How is spatial proteomics used in drug discovery?

Spatial proteomics maps which proteins exist in a specific tissue microenvironment, where and how a therapeutic acts once it reaches tissue, and whether a drug successfully engages its intended target. Applications include diagnostic-grade protein signature classification, mapping pathway activation to identify treatment mechanisms, and measuring a drug’s spatially localized pharmacodynamic effects on specific cell types.

Can you map drug targets as proteins in tissue?

Yes. Mass spectrometry imaging has been used to classify thyroid nodules by protein signature with 100% sensitivity and 96% specificity, and multiplexed imaging has identified specific cell populations, such as CD68-positive vascular smooth muscle cells and foam cells in atherosclerotic plaque, as key drug-relevant mediators of disease directly in tissue.

What does spatial proteomics add over transcriptomics?

Spatial proteomics measures the actual protein a drug binds, including its post-translational modification state and precise tissue location, information transcript abundance alone cannot reliably predict. In one case, spatial proteomic mapping of pathway activation in a severe drug reaction identified a specific molecular mechanism that led directly to a targeted drug being used clinically for the condition.

Spatial proteomics in drug discovery

Highlights in this article

Why map proteins in drug discovery
Spatial protein targets
Pathway activity in tissue
Measuring drug effects spatially
Methods and trade-offs

Applying spatial proteomics to a drug discovery program addresses a gap no amount of transcriptomic data can fully close—a drug binds a protein, not the messenger RNA that encoded it—and mRNA abundance is frequently a poor proxy for the protein level, post-translational modification state, or subcellular localization that actually determines whether a target is druggable in a given tissue location. A 2026 review in Precision Clinical Medicine frames the field’s drug discovery relevance around three specific questions worth stating precisely, since they organize everything spatial proteomics is actually used for: what target proteins exist within the tissue microenvironment?, where and how do therapeutic interventions act once administered?, and does a drug successfully reach and engage its intended target?


Key takeaways Automated, protein-signature-based classification of thyroid nodules by mass spectrometry imaging has achieved 100% sensitivity and 96% specificity, demonstrating clinical-diagnostic-grade performance rather than only research-stage utility. Deep visual proteomics mapped more than 5,000 proteins at single-cell resolution in a rare, severe drug reaction, identifying pathway activation that led directly to a specific approved drug being used clinically for the condition. Spatial protein imaging has identified specific macrophage and vascular smooth muscle cell populations as key mediators of atherosclerotic plaque instability, extending spatial proteomics into cardiovascular disease. A kinase inhibitor’s spatially resolved effect on immune checkpoint protein expression in a specific cell type has been measured directly in pancreatic islet tissue, demonstrating pharmacodynamic assessment beyond primary target engagement. Three structurally different spatial proteomics approaches, DNA-barcoding, fluorophore-based imaging, and mass spectrometry, trade plex capacity, spatial resolution, and clinical readiness against each other, with no single method dominating every use case.

Key takeaways

Automated, protein-signature-based classification of thyroid nodules by mass spectrometry imaging has achieved 100% sensitivity and 96% specificity, demonstrating clinical-diagnostic-grade performance rather than only research-stage utility.
Deep visual proteomics mapped more than 5,000 proteins at single-cell resolution in a rare, severe drug reaction, identifying pathway activation that led directly to a specific approved drug being used clinically for the condition.
Spatial protein imaging has identified specific macrophage and vascular smooth muscle cell populations as key mediators of atherosclerotic plaque instability, extending spatial proteomics into cardiovascular disease.
A kinase inhibitor’s spatially resolved effect on immune checkpoint protein expression in a specific cell type has been measured directly in pancreatic islet tissue, demonstrating pharmacodynamic assessment beyond primary target engagement.
Three structurally different spatial proteomics approaches, DNA-barcoding, fluorophore-based imaging, and mass spectrometry, trade plex capacity, spatial resolution, and clinical readiness against each other, with no single method dominating every use case.

