Transcriptome-Derived Immune Activity in Breast Cancer Reveals Prognostic CCL19–IL10 Expression States with External Evaluation in SCAN-B
Seyed Mostafa Rahimi,1Fatemeh Arab Mirrahmani,2Mohsen Najafi,3,*
1. Cellular and Molecular Biology Research Center, Health Institute, Babol University of Medical Sciences 2. Department of Human Genetics, Negin Genetics & Pathobiology Laboratory, Negin Medical Complex 3. Medical Genomics Research Center, Tehran Medical Sciences Islamic Azad University
Introduction: Breast-cancer immune landscapes are heterogeneous, and broad measures of immune infiltration may not fully capture prognostically relevant immune states. We investigated whether transcriptome-derived immune activity could identify biologically and clinically informative patterns.
Methods: RNA-seq data from 1,106 TCGA-BRCA primary tumors were analyzed using 17 prespecified ssGSEA signatures, ESTIMATE, MCP-counter, and xCell. Principal-component analysis generated a continuous immune-activity axis divided into Low, Mid, and High tertiles. Differential expression, functional enrichment, and an immune-focused STRING/CytoHubba network were used to prioritize candidate genes. Prognostic associations of hub genes and joint CCL19–IL10 expression states were evaluated in TCGA and externally assessed in SCAN-B/GSE96058.
Results: PC1 explained 48.1% of signature variance and captured marked differences in immune/stromal activity and tumor purity. Immune-High versus Immune-Low tumors showed 2,830 differentially expressed genes and 19 FDR-significant Hallmark pathways, all enriched toward Immune-High. However, immune-activity tertiles did not significantly stratify TCGA overall survival (P=0.241). CCL19 and IL10 showed opposing adjusted associations (HR=0.592 and 1.653, respectively). The Low-CCL19/High-IL10 state was associated with poorer survival in TCGA (HR=2.445, P=0.001). In SCAN-B (n=3,069), the favorable CCL19 association and unadjusted joint-state pattern were reproduced, whereas IL10 alone and the adjusted state contrast were not significant.
Conclusion: Global immune activity characterized breast-cancer TME biology but was not independently prognostic. Specific immune-network expression states, particularly involving CCL19 and IL10, warrant further mechanistic and prospective evaluation.
Keywords: Breast Cancer; Tumor Microenvironment; Transcriptomics; CCL19; IL10
به خانواده بزرگ کنسر ژنتیکس و ژنومیکس سرطان بپیوندید!