Comparative Landscapes of the Tumor Immune Microenvironment Across Cancer Types

Tumor immune microenvironment (TIME) show significant differences in different cancer types. The heterogeneity of TIME in different cancer types and its impact on immunotherapy response and prognosis is a complex and important research area. This article will explore this issue in detail from multiple perspectives.

Hot vs. Cold Tumors of the TIME

Definitions and representative tumor types

In the tumor immune microenvironment, immune phenotypes are roughly divided into two categories: "hot tumors" and "cold tumors". This classification is based on a variety of factors, especially the degree of immune cell infiltration and tumor mutation burden (TMB). More specifically, based on the spatial distribution of cytotoxic immune cells in tumor tissues, tumors can be further divided into three main phenotypes: immune-inflammatory, immune-rejection, and immune-desert (see Figure 1). Immunoinflammatory tumors, commonly known as "hot tumors", such as melanoma and lung cancer, are characterized by abundant T cell infiltration, upregulated IFN-γ signaling, increased PD-L1 expression, and high TMB. These characteristics enhance the ability of the immune system to recognize and eliminate cancer cells. Clinical studies have shown that such tumors usually respond better to treatment with immune checkpoint inhibitors (ICIs) and have better therapeutic effects (Galon et al. 2019) . In contrast, immune-rejection and immune-desert tumors are collectively referred to as "cold tumors", such as breast cancer and primary liver cancer, which are characterized by CD8+ T cells being mainly confined to the tumor margins, with minimal or even absent infiltration inside the tumor and adjacent tissues. In addition, cold tumors are often accompanied by decreased TMB, reduced expression of MHC-I and PD-L1, and accumulation of immunosuppressive cells such as regulatory T cells (Tregs), reflecting their inherent deficiency in anti-tumor immune response. This immune deficiency is an important reason why some patients do not respond well to ICIs therapy. Therefore, it is of great significance to develop specific immune checkpoint blockade strategies based on the unique characteristics of cold tumors to improve clinical treatment effects.

Figure 1. Tumor immune phenotypes.

How do cold tumors evade immune surveillance?

Zhang et al. proposed the concept of effector immune cell deployment (EICD ) , which includes the success of immune cell priming, trafficking, infiltration, survival, and recognition and killing of tumors. Failure of any link in EICD will lead to tumor escape ( Zhang et al. 2022) .

Lack of immunogenicity

Tumor-specific antigens are key factors in activating T cell immune responses. Neoantigens are mainly derived from non-synonymous mutations in tumor cells, and their number is measured by TMB. The generation of neoantigens and their presentation through HLA molecules jointly determine the immunogenicity of the tumor. Patients with high TMB usually respond better to immune checkpoint blockade (ICB) therapy. Tumor cells reduce the generation of neoantigens or escape immune surveillance through a variety of mechanisms, including: First, at the genomic level, about 17% of tumors have HLA-I gene loss, which is a common mechanism for escaping neoantigen immune surveillance; second, epigenetic modification silences HLA-I expression; third, in transcriptional regulation, specific transcription factors negatively regulate MHC-I expression; fourth, promote the degradation of MHC-I through autophagy and proteasome pathways, thereby reducing tumor immunogenicity.

Antigen presentation defect

Antigen presenting cells (APCs), mainly dendritic cells (DCs), play an important role in initiating anti-tumor immune responses by taking up and processing tumor antigens, cross-presenting and activating naive T cells. DCs are recruited by chemokines, such as CCL5 and XCL1, and activated by so-called "danger signals". These "danger signals" include pathogen-associated molecular patterns (PAMPs), such as cytoplasmic DNA and RNA, and damage-associated molecular patterns (DAMPs), including ATP, calreticulin (CRT) and HMGB1, which are released from DNA damage and immunogenic cell death of tumor cells. Tumors hide "danger signals" through multiple mechanisms to avoid DC phagocytosis.

Impaired T cell trafficking and infiltration

After priming and activation of T cells in lymph nodes, T cells are recruited and infiltrated into tumors through a variety of cytokines, chemokines, and adhesion molecules, which are tightly coordinated processes including trafficking, extravasation of tumor blood vessels, and penetration into the stroma. Chemokines and their corresponding receptors determine T cell recruitment, while blood vessels and tumor stroma are physical barriers to T cell penetration and survival. The cellular stroma, which is mainly composed of fibroblasts and extracellular matrix, can prevent T cells from infiltrating into the tumor core.

T cell dysfunction

T cell dysfunction or exhaustion is manifested by downregulation of effector molecules and upregulation of inhibitory immune checkpoint receptors such as PD-1, TIM3, TIGIT, and LAG3.

T cell death

Tumor endothelial cells selectively kill CD8+ T cells but not Treg cells through FasL-Fas interaction; T cells undergo ferroptosis.

Cross-Cancer Analysis of TIME Profiles

Tumor-infiltrating B cells play an important role in tumor development, progression, and prognosis. However, relatively little research has been done on the B cell component of the TIME. et al. analyzed and identified 10 different B cell and plasma cell subsets in seven human tumor types, each with its own unique gene expression profile and biological function, thus revealing the complexity of B cells in tumors. Trajectory analysis revealed multiple developmental pathways from naive B cells to plasma cells. In addition, the study found that the abundance of some B cell subsets was significantly correlated with the response to immune checkpoint inhibitor treatment, suggesting that they may play an important role in tumor immunotherapy. Further combined with spatial transcriptome data, the spatial localization of specific subsets and the ligand-receptor interaction between B cells and T cells were verified, strengthening their potential function in immune regulation.

