Spatial Omics Solutions for Cancer and Oncology Research
Integrate single-cell sequencing, spatial transcriptomics, and spatial proteomics to resolve tumor heterogeneity, immune niches, and cell-cell interactions in FFPE or fresh-frozen tissue. Build a study around the biological question, sample type, resolution, and evidence depth your oncology project requires.
Study-design-driven multi-omics integration
FFPE and fresh-frozen workflow options
Cell states, spatial niches, and protein evidence
Why Single-Method Approaches Are Not Enough for Cancer Research
Cancer develops within spatially organized ecosystems in which tumor, immune, stromal, and vascular cells form functional niches. Methods that dissociate tissue or average signals across a specimen can obscure these relationships. Three common evidence gaps shape oncology study design:
1
Bulk and single-cell methods lose spatial context
Bulk RNA-seq averages signal across mixed populations; scRNA-seq and snRNA-seq dissociate tissue, destroying the very cell-cell neighborhoods and niche boundaries that govern tumor behavior. Researchers can identify cell states but cannot determine which states co-localize in functionally relevant regions.
2
Transcript-level data alone cannot confirm protein phenotypes
RNA abundance does not reliably predict protein expression in tissue. Key immune checkpoint markers, signaling receptors, and functional phenotypes may be post-transcriptionally regulated β a gap that transcript-only profiling cannot close.
3
Without multi-layer spatial evidence, key features remain inferred
Tumor heterogeneity, immune niches, and cell-cell communication may otherwise depend heavily on computational inference. Spatial measurements provide direct positional evidence that can strengthen and refine these hypotheses.
Solution Overview: The Multi-Omics Approach to Cancer
CD Genomics provides an integrated spatial omics solution for cancer and oncology research that combines single-cell discovery, spatial mapping, and protein-level validation into a unified evidence framework. Core advantages:
Multi-layer evidence integration
Cell states identified by single-cell RNA-seq are mapped to tissue coordinates by spatial transcriptomics, then validated at the protein level by spatial proteomics β each layer addresses the limitation of the previous, reducing inference gaps.
Cross-cancer-type applicability
The same three-layer framework applies to breast, lung, colorectal, prostate, cervical, and other cancer types, with configurable panels and analysis routes. Both FFPE and fresh-frozen tissue are supported.
Configurable project routes
Choose a Discovery route (cell-state survey), a Mechanism route (multi-layer characterization with protein validation), or a Validation route (cohort replication with custom panels) β depending on your research stage and evidence depth requirements.
Validated analytical framework
Cell-state maps, spatial niche annotations, communication hypotheses, and protein measurements can be evaluated across complementary data layers with transparent analysis and quality-control steps.
The Three-Layer Technology Stack for Cancer Research
Each layer answers a different question; no single technology provides all three. Use the comparison table to decide which combination fits your project.
Layer 1
Single-Cell Discovery
scRNA-seq / snRNA-seq (10x Chromium, 10x Flex for FFPE)
Discovers cell states, subtypes, and rare populations at single-cell resolution. Produces a clustered cell-type map + marker gene panels + UMAP.
Best for: Initial discovery of cell populations and state transitions.
Limitation: No spatial context; tissue dissociation may alter expression.
Layer 2
Spatial Mapping
Spatial transcriptomics (10x Visium/Visium HD, Stereo-seq, 10x Xenium, Bruker CosMx SMI)
Maps transcript-defined cell states to tissue coordinates; resolves niche boundaries. Produces spatial gene expression map with niche annotation.
Best for: Spatial validation of scRNA-seq findings; niche and boundary analysis.
Limitation: Resolution varies by platform; protein validation recommended for key markers.
Layer 3
Protein Validation
Spatial proteomics (Bruker GeoMx DSP, Akoya PhenoCycler, mIHC)
Validates RNA-defined phenotypes at protein level in intact tissue; quantifies protein expression per niche. Produces protein expression overlay on spatial map.
Best for: Protein confirmation of immune markers, checkpoints, or signaling receptors.
