Sidedness matters: single-cell perspectives on left- and right-sided colorectal cancer

Colorectal cancer (CRC) remains one of the leading causes of cancer-related morbidity and mortality worldwide. Tumor sidedness, distinguishing left- (LCRC) and right-sided (RCRC) cancers, has emerged as a critical clinical determinant, influencing patient prognosis and therapeutic response. However, the cellular and molecular mechanisms underlying these differences remain poorly understood. Recent advances in single-cell RNA sequencing (sc-seq) provide high-resolution insights into CRC heterogeneity, revealing distinct tumor, immune, and stromal cell populations and their context-specific interactions. In this review, we synthesize sc-seq studies that dissect the molecular programs driving progression, therapy resistance, and metastasis in CRC. We highlight malignant subclusters characterized by metabolic reprogramming and spatially organized oncogenic signaling; specialized immune cell states, including macrophage subsets, exhausted T cells, and mast cells, that shape tumor immunity; and stromal elements such as cancer-associated fibroblasts and endothelial tip cells that remodel the extracellular matrix, promote angiogenesis, and foster immune evasion. Importantly, sc-seq demonstrates that LCRC and RCRC represent distinct multicellular ecosystems with differential immune recruitment and stromal signaling, underscoring the need for sidedness-informed therapeutic strategies. We propose that future interventions should target cell-cell communication networks and spatially defined tumor–microenvironment interactions to overcome heterogeneity and improve clinical outcomes.

1 Introduction

Colorectal cancer (CRC) ranks as the third most common and second most fatal cancer globally, with approximately 1.9 million new cases and 905,000 deaths recorded in 2022 (Bray et al., 2024). At diagnosis, one-fifth of newly diagnosed patients present with metastatic CRC (mCRC) (Biller and Schrag, 2021). Moreover, up to 50% of patients initially diagnosed with localized CRC later develop metastases (Ciardiello et al., 2022). While the prognosis for mCRC has improved due to new treatment methods and the diagnostic process, the five-year survival rate remains critically low (under 13%) (Siegel et al., 2025). This creates a pressing clinical need to identify biomarkers and therapeutic targets to improve the effectiveness of treatment and predict the course of the disease.

This search is significantly complicated by pronounced tissue heterogeneity in CRC. Indeed, CRC is known to be highly heterogeneous in both primary tumors and metastatic lesions, which arises from multiple sources (Blank et al., 2018). Clinical studies have revealed significant differences in the incidence, outcomes, molecular and genetic profiles between anatomically distinct left-sided and right-sided CRC (LCRC and RCRC), first described by Bufill (1990). Although there is no clear distinction on the classification of CRC into LCRC and RCRC, the common definition is that RCRC refers to cancer proximal to the splenic flexure, whereas LCRC refers to cancer distal to the splenic flexure. This cutoff point is widely applied due to the embryonic development of the distal one-third from the hindgut, and the two-thirds of the transverse colon from the midgut. Additionally, the vascular supply is used to mark the embryologic origin with the superior mesenteric arteries supplying the midgut and the inferior mesenteric arteries supplying the hindgut (Ahmad Zawawi and Musa, 2022). Moreover, in some studies rectal carcinomas are classified together with LCRC due to similar embryological origin (Chen et al., 2021). Notably, despite the absence of defined molecular markers identifying phenotypically distinct CRC subtypes, classification by site of origin remains clinically useful - though not indicative of mutually exclusive signaling pathways (Cervantes et al., 2023). For example, left- or right-sided tumor location may help guide the choice between anti-EGFR and anti-VEGF therapy. It has been established that LCRC with unmutated RAS is more responsive to anti-EGFR treatment than RCRC (Cervantes et al., 2023). In contrast, for RCRC, anti-VEGF therapy is generally recommended regardless of pan-RAS status (Abdel Hamid et al., 2025). Consequently, tumor sidedness is now recognized as a critical prognostic and therapeutic determinant in CRC.