Figure 2. A pan-cancer sc RNA-seq atlas of intratumoral B cells.

Public Databases and Computational Tools for TIME Characterization

TCGA immune analysis

The Cancer Genome Atlas (TCGA) database provides a wealth of immune-related omics data for the study of the tumor immune microenvironment, including RNA-seq, protein expression, and mutation spectra. Researchers can use the TCGA data portal to screen specific cancer types and sample types according to research needs and obtain the required data. For RNA-seq data, standardized processes can be used for processing to analyze gene expression levels; protein expression data can be obtained through techniques such as immunohistochemistry; mutation spectrum data can help identify gene mutations in tumor cells. In order to more comprehensively analyze the immune microenvironment, these multi-omics data need to be integrated. Bioinformatics methods, such as constructing gene co-expression networks and performing pathway enrichment analysis, can be used to explore the intrinsic connections between data.

TCGA immune scores, such as the ESTIMATE algorithm, evaluate the characteristics of the tumor immune microenvironment by calculating the ratio of immune cells and stromal cells in tumor samples. Its calculation logic is based on gene expression data, selects gene sets related to immune cells and stromal cells, and calculates the score through a specific algorithm. Multiple clinical studies have verified the effectiveness of the ESTIMATE algorithm, which is closely related to the prognosis and treatment response of tumor patients. For example, in breast cancer studies, patients with high ESTIMATE scores have a better prognosis and a better response to immunotherapy.

Figure 3. The violin plot shows the difference in the ESTIMATE, Immune, and Stromal Scores.

Integration of bulk and single-cell data

Bulk RNA-seq provides an average measurement of gene expression across a population of cells, presenting a comprehensive view of the transcriptional landscape, making it widely used in tumor diagnosis, discovering prognostic biomarkers, identifying novel gene fusions, and guiding therapeutic interventions. scRNA-seq performs sequencing at the single-cell level, which can capture subtle differences between cells, reveal cellular heterogeneity, identify rare cell subpopulations, and changes in transcriptional states . Although scRNA-seq has the advantage of high resolution, it also faces challenges in experimental cost, data quality, and analysis complexity. For example, the sequencing depth of scRNA-seq is generally not as good as that of bulk RNA-seq, and data processing and analysis are more complex. In addition, scRNA-seq may result in the loss of spatial and/or temporal information when extracting RNA from a single cell, while spatial transcriptomics (ST) can simultaneously obtain gene expression and spatial location information, making up for this deficiency .

Application in Precision Oncology

Predicting response to checkpoint inhibitors

In precision oncology, building a predictive model based on TIME features is crucial for accurately predicting the therapeutic response to ICIs. Currently, tumor infiltrating lymphocytes (TILs), programmed death ligand 1 (PD-L1) expression, and TMB are key TIME features for building predictive models.

When building a prediction model, we first collect TILs, PD-L1 expression, and TMB data from a large number of patients and combine them with their ICI treatment response. These data are analyzed and modeled using machine learning algorithms such as logistic regression and support vector machines. For example, the logistic regression algorithm can be used to determine the weight of each feature for ICI response, thereby building a prediction model.

Tumor immune subtyping as a biomarker

Tumor immune subtypes can be used as important biomarkers in precision oncology to guide the selection of treatment strategies. Common immune subtypes, such as C1-C6 subtypes, each have their own unique immune characteristics, which match different treatment strategies. For example, the C1 (wound healing type) subtype has a high immunosuppression, and the use of ICI alone may not be effective. At this time, ICI combined with chemotherapy can be considered to enhance immune activation and improve the treatment effect. The C2 (IFN-γ dominant type) subtype has a high immune cell infiltration and immune activation state, and may have a better response to ICI monotherapy. Therefore, this type of patient can give priority to ICI monotherapy. The immune microenvironment of the C3 (inflammatory type) subtype is rich in inflammatory factors, and ICI combined with targeted therapy may be a suitable choice. Targeted therapy can target specific molecular targets of tumor cells and jointly regulate the immune microenvironment with ICI to enhance the anti-tumor effect. The immune cell function of the C4 (lymphocyte depletion) subtype is suppressed, and ICI combined with immunomodulators can be tried to restore the activity of immune cells. Immunomodulators can regulate the function of immune cells and enhance their ability to recognize and kill tumor cells. The C5 (immune silent) subtype has less immune cell infiltration and low immune activation. Chemosensitization may be the main treatment strategy, which improves the immunogenicity of tumor cells through chemotherapy and creates conditions for subsequent immunotherapy. The C6 (TGF-β-dominant) subtype has a high immunosuppression, and ICI combined with anti-TGF-β treatment can be considered to relieve immunosuppression and enhance the effect of immunotherapy .

CD Genomics offers Tumor Microenvironment Profiling services for comprehensive immune phenotyping, biomarker discovery, and tumor-immune interaction studies, enabling researchers to explore the complexity of the tumor immune landscape in various research contexts.

Conclusion

The TIME is a key link between tumor development and immune response. It shows a high degree of heterogeneity among cancer types and has a profound impact on the effect of immunotherapy. This article systematically compares the TIME characteristics in different cancer types and integrates multi-omics data to reveal the immune infiltration pattern, immunosuppression mechanism and its relationship with treatment response, highlighting the essential differences between "hot tumors" and "cold tumors" in immune response. These findings not only deepen our overall understanding of the tumor immune ecology, but also provide a theoretical basis for the development of more targeted immunotherapy strategies and accurate prediction models. In the future, comprehensive and dynamic characterization of TIME characteristics will be a key direction to promote the individualization and precision of tumor immunotherapy.