Limitation: Limited panel coverage (targeted); complement with transcriptomics for discovery.
| Technology | Analytical Role | Key Output | Sample Suitability | When to Choose | Limitation |
|---|---|---|---|---|---|
| scRNA-seq / snRNA-seq (10x Chromium) |
Discover cell states, subtypes, rare populations at single-cell resolution | Clustered cell-type map + marker gene panels + UMAP | Fresh tissue (scRNA-seq); FFPE/frozen (snRNA-seq, 10x Flex) | Initial discovery of cell populations and state transitions | No spatial context; tissue dissociation may alter expression |
| Spatial transcriptomics (10x Visium/Visium HD, Stereo-seq, Xenium, CosMx SMI) |
Map transcript-defined cell states to tissue coordinates; resolve niche boundaries | Spatial gene expression map with niche annotation; cell-type deconvolution overlay | FFPE or fresh-frozen tissue sections | Spatial validation of scRNA-seq findings; niche and boundary analysis | Resolution varies by platform; protein validation recommended for key markers |
| Spatial proteomics (GeoMx DSP, PhenoCycler, mIHC) |
Validate RNA-defined phenotypes at protein level; quantify protein expression per niche | Protein expression overlay on spatial map; per-region protein quantification | FFPE tissue sections | Protein confirmation of immune markers, checkpoints, or signaling receptors | Limited panel coverage; complement with transcriptomics for discovery |
Integration logic: Single-cell data can provide a reference catalog of cell states. Spatial transcriptomics maps those states and expression programs back to tissue coordinates, while spatial proteomics supplies orthogonal evidence for selected protein phenotypes. The layers are modular: the appropriate combination depends on the question, tissue, and available material.
Demo: Representative Spatial Omics Outputs for Oncology
These example views illustrate how integrated oncology projects can organize cell-state discovery, tissue localization, neighborhood analysis, and communication hypotheses into interpretable visual outputs. Final figures vary with platform, tissue quality, cohort design, and analysis scope.
Multi-Modal Oncology Analysis Dashboard
A coordinated view of single-cell clusters, spatial tissue domains, marker expression, and selected protein evidence for reviewing findings across complementary data layers.
TumorβImmune Neighborhood Map
Example visualization of tumor regions, stromal boundaries, immune-rich niches, and local cellular composition within intact tissue.
Spatially Supported Communication Network
Candidate ligandβreceptor relationships prioritized using cell identity, expression patterns, tissue location, and neighborhood proximity.
Research Directions: Where to Apply Spatial Omics in Cancer
Start with the biological question. Each direction below highlights the spatial relationships, evidence layers, and downstream analyses most relevant to that oncology research goal.
TME
Tumor Microenvironment
The tumor microenvironment is the cellular and molecular ecosystem surrounding tumor cells β including immune cells, fibroblasts, vasculature, and extracellular matrix. Resolving TME composition requires spatial technologies that preserve tissue architecture.
- Multi-layer niche characterization: Immune niches defined by both transcript-level cell states and protein-level functional markers
- Spatial boundary resolution: Tumor-stroma interfaces mapped at tissue-region resolution
- Communication network mapping: Ligand-receptor interactions between spatially co-localized populations
Heterogeneity
Cancer Heterogeneity
Tumor heterogeneity β both intertumoral and intratumoral β drives therapy resistance, metastasis, and relapse. Spatial omics reveals whether different subclones occupy distinct spatial niches or intermingle.
- Subclone spatial mapping: Transcriptionally distinct subclones mapped to specific tumor regions
- Transition zone identification: Boundary regions between subclones identified at tissue resolution
- Phenotype-location correlation: Subclone identity correlated with local microenvironment composition
Progression
Tumor Progression
Tumor progression β from in situ to invasive, from primary to metastatic β involves spatial reorganization of both tumor cells and their microenvironment. Spatial omics captures this reorganization directly.
- Invasion front mapping: Cell-state transitions at the tumor-normal tissue boundary
- Progression-stage comparison: Serial sections reveal spatial reorganization of niches
- Metastatic niche characterization: Pre-metastatic and metastatic sites profiled
Morphology
Tumor Morphology
Tumor morphology reflects underlying molecular states, but the relationship is not always intuitive. Spatial omics bridges morphology and molecular profiling by overlaying gene expression on tissue structure.