CRC heterogeneity manifests not only through genetic, transcriptional and translational alterations within cancer cells but also from the diverse cellular and extracellular matrix (ECM) composition of the tumor microenvironment (TME) (Blank et al., 2018). Single-cell analyses reveal that RCRC and LCRC are distinct multicellular ecosystems, characterized by pronounced immunologic and stromal differences (Liu B et al., 2024). These differences have direct therapeutic relevance, as the TME is increasingly recognized as a driver of tumor progression and a target for novel interventions (Li et al., 2022a). Biomarker development and treatment strategies should therefore integrate both tumor sidedness and intratumoral cellular heterogeneity to optimize clinical outcomes.

Bulk-tissue sequencing, one of the methods traditionally employed for consequent biomarker search and discovery, provides an overview of the cancerous tissue state as a whole, but fails to resolve tissue heterogeneity. In contrast, single-cell sequencing (sc-seq) allows to overcome this limitation by enabling the analysis of genomic or transcriptional data at individual cell resolution (Sun et al., 2025). This approach helps to point out the differences within the diverse cell populations comprising a tumor. Widely applied in the field of cancer research (e.g., colorectal, breast, brain, liver and lung cancers (Tang et al., 2019)), sc-seq provides critical insights into tumor heterogeneity and signaling pathways, including those promoting tumor progression and metastasis.

A key application of sc-seq is the analysis of intercellular signaling–a primary coordinator of cellular activities - predicted from gene expression data. In this context, the analysis of ligand-receptor pairs has become a useful technique: by detecting the coordinated expression of corresponding genes, researchers can characterize cellular communication and propose its potential influence on tumor growth and progression (Armingol et al., 2021). A typical workflow for such an analysis includes data acquisition, the construction of cell-cell communication networks, their computational inference, and, critically, experimental validation (Shao et al., 2020). Currently, several ligand/receptor tools exist to facilitate this process. The most common group of tools, including CellPhoneDB and CellChat, utilize curated ligand-receptor databases to assess co-expression across existing cell populations, employing permutation tests to identify significant interactions between cell types. The operational principles of current ligand-receptor analysis algorithms are reviewed in several publications (Armingol et al., 2021; 2024; Ma et al., 2021). Another group of tools includes NicheNet (Browaeys et al., 2020). Unlike methods that rely solely on ligand-receptor co-expression, NicheNet assumes that signaling to receiver cells causes downstream effects, such as changes in transcription factor activity. It therefore incorporates prior knowledge of signaling pathways to measure communication based on the enrichment of a receiver cell’s target genes. This approach effectively prioritizes cell-cell communication inferences based on downstream biological effects, thus helping to reduce the large number of interactions identified by other methods. However, this focus may also overlook genuine interactions that lack the expected downstream signature in the database.

Despite their utility, such computational methods have several limitations. A primary concern is their foundational assumption that transcriptomic data is a suitable proxy for cell-cell communication events. In reality, cells communicate via proteins (not RNA), and signaling is spatially constrained - neither of which can be fully captured by sc-seq. Furthermore, while these tools rely on curated ligand-receptor databases, true biological communication extends beyond the database-listed interactions. Consequently, the interactions they identify remain theoretical, making experimental validation essential. For these reasons, although these methods are widely applied in single-cell research due to their accessibility, their results must be interpreted with caution and confirmed experimentally.

This review synthesizes sc-seq findings in CRC to provide an overview of the molecular mechanisms underlying tumor therapy resistance, progression and metastasis in both overall CRC and the LCRC/RCLR subtypes, with an emphasis on context-specific interactions of cell populations inside tumors. Critically, sc-seq, in accordance with previous research, has identified that LCRC and RCRC are differing multicellular ecosystems, each defined by unique and varied immune and stromal signaling patterns, supporting the need for sidedness-informed therapeutic strategies. Consequently, further research into these spatially-defined intercellular communication networks and their subsequent therapeutic targeting holds promise for improving the clinical outcomes in CRC patients.

2 Prevalent cell types in colorectal cancer identified by single-cell sequencing

CRC is known to be comprised from a variety of cells participating in processes such as angiogenesis, tumorigenesis, immune evasion and metastasis (Fridman et al., 2020). Understanding the spatial and temporal distribution of these cellular components is therefore critical for developing precise therapeutic strategies. Sc-seq studies have recently enabled high-resolution mapping of CRC’s cellular landscape. The distribution of cells in overall and sided CRC is presented in Table 1 and will be discussed in more detail in the upcoming sections.

