J Leukoc Biol. 2026 Jun 8;118(6):qiag062. doi: 10.1093/jleuko/qiag062.
ABSTRACT
The immunotherapy represents the most promising new cancer treatment approach but due to the high heterogeneity of the tumor microenvironment among patients, only a small portion of patients respond to a certain immunotherapy. Therefore, identifying the immune-related subtypes and immune molecular characteristics is of great guiding significance for improving the response and results of immunotherapy and personalized treatment of cancer. This study aims to identify immune subtypes across various cancers based on immune cell infiltration profiles and assess their clinical significance. We analyzed immune cell infiltration data from The Cancer Genome Atlas (TCGA) using Tumor IMmune Estimation Resource (TIMER), Microenvironment Cell Populations-counter (MCP), and Estimate the Proportion of Immune and Cancer cells (EPIC) algorithms to ensure consistency. Clustering analysis categorized 9,648 tumor samples into 5 immune subtypes (I to V) based on their infiltration profiles. We evaluated the relationship between these subtypes and patient survival, tumor characteristics, and genetic mutations. Five distinct immune subtypes were identified as cold tumors with low immune infiltration and hot tumors with high immune infiltration. Survival analysis revealed significant differences, with hot tumors associated with better prognosis except subtype IV. Mutational analysis identified isocitrate dehydrogenase 1 (IDH1) mutations as a significant prognostic biomarker for cold tumors, accounting for 85.3% of mutations in this subtype. We also identified 35 differentially expressed genes, leading to the construction of classifiers with high accuracy for predicting treatment responses. These findings highlight the heterogeneity of immune cell infiltration in tumors and its impact on patient prognosis and treatment sensitivity. The findings suggest that immune subtypes can serve as valuable biomarkers, particularly IDH1 mutations in cold tumors, paving the way for personalized cancer therapies.
PMID:42367149 | DOI:10.1093/jleuko/qiag062