Psoriasis is a chronic inflammatory skin disease driven by complex, spatially organized interactions among keratinocytes, immune cells, and stromal cells (e.g., fibroblasts). Conventional bulk profiling often blurs "who is doing what, where," making it hard to connect molecular programs to tissue microanatomy. Spatial omics—especially spatial transcriptomics integrated with single-cell and imaging—enables in situ mapping of gene programs, cell states, and neighborhood-level communication, accelerating discovery of disease mechanisms, severity stratification signals, and candidate therapeutic targets.

This page curates five influential recent studies that used spatial and single-cell multi-omics to decode psoriasis, and summarizes practical, reusable study-design lessons.

TL;DR

• Spatial omics reveals where IL-17/TNF/IL-36 programs localize in psoriatic skin and how immune–epithelial proximity shapes outcomes.
• Across key papers, fibroblast and myeloid states repeatedly act as inflammation amplifiers via chemokine and crosstalk networks.
• Spatial "cellular ecosystems" can stratify severity, and non-lesional skin may already show early remodeling in more severe disease.
• Therapy response is often a spatial-threshold phenomenon: small residual IL-17 activity near keratinocytes can separate responders from non-responders.
• This roundup includes a one-page study design checklist you can reuse for mechanism, severity, or treatment-response projects.

If you're new to platform selection or analysis, start with our guide to choose the right spatial transcriptomics technology and a spatial transcriptomics data analysis workflow to see how studies like these are executed end to end.

For multi-modal projects, see our overview on spatial multi-omics integration.

Why Psoriasis Needs Spatial Omics

This section explains why psoriasis biology is inherently spatial—linking epidermal–dermal organization and immune–stromal interactions to localized pathway activation that bulk assays often miss.

Psoriasis is not just "more inflammation." It is a layered, neighborhood-driven tissue disease in which keratinocytes, immune cells, and stromal cells activate distinct programs depending on where they sit in the epidermis–dermis architecture. Lesions typically combine hyperproliferative epidermal states with immune infiltration and stromal remodeling in the dermis, and many key signals are strongest in specific microanatomical zones rather than uniformly across the biopsy.

Bulk RNA profiling can identify upregulated pathways (e.g., IL-17/TNF-related programs), but it cannot reliably answer the questions that matter most for mechanism and translation: Which cell state is driving the signal? Where is it concentrated (suprabasal epidermis vs basal layer vs dermal clusters)? Which cell types are physically adjacent, enabling paracrine loops? In psoriasis, proximity is often the mechanism—immune cells positioned next to keratinocytes can trigger high local cytokine responses, while fibroblast–myeloid neighborhoods can amplify recruitment and sustain inflammation.

Spatial omics addresses these gaps by mapping gene programs in situ, preserving tissue context. Spatial transcriptomics can reveal epidermal inflammatory hotspots, dermal immune hubs, and gradients across lesion borders, while integration with single-cell data sharpens cell-type and cell-state interpretation. Importantly, spatial approaches also elevate the value of non-lesional skin, which may already show early remodeling in more severe disease—signals that can be missed when non-lesional samples are treated as "normal."

In short, spatial omics turns psoriasis from a list of differentially expressed genes into an interpretable tissue model: cell states + locations + neighborhoods + communication, enabling clearer mechanism hypotheses, severity stratification features, and more actionable targets for validation.

For a practical decision framework, compare Bulk RNA-seq vs single-cell vs spatial transcriptomics and review key concepts in understanding spatial genomics principles and techniques.

At-a-Glance Literature Map

Featured Papers and What They Add

In this section, we break down five representative studies and highlight what each contributes to understanding psoriatic tissue ecosystems and actionable mechanisms.

If you want deeper guidance on analysis modules mentioned here (domain detection, deconvolution, neighborhood signaling), see computational strategies and machine learning for spatial genomics data and our practical notes on cell segmentation methods (StarDist, Cellpose, CellBin and SCS).

