JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Integrative multi-omic and machine learning approach for prognostic stratification and therapeutic targeting in lung squamous cell carcinoma

  • 0Department of Thoracic Surgery, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China.
Biofactors +

|

October 11, 2024

|

October 11, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

This study developed a new multi-omics signature (LMS) to predict prognosis and immunotherapy response in lung squamous cell carcinoma (LUSC). The LMS model improves patient stratification and identifies potential therapeutic targets for LUSC.

Area Of Science

  • Oncology
  • Bioinformatics
  • Genomics

Background

  • Lung squamous cell carcinoma (LUSC) presents challenges in treatment due to cancer cell proliferation, metastasis, and drug resistance.
  • Optimal predictive models for LUSC prognosis and targeted therapy selection are currently lacking.
  • Multi-omic data offers a unique perspective for understanding cancer biology and identifying prognostic indicators.

Purpose Of The Study

  • To integrate multi-omic data for LUSC patient classification.
  • To develop an optimal machine learning model for predicting prognosis and guiding therapy.
  • To explore the tumor microenvironment and immunotherapy response in different risk groups.

Main Methods

  • Integration of gene expression, DNA methylation, genomic mutations, and clinical data from LUSC patients.
  • Application of 10 consensus clustering algorithms to identify subtypes.
  • Development of a machine learning model (LUSC multi-omics signature - LMS) using 7 key genes.

Main Results

  • Identification of two prognostically relevant LUSC subtypes, with CS1 showing a better prognosis.
  • The LMS score demonstrated superior predictive performance compared to existing LUSC biomarkers.
  • Lower LMS scores correlated with higher overall survival and improved immunotherapy response.
  • High LMS group characterized by 'cold' tumors, suggesting potential benefit from drugs like dasatinib.
  • In vitro validation of SERPINB13 as a potential oncogene and therapeutic target.

Conclusions

  • The developed LMS model refines molecular classification of LUSC.
  • The LMS score aids in predicting patient prognosis and immunotherapy response.
  • Findings provide insights for optimizing LUSC treatment strategies and immunotherapy.

Keywords: