Medicine, Guilin Medical University
Guohui Ma
School of Artificial Intelligent Medicine, Guilin Medical University
Wei Shu
School of Artificial Intelligent Medicine, Guilin Medical University
DOI: https://doi.org/10.59429/esta.v13i1.13387
Keywords: lung squamous cell carcinoma; dual PASNet; prognostic signature; signaling pathway; tumor immune microenvironment; single-cell transcriptome
Abstract
To identify core prognostic markers and key pathways of lung squamous cell carcinoma (LSCC), this study adopted the interpretable Dual PASNet deep learning model with a dynamic pathway mask mechanism to screen survival-related pathways/genes, and systematically validated their prognostic value through multi-cohort verification, feature selection, and functional analysis. The model identified 10 survival-related core pathways in LSCC (P<0.01), with p53 signaling, cell cycle, and PI3K-Akt signaling pathways having the highest weights. Sixteen core genes showed significant expression differences between high- and low-risk groups in both TCGA-LUSC (internal) and GSE19804 (external) cohorts (*P<0.001), and 10 prognostic signature genes were further screened by LASSO regression. The 10-gene prognostic model exhibited robust risk stratification (Log-rank P=0.000765), with time-dependent ROC AUC of 0.709–0.780 at 6–60 months. Functional enrichment indicated significant enrichment in tumor malignant phenotype-related pathways (cell cycle, DNA replication). Immune analysis showed a close association with LSCC immunosuppressive microenvironment (more prominent in high-risk group), which was validated by single-cell transcriptome analysis (GSE131907) showing specific expression of signature genes in myeloid cells and T lymphocytes. This 10-gene prognostic signature has reliable prognostic predictive value for LSCC, reveals key molecular regulatory pathways and immune microenvironment characteristics, and provides potential molecular markers and a theoretical basis for precise prognostic evaluation and targeted therapy of LSCC.
References
[1] Siegel R L, Miller K D, Goding Sauer A, et al. Cancer Statistics, 2024[J]. CA Cancer J Clin, 2024, 74(1): 3-48.
[2] Zhang S, Li Y, Wang Y, et al. Explainable deep learning for cancer prognosis: A systematic review and metaanalysis[J]. Brief Bioinform, 2023, 24(3): bbac125.
[3] Wang H, Chen Y, Liu J, et al. GSE19804 dataset reanalysis: Identifying lung squamous cell carcinoma
prognostic biomarkers via integrated multi-omics[J]. Comput Biol Chem, 2022, 98: 107568.
[4] Tibshirani R, Hastie T, Friedman J H. Extended lasso methods for high-dimensional survival analysis[J]. Stat
Methods Med Res, 2022, 31(5): 985-1001.
[5] Chen L, Yang C, Zhang M, et al. Cox proportional hazards model with adaptive L1 regularization for lung
cancer prognostic signature construction[J]. BMC Bioinform, 2021, 22(1): 563.
[6] Kaplan E L, Meier P. Nonparametric estimation from incomplete observations—50 years later[J]. J R Stat Soc
Ser B (Methodol), 2021, 83(1): 3-19.
[7] Subramanian A, Tamayo P, Mootha V K. Gene set enrichment analysis (GSEA) 4.0: New features and
applications[J]. Nucleic Acids Res, 2023, 51(12): 6388-6396.
[8] Han X, Liu Y, Zhang J, et al. Single-cell RNA sequencing reveals myeloid cell heterogeneity and immune
suppression in lung squamous cell carcinoma[J]. Cell Rep, 2022, 38(10): 110765.
[9] Becht E, McInnes L, Hechtlinger Y, et al. UMAP for single-cell data visualization and interpretation[J]. Nat
Protoc, 2021, 16(11): 5419-5456.
[10] Vogelstein B, Lane D P, Levine A J. The p53 network in 2023: From basic biology to clinical applications[J].
Cell, 2023, 186(12): 2568-2586.
[11] Malumbres M, Barbacid M. Cell cycle kinases in lung cancer: Therapeutic opportunities and challenges[J].
Nat Rev Cancer, 2022, 22(8): 486-504.
[12] Fruman D A, Rommel C. PI3Kδ/γ inhibition in lung cancer: Targeting the tumor microenvironment[J].
Immunity, 2021, 54(7): 1388-1402.
[13] Cargnello M, Roux P P. MAPK signaling pathways in lung squamous cell carcinoma progression and therapy
resistance[J]. Semin Cancer Biol, 2023, 89: 157-170.
[14] Clevers H, Nusse R. Wnt signaling in cancer stem cells and metastasis[J]. Cell Res, 2021, 31(1): 26-46.
[15] Oppenheim J J, Yang D, Howard O M. Cytokine networks in lung cancer immune microenvironment[J].
Immunol Rev, 2022, 306(1): 123-138.
[16] Kanchanawong P. Focal adhesion signaling in lung cancer invasion and metastasis[J]. Essays Biochem,
2023, 67(3): 245-256.
[17] Zhao X, Wang Y, Zhang L, et al. Antifolate resistance mechanisms in lung squamous cell carcinoma:
Implications for precision therapy[J]. Drug Resist Updat, 2023, 64: 100932.
[18] Hynes R O. Integrins and ECM-receptor interactions in tumor progression[J]. Cold Spring Harb Perspect
Biol, 2022, 14(9): a040287.
[19] Lee D K, Cheng R, Carbone C. Apelin signaling in lung cancer: From proliferation to immune regulation[J].
Pharmacol Rev, 2023, 75(2): 312-336.