An Immune Subtype Classification System Enables the Development of Strategies to Predict and Enhance Immunotherapy...

Area of Science:

• Oncology
• Immunology
• Genomics

Background:

• Colorectal cancer (CRC) is a leading global cause of cancer mortality.
• Current immunotherapies benefit only a subset of CRC patients, necessitating improved predictive models.
• Understanding tumor immune microenvironment heterogeneity is crucial for effective CRC treatment.

Purpose of the Study:

• To develop an immune subtype classification system for colorectal cancer.
• To create a machine learning model for predicting immunotherapy sensitivity in CRC patients.
• To identify novel therapeutic strategies to enhance CRC immunotherapy response.

Main Methods:

• Developed an immune subtype classification system using diverse immune cells and pathways.
• Constructed a machine learning model for predicting immunotherapy sensitivity.
• Validated the model using public datasets and clinical samples.
• Performed drug screening based on immune subtypes and conducted mechanistic investigations.

Main Results:

• The developed immune subtype classification system accurately predicted colorectal cancer immunotherapy sensitivity.
• The insulin-like growth factor 1 receptor inhibitor AG-538 was identified as a potent enhancer of antitumor immunity.
• AG-538 induced DNA damage and upregulated type I IFN signaling, increasing tumor immunogenicity.
• Combined AG-538 and anti-PD-1 antibodies demonstrated potential as a chemoimmunotherapeutic strategy.

Conclusions:

• The study presents a predictive model for immune response in colorectal cancer.
• Immune subtype classification facilitates personalized immunotherapy selection.
• AG-538 combined with anti-PD-1 antibodies offers a promising chemoimmunotherapeutic approach for colorectal cancer.
• These findings pave the way for personalized treatment strategies in colorectal cancer.