Table 1

Table 1. Cellular heterogeneity in overall and sided CRC.

2.1 Overall colorectal cancer cell composition

As summarized in Table 1, the TME of both overall CRC and sided subtypes (LCRC/RCRC) comprises heterogenous cell populations. These include epithelial cells, immune populations (T cells, B cells, monocytes/macrophages and mast cells) and stromal cells (cancer-associated fibroblasts (CAFs), endothelial cells and pericytes).

It should be noted that epithelial cells consistently dominate overall CRC tumors, although their abundance varies widely (24%–67%) (Lu et al., 2025; Qin et al., 2023; Wang et al., 2022). This variation likely reflects biological context (such as the tumor stage) and/or methodological differences in sample preparation and processing. Collectively, the analyzed studies demonstrate the diverse cellular composition of tumors and point to significant immune and stromal infiltration. Thus, a substantial fraction of the TME is composed of immune cells, with infiltrating T cells representing the predominant immune subset (8%–32% abundance). Stromal components, such as CAFs and endothelial cells, implicated in metastasis and therapy resistance are universally present, albeit at lower frequencies (typically <10% abundance) (Lu et al., 2025; Qin et al., 2023; Wang et al., 2022). Rare cell populations, such as mast cells and pericytes, are scarce but may exert a significant functional influence within the tumor.

Therefore, this cellular architecture of CRC highlights three potential therapeutic targets: 1) malignant epithelial populations, 2) immunomodulatory cell networks and 3) stromal mediators. Frequently, targeting a specific cell niche results in only partial, or even no, response to therapy (Fridman et al., 2020; Gallo et al., 2021; Leonard et al., 2021; Meyiah et al., 2025; Sphyris et al., 2021; Tauriello and Batlle, 2016). This limited efficacy may result from complex signaling pathways activated via cell-cell interactions. Sc-seq brings novel insights not only into cell-cell interactions, but also into the differences in such interactions across distinct anatomical locations. Consequently, targeting the newly identified pathways activated within these cellular components may represent a promising strategy to overcome therapy resistance and metastasis in CRC.

2.2 Left/right colorectal cancer cell composition

While LCRC and RCRC share the core cellular components of total CRC, their cell distribution differs (Table 1). Notably, RCRC demonstrates increased epithelial cell abundance in comparison with LCRC (20%–38% vs. 11%–30%). Such a difference may originate from their distinct embryonic development–while RCRC is midgut-derived, LCRC originates from the hindgut (Abdel Hamid et al., 2025). Interestingly, evidence from Drosophila models, which share structural and functional similarities with the mammalian colon and small intestine (Beumer and Clevers, 2021; Colombani and Andersen, 2020), has revealed that the midgut is characterized by a high rate of intestinal stem cell division and turnover, whereas the hindgut lacks constitutively active stem cells (Fox and Spradling, 2009; Micchelli and Perrimon, 2006). These findings may shed light onto the difference between TME in sided CRC.

Sidedness also shapes the immune landscape: according to sc-seq data RCRC demonstrates a slight B-cell enrichment (up to 26% vs. 20% in LCRC), while LCRC exhibits T cell dominance (up to 65% vs. 49% in RCRC). Interestingly, a recent study has revealed that patients with RCRC, but not LCRC have a more favorable clinical outcome with a higher density of tumoral B cells (Berntsson et al., 2016), suggesting distinct B-cell recruitment mechanisms. On the other hand, T cell populations can differentially influence CRC cell survival, as their effects are dictated by functionally distinct subsets; some are tumor-suppressive while others promote tumor progression (Guo et al., 2020). This suggests possible distinct molecular mechanisms for immune cell recruitment in the tumor.

The stromal components are present at similar levels in both subtypes (Table 1), constituting no more than 4% of the total cell population. Such a distribution points to their importance in the signaling of CRC tumors regardless of the anatomical site, although cellular differences and cell-cell interactions in stromal populations may be altered.

Collectively, the analyzed studies support the characterization of LCRC and RCRC as distinct multicellular ecosystems and suggest that left- and right-sided tumors employ different molecular mechanisms for immune cell recruitment.

