Core Framework & Diagram What Is Spatial Immunology?
7 月 23, 20261 Min Read What Is the Tumor Immune Atlas?
7 月 23, 2026Knowing what an immune cell 'is' isn't enough — you also need to know where it is
—— Spatial Immunology: combining gene expression information with cells' physical locations to read the geography of the immune system.
I. Birth of spatial immunology: a Nobel-level technology from 2016
Spatial transcriptomics' core technology was first systematically described in a 2016 Science paper by Jonas Frisén and Joakim Lundeberg's team at KTH Royal Institute of Technology and Stockholm University. Their method was later commercialized as 10x Genomics' Visium platform, rapidly becoming the most widely used technology in the field. This work's importance was quickly recognized: Nature Methods selected spatial transcriptomics as 2020 Method of the Year — an award historically only given to technologies that truly changed research paradigms. 10x Genomics Visium has spatial resolution of approximately fifty-five micrometers per capture point, simultaneously covering an entire tissue section (~6.5mm × 6.5mm), measuring thousands of genes. The more high-resolution Slide-seq (developed at Harvard) improves resolution to approximately ten micrometers, approaching single-cell resolution.
2. Three technical routes of spatial immunology
Route 1 — Sequencing-based spatial transcriptomics (Visium, Slide-seq): placing capture probe arrays with 'barcodes' on tissue sections — each position's probe has a unique spatial barcode sequence. mRNA on the section hybridizes with probes in-situ, then gene expression information at each position is read through sequencing, and spatial coordinates are recovered using barcodes. Advantage: large coverage area, comprehensive gene coverage. Limitation: resolution not reaching single-cell level (one fifty-five-micrometer spot may contain multiple cells). Route 2 — In-situ fluorescence hybridization (MERFISH, seqFISH): detecting expression of hundreds to thousands of specific genes in-situ through multiple rounds of fluorescence hybridization while preserving cells' original positions. This type's greatest advantage: close to single-cell resolution, able to precisely locate each mRNA molecule's position within the cell. Long Cai's team at Caltech (inventor of seqFISH) published research in 2019 Nature using seqFISH+ to in-situ detect expression of 10,000 genes in mouse brain cortex, precisely mapping different neuron subtypes' and glial cells' spatial distribution across cortical layers.
Route 3 — Spatial proteomics (CODEX, IMC): technologies including Akoya Biosciences' CODEX (co-detection by indexing, based on multiple rounds of fluorescent antibody labeling) and Fluidigm's IMC (Imaging Mass Cytometry) simultaneously detect dozens to hundreds of proteins' spatial distribution on tissue sections. Stanford University's Garry Nolan team is one of CODEX technology's core developers. They used CODEX to perform detailed spatial proteomic analysis of human tonsils, drawing an immune cell map at single-cell precision including over twenty types of immune cells — the first time humans could 'see' lymphoid organ internal immune cell organization at this precision.
|
The three technical routes are complementary: Sequencing-type (Visium): large coverage, comprehensive genes — for tissue-level panoramic scanning In-situ hybridization (MERFISH): single-cell resolution — for precise inter-cellular spacing and sub-cellular localization Spatial proteomics (CODEX): protein level — for immune cell surface markers and functional protein spatial analysis |
3. Key findings: T cell 'location' matters more than 'quantity'
Multiple spatial transcriptomics studies found T cells' 'position' has far greater prognostic predictive power than T cell 'count.' Specifically: T cell infiltration in tumor cores versus only clustering at tumor margins has completely different prognosis — patients with core infiltration have significantly longer survival. T cells' actual contact distance with tumor cells (immune synapse formation potential) is a key spatial parameter for predicting immunotherapy response — even if T cell numbers are high, if there's large amounts of CAF physical separation between T cells and tumor cells, effective killing can't be achieved. And Tertiary Lymphoid Structures (TLS) — in-situ spontaneously formed lymph node-like immune cell aggregates inside or around tumors — their presence and maturity is highly positively correlated with multiple tumor immunotherapy response rates and long-term survival.
Yale University's Chen Chen team and Institut Curie's Wolf Fridman team most systematically proved TLS's clinical predictive value — in renal cell carcinoma, soft tissue sarcoma, and melanoma, patients with mature TLS (containing germinal centers) inside tumors have two to four times higher PD-1 inhibitor response rates than patients without TLS. More broadly, the Harvard Medical School Nir Hacohen team found that in non-small cell lung cancer, based on tumor microenvironment's spatial cell organization (not just cell type or gene mutation), patients can be divided into clearly different 'Ecotypes' with different immunotherapy response rates and prognoses. This 'Ecotype' classification has stronger predictive power than any single biomarker (TMB, PD-L1).
4. Autoimmune disease applications
Beyond tumor immunity, spatial immunology is beginning to produce important findings in autoimmune disease research. The rheumatoid arthritis joint synovium is the core battlefield of autoimmune attacks. Spatial transcriptomics let researchers for the first time see clearly how self-reactive T cells, plasma cells, and synovial fibroblasts spatially interact in RA patients' synovium. University of Michigan and Stanford University teams found RA synovium can be divided into different 'Pathotypes' based on spatial immune cell organization, with these pathotypes significantly correlated with patient response to specific biologics (TNF inhibitors vs. rituximab B-cell depletion treatment) — early proof of 'spatial immune typing guiding treatment selection.'
5. Human Cell Atlas and spatial components
Spatial immunology is a core component of the Human Cell Atlas (HCA) project — a global scientific collaboration aiming to map all human cell types at single-cell resolution, including their molecular characteristics and (through spatial technology) location distribution in each organ. HCA's immunology components have produced several milestones: human thymus development cell atlas (2020 Science) — using scRNA-seq and spatial technology jointly to describe human thymus cell composition changes from fetal period to old age, directly demonstrating thymus atrophy with age and providing a cellular-resolution 'map' for immune aging research; human gut immune atlas — spatially describing immune cell distribution differences across different gut segments (duodenum to rectum), revealing why IBD's immunopathology is highly segmented; and the COVID-19 lung spatial analysis — rapidly revealing the precise spatial origin of 'cytokine storms' in severe patients' lungs.
Frequently Asked Questions