Multimodal data fusion AI model uncovers tumor microenvironment immunotyping heterogeneity and enhanced risk stratification of breast cancer

  • 0Guangdong Provincial Key Laboratory of Malignant Tumor Epigenetics and Gene Regulation, Department of Medical Oncology Breast Tumor Centre, Phase I Clinical Trial Centre, Sun Yat-sen Memorial Hospital, Sun Yat-sen University Guangzhou China.

|

|

Summary

This summary is machine-generated.

This study reveals four immune-metabolic subtypes in breast cancer using long noncoding RNAs (lncRNAs) and AI. These findings enhance prediction of immunotherapy response and patient prognosis.

Area Of Science

  • Oncology
  • Genomics
  • Bioinformatics

Background

  • Breast cancer is a leading cause of death in women, often involving recurrence and metastasis.
  • Understanding factors influencing immunotherapy response is crucial for improving survival rates.

Purpose Of The Study

  • To investigate the role of long noncoding RNAs (lncRNAs) in breast cancer immunotherapy response.
  • To develop AI-driven models for predicting immune-metabolic subtypes and patient prognosis.

Main Methods

  • Analysis of RNA sequencing and pathology data from 1027 breast cancer patients.
  • Unsupervised clustering to identify lncRNA expression patterns.
  • Development of AI-based pathology and multimodal models for subtype prediction and prognostication.

Main Results

  • Identification of four distinct immune-metabolic subtypes.
  • AI models demonstrated high accuracy in predicting subtypes and prognostic performance.
  • lncRNAs significantly impact antitumor immunity and metabolic states in the tumor microenvironment.

Conclusions

  • AI models, DeepClinMed-IM and DeepClinMed-PGM, offer robust tools for precise breast cancer prognostication.
  • These models can aid in identifying suitable candidates for immunotherapy.
  • The study advances breast cancer research and treatment strategies through AI and lncRNA analysis.