- Morphology-molecule alignment: H&E annotations registered with spatial transcriptomics data
- Hidden heterogeneity detection: Molecular differences within morphologically identical regions
- Pathologist-informed analysis: Morphological annotations guide spatial region selection
Cancer-Type Applications
The multi-omics approach applies across cancer types. Each cancer type has distinct biology, dominant immune components, and clinical questions β the technology stack is configured accordingly.
| Cancer Type | Key Research Question | Recommended Technology Combination | Dedicated Solution |
|---|---|---|---|
| Breast cancer | Tumor subtype heterogeneity; immune niche composition; DCIS-to-invasive transition | Single-cell profiling + spatial discovery + targeted in situ or protein validation, selected according to tissue and study goal | Breast Cancer β |
| Lung cancer | Immune exclusion mechanisms; EGFR-TKI resistance microenvironment; tumor-stroma interface | Single-cell profiling + spatial transcriptomics + region- or cell-resolved immune phenotyping | Lung Cancer β |
| Colorectal cancer | Tumor heterogeneity across CMS subtypes; immune infiltration gradients; CAF-tumor interactions | Single-cell profiling + large-area spatial mapping + selected protein phenotyping | Colorectal Cancer β |
| Prostate cancer | Intratumoral heterogeneity; neuroendocrine transdifferentiation; bone microenvironment interactions | Single-cell or nuclei profiling + spatial transcriptomics + multiplex protein imaging | Prostate Cancer β |
| Cervical cancer | HPV-associated immune modulation; tumor progression from CIN; immune niche evolution | Single-cell profiling + spatial transcriptomics + region-resolved multi-omic profiling | Cervical Cancer β |
If your cancer type is not listed, the multi-omics framework is configurable β contact us to discuss your specific project.
Integrated Approach vs. Single-Method: What Changes
| Dimension | Single-Method Approach | CD Genomics Integrated Multi-Omics |
|---|---|---|
| Cell-state resolution | scRNA-seq identifies states but loses tissue context | Cell states discovered by scRNA-seq and mapped to tissue coordinates by spatial transcriptomics |
| Spatial information | Bulk assays provide no spatial context; conventional low-plex IHC is limited to a small marker set | Discovery-scale or targeted spatial assays can resolve expression patterns across tissue at platform-dependent coverage and resolution |
| Protein validation | May require a separate experiment and manual alignment with transcript data | Spatial proteomics provides orthogonal evidence for selected RNA-defined phenotypes in matched tissue context |
| Evidence depth | One measurement layer limits cross-modal interpretation | Complementary layers reduce inference gaps while retaining explicit computational assumptions and QC |
| Project coordination | Multiple vendors, inconsistent data formats, manual integration burden | Single project team, unified data standards, integrated analysis pipeline |
| Sample compatibility | Fresh-only for scRNA-seq; FFPE-only for IHC; no unified workflow | scRNA-seq (fresh), snRNA-seq/10x Flex (FFPE), spatial transcriptomics (FFPE and fresh), spatial proteomics (FFPE) β all from the same project |
Project Routes: Choose Your Evidence Depth
These routes are study-design starting points rather than fixed packages. Final platform selection, sample numbers, and deliverables are defined after tissue, feasibility, and analysis review.
Discovery Route
Initial exploration; limited samples; broad cell-state survey; identifying what cell types are present
Core Technologies
scRNA-seq or snRNA-seq + spatial transcriptomics (10x Visium or Stereo-seq)
Key Deliverables
Cell-type annotation + UMAP; spatial localization of major cell populations; preliminary niche identification report
Not Included
Protein validation; cohort-level statistics; custom panel design
Multi-Layer
Mechanism Route
Hypothesis-driven; needs multi-layer evidence at tissue scale; confirming protein-level phenotypes
Core Technologies
Discovery Route + spatial proteomics (GeoMx DSP, PhenoCycler, or mIHC) + integration analysis
Key Deliverables
Integrated niche characterization; protein-validated spatial map; cell-cell communication network; multi-layer evidence report
Not Included
Cohort replication; multi-sample statistical comparison
Validation Route
Translational validation; multi-sample cohort; publication-ready spatial signatures
Core Technologies
Mechanism Route + custom targeted panel (Xenium or CosMx) + cohort replication
Key Deliverables
Reproducible spatial signatures across samples; cohort-level statistical report; publication-ready figures and data
Not Included
Defined after cohort design and statistical review
How to choose: If you are asking "What cell types are in my tumor?" β Discovery. If you are asking "Do the immune niches I found at RNA level hold up at protein level?" β Mechanism. If you are asking "Is this spatial signature reproducible across my cohort?" β Validation.
