Low CEBPD Expression Tied to Poor Breast Cancer Survival, Study Finds

A gene tied to fat cell formation may hold important clues about breast cancer outcomes. In a study published in Cell Division, researchers report that CEBPD, a transcription factor involved in adipogenesis and inflammation, is expressed at significantly lower levels in breast tumors than in normal tissue, and that patients with reduced CEBPD show poorer survival. The team, led by Shuang Qiu and Jianwei Sun, combined large-scale bioinformatic analysis of public datasets with laboratory experiments in breast cancer cell lines to build their case.

The investigation began with the TCGA-BRCA dataset, which provided mRNA sequencing data and clinical information for 1,109 breast cancer tumor samples and 113 adjacent noncancerous samples. From the Molecular Signatures Database, the researchers collected a set of adipogenesis-related genes, or ARGs, and used the limma software package to identify which of these genes were differentially expressed between tumor and normal tissue. The criteria were strict: an adjusted p-value below 0.05 and an absolute log fold change of at least 0.5. The screen revealed 26 significantly upregulated ARGs and 57 significantly downregulated ARGs in breast tumors. A Gene Set Enrichment Analysis reinforced the picture, showing a negative normalized enrichment score of minus 2.126 for adipogenesis pathways, with adjusted p-values and false discovery rates both below 0.001, indicating that adipogenesis-related programs are broadly downregulated in tumor tissue.

To find which of these altered genes mattered most for patients, the team constructed a protein-protein interaction network using the STRING database, focusing on interactions with confidence scores above 0.7, and ranked genes by degree centrality with the CytoHubba plugin in Cytoscape. A Venn diagram analysis then identified CEBPD as a central gene intersecting both the prognostic gene set and the interaction network. The choice was biologically motivated as well: CEBPD encodes CCAAT/enhancer-binding protein delta, a transcription factor known to regulate inflammatory responses and adipogenesis, two processes implicated in the crosstalk between cancer cells and their surrounding stroma.

The expression and survival data were striking. CEBPD levels were significantly lower in breast cancer tumor tissues than in normal tissues, both in the TCGA-BRCA dataset and in an independent analysis using the TNMplot tool, with p-values below 0.001. When patients were split into high and low CEBPD groups at the median expression value, those with reduced CEBPD showed lower overall survival, shorter disease-specific survival, and shorter progression-free intervals. Crucially, univariate and multivariate Cox regression analyses indicated that CEBPD expression remained an independent prognostic indicator after accounting for clinicopathological parameters. The finding was corroborated in an external dataset, GSE20685, and downregulation was observed across pathologic T, N, and M stages, across disease stages I through IV, and in both White and Asian patient subgroups. HER2-positive tumors showed significantly lower CEBPD expression than HER2-negative tumors.

To translate these associations into a potentially usable tool, the researchers built a nomogram predicting one-year, three-year, and five-year survival probabilities based on CEBPD expression. Receiver operating characteristic analysis yielded an area under the curve of 0.705, which the authors characterize as a moderate ability to discriminate survival outcomes. Calibration curves aligned reasonably well with observed outcomes at one and three years, but deviated noticeably for five-year predictions, a limitation the authors flag explicitly. They note that validation in independent, prospective cohorts is essential before any clinical application, and that long-term predictions from the current model require cautious interpretation.

Beyond prognosis, the study probed what CEBPD might actually be doing in tumors. Comparing high and low CEBPD expression groups uncovered 7,197 differentially expressed genes, of which 5,257 were downregulated and 1,940 upregulated. Enrichment analysis of these genes pointed to involvement in the cell cycle, DNA replication, ubiquitin-mediated proteolysis, the VEGF signaling pathway, the PPAR signaling pathway, and complement and coagulation cascades. Gene Set Variation Analysis then quantified correlations between CEBPD expression and cancer-related processes: positive associations with angiogenesis (R = 0.19), apoptosis (R = 0.19), hypoxia (R = 0.28), and inflammation (R = 0.32), alongside negative correlations with the cell cycle (R = minus 0.24), DNA damage (R = minus 0.18), and DNA repair (R = minus 0.26). The authors propose that CEBPD-low tumors may therefore carry features of genomic instability and metabolic stress adaptation, while paradoxically, even as CEBPD suppresses tumor cells directly, it may foster a pro-inflammatory microenvironment.

The tumor immune microenvironment also differed by CEBPD status. Using the xCell, CIBERSORT, TIMER, MCPCOUNTER, and EPIC algorithms, the team found significant differences in immune cell infiltration between high and low CEBPD subgroups. CEBPD expression correlated positively with stroma and microenvironment scores, T cell NK populations, endothelial cells, effector and central memory CD4-positive T cells, and cancer-associated fibroblasts, while correlating negatively with common lymphoid progenitors, mast cells, and several other CD4-positive T cell subsets. CEBPD expression was also positively correlated with several immune checkpoint molecules, including PDCD1, VSIR, VTCN1, CD40, TNFRSF18, TNFSF9, and TNFRSF4, and negatively correlated with CD80 and TNFSF4. Single-cell RNA sequencing data from the TISCH repository, analyzed in the BRCA GSE114727 dataset, showed that CEBPD expression varied by cell type, with higher levels in monocytes, macrophages, dendritic cells, NK cells, and neutrophils, and lower levels in B cells and T cells. The authors suggest this pattern could eventually help stratify patients for immunotherapy, though they stress this remains a hypothesis for future testing.

The most direct evidence came from the laboratory. The team obtained ten tumor specimens and ten matched adjacent non-neoplastic tissues from patients at the First People’s Hospital of Yunnan Province, all of whom had not received radiotherapy or chemotherapy before surgery, and confirmed by quantitative reverse transcription PCR that CEBPD was downregulated in the tumors, consistent with the bioinformatic findings. In cell lines, CEBPD mRNA was abundant in the normal breast epithelial line HBL-100 but markedly reduced in the breast cancer lines MCF-7, BT-474, and HCC1806. The researchers then created stable CEBPD-overexpressing versions of MCF-7 and BT-474 cells using a lentiviral vector carrying the full-length human CEBPD cDNA, confirming successful overexpression at both the mRNA and protein levels by PCR and Western blot.

Functional assays using the CCK-8 proliferation kit showed that enforced CEBPD expression significantly inhibited cell proliferation in both MCF-7 and BT-474 cells compared with empty-vector controls over a 48-hour period, with measurements taken at 0, 12, 24, and 48 hours and each experiment performed in triplicate and repeated three times independently. These results provide direct experimental support for a tumor-suppressive role of CEBPD in breast cancer cells, though the authors caution that their functional validation is confined to in vitro overexpression models. They call for loss-of-function studies in CEBPD-high cell lines and in vivo models to confirm the mechanism and to separate cell-intrinsic effects from microenvironmental ones.

The study also explored therapeutic possibilities, though only computationally. Screening the Connectivity Map database identified a compound labeled W-13 with a significantly negative connectivity score, suggesting a predicted inverse relationship with CEBPD expression, and molecular docking analysis using AutoDock provided a structural hypothesis for how W-13 might interact with the CEBPD protein. The authors emphasize that these findings derive solely from computational analyses and that any effect of W-13 on breast cancer cells with differing CEBPD levels remains to be experimentally validated. Taken together, the work positions CEBPD as an independent prognostic marker whose decreased expression correlates with poor outcomes in breast cancer, supported by cell-based evidence of tumor-suppressive function, while leaving open the mechanistic questions, prospective clinical validation, and therapeutic exploration that the authors themselves identify as the necessary next steps.

Subject of Research: The role of the adipogenesis-related gene CEBPD as a prognostic biomarker and tumor suppressor in breast cancer

Article Title: The adipogenesis-related gene CEBPD as a prognostic biomarker and functional tumor suppressor in breast cancer

Article References: Qiu, S., Chen, X., Yang, J., Sun, Y., Liu, Y., Wu, D., & Sun, J. (2026). The adipogenesis-related gene CEBPD as a prognostic biomarker and functional tumor suppressor in breast cancer. Cell Division, 21(1), Article 13. https://doi.org/10.1186/s13008-026-00183-3

Image Credits: AI Generated

DOI: 10.1186/s13008-026-00183-3

Keywords: CEBPD, breast cancer, prognostic biomarker, tumor suppressor, adipogenesis, TCGA-BRCA, tumor microenvironment, immune infiltration, nomogram, gene expression, cell proliferation, bioinformatics