Case 1 — Single-Cell + Spatial Profiling Defines Inflammatory Amplification by Keratinocytes and Fibroblasts

Ma F. et al. Nature Communications (2023-06-12). DOI: 10.1038/s41467-023-39020-4

What they asked

Psoriasis biology is often framed around immune cells (e.g., T cells), but stromal contributions—especially fibroblast roles and their spatial coordination with epidermal programs—remain under-characterized. This study asked how keratinocytes and fibroblasts contribute to inflammatory amplification, and where these programs localize within tissue.

Study design & tech stack

• 14 plaque psoriasis patients and 11 healthy controls; lesional and non-lesional biopsies (33 total)
• scRNA-seq + high-resolution spatial transcriptomics (Seq-Scope / spatial-seq)
• CRISPR-Cas9 perturbation and tissue validation (IHC/IF, qRT-PCR)

Key findings (spatial + cell states)

• Identified 10 major skin cell types and their spatial distributions.
• Lesional epidermis showed striking keratinocyte transcriptome shifts and increased immune-cell fractions (T cells, myeloid).
• The suprabasal epidermis emerged as a central inflammatory zone enriched for IL-17A, TNF, and IL-36 responses.
• Highlighted IL-36G/IL-36R signaling as an inflammation amplifier that can operate without neutrophil proteases (context-specific amplification logic).
• Identified an SFRP2⁺ fibroblast subset capable of shifting from pro-fibrotic to pro-inflammatory behavior, secreting chemokines (e.g., CCL13/CCL19/CXCL12) and proteases to shape neighborhood signaling networks.
• CD8⁺ Tc17 cells were a major IL-17A source in lesions, and distinct myeloid subsets (e.g., cDC2A, CD16⁺ DC) contributed to disease programs.

Figure 2. Single-cell atlas and spatial localization of major skin cell types across healthy, non-lesional, and lesional psoriasis samples (scRNA-seq + spatial-seq).

Methods worth copying

• Joint scRNA + spatial cell-state mapping
• Trajectory analysis to interpret state transitions
• Ligand–receptor inference anchored to spatial neighborhoods (not just cell lists)

Takeaways for your project

If your hypothesis involves "amplification loops," pair epidermal state mapping with fibroblast-state resolution, and validate top axes (e.g., IL-36, chemokine programs) with orthogonal assays.

Case 2 — Spatial Transcriptomics Stratifies Disease Severity via Emergent Cellular Ecosystems

Castillo R.L. et al. Science Immunology (2023-06-02). DOI: 10.1126/sciimmunol.abq7991

What they asked

Psoriasis exhibits heterogeneous severity and systemic associations (metabolic syndrome, cardiovascular disease, PsA). The authors asked whether spatial tissue ecosystems can stratify severity and reveal early molecular remodeling in non-lesional skin.

Study design & tech stack

• Healthy controls and PsO/PsA patients; lesional vs non-lesional biopsies
• 10x Genomics Visium spatial transcriptomics
• scRNA-seq integration with public datasets + multiplex IF
• BayesPrism deconvolution, SpaceFold localization, NicheNet signaling inference

Key findings

• Built spatial atlases for healthy and psoriatic skin; highlighted immune surveillance niches around hair follicles and vessels in healthy tissue.
• Identified lesion-specific epidermal and dermal inflammatory clusters enriched for IL-17/IL-22/STAT programs.
• Observed B cell infiltration (CD20⁺) in lesions (spatially relevant immune ecology).
• Unsupervised clustering revealed that both lesional and non-lesional samples could stratify by severity, more strongly than by PsA status.
• In more severe disease, non-lesional skin already carried molecular changes, including upregulation of metabolic genes (e.g., DGAT2, FGFR3), suggesting a broader remodeling of tissue microenvironment.

Methods worth copying

• Severity-aware clustering and spatial domain definition
• Deconvolution + spatial localization for cell-type mapping
• Signaling inference linked to emergent ecosystems

Takeaways for your project

If you care about severity or systemic links, include non-lesional matched biopsies and treat them as biologically informative—not just "controls."