3 Insights from single-cell sequencing in overall colorectal cancer

While the clinical significance of sidedness is of importance, single-cell technologies first aimed to clarify the profound heterogeneity inherent to all CRC tumors. Thus, this section of our review focuses on the foundational insights from sc-seq into the universal mechanisms of tumor therapy resistance, progression and metastasis in overall CRC, establishing the core cell types and corresponding interactions that are then differentially co-opted in left- and right-sided CRC tumors. Pathways specific to tumor sidedness will be discussed in subsequent sections.

3.1 Single-cell sequencing to elucidate tumor therapy resistance and progression in overall colorectal cancer

Despite advances in the treatment of CRC including chemotherapy, targeted therapy and immunotherapy, many patients exhibit either intrinsic or acquired resistance to tumor treatment. This ultimately leads to limited therapeutic efficacy and high rates of tumor recurrence. This challenge stems from the considerable diversity of resistance mechanisms and tumor progression pathways in CRC, which involve complex signaling interactions among tumor, immune, and stromal cells (Table 2). As the detailed study of these pathways is ongoing, sc-seq has provided insights into their function. The contribution of single-cell technologies will be detailed throughout this review.

Table 2

Table 2. Cell-cell communication in colorectal cancer: predicted ligand-receptor interactions and their functional roles in overall and sided colorectal cancer.

3.1.1 Mechanisms of tumor cell signaling leading to colorectal cancer progression

The progression of CRC is driven not only by the intrinsic properties of tumor cells, but also involves their dynamic interactions with the TME. The tumor mass itself is a complex ecosystem, primarily composed of tumor cells, with stromal cells providing essential support (Liu and Dilger, 2025). Immune cells within this niche add another layer of complexity, exerting opposing influences on tumor progression through pro- and anti-tumor mechanisms. While foundational therapies like surgery, radiotherapy, and chemotherapy directly target tumor cells, the disruption of the TME has become a major therapeutic focus. The pursuit of more effective treatments is now being guided by insights from sc-seq and spatial transcriptomic studies, which have begun to unravel these complex mechanisms (Table 3).

Table 3

Table 3. Transcriptomic features of tumor cells in overall and left- and right-sided CRC progression revealed by single-cell sequencing and spatial transcriptomics.

An example of a therapeutically targetable pathway intrinsic to tumor cells is the peroxisome proliferator-activated receptor (PPAR) signaling pathway. Wang et al. demonstrated marked upregulation of the PPAR signaling in tumor epithelial cells relative to normal tissue (Wang et al., 2022). Functionally, inhibition of this pathway suppressed growth and activated apoptosis in patient-derived CRC organoids - an effect comparable to the standard chemotherapy drug 5-fluorouracil (5-FU). This suggests PPAR is a promising therapeutic target for CRC tumor cells, particularly relevant given that, despite substantial preclinical evidence, PPAR inhibitors have not yet seen widespread effective use in CRC patient treatment. Beyond obtaining data in support of well-studied pathways, new signaling networks have been described with the help of single-cell transcriptomic technologies. One such network was identified by utilizing bulk RNA sequencing and sc-seq findings to build a four-gene prognostic model using membrane-tension related genes (MTRGs) in patients from the TCGA cohort (Li et al., 2022b). Among them, the expression of tissue inhibitor of metalloproteinase 1 (TIMP1) notably increased during epithelial pseudotime trajectory analysis, potentially linking it to CRC progression. Furthermore, high-risk TCGA patients showed increased sensitivity (lower IC50) to the drugs elesclomol, shikonin, and bryostatin-1 - agents predicted to have high affinity for TIMP1. Given that previous studies have implicated TIMP1 in promoting tumor proliferative abilities and metastasis both in vitro and in vivo (B. Ma et al., 2022; Rao et al., 2022; Song et al., 2016), the findings position the TIMP1-centered signaling axis as a potential therapeutic target in CRC.

