EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LDHA, BIK, and CNIH4 Are Diagnostic Markers of Endoplasmic Reticulum Stress in Lung Cancer Comorbid With Sepsis: Integrating Machine Learning and Single‐Cell Analysis of Immune Signaling

Yifeng Pan iD, Xiaoru Liu iD, Jiawen Wang iD, Jing Yan iD

Mediators of Inflammation · 2026-01 · 2026 (1)

DOI: 10.1155/mi/2437852 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Mediators of Inflammation, online erschienen am 11.07.2026. Verfasst von Yifeng Pan und 3 weiteren Autor:innen. Beteiligte Einrichtung: The Eighth Clinical Medical College of Guangzhou University of Chinese Medicine, Foshan, 528000, Guangdong, China. Hauptthemen laut MEDLINE: Lung Neoplasms, Sepsis, Machine Learning, Endoplasmic Reticulum Stress. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Joint University-Affiliated Hospital Scientific and Technological Innovation Fund of Guangzhou University of Chinese Medicine.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Lung cancer is intricately associated with the onset of sepsis. Endoplasmic reticulum (ER) stress (ERS) is a cellular stress response to aberrant protein folding in the ER, closely associated with the cellular immune response. Currently, numerous research have elucidated the correlation between ERS and lung cancer, as well as sepsis. The mechanism of ERS in lung cancer comorbid with sepsis requires more investigation.

Objectives

This study aimed to investigate the interacting mechanisms between ERS and the immune response, explore prospective ERS‐related diagnostic biomarkers for lung cancer comorbid with sepsis, and elucidate their underlying pathological roles.

Methods

Datasets for lung cancer and sepsis were sourced from the Gene Expression Omnibus (GEO). Differentially expressed genes (DEGs) and weighted gene coexpression network analysis (WGCNA) modules were intersected with ERS‐related genes. Protein–protein interaction (PPI) and enrichment analyses were conducted. A genetic diagnostic model was developed using exhaustive machine learning algorithms, with accuracy assessed by receiver operating characteristic (ROC) curves and confusion matrices. Hub genes (area under the curve [AUC] ≥ 0.7) were analyzed for immune cell infiltration and cross‐validated using single‐cell RNA sequencing datasets. Crucially, the expression and functional roles of the hub genes were experimentally validated by western blot in clinical tissue cohorts (adjacent normal, lung cancer, and lung cancer with sepsis) and via wound healing and Transwell assays in PC9 lung cancer cells. Finally, prospective therapeutic agents were identified through molecular docking.

Results

Machine learning identified lactate dehydrogenase A (LDHA), Bcl‐2 interacting killer (BIK), and cornichon homolog 4 (CNIH4) as robust diagnostic biomarkers. Western blot analysis confirmed that the protein expression levels of LDHA, BIK, and CNIH4 were significantly upregulated in lung cancer and further elevated in the lung cancer comorbid with sepsis group. In vitro functional assays demonstrated that silencing these genes significantly inhibited the migration and invasion capabilities of PC9 cells. Single‐cell analysis revealed that these markers exhibit cell‐type‐specific expression in malignant cells and regulate immune dysregulation, particularly correlating with the functions of plasma cells and monocytes. Molecular docking indicated that tetrahydro‐NAD and amikacin are promising therapeutic candidates.

Conclusions

We identified and experimentally validated LDHA, BIK, and CNIH4 as specific ERS‐associated diagnostic biomarkers for lung cancer comorbid with sepsis. These markers drive tumor progression and modulate cellular immune responses, providing novel insights and therapeutic targets for this comorbidity.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Lung NeoplasmsSepsisMachine LearningEndoplasmic Reticulum StressHumansProtein Interaction MapsSignal Transduction

Stichwörter der Autor:innen

endoplasmic reticulum stress diagnostic markersimmune cell infiltrationlung cancersepsis

Autor:innen und Institutionen

  • The Eighth Clinical Medical College of Guangzhou University of Chinese Medicine, Foshan, 528000, Guangdong, China.
  • Foshan Hospital of Traditional Chinese Medicine, Foshan, 528000, Guangdong, China, foshan.gov.cn.
  • Graduate School of Guangxi University of Chinese Medicine, Nanning, 530200, Guangxi, China.
  • The First Affiliated Hospital of Guangxi University of Chinese Medicine, Nanning, 530200, Guangxi, China, gxtcm.com.

ORCID-Profile: Yifeng Pan, Xiaoru Liu, Jiawen Wang, Jing Yan

Förderung

  • Joint University-Affiliated Hospital Scientific and Technological Innovation Fund of Guangzhou University of Chinese Medicine

Zitieren

APA

Pan, Y., Liu, X., Wang, J., & Yan, J. (2026). LDHA, BIK, and CNIH4 Are Diagnostic Markers of Endoplasmic Reticulum Stress in Lung Cancer Comorbid With Sepsis: Integrating Machine Learning and Single‐Cell Analysis of Immune Signaling. Mediators of Inflammation, 2026(1). https://doi.org/10.1155/mi/2437852

Vancouver

Pan Y, Liu X, Wang J, Yan J. LDHA, BIK, and CNIH4 Are Diagnostic Markers of Endoplasmic Reticulum Stress in Lung Cancer Comorbid With Sepsis: Integrating Machine Learning and Single‐Cell Analysis of Immune Signaling. Mediators of Inflammation. 2026;2026(1). doi:10.1155/mi/2437852

BibTeX
@article{Pan2026,
  author = {Pan, Yifeng and Liu, Xiaoru and Wang, Jiawen and Yan, Jing},
  title = {LDHA, BIK, and CNIH4 Are Diagnostic Markers of Endoplasmic Reticulum Stress in Lung Cancer Comorbid With Sepsis: Integrating Machine Learning and Single‐Cell Analysis of Immune Signaling},
  journal = {Mediators of Inflammation},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/mi/2437852}
}
Kostenlos starten →