If you're evaluating Visium for whole-section ecosystem mapping, see our 10x Genomics Visium platform overview (and Introduction to 10x Spatial Transcriptomics for a quick primer).

Case 3 — A Single-Cell Atlas of IL-23 Inhibition Distinguishes Clinical Response in Cutaneous Psoriasis

Wu D. et al. Science Immunology (2024-01-26). DOI: 10.1126/sciimmunol.adi2848

What they asked

IL-23 blockade is widely used, but ~10–20% show poor response. The study asked what cellular and spatial programs separate strong responders from weak responders, focusing on skin-resident immune states.

Study design & tech stack

• 15 moderate–severe untreated patients (PASI ≥ 7)
• IL-23 inhibitor treatment; strong vs weak responders; longitudinal sampling (pre vs mid-treatment)
• scRNA-seq + CITE-seq; Visium spatial transcriptomics
• PhenoCycler/CODEX multiplex imaging

Key findings

• In strong responders, skin resident memory T cell (Trm) psoriasis signatures dropped substantially, and T17 populations declined sharply with cytokine downregulation (IL-17A/IL-17F/IL-26).
• Weak response was primarily characterized by persistent T17 activation, rather than keratinocyte-only resistance or alternative cytokine compensation.
• Even strong responders retained some disease-associated transcriptional abnormalities (suggesting need for maintenance therapy in real-world settings).
• Spatially, clinical remission required suppressing IL-17-induced responses in keratinocytes near lymphocytes to a very high degree; weak responders retained a small but meaningful residual IL-17 program.

Methods worth copying

• Responder vs non-responder stratified analysis
• Spatial pathway scoring anchored to epithelial–immune proximity
• Multiplex imaging to validate spatial immune–epithelial neighborhoods

Takeaways for your project

For therapy-response studies, design for clinical stratification and model spatial thresholds (how much pathway activity remains, and where).

Case 4 — IL-26⁺ TH17 Intermediates Differentiate into IL-17A Producers via Epithelial Crosstalk in Psoriasis

Fries A. et al. Nature Communications (2023-06-30). DOI: 10.1038/s41467-023-39484-4

What they asked

IL-26 is a TH17-associated cytokine, but its role in generating pathogenic IL-17A producers in psoriasis was unclear. The authors asked whether IL-26⁺ intermediates represent a differentiation stage and how epithelial cues shape this trajectory in tissue.

Study design & tech stack

• PBMCs + psoriasis lesional/non-lesional skin; AD comparator tissue
• scRNA-seq + scTCR-seq; spatial transcriptomics
• Flow-FISH, ELISA, RT-qPCR, confocal microscopy, CRISPR perturbation
• Pseudotime and clonal lineage evidence

Key findings

• Identified abundant IL-26 single-positive TH17-like cells in blood—appearing early in differentiation, without TGF-β1.
• IL-17A⁺ cells emerged later and depended on TGF-β1.
• In psoriasis lesions, IL-26⁺ intermediates localized near keratinocytes expressing TGF-β1, supporting a spatially driven differentiation loop.
• IL-26 signaling via IL-26 receptor components induced keratinocyte TGF-β1, reinforcing a feedback mechanism.
• The conversion was more pronounced in psoriasis vs AD, consistent with disease-specific infiltration patterns.

Figure 3. Spatial proximity of IL26+ T-cell spots to TGFB1-expressing keratinocyte spots in psoriatic lesions, supporting an epithelial–T cell paracrine loop.

Methods worth copying

• Combine pseudotime with spatial adjacency and cytokine-axis validation
• Use disease comparators (e.g., AD) to highlight psoriasis-specific mechanisms

Takeaways for your project

If you hypothesize "immune differentiation in tissue," design analyses around trajectory + neighborhood context, not just cell proportions.