The use of single-cell technologies extends beyond identifying single potential therapeutic targets/signaling pathways - they allow for the identification of functional heterogeneity within the tumor cell population itself. As opposed to being a uniform mass, CRC tumors contain distinct subclusters of malignant cells characterized by differing molecular programs driving tumor progression. For instance, sc-seq has uncovered malignant cell subclusters enriched in genes associated with histone deacetylase (HDAC) inhibitor response (ATF3, CAV1), inflammation (LXN, PGM1), and hypoxia-linked tumor metastasis (WSB1) (Xiao et al., 2024), highlighting their importance in tumor progression. This cellular diversity, which allows the tumor to adapt and resist therapy, is further organized in a specific spatial context. Spatial transcriptomics of CRC tumor cryosections revealed high expression of thymosin beta-4 (TMSB4X) in tumor cells - a protein involved in actin binding, cell migration, and inflammation (Gemoll et al., 2015; Nemolato et al., 2012). Tumor subregions further exhibited upregulated pathways related to focal adhesion dynamics, ECM, metabolism, and hypoxia response, implicating these processes in tumor progression. This spatial organization is additionally characterized by active signaling crosstalk, such as the crosstalk between stromal and tumor regions via the tumor progression and immune evasion (C5AR1-RPS19) signaling axis. These findings demonstrate that CRC progression is driven by functional heterogeneity among malignant cells, their spatially organized molecular gene expression upregulation (TMSB4X in tumor cells), regional activation of key oncogenic pathways and active signaling crosstalk between tumor and stromal compartments.

Metabolic reprogramming, specifically regarding nucleotide synthesis, was identified by single-cell analysis as a driver of CRC progression, with elevated levels detected in tumor versus normal tissue (Liu J et al., 2024). Tumor cells exhibiting high nucleotide metabolism (termed NU^high cells) showed extensive communication with stromal and immune cells, acting as coordinators of the tumor-stromal-immune interactions (Table 2). Interestingly, NU^high cells signal via macrophage migration inhibitory factor (MIF) and secreted phospholipase A2 (PLA2) to engage TNF receptors/integrins on stromal/immune cells, whereas NU^high cells receive signals from fibroblast- and macrophage-secreted B cell-activating factor (BAFF) through the transferrin receptor (TFRC) (Liu J et al., 2024). Spatial transcriptomics, in turn, confirmed elevated nucleotide metabolism in CRC tumors alongside increased EGFR, hypoxia, MAPK, and TGF-β signaling. Importantly, the nucleoside diphosphate kinase A (NME1) gene was identified as a key regulator of nucleotide metabolism. Functional validation of such findings in CRC xenograft models demonstrated that NME1 inhibits migration and reduces stemness, confirming its importance in tumor biology. Thus, nucleotide metabolism is not only a metabolic hallmark of CRC but acts as a central focal point that coordinates tumor-stromal-immune signaling (via MIF/PLA2/BAFF/TFRC) and spatially co-localizes with oncogenic pathways. The identification of NME1 as a key regulator whose activity suppresses migration and stemness highlights nucleotide metabolism (and NME1 specifically) as a therapeutically relevant pathway in CRC.

In total, overall CRC progression is driven by dysregulated signaling within tumor cells that combine both intrinsic oncogenic pathways and the intercellular signaling within the TME. The newly identified important mechanisms include, firstly, metabolic reprogramming with elevated nucleotide metabolism (via NME1) that allows the communication of tumor-stromal-immune cells through both outgoing (MIF-PLA2) and incoming (BAFF-TFRC) signals, all while being spatially colocalized with EGFR/hypoxia/MAPK/TGF-β pathways. Secondly, subregions within the tumor demonstrate spatially organized signaling through PPAR, focal adhesion/ECM remodeling and hypoxia-driven metastasis. Lastly, the novel mechanisms include functional heterogeneity where malignant subclusters in CRC tumors drive progression via HDAC inhibitor-responsive genes (ATF3/CAV1), inflammatory mediators (LXN/PGM1, TMSB4X) and the stromal-tumor crosstalk for tumor progression and immune evasion (C5AR1-RPS19 axis). Targeting the mentioned pathways, in particular PPAR, TIMP1, and NME1-regulated pathways, serves as a promising strategy for CRC patient treatment.

3.1.2 Mechanisms of immune cell signaling in CRC progression

The immune landscape within CRC is a complex network of specialized cells whose functions are defined by their identity, location within the tumor and cellular communication networks. Single-cell technologies are helping to navigate through this complexity, revealing how specific immune subsets influence progression and helping to guide the direction for novel therapeutic strategies development (findings summarized in Table 4).

