Charting the future of spatial biology in drug discovery means looking past oncology, the disease area where nearly every finding in this cluster so far has been demonstrated, toward the therapeutic areas and technologies where spatial methods are only beginning to establish themselves. Four areas illustrate that expansion concretely: a disease area outside cancer entirely, a drug modality this cluster has not yet addressed, a molecular layer beyond transcript and protein, and an experimental design that does more than observe.
Key takeaways
|
Beyond oncology: New therapeutic areas
Oncology’s head start is easy to explain: tumor tissue is comparatively abundant, the clinical need is urgent, and the therapeutic categories, checkpoint inhibitors and targeted therapies, map naturally onto spatially resolved biomarkers. None of that logic is exclusive to cancer, and autoimmune disease is one of the clearest examples of the same spatial approach extending cleanly to a different organ system entirely.
A 2025 review of spatial transcriptomics applications in autoimmune rheumatic disease describes a study establishing distinct pathogenic roles for resident and monocyte-derived macrophages in lupus nephritis, an autoimmune kidney disease and a serious complication of systemic lupus erythematosus. That distinction, between two macrophage populations with different origins and, the study found, different pathogenic roles, is precisely the kind of finding dissociated single-cell data cannot reliably establish on its own, since it depends on knowing where within the kidney each population sits and what tissue damage surrounds it, not simply which transcriptional markers each population expresses.
That single example generalizes to a broader pattern worth naming directly: any disease involving tissue-infiltrating immune cells with heterogeneous origin and function, which describes a large share of autoimmune and inflammatory conditions well beyond lupus specifically, is a plausible candidate for the same kind of spatially resolved distinction. The specific mechanistic and drug-target detail for immunology and autoimmune disease is developed in full in Spatial Biology in Immunology and Autoimmune Disease.
Neuroscience and central nervous system drug discovery represent a second, distinct area of expansion, for reasons specific to that tissue rather than shared with autoimmune disease: brain tissue architecture is unusually dependent on precise spatial organization for normal function, and neurodegenerative and psychiatric conditions frequently involve region-specific vulnerability that a dissociated assay, discarding which brain region a cell came from, cannot preserve. The specific applications and drug discovery implications for this therapeutic area are developed in full in Spatial Biology in Neuroscience and CNS Drug Discovery.
Spatial proteomics in discovery
Spatial proteomics’ trajectory as a discovery platform is already well established elsewhere in this cluster, most notably through its recognition as Nature Methods’ 2024 Method of the Year, credited with transforming understanding of tumor microenvironment biology and cell-to-cell interaction. Rather than repeating that ground, this section simply notes where the platform is heading: toward broader application outside oncology, toward tighter integration with transcriptomic and, increasingly, epigenomic data on the same tissue, and toward the discovery workflows and validation standards covered throughout this cluster’s biomarker development guides. The full discovery-stage trajectory for spatial proteomics specifically is developed in Spatial Proteomics in Drug Discovery.
Spatial biology for advanced therapies
Cell and gene therapy is territory this cluster has not yet addressed directly, and it is a genuinely different application of spatial biology from anything covered so far: rather than characterizing disease biology or predicting drug response, spatial methods here are being applied to a therapy’s own safety profile after it has been administered.
Mapping where a gene therapy’s side effect actually originatesAdeno-associated virus vector-mediated gene therapies are already in clinical use for inherited retinal disease, including an FDA-approved treatment for retinal degeneration linked to biallelic RPE65 mutations. A 2025 study applied single-cell and spatial transcriptomics to characterize gene therapy-associated retinal inflammation in non-human primates following subretinal administration of two clinically relevant AAV vectors. The results demonstrated that subretinal AAV administration activates microglia and drives their migration to the subretinal space, where monocyte-derived phagocyte pro-inflammatory signaling recruits a chronic, cell-mediated antiviral response characterized specifically by CD8-positive effector memory T cells and myeloid cells. That is a spatially specific answer to where and how a gene therapy’s inflammatory side effect actually originates, information a bulk or dissociated assessment of inflamed tissue could describe only in aggregate, not as a traceable sequence starting from a specific retinal compartment. |
That kind of spatially resolved safety characterization is directly actionable for gene therapy development specifically, since it identifies which cell populations and which tissue compartment to target with an adjunct anti-inflammatory strategy, rather than only confirming that inflammation occurred somewhere in the treated tissue. The fuller treatment of spatial biology across cell and gene therapy development, including applications beyond gene therapy safety specifically, is developed in Spatial Biology in Cell and Gene Therapy Development.
Technology on the horizon
Two developments illustrate where spatial biology’s technical frontier actually sits right now: extending spatial measurement to an entirely new molecular layer, and extending spatial experiments from purely observational to genuinely functional.
Spatial-CUT&Tag, published in Science in 2022, performs genome-wide profiling of specific histone modifications directly on a frozen tissue section without dissociation, resolving chromatin state at pixels roughly 20 micrometers across, small enough that many pixels contain only a single nucleus, yielding effectively single-cell epigenomic profiles. Applied to mouse embryonic organogenesis and postnatal brain development, the method revealed epigenetic control of cortical layer development and spatial patterning of cell types determined specifically by histone modification state, in concordance with independent reference datasets. This is the field’s extension into chromatin biology specifically, a molecular layer beyond the transcript and protein measurements that have defined spatial biology’s first wave.
A second development extends spatial experiments in a different direction entirely: from observation to function. A 2025 Cell paper introduces Perturb-FISH, combining imaging-based spatial transcriptomics with parallel optical detection of in situ amplified guide RNAs, enabling a pooled CRISPR perturbation screen read out with single-cell spatial resolution rather than only in dissociated cells. Validated against conventional Perturb-seq for intracellular effects in a screen of monocyte inflammatory response, Perturb-FISH additionally revealed intercellular and cell-density-dependent regulation of that same response, information no dissociated screen can capture because it depends entirely on knowing which cells sat near which others. The same approach was further applied to three-dimensional tumor-immune xenograft models and to a functional screen of autism spectrum disorder risk genes in patient-derived cells, demonstrating the method’s reach well beyond its original validation context.
That shift, from spatial biology as a descriptive readout to spatial biology as the output of a functional perturbation experiment, is arguably the more consequential of the two developments for drug discovery specifically, since it means a target hypothesis can be tested directly, with spatial and single-cell resolution, rather than inferred from an observational correlation alone.
What to watch
Three signals are worth tracking as indicators of how quickly this expansion actually translates into drug discovery practice, rather than remaining confined to methods papers.
- Whether non-oncology disease areas produce druggable, spatially defined targets, not just descriptive findings. The lupus nephritis macrophage distinction is a genuine biological finding; whether it, or similar findings in other autoimmune and inflammatory diseases, yields an actual therapeutic hypothesis a program pursues is the practical test of whether this expansion matters commercially, not only scientifically.
- Whether gene and cell therapy developers adopt spatial safety assessment as standard practice, not a one-off academic study. The retinal AAV inflammation study is a single, well-designed demonstration; whether spatial profiling becomes a routine part of gene therapy preclinical safety packages is the signal that determines whether this application scales.
- Whether functional spatial genomics methods like Perturb-FISH reach a throughput and cost point comparable to conventional pooled screens. A method’s scientific capability and its practical adoption in a drug discovery pipeline are separate questions, and the throughput and automation considerations covered elsewhere in this cluster apply directly to whether a functional spatial screening method becomes routine or remains a specialized, low-throughput technique.
For where these emerging directions sit within the full spatial biology pipeline, from target discovery through the tumor microenvironment, biomarkers, and clinical translation, see Spatial Biology in Drug Discovery: From Target Discovery to Translational Medicine.
This article was produced under Drug Discovery News’s AI editorial policies.