Why map proteins in drug discovery

Our earlier coverage in The Promise of Spatial Proteomics introduces the field’s general case. This guide develops that case specifically for drug discovery, using the framing a 2026 Precision Clinical Medicine review states directly: protein function is inherently spatial, since the same molecule can produce entirely different biological outcomes depending on its localization, its interacting partners, and the surrounding tissue microenvironment. Spatial proteomics enables systematic in situ characterization of protein localization, abundance, and interaction across scales from subcellular structures to whole tissue, surpassing what conventional, lysate-based bulk proteomics can resolve.

That review organizes spatial proteomics’ translational value around three specific questions a drug discovery program actually needs answered: what target proteins exist within the tissue microenvironment, where and how do therapeutic interventions act once they reach tissue, and does a drug successfully reach and engage its intended target. Each of the following sections addresses one of those three questions directly, using a specific, verifiable example rather than treating spatial proteomics’ value as self-evident.

Spatial protein targets

The first question, what target proteins exist within the tissue microenvironment, is where spatial proteomics has already demonstrated clinical-diagnostic-grade performance, not merely research-stage promise.

The same review reports automated, protein-signature-based classification of thyroid nodules using matrix-assisted laser desorption/ionization mass spectrometry imaging, achieving 100% sensitivity and 96% specificity in a manner that complements traditional cytology. That level of diagnostic accuracy, achieved through a spatially resolved protein signature rather than a single-marker assay, is a concrete demonstration that spatial proteomics can already meet the bar clinical diagnostics require, not only the bar discovery-stage research requires.

A second example extends spatial protein target identification into cardiovascular disease specifically, a therapeutic area this cluster has not otherwise used as a primary example. PhenoCycler imaging identified CD68-positive vascular smooth muscle cells and foam cells as key mediators of atherosclerotic plaque instability, while a separate near-infrared photoacoustic imaging approach linked the macrophage markers CD74 and CD163 to plaque vulnerability specifically. Both findings tie a spatially defined cellular population directly to a clinically meaningful risk, offering a path toward imaging-based biomarkers for stroke or myocardial infarction risk assessment, and demonstrating that the tissue-organization principle established repeatedly in oncology throughout this cluster extends cleanly to vascular disease as well.

Pathway activity in tissue

The second question, where and how do therapeutic interventions act, is best answered not by measuring a single protein’s abundance but by mapping an entire signaling pathway’s activation state directly in diseased tissue. The clearest available demonstration of this, and arguably the clearest example in this entire cluster of a complete pipeline from spatial observation to clinical drug use, comes from a rare and severe drug reaction rather than from oncology.


From a spatial protein map to a specific drug, in one continuous chain Deep visual proteomics was applied to archived skin tissue from patients with toxic epidermal necrolysis, a rare, severe, and frequently fatal drug reaction, mapping more than 5,000 proteins at single-cell resolution directly in the affected tissue. The analysis revealed strong activation of the JAK/STAT and interferon signaling pathways as the driver of disease pathology, and further identified local signal transducer and activator of transcription 1 phosphorylation as the specific key molecular mechanism responsible. That finding led directly to a targeted clinical intervention: tofacitinib, a pan-Janus kinase inhibitor already approved for other indications, was used to treat the condition based specifically on the pathway the spatial protein map identified as active. This is the complete pipeline in a single case: a spatial proteomic map identifies which pathway is actually driving disease in the affected tissue, that pathway points to an existing drug class with a known mechanism, and the drug is deployed clinically as a direct consequence of the spatial finding, not a coincidental afterthought.

From a spatial protein map to a specific drug, in one continuous chain

Deep visual proteomics was applied to archived skin tissue from patients with toxic epidermal necrolysis, a rare, severe, and frequently fatal drug reaction, mapping more than 5,000 proteins at single-cell resolution directly in the affected tissue. The analysis revealed strong activation of the JAK/STAT and interferon signaling pathways as the driver of disease pathology, and further identified local signal transducer and activator of transcription 1 phosphorylation as the specific key molecular mechanism responsible.

That finding led directly to a targeted clinical intervention: tofacitinib, a pan-Janus kinase inhibitor already approved for other indications, was used to treat the condition based specifically on the pathway the spatial protein map identified as active. This is the complete pipeline in a single case: a spatial proteomic map identifies which pathway is actually driving disease in the affected tissue, that pathway points to an existing drug class with a known mechanism, and the drug is deployed clinically as a direct consequence of the spatial finding, not a coincidental afterthought.

Measuring drug effects spatially

The third question, does a drug successfully reach and engage its intended target, extends naturally into a related and equally important measurement: what does the drug actually do, spatially, to the specific cell types it reaches, beyond simply confirming binding occurred.

The same review describes a pharmacodynamic finding of exactly this kind. In mouse pancreatic islets treated with inhibitors of protein kinase R-like endoplasmic reticulum kinase, surface expression of the immune checkpoint protein PD-L1 on insulin-producing beta cells was significantly upregulated, enhancing beta cell immune tolerance and offering a strategy to delay type 1 diabetes onset. That is a spatially resolved pharmacodynamic readout in the fullest sense: not confirming that the drug bound its primary target, but measuring a specific, spatially localized downstream consequence, elevated checkpoint protein expression on a specific cell type in its native tissue location, that constitutes the actual therapeutic hypothesis being tested.

That distinction, between confirming target engagement and measuring the resulting spatial phenotype, matters for how a program designs its own pharmacodynamic biomarker strategy. A target engagement assay alone answers whether the drug reached its target; a spatial phenotypic readout, of the kind described here, answers whether reaching that target produced the specific, spatially localized biological change the drug was designed to cause.

Methods and trade-offs

No single spatial proteomics method dominates every application, and choosing among them means understanding what each specifically trades away.

Approach	Strength	Trade-off
DNA-barcoding	Highest plex capacity, typically 50 to 100-plus markers, with subcellular spatial resolution	High cost and complexity; antibody and epitope dependence across many hybridization cycles
Fluorophore-based imaging	Highest clinical readiness and moderate cost; compatible with standard fluorescence microscopy	Moderate plex capacity, typically 20 to 60 markers; cannot detect post-translational modifications
Mass spectrometry-based	Highest overall proteome coverage and the only approach able to detect post-translational modifications	Lowest clinical readiness currently and no fixed marker panel; typically coarser spatial resolution, on the order of 5 to 50 micrometers

Table 1. Three structurally different spatial proteomics approaches and their central trade-off. Selecting among them depends on whether a project prioritizes plex, clinical translatability, or post-translational modification detection specifically.

That final trade-off, mass spectrometry’s unique ability to detect post-translational modifications, including the phosphorylation state central to the toxic epidermal necrolysis case above, is precisely why mass spectrometry-based spatial proteomics remains the preferred approach for pathway activity and mechanism-of-action questions specifically, even though it currently lags fluorophore-based imaging in clinical readiness. The deep methodological detail behind mass spectrometry-based spatial proteomics specifically, including laser capture microdissection and single-cell mass spectrometry workflows, is covered in full by our colleagues at Separation Science in Spatial Proteomics by Mass Spectrometry: LCM, Single-Cell and Imaging Approaches, and the broader multiplexed tissue imaging methodology this section surveys at a comparative level is covered by Technology Networks in Spatial Proteomics and Multiplexed Tissue Imaging: A Methods Guide.

For where spatial proteomics sits within spatial biology’s broader emerging trajectory, see Where Spatial Biology is Headed in Drug Discovery, and for where this fits within the full spatial biology pipeline this cluster has described from target discovery through clinical translation, Spatial Biology in Drug Discovery: From Target Discovery to Translational Medicine.

This article was produced under Drug Discovery News’s AI editorial policies.