Table 4

Table 4. Transcriptomic features of immune cells in overall and sided CRC progression revealed by single-cell sequencing and spatial transcriptomics.

The role of tumor-associated macrophages (TAMs) in CRC is defined by functional heterogeneity in a number of studies employing sc-seq. Zhang et al. identified two transcriptionally distinct TAM populations with opposing roles: 1) C-chain polypeptide of serum complement subcomponent C1q-positive (C1QC⁺) TAMs expressing genes related to phagocytosis and antigen presentation and predicted to interact primarily with immune cells (particularly T cells) via enriched immune cell recruitment CXCL10-CXCR3 signaling and 2) osteopontin-positive (SPP1⁺) TAMs upregulating regulators of angiogenesis and metastatic pathways, engaging with CAFs and myofibroblasts and participating in ECM degradation SDC2-MMP2 signaling crosstalk with endothelial cells (Zhang et al., 2020). This TAM plasticity represents a clinical challenge, as demonstrated in Renca tumor–bearing mice, where anti-colony stimulating factor 1 receptor (CSF1R) therapy - which disrupts a pathway essential for macrophage development, differentiation, and survival (O’Brien et al., 2021; Wen et al., 2023) - reduced total TAM numbers but revealed a treatment-resistant TAM subset (Zhang et al., 2020). These resistant TAMs preferentially express angiogenic (VEGFA) and immunosuppressive (CD274, ARG1) genes and promote tumor vascularization. Hence, the TME in CRC is composed of specialized TAM subpopulations with opposing roles: immune coordination (C1QC⁺) with anti-tumor functions versus stromal activation (SPP1⁺) involved in tumor progression signaling. Notably, resistance to TAM-depleting therapies arises from a vasculature-promoting TAM subset, suggesting the need for macrophage subtype-specific targeting strategies in CRC treatment and, potentially, adjust therapy based on tumor sidedness. Further complexity of the immune cell niche is revealed by the gut’s highly organized macrophage network, where distinct subsets are specific to their anatomical niche (Domanska et al., 2022). Lamina propria macrophages (LpMs) in the subepithelial region include proinflammatory and antigen-presenting/phagocytic subsets highly expressing proinflammatory (e.g., S100A9) and antigen presentation genes (e.g., HLA-DRA). Notably, MHC II-high LpM subsets interact with T cells via both immune cell co-stimulatory and co-inhibitory pairs like CD28/CTLA4-CD80/CD86 and recruit CD8⁺ T cells via chemokine CXCR3 ligands. Furthermore, submucosal macrophages (SmMs), particularly lymphatic vessel endothelial hyaluronan receptor 1-positive (LYVE1⁺) SmMs, exhibit low antigen presentation but high chemotactic and tissue-protective capacity. Finally, muscularis macrophages (MMs) express homeostatic genes and engage in the crosstalk with neurons: neuron-derived Notch ligands DLL1/DLL3/JAG2-NOTCH2 and IL34-CSF1R regulate MM survival and differentiation. Overall, such cellular organization suggests that the development of targeted therapeutic approaches accounting for not only immune cell subtypes, but additionally location-specific functions is preferred.

A promising alternative to broad macrophage depletion is the precision targeting of specific immunosuppressive receptors on TAMs. In support of this approach, Huang et al. reported that TAMs upregulate immunosuppressive receptors bearing immunoreceptor tyrosine-based activation motifs (ITAMs) (Huang et al., 2024). These include signal regulatory protein α (SIRPA), which showed significantly higher expression in tumor tissues versus paracancerous/normal tissues. SIRPA/Sirpα elevated levels are associated with tumor progression, as genetically engineered SIRPA^−/− mice were demonstrated to have developed smaller subcutaneous MC38 tumors and exhibited prolonged survival, while rechallenged tumors were completely rejected. Mechanistically, Sirpα knockout (KO) reprograms macrophages to recruit T cells to tumors via spleen tyrosine kinase/Bruton tyrosine kinase-dependent (SYK/BTK-dependent) C-C motif chemokine ligