Integrated Multi-Omics and Machine Learning Reveal Immunogenic Cell Death-Related Signature for Colorectal Cancer Prognosis and Therapy

Strategic Insights -

Researchers at Suzhou University of Science and Technology have made significant strides in understanding and combating colorectal cancer (CRC) through the discovery of an immunogenic cell death-related signature. By employing an integrated multi-omics and machine learning approach, the team identified 11 genes with prognostic significance in CRC, which were used to construct a novel computational framework. This framework, known as the ICDRS, exhibited strong predictive performance for overall survival in CRC patients and revealed significant associations with distinct immune infiltration patterns, immunotherapy response, and tumor microenvironment characteristics.

Key Takeaways:

  • The research identified 11 ICD-related genes with prognostic significance in CRC through single-cell RNA sequencing and bulk transcriptomic data analysis.
  • A comprehensive computational framework was employed to evaluate 101 machine learning combinations and construct an optimized 11-gene ICD-related signature (ICDRS).
  • The ICDRS demonstrated strong predictive performance for overall survival in CRC patients, achieving outstanding time-dependent AUCs (0.90) for 1- to 3-year survival prediction.
  • The study revealed significant associations between ICDRS-derived risk score and distinct immune infiltration patterns, immunotherapy response, and tumor microenvironment characteristics.
  • A novel macrophage subtype, SPP1+/SLC11A1+, was discovered and characterized by high infiltration levels, indicating its potential role in CRC progression.
  • The research suggested Olaparib as a potential therapeutic candidate for high-risk CRC patients through drug repurposing analysis.
  • The ICDRS-based nomogram could serve as a promising tool for CRC prognosis and immunotherapy, guiding personalized treatment strategies.

Statistics:

  • 11 ICD-related genes were identified with prognostic significance in CRC.
  • The ICDRS exhibited a predictive performance consistency of 0.90 for 1- to 3-year survival prediction.
  • 101 machine learning combinations were evaluated to construct the ICDRS.
  • 3 years of survival prediction were achieved with outstanding time-dependent AUCs.
  • 15% of patients had high levels of SPP1+/SLC11A1+ macrophage subtype infiltration.

Sources:

  • Integrated multi-omics and machine learning reveal an immunogenic cell death-related signature for prognostic stratification and therapeutic optimization in colorectal cancer. Frontiers in Immunology, 2025,16.
  • Suzhou University of Science and Technology Researchers Have Provided New Data on Personalized Medicine (Integrated multi-omics and machine learning reveal an immunogenic cell death-related signature for prognostic stratification and ...). Immunotherapy Weekly. July 30, 2025; p 5250.