Why Choose CD Genomics
Single-Cell + Spatial Integration Pipeline
Aligns single-cell reference states with Visium, Xenium, CosMx SMI, or Stereo-seq spatial measurements through a documented integration and QC workflow. Outputs can connect transcriptomic identity with tissue position while keeping mapping confidence and analytical assumptions visible.
Multi-Platform Spatial Proteomics
GeoMx DSP, PhenoCycler, and mIHC span region-level profiling and single-cell spatial phenotyping. Platform and panel selection are matched to the transcriptomic finding, tissue area, cohort size, and required plex.
FFPE-Compatible Workflow
FFPE-compatible options include 10x Flex for isolated cells or nuclei from qualified FFPE material, Visium/Visium HD, Xenium, CosMx SMI, GeoMx DSP, and multiplex protein imaging. Feasibility depends on tissue quality, fixation history, section area, and assay-specific QC.
Configurable Analysis Routes
Discovery, Mechanism, and Validation routes connect research questions with suitable evidence layers and expected outputs. The final scope is documented after sample review and technical consultation.
Cross-Cancer-Type Experience
The integration framework can be configured for breast, lung, colorectal, prostate, cervical, and other cancer studies, with cancer-type-specific marker selection and analysis parameters. Related service pages describe the platforms available for each modality.
Published Research Examples
The following published studies illustrate how multi-layer spatial omics approaches have been applied in oncology research. The summaries focus on study design, findings, and methodological relevance, and the accompanying illustrations are original conceptual visuals rather than reproductions or adaptations of published figures.
Example 1: Integrated Single-Cell, Spatial, and In Situ Analysis of Breast Cancer Heterogeneity
Janesick A, Shelansky R, Gottscho AD, et al. Nature Communications. 2023;14(1):8353. doi:10.1038/s41467-023-43458-x
Research Question
Can combining whole-transcriptome single-cell, whole-transcriptome spatial, and targeted in situ analysis reveal tumor heterogeneity that no single technology detects alone?
Study Design
FFPE breast cancer tissue sections were profiled with three technologies on serial sections: scFFPE-seq (10x Chromium Single Cell Gene Expression Flex) for whole-transcriptome single-cell profiling, Visium CytAssist for whole-transcriptome spatial profiling, and Xenium In Situ for subcellular-resolution targeted analysis of 313 genes. Two tumor blocks representing different breast cancer subtypes were analyzed.
Key Findings
- Xenium localized a small ERBB2+/ESR1+/PGR+ region that was not resolved in the scFFPE-seq data. Registration with Visium then enabled whole-transcriptome characterization of that region.
- A rare population of "boundary cells" co-expressing tumor and myoepithelial markers was identified at a deteriorating myoepithelial boundary.
- The study illustrates how targeted high-resolution mapping and whole-transcriptome measurements can provide complementary evidence.
Relevance to This Solution
This study demonstrates how single-cell reference data, whole-transcriptome spatial profiling, and targeted subcellular mapping can be integrated to characterize small or rare tissue regions. A similar design can be adapted after confirming tissue quality and assay compatibility.
Boundary
The study analyzed two breast cancer tumor blocks. The biological findings are context-specific and require validation in larger, independent cohorts.
Example 2: Single-Cell and Spatially Resolved Atlas of Human Breast Cancers
Wu SZ, Al-Eryani G, Roden DL, et al. Nature Genetics. 2021;53:1334-1347. doi:10.1038/s41588-021-00911-1
Research Question
What is the cellular architecture of breast cancer, and how are heterogeneous cell populations spatially organized within tumors?
Study Design
26 breast cancer tumors were profiled using scRNA-seq (10x Chromium) combined with CITE-seq for simultaneous surface protein measurement, and spatial transcriptomics (Visium) for tissue-level gene expression mapping. SCSubtype classification was developed to identify intrinsic subtype heterogeneity. Spatial deconvolution using single-cell signatures was applied to identify stromal-immune niches.
Key Findings
- Intrinsic subtype heterogeneity was common β multiple SCSubtypes co-existed within single tumors, revealing breast cancer as a mosaic of subclonal states.
