Transcriptome mapping of renal clear cell carcinoma revealed by machine learning algorithm based on enhanced computed...

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Inflammation-related components in the tumor microenvironment (TME) significantly influence cancer patient outcomes.
• Radiomics offers potential for predicting survival and prognosis in cancer patients.

Purpose of the Study:

• To systematically analyze inflammation-related genes (IRGs) in clear cell renal cell carcinoma (ccRCC).
• To assess the relationship between differentially expressed IRGs (DEIRGs) and inflammation.
• To validate the prognostic value of IRGs and radiomics in ccRCC.

Main Methods:

• Systematic analysis of IRGs in ccRCC using TCGA and GEO datasets.
• Construction and validation of an IRG-related risk score.
• Extraction of radiomics signatures from CT images for prognostic modeling.

Main Results:

• Prognostic IRGs correlate positively with specific inflammatory cells (CD8+ T cells, MDSCs, neutrophils) in the TME.
• A validated risk signature based on DEIRGs demonstrated good prognostic assessment.
• Radiomics-based models outperformed risk signatures and clinical characteristics in predicting ccRCC prognosis.

Conclusions:

• IRG-related risk scores are crucial for ccRCC prognosis assessment and patient management.
• IRGs can predict immune cell infiltration within the TME.
• Non-invasive radiomics signatures effectively predict ccRCC prognosis.