Case 5 — Celastrol Targets LRP1 to Block Fibroblast–Macrophage Crosstalk and Ameliorate Psoriasis Progression

Zhu Y. et al. Acta Pharmaceutica Sinica B (2025-01-04). DOI: 10.1016/j.apsb.2024.12.041

What they asked

Beyond the IL-23/TH17 axis, could fibroblast-led signaling networks be targeted to reduce disease? The authors investigated fibroblast–macrophage crosstalk and evaluated celastrol as a candidate modulator.

Study design & tech stack

• Human psoriasis tissue + IMQ mouse psoriasis-like model + macaque evidence
• Fibroblast-specific LRP1 knockout mouse
• scRNA-seq + scATAC-seq + spatial transcriptomics
• Mechanistic validation: WB/IF, immunoprecipitation, CETSA/MST for target engagement

Key findings

• Fibroblasts secreted chemokines (notably CCL2) that recruited macrophages and exacerbated inflammation; an HLA⁺ fibroblast subset was a major CCL2 source.
• Celastrol reduced fibroblast CCL2 secretion and improved lesions across models.
• Mechanistically, celastrol bound LRP1 β chain, disrupted interaction with c-Jun, and reduced CCL2 transcription.
• Fibroblast-specific LRP1 loss reduced inflammation and dampened macrophage activation, linking target → signaling → phenotype.

Methods worth copying

• Multi-omics integration for mechanism discovery + direct target engagement assays
• Genetic specificity (cell-type-specific KO) to validate causality

Takeaways for your project

If you want to claim a "targetable axis," build an evidence chain from omics → spatial context → perturbation → phenotype.

Cross-Paper Synthesis: Recurring Spatial Patterns in Psoriasis

Across the curated literature, this section distills recurring spatial motifs—epidermal hotspots, dermal immune hubs, neighborhood-driven signaling, and non-lesional changes linked to severity.

Across these studies, several spatial themes repeat:

1. Epidermal inflammatory zones matter
   Lesional epidermis—often suprabasal layers—shows concentrated inflammatory programs (IL-17/TNF/IL-36), reinforcing that "where" inflammation sits can be as important as "how much."
2. Dermal immune hubs and stromal amplifiers are not passive
   Fibroblast states can shift toward pro-inflammatory roles and shape immune recruitment and activation, especially via chemokine programs and macrophage interactions.
3. Neighborhood-driven signaling provides mechanistic leverage
   Ligand–receptor predictions become more convincing when anchored to spatial proximity (e.g., immune cells adjacent to keratinocytes; IL-26⁺ T cells near TGF-β1⁺ epidermis).
4. Non-lesional skin is informative, especially in severe disease
   Non-lesional tissue may carry early molecular remodeling that associates with severity and systemic features—useful for biomarker discovery and stratified study design.
5. Therapy response can be a spatial threshold phenomenon
   Response is not just "cell counts"; the residual spatial activation of key pathways (especially IL-17 programs near keratinocytes) can separate responders from non-responders.

If You're Planning a Psoriasis Spatial Omics Study (Practical Checklist)

This practical checklist translates the literature into actionable guidance for cohort design, platform pairing, and analysis strategies for mechanism, severity, and treatment-response studies.

Cohort and grouping

• Core comparison: Lesional vs non-lesional matched biopsies ± healthy controls
• Stratify intentionally: severity (e.g., PASI), therapy response (responder vs non-responder), optional PsO vs PsA
• Include metadata: treatment history, comorbidities (metabolic/cardiovascular), lesion location

Platform pairing strategy (decision logic)

• Visium (whole-section domains): best for broad ecosystem mapping and severity stratification
• Higher-resolution spatial (if needed): better for epidermal microdomains, immune–epithelial adjacency, or rare populations
• scRNA-seq integration: improves cell type mapping, deconvolution, and differential state analysis
• Imaging (mIF/CODEX): strengthens spatial validation at protein level and supports reviewer expectations

Analysis checklist (tool-agnostic)

• Spatial QC and normalization
• Domain discovery (spatial clustering) + layer-aware interpretation (epidermis vs dermis)
• Cell-type mapping (reference integration / deconvolution)
• Pathway scoring (IL-17/IL-23, IL-36, STAT, chemokine axes, metabolic signatures)
• Cell–cell communication inference anchored to neighborhoods
• Differential comparisons: lesion vs non-lesion; severity; response; timepoints

