Situation
A biotechnology company investigating novel mechanisms of metastasis sought to determine the role of a rare “hybrid” cell population across multiple tumor types. Prior studies in colorectal cancer had suggested these cells may facilitate immune evasion and tumor spread, but it was unclear whether the same phenomenon extended beyond the colorectal setting.
The company’s central questions were:
- Do these cells exist in other solid tumors?
- If so, do they share a conserved transcriptional program that could inform therapeutic targeting?
Given the complexity of integrating heterogeneous single-cell RNA sequencing (scRNA-seq) datasets across cancer types, the company turned to Bridge Informatics (BI) for a rigorous, multi-phase analysis.
Strategy
We designed a pan-cancer discovery framework combining differential expression, pathway analysis, and interaction modeling to identify conserved molecular programs of this rare cell population. This was our approach:
Cross-Cancer Signature Analysis
- Curated scRNA-seq datasets from colon, pancreatic, lung, and breast cancers.
- Applied standardized quality-control pipelines to harmonize data across sources.
- Performed pseudo-bulk differential gene expression analyses to detect consistently dysregulated genes in the target cells across tumor types.
Pathway and Network Enrichment
- Focused on 135 candidate genes identified in colorectal cancer.
- Used KEGG, Reactome, and WikiPathways to identify shared biological programs.
- Highlighted recurrent features including:
- WNT signaling and epithelial-to-mesenchymal transition (EMT),
- ferroptosis resistance pathways,
- and chemokine ligand expression relevant to immune evasion.
Ligand–Receptor and Transition Modeling
- Applied CellChat to map signaling interactions between hybrid cells and their microenvironment.
Used Slingshot trajectory inference to explore lineage transitions and identify unique differentiation pathways specific to the hybrid phenotype.
Upon completion of the project, we delivered the following:
- High-resolution network diagrams showing signaling interactions.
- Dot plots and heatmaps illustrating conserved transcriptional signatures.
- Annotated gene lists prioritized by cross-cancer relevance.
- A summary report contextualizing results for translational and therapeutic planning.
Results
This project demonstrated how systematic cross-cancer mining and integrative pathway analysis can reveal conserved metastatic mechanisms overlooked by single-disease studies.
Pan-Cancer Signature Validated:
A conserved gene program was detected across colon, pancreatic, lung, and breast cancer datasets—providing evidence that this rare cell population is not tumor-type specific, but rather a generalizable feature of metastatic biology.
Targetable Pathways Highlighted:
Several genes and pathways emerged as actionable therapeutic candidates, including those tied to EMT, ferroptosis resistance, and immune evasion signaling.
Actionable Next Steps Enabled:
- Prioritize preclinical studies exploring inhibitors of EMT and WNT pathways,
- Develop hypotheses for pan-cancer therapeutic strategies,
- and build a foundation for target validation across multiple tumor indications.
Bridge Informatics delivered not just results, but a scalable framework for pan-cancer target discovery—empowering the client to accelerate therapeutic development with clear, biologically grounded leads.
Ready to turn complex single-cell data into real-world oncology insights? Let’s build your next discovery strategy together.Click here to schedule a free introductory call with a member of our team.