- CITE-seq immune profiling identified a new PD-L1/PD-L2+ macrophage population associated with clinical outcome β a protein-level finding that transcript-only profiling would have missed.
- Spatial transcriptomics revealed that stromal-immune niches were spatially organized, not randomly distributed.
- Single-cell signatures were used to deconvolve larger breast cancer cohorts into nine ecotype groups with distinct cellular compositions and clinical associations.
Relevance to This Solution
This study demonstrates two complementary principles: CITE-seq added protein-level immune phenotyping, while spatial transcriptomics revealed organized stromal-immune niches. Separately, cohort deconvolution showed how single-cell signatures can support clinically relevant patient stratification.
Boundary
The study focused on breast cancer, and its ecotype and macrophage findings require validation in other cohorts and cancer types. Its Visium data used the spot-based generation available at the time and should not be generalized to Visium HD.
Frequently Asked Questions
Yes. FFPE-compatible options include 10x Flex for qualified FFPE-derived cells or nuclei, Visium/Visium HD, Xenium, CosMx SMI, GeoMx DSP, and multiplex protein imaging. Suitability still depends on fixation history, RNA quality, tissue area, morphology, and platform-specific QC. A pathology and feasibility review should precede final study design.
Platform choice balances gene coverage, spatial resolution, tissue area, sample preservation, cohort size, and whether the study is exploratory or targeted. Visium generations should be distinguished: classic spot-based workflows and Visium HD do not provide the same resolution. Xenium and CosMx SMI offer subcellular in situ measurements, while large-area platforms may be useful for broader tissue surveys. The best route is selected after reviewing the biological question and tissue.
This oncology page helps match a cancer research question to an integrated study route. The TME solution focuses specifically on immune niches, tumor-stroma boundaries, and spatially informed cell-cell communication. Use this page for broad oncology study planning and the TME page when the microenvironment is the central question.
There is no universal minimum. Sample number depends on the biological unit, paired or longitudinal structure, expected effect size, tissue heterogeneity, technical replication, and planned statistical model. A small feasibility pilot can establish tissue and platform performance, but confirmatory cohort size should be determined through study-specific power and design review.
Yes. If you already have scRNA-seq data from your cancer samples, we can use it as a reference cell-type catalog for spatial transcriptomics deconvolution. This means you do not need to repeat single-cell profiling β we run spatial transcriptomics on new tissue sections and map your existing cell-type signatures onto the spatial data. This approach is cost-effective and particularly useful when fresh tissue for scRNA-seq is no longer available but FFPE blocks exist for spatial profiling.
Deliverables are defined in the project scope and may include raw or platform-native data, processed matrices, QC summaries, cell-type annotations, spatial expression maps, niche or neighborhood analyses, integration outputs, and a project report. Compatible downstream formats and publication-oriented figures can be agreed during project design.
Timeline depends on tissue QC, platform availability, assay optimization, panel design, sample number, sequencing or imaging depth, and analysis scope. A project-specific schedule is provided after feasibility review; samples requiring optimization or replacement may extend the timeline.
Explore Spatial Omics Solutions in Other Research Areas
Beyond cancer and oncology, CD Genomics applies the same multi-omics framework across diverse research domains. Explore how spatial technologies address questions in these fields.
References
Janesick A, Shelansky R, Gottscho AD, et al. High resolution mapping of the tumor microenvironment using integrated single-cell, spatial and in situ analysis. Nature Communications. 2023;14(1):8353. doi:10.1038/s41467-023-43458-x
Wu SZ, Al-Eryani G, Roden DL, et al. A single-cell and spatially resolved atlas of human breast cancers. Nature Genetics. 2021;53:1334-1347. doi:10.1038/s41588-021-00911-1
Kumar V, Ramnarayanan K, Sundar R, et al. Single-cell atlas of lineage states, tumor microenvironment, and subtype-specific expression programs in gastric cancer. Cancer Discovery. 2022;12(3):670-691. doi:10.1158/2159-8290.CD-21-0683
Liu X, Zhao S, Wang K, et al. Spatial transcriptomics analysis of esophageal squamous precancerous lesions and their progression to esophageal cancer. Nature Communications. 2023;14(1):4779. doi:10.1038/s41467-023-40343-5
For research purposes only, not intended for clinical diagnosis, treatment, or individual health assessments.
Online Inquiry