Tumor Biology and Immuno-oncology

The tumor as an ecosystem

Tumor biology is the most active area of biomedical research today. Cancer is not a cell-autonomous phenomenon and a tumor is an ecosystem of malignant cells at various stages of clonal evolution, immune cells in active surveillance to deep exhaustion states, cancer-associated fibroblasts remodeling the extracellular matrix, endothelial cells forming abnormal vasculature, and a dense network of signaling interactions among all of them. Bulk sequencing blends this complexity while single-cell and spatial methods preserve it.

Common questions in the field include which immune cell states predict response to treatments such as checkpoint blockade? Where in the tumor are therapy-resistant tumor cells located? What signaling interactions at the tumor-immune boundary drive immune evasion? These questions require single-cell resolution and spatial context.

Tumor heterogeneity and clonal architecture

Intratumoral heterogeneity — the coexistence of genetically and transcriptionally distinct subpopulations within a single tumor — is a primary driver of treatment resistance. Cells with different copy number profiles, transcriptional states, or epigenetic programs respond differently to therapy, and resistant clones expand under selective pressure. Single-cell transcriptomics resolves this heterogeneity at the level of individual cells rather than averaged populations.

Copy number variation inference from expression data identifies malignant subclones without requiring matched DNA sequencing. Numbat integrates allelic ratios and population haplotype information for accurate allele-specific CNV calling. CopyKAT offers a faster alternative for samples without germline reference data. Both tools reconstruct clonal architecture and can be combined with trajectory inference to map the evolutionary path from normal to malignant cell states.

For FFPE archival samples, single-nucleus RNA-seq (snRNA-seq) enables analysis of fixed tissue that cannot be dissociated into viable single cells, greatly expanding access to clinical biobank material.

The tumor microenvironment

The tumor microenvironment (TME) encompasses all non-malignant cells within and surrounding the tumor mass. Its composition and functional state are now understood to be as important as the intrinsic properties of the malignant cells in determining clinical outcomes — particularly response to immunotherapy.

Key TME compartments resolvable by single-cell transcriptomics include: cytotoxic T cells across a spectrum from nave to exhausted; regulatory T cells (Tregs) that suppress anti-tumor immunity; diverse macrophage states from M1-like inflammatory to M2-like immunosuppressive; cancer-associated fibroblast subtypes with distinct roles in matrix remodeling and immunosuppression; dendritic cell subsets including the cDC1 population critical for cross-presentation and response to checkpoint blockade; and NK cells, B cells, and innate lymphoid cells that are increasingly recognized as key determinants of immunotherapy response.

T cell states and immunotherapy response

T cell biology in tumors is one of the most intensively studied areas in single-cell transcriptomics. The transition from functional effector T cells to dysfunctional exhausted states characterized by expression of checkpoint molecules including PD-1, TIM-3, LAG-3, and CTLA-4 is resolved at single-cell resolution into a spectrum of intermediate states with distinct therapeutic implications.

Progenitor exhausted T cells (Tpex) which retain proliferative capacity and express TCF7 are PD-1 blockade responsive cell populations that expand upon successful immunotherapy, whereas terminal exhausted T cells (Tex) do not respond. The ratio and spatial localization of these populations within the TME are increasingly used as predictive biomarkers. TCR sequencing (scTCR-seq), available as an add-on to standard 10x Chromium single-cell RNA-seq, links T cell transcriptional state to clonal identity, enabling tracking of specific clones across tissue compartments and time points.

Tertiary lymphoid structures

Tertiary lymphoid structures (TLS) are ectopic lymphoid aggregates that form within tumors and represent an organized immune response against the tumor. Their presence is associated with improved prognosis and response to immunotherapy across multiple cancer types. Single-cell and spatial transcriptomics are now the primary tools for characterizing TLS composition and maturation state, identifying the B cell, T cell, and dendritic cell subpopulations that define functional TLS, and mapping their spatial relationship to malignant cells and the tumor invasive front.

Cancer-associated fibroblasts

Cancer-associated fibroblasts (CAFs) are a heterogeneous population that has historically been treated as a single immunosuppressive compartment. Single-cell transcriptomics has resolved CAFs into multiple distinct subtypes with opposing functional roles: myofibroblastic CAFs (myCAFs) that are associated with matrix deposition and physical exclusion of immune cells from the tumor core; inflammatory CAFs (iCAFs) that secrete cytokines supporting immunosuppression; and antigen-presenting CAFs (apCAFs) that interact directly with T cells. The relative abundance of these subtypes is a determinant of immune exclusion and immunotherapy resistance.

Spatial tumor architecture

The spatial organization of the TME adds critical information that cell-type composition alone cannot capture. Tumor-immune boundary topology — whether immune cells infiltrate the tumor core or are excluded to the periphery — predicts immunotherapy response independently of immune cell abundance. Spatial transcriptomics maps these organizational features onto tissue architecture, enabling quantification of immune infiltration patterns, distance from tumor cells to immune populations, and the co-localization of signaling pairs.

VisiumHD provides whole-transcriptome coverage across the full tissue section at near-cellular resolution, enabling unbiased discovery of spatially variable genes and cell type distributions. For targeted validation at higher resolution, Xenium maps individual transcripts at subcellular resolution within a defined gene panel. The recently announced Atera platform (10x Genomics, announced April 2026, shipping H2 2026) will combine whole-transcriptome coverage with in situ single-cell resolution, collapsing the tradeoff between VisiumHD and Xenium into a single instrument. 3DG will onboard Atera as it becomes commercially available.

Multicellular programs and cross-condition comparison

Individual cell type analysis misses coordinated programs that span multiple cell types simultaneously. DIALOGUE identifies multicellular programs — sets of genes that are co-regulated across different cell types within the same spatial niche — capturing emergent properties of the TME that are invisible in single-cell-type analyses. MultiNicheNet (v2.0, 2024) enables systematic comparison of cell type compositions and intercellular communication programs across conditions, treatment time points, or responder versus non-responder groups, with prioritization of differential ligand-receptor interactions based on statistical evidence across samples.

Immune checkpoint and treatment response

Mechanistic understanding of why tumors respond or resist immune checkpoint blockade is one of the primary applications of single-cell transcriptomics in oncology. The field has moved beyond simple marker-based characterization toward systematic comparison of pre- and post-treatment samples, identification of clonal T cell expansion, mapping of resistance mechanisms including upregulation of alternative checkpoint pathways and recruitment of immunosuppressive populations, and spatial analysis of treatment-induced TME remodeling.

For pharmaceutical and biotech clients working on immuno-oncology programs, 3DG applies these analytical frameworks to clinical samples and patient-derived material with full attention to data confidentiality and compliance with applicable data use agreements.

Tumor types and tissue considerations

Single-cell and spatial transcriptomics have been applied across virtually all solid tumor types, with particularly deep literature in melanoma, lung adenocarcinoma, breast cancer, colorectal cancer, glioblastoma, pancreatic ductal adenocarcinoma, and hepatocellular carcinoma. Each tissue type presents distinct technical challenges: brain tumors require nuclei isolation from frozen tissue; pancreatic cancer has notoriously low RNA quality from FFPE blocks; melanoma has high immune infiltration that can overwhelm malignant cell signal. 3DG has experience adapting workflows to these tissue-specific challenges.