The immune system can identify and destroy abnormal cells, but this response depends on whether immune cells can recognize the unique features of a tumor. Cancer cells often produce altered proteins called neoantigens, which arise from tumor-specific genetic mutations. These neoantigens can act like molecular warning signs, allowing certain T cells to distinguish cancer cells from healthy tissue.

Finding the T cells that recognize a specific neoantigen is important for developing personalized cancer treatments. However, most existing methods analyze genetic mutations, gene activity, and immune cells separately. This makes it difficult to determine where neoantigen-producing tumor cells are located and whether the appropriate T cells are nearby.

Researchers from the Department of Medical Oncology at Dana-Farber Cancer Institute, Harvard Medical School, and the Broad Institute of MIT and Harvard, developed a new spatial transcriptomics method called Slide-GoTags. The technique analyzes frozen tumor tissue while preserving information about where individual cells were located in the original sample.

Slide-GoTags for genotyping of spatially mapped snRNA-seq

a, Schematic of the Slide-GoTags workflow. Spatially barcoded nuclei from a 20-μm fresh-frozen tissue section were profiled using 10x Genomics droplet-based snRNA-seq as per the Slide-tags protocol³. The cDNA library was used for whole-transcriptome amplification, as well as TCR amplification and genotyping of mutations. b, Schematic of the syngeneic tumor model. A single mouse was implanted subcutaneously with 1 × 10⁶ MC38 cells composed of a mixture of WT and SIINFEKL-expressing MC38 cells. Then, 2 days later, the mouse received 2 × 10⁵ CD8⁺ OT-I T cells through intravenous injection. Tumors were isolated on day 9 after implantation. c, Serial section of the MC38 tumor stained with H&E. This staining was performed once on the section shown. d, Spatial mapping of snRNA-seq profiles, colored by cell type. e, Bar plot showing the frequency of TCRα and TCRβ CDR3 sequences in snRNA-seq T cells. f, Proportion of tumor cells genotyped for SIINFEKL, based on SIINFEKL expression. g, Spatial mapping of tumor cells (light blue), SIINFEKL-genotyped tumor cells (blue), OT-I T cells expressing CDR3 TRA/TRB (red) and other T cells with non-OT-I CDR3 sequences (pink). h,i, Immunofluorescence staining showing CD45.2⁺ OT-I T cells (red) and SIINFEKL-H-2Kᵇ positive cells (light blue) in the same tumor. This staining was performed once on the section shown.

Slide-GoTags combines three types of information from the same tissue section. It uses single-nucleus RNA sequencing to measure gene activity, targeted transcript genotyping to identify tumor cells carrying specific mutations, and T cell receptor sequencing to identify individual groups of T cells. By integrating these measurements, researchers can determine which tumor cells express a neoantigen, which T cells may recognize it, and whether those cells are positioned close together.

Connecting tumor mutations with immune responses

Neoantigens are created when cancer-associated mutations alter the proteins made by tumor cells. Fragments of these altered proteins may be displayed on the tumor cell surface, where they can be recognized by T cell receptors.

Each T cell carries a distinctive receptor that recognizes a limited set of molecular targets. When a T cell encounters a matching neoantigen, it may become activated and multiply, producing a group of genetically related T cells known as a clonotype.

Slide-GoTags allows researchers to map both sides of this interaction within the tumor. It can identify tumor cells expressing a particular neoantigen and locate expanded T cell clonotypes carrying receptors that recognize that target.

When the researchers applied Slide-GoTags to mouse and human tumors, they found that neoantigen-specific T cells were spatially concentrated near tumor cells expressing the corresponding neoantigens. This physical proximity provides evidence that the immune cells were responding to specific tumor targets rather than simply being present within the surrounding tissue.

Examining responses to immune checkpoint therapy

The researchers also used Slide-GoTags to examine mouse colorectal tumors treated with immune checkpoint inhibitors. These therapies remove molecular signals that normally restrain T cell activity, allowing immune cells to mount a stronger response against cancer.

Tumors treated with anti-PD1 and anti-CTLA4 therapies developed different spatial immune environments. These differences suggest that each treatment may influence how T cells, tumor cells, and other cells within the tumor microenvironment organize and communicate.

Understanding these patterns could help explain why some tumors respond well to one checkpoint inhibitor but not another. It may also help researchers identify combinations of treatments that produce a stronger or more coordinated immune response.

Identifying immunogenicity niches

In human tumor samples, Slide-GoTags identified regions called immunogenicity niches. These were localized areas where immune activity was particularly strong and where tumor cells, T cells, and interferon-related gene activity occurred together.

Interferons are signaling molecules that help activate immune defenses. The researchers found more interferon-driven niches in immunologically hot tumors, which contain many active immune cells, than in cold tumors, which tend to have weaker immune responses.

Within these niches, the researchers identified three T cell clonotypes positioned near tumor cells expressing their matching neoantigens. This finding suggests that effective antitumor immunity may be organized into specific neighborhoods within a tumor rather than occurring evenly throughout the tissue.

Why spatial information is important

Traditional RNA sequencing can reveal which genes are active within a tumor, but it does not always preserve the physical relationships between cells. Spatial transcriptomics adds this missing layer by showing where gene activity occurs within the tissue.

This is particularly important for cancer immunology because the location of a T cell can be as important as its molecular identity. A T cell capable of recognizing a tumor neoantigen may have little effect if it cannot enter the tumor or reach the cells expressing that target.

By combining spatial location with RNA sequencing, mutation detection, and T cell receptor analysis, Slide-GoTags provides a more complete view of the immune response. It shows not only which cells are present, but also which tumor mutations they may recognize and where those interactions are occurring.

Supporting personalized cancer immunotherapy

Slide-GoTags could help researchers identify therapeutically relevant neoantigens and the T cells that recognize them directly from an individual tumor sample. This information may support the development of personalized cancer vaccines, engineered T cell therapies, and other treatments designed around a patient’s specific tumor mutations.

The approach may also help researchers evaluate whether an immune therapy is producing the intended response. By comparing tumor samples before and after treatment, scientists could examine whether neoantigen-specific T cells are expanding, entering the tumor, and reaching the appropriate cancer cells.

Additional research will be needed to determine how easily Slide-GoTags can be incorporated into clinical workflows. However, its ability to connect tumor genetics, immune cell identity, gene activity, and tissue location provides a valuable framework for studying how the immune system recognizes cancer. The article was published online in Nature Biotechnology on July 22, 2026.