JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Construction and validation of immune prognosis model for lung adenocarcinoma based on machine learning

  • 0Department of Cardiothoracic Surgery, Affiliated Jinhua Hospital, Zhejiang University School of Medicine, Jinhua, China.
Frontiers in Oncology +

|

August 6, 2025

|

August 6, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

This study developed a new prognostic model for lung adenocarcinoma (LUAD) using immune-related genes. The model identifies key biomarkers to improve personalized diagnosis and treatment strategies for LUAD patients.

Area Of Science

  • Oncology
  • Genomics
  • Immunology

Background

  • Lung adenocarcinoma (LUAD) is a major cause of cancer mortality with frequent recurrence and metastasis.
  • Identifying reliable prognostic biomarkers is crucial for enhancing patient outcomes in LUAD.

Purpose Of The Study

  • To construct and validate a robust immune-related prognostic model for LUAD.
  • To identify key hub genes associated with immune infiltration and patient survival.
  • To explore immune subtypes within LUAD for personalized treatment strategies.

Main Methods

  • Transcriptomic data from TCGA and an external cohort were analyzed.
  • Weighted Gene Co-expression Network Analysis (WGCNA) integrated differentially expressed and immune-related genes.
  • Machine learning algorithms (Random Forest, LASSO, SVM-RFE) identified hub genes.
  • Multivariate Cox regression, ROC, and ANN models assessed prognostic performance.
  • Immune infiltration and subtype analyses were performed using TIMER, ssGSEA, and consensus clustering.

Main Results

  • A prognostic model was built using four hub genes: CBLC, GDF10, LTBP4, and FABP4.
  • The model demonstrated strong predictive accuracy in internal and external validation.
  • Elevated levels of CD4+ T cells, macrophages, and dendritic cells were observed in LUAD.
  • Two distinct immune subtypes with differing prognoses and immune landscapes were identified.

Conclusions

  • A validated, immune-related prognostic model for LUAD was established.
  • Key biomarkers correlating with immune infiltration and survival were identified.
  • The findings provide a foundation for personalized diagnostic and therapeutic approaches in LUAD.

Keywords: