Inflammatory bowel disease (IBD) is a group of chronic conditions that cause ongoing inflammation in the digestive tract. It mainly encompasses two diseases: Crohn's disease (CD), which can affect any part of the gut but most commonly impacts the end of the small intestine and the colon, and ulcerative colitis (UC), which is confined to the colon and rectum. Worldwide, approximately ten million people live with IBD.
One of the most serious long-term risks for people with IBD is that the chronic inflammation significantly increases the likelihood of developing colorectal cancer (CRC). This risk is particularly elevated for patients who have had IBD for many years, who have extensive colon involvement, or who have a family history of CRC. Regular colonoscopy surveillance is therefore an essential part of IBD management.
Despite this well-established clinical connection, the precise molecular mechanisms by which chronic intestinal inflammation triggers malignant transformation are not fully understood. What specific immune cells and gene expression changes mediate the transition from inflamed but non-cancerous tissue to cancerous tissue? Answering this question could reveal biomarkers for earlier cancer detection and new therapeutic targets.
This study tackled this question using a uniquely comprehensive approach: combining multi-omics technologies - bulk RNA sequencing, single-cell RNA sequencing, and spatial transcriptomics - together with clinical data from 1,000 patients to map the molecular landscape of IBD and its progression toward CRC in unprecedented detail.
The researchers assembled an unprecedented dataset combining three complementary types of RNA analysis. Bulk RNA sequencing of 415 samples measured average gene activity across entire tissue sections. Single-cell RNA sequencing (scRNA-seq) of 18 samples analyzed gene expression in nearly 60,000 individual cells, revealing which specific cell types are most active. Spatial transcriptomics of 3 colorectal cancer tissue sections mapped where different cell populations are physically located within the tumor architecture.
These molecular analyses were complemented by a large clinical cohort study of 1,000 patients: 735 with IBD (320 Crohn's disease, 415 ulcerative colitis) and 265 with colorectal cancer across all clinical stages. In this group, blood levels of carcinoembryonic antigen (CEA) - a protein produced by many cancers and elevated in some inflammatory conditions - were measured to assess its diagnostic value for distinguishing CRC from IBD.
The bulk RNA sequencing data came from four publicly available datasets from multiple institutions, representing 90 healthy control samples alongside hundreds of IBD tissue samples from various intestinal regions. This allowed comprehensive identification of genes that are consistently altered across different IBD subtypes and anatomical locations.
For single-cell analysis, the team used sophisticated computational tools including UMAP visualization (which reduces complex data to two dimensions to reveal cell clusters), pseudotime trajectory analysis (which orders cells along a developmental timeline to infer biological progression), and cell-cell communication analysis to map how different cell types signal to each other in diseased versus healthy tissue.
Single-cell RNA sequencing of nearly 60,000 cells identified 11 distinct cell types in IBD tissues, including epithelial cells, multiple T cell subtypes, B cell subtypes, tissue stem cells, macrophages, monocytes, neutrophils, NK cells, and dendritic cells. This resolution allowed the researchers to see exactly which cell populations change in diseased versus healthy intestinal tissue.
In Crohn's disease, epithelial cells, B cells, and tissue stem cells were significantly reduced compared to healthy tissue, while T cells, macrophages, monocytes, and neutrophils were elevated. This shift reflects the intense immune activation that characterizes active CD - the tissue becomes dominated by inflammatory immune cells at the expense of the normal epithelial lining.
Further analysis of T cell subtypes revealed important shifts in CD4+ and CD8+ T cell populations between healthy, CD, and UC tissues. These immune cell population changes provide insight into why some IBD patients are more susceptible to progression toward cancer - certain immune profiles may be less effective at clearing abnormal cells before they become cancerous.
The increase in tissue stem cells in diseased tissues was a notable finding with potential implications for cancer risk. Stem cells are responsible for regenerating the intestinal lining after damage, but their dysregulation can also contribute to abnormal cell growth that precedes cancer development.
Among all the genes found to be altered across both IBD and CRC datasets, CEACAM5 (the gene that encodes the carcinoembryonic antigen protein, CEA) consistently stood out as one of the most significantly and consistently upregulated across Crohn's disease and ulcerative colitis samples. This places CEACAM5 at the molecular intersection of inflammation and cancer.
Pseudotime trajectory analysis - a computational method that orders individual cells along a developmental timeline without needing to observe them over real time - revealed that CEACAM5 expression increases progressively during the later stages of epithelial cell differentiation. This suggests CEACAM5 is not just a passive marker of disease but is actively involved in how epithelial cells respond to chronic inflammatory stress.
Spatial transcriptomics mapping of colorectal cancer tissues confirmed that CEACAM5 is specifically and highly expressed in the epithelial regions of tumor tissue, with a localization pattern consistent with its role in the cancer cells themselves rather than in the surrounding immune cells or stroma. This spatial specificity strengthens its value as a cancer-specific marker.
Analysis of the TCGA cancer genome database confirmed that CEACAM5 expression is significantly elevated in colorectal cancer tissue compared to adjacent normal tissue, and that higher expression correlates with worse patient outcomes. Together, these multiple lines of evidence establish CEACAM5 as a key molecular link between IBD-associated inflammation and cancer progression.
In the clinical arm of the study, serum CEA levels - the blood test version of CEACAM5 protein - were measured in 1,000 patients. This provided a direct test of whether the molecular findings translated to a clinically useful difference in blood tests between the IBD and CRC groups.
CRC patients had significantly higher CEA levels than IBD patients (p less than 0.05). This difference was large enough to have diagnostic value: CEA measurement could help clinicians identify which patients with gastrointestinal complaints might have cancer rather than inflammatory disease alone.
Interestingly, within the IBD group, Crohn's disease patients had higher CEA levels than ulcerative colitis patients. This is consistent with the known higher cancer risk associated with CD in certain locations and supports the idea that CEA elevation tracks with the degree of mucosal inflammation and epithelial disruption.
The finding that a simple, widely available blood test - CEA measurement - can discriminate between CRC and IBD has immediate practical implications for clinicians managing patients with ambiguous presentations or symptoms that overlap between the two conditions. Combined with other clinical information, CEA levels could help guide decisions about whether to pursue more invasive investigations like colonoscopy.
Using the CellChat computational tool, the researchers mapped cell-to-cell communication networks in healthy versus diseased intestinal tissue. This analysis identified key signaling pathways through which different cell types interact - revealing how the inflammatory microenvironment in IBD rewires intercellular communication in ways that may predispose the tissue to cancer.
The transition from IBD to CRC involves not just changes in individual cells but a comprehensive remodeling of the entire tissue ecosystem. IBD-associated immune cells communicate with epithelial cells in ways that promote epithelial regeneration - a necessary response to inflammation-induced damage. However, this same regenerative signaling, if sustained, can eventually push cells toward uncontrolled growth.
The spatial transcriptomics data added crucial context by showing where these cellular changes occur within the tissue. Rather than all cells being equally affected, specific anatomical regions within IBD and CRC tissues showed concentrated signals of particular cell populations, suggesting the disease process is spatially organized rather than uniform across the tissue.
Together, these findings support a model where chronic IBD-driven inflammation progressively alters the intestinal epithelium through sustained CEACAM5 upregulation and epithelial-stem cell dysregulation, eventually reaching a tipping point where cells acquire cancer-like properties. Understanding these sequential molecular events could enable earlier intervention before the transition to overt CRC.
This study establishes CEACAM5 (CEA) as a key molecular marker that links epithelial activation in IBD to tumorigenic processes in colorectal cancer. The convergent evidence from bulk RNA-seq, single-cell analysis, spatial transcriptomics, and clinical blood testing all point to CEACAM5 as a robust biomarker for the IBD-CRC transition.
From a diagnostic standpoint, CEA blood measurement already exists as a clinical test and is approved for monitoring CRC treatment response. This study's findings support its use in distinguishing CRC from IBD and potentially for monitoring IBD patients who are at elevated cancer risk, where serial CEA measurements might help detect malignant transformation earlier.
The comprehensive multi-omics approach demonstrated that combining technologies provides much richer insights than any single method alone. The spatial context provided by spatial transcriptomics - knowing not just what genes are active but where in the tissue they are active - added a dimension of understanding that bulk or even single-cell analysis cannot capture.
Future directions include using these molecular signatures to develop better risk stratification tools for IBD patients - identifying which individuals need more intensive colonoscopy surveillance and which might benefit from preventive interventions to reduce their CRC risk before cancer develops.