Common pitfalls to avoid

• Treating non-lesional skin as "normal baseline" without analysis
• Over-interpreting ligand–receptor outputs without spatial anchoring and validation
• Underpowered subgrouping for severity/response claims
• Batch effects across slides, regions, or collection sites

FAQs

1) Do I need single-cell data to interpret spatial transcriptomics in psoriasis?
Not strictly, but scRNA-seq integration typically improves cell-state resolution and supports more confident ecosystem interpretation.

2) Is non-lesional skin worth profiling?
Yes—multiple studies show non-lesional tissue can carry severity-associated molecular remodeling.

3) What are the most informative spatial readouts for psoriasis?
Spatial domains (epidermal vs dermal programs), immune–epithelial neighborhoods, and pathway activity gradients (IL-17/IL-36/STAT-related programs).

4) How should I design a therapy-response study?
Plan pre- and on-treatment timepoints, define responder groups up front, and analyze spatial pathway suppression near keratinocytes rather than relying only on cell counts.

5) How do I validate spatial signaling hypotheses?
Common approaches include multiplex imaging, targeted qPCR panels, and functional perturbations (where feasible) to connect mechanism to phenotype.

6) What's a strong reviewer-friendly story structure?
Mechanism discovery from spatial + single-cell → neighborhood-based hypothesis → orthogonal validation → (optional) perturbation evidence.

Conclusion

Figure 4. What spatial omics enables in psoriasis: from atlases to inflammatory niches, severity stratification, response mapping, and target discovery.

Spatial omics is reshaping psoriasis research by reconnecting gene programs to the tissue microenvironment—revealing where inflammatory pathways ignite, which cell states act as amplifiers, and how neighborhood-level interactions sustain or resolve disease. Across the studies summarized here, several consistent themes emerge: inflammation concentrates in specific epidermal and dermal spatial domains; fibroblast and myeloid programs can form powerful amplification loops; non-lesional skin may already carry severity-linked molecular remodeling; and therapeutic success can depend on reaching a high spatial threshold of pathway suppression (especially IL-17–driven responses near keratinocytes).

Looking ahead, the most impactful psoriasis projects will likely combine whole-section spatial profiling with higher-resolution mapping, integrate single-cell and imaging validation, and adopt severity- or response-aware designs that treat spatial context as a first-class biological variable. As spatial resolution improves and multi-omics integration becomes more routine, spatially informed biomarkers and mechanistic targets—grounded in reproducible tissue niches—should accelerate both discovery and translational progress in psoriasis.

Explore our Spatial Transcriptomics Services, and if your project needs chromatin context or immune validation, consider Spatial ATAC-seq and mIHC spatial immune profiling.

Reference

1. Ma F, Plazyo O, Billi AC, et al. Single cell and spatial sequencing define processes by which keratinocytes and fibroblasts amplify inflammatory responses in psoriasis. Nat Commun. 2023;14(1):3455.
2. Castillo RL, Sidhu I, Dolgalev I, et al. Spatial transcriptomics stratifies psoriatic disease severity by emergent cellular ecosystems. Sci Immunol. 2023;8(84):eabq7991.
3. Wu D, Hailer AA, Wang S, et al. A single-cell atlas of IL-23 inhibition in cutaneous psoriasis distinguishes clinical response. Sci Immunol. 2024;9(91):eadi2848.
4. Fries A, Saidoune F, Kuonen F, et al. Differentiation of IL-26+ TH17 intermediates into IL-17A producers via epithelial crosstalk in psoriasis. Nat Commun. 2023;14(1):3878.
5. Zhu Y, Zhao L, Yan W, et al. Celastrol directly targets LRP1 to inhibit fibroblast-macrophage crosstalk and ameliorates psoriasis progression. Acta Pharm Sin B. 2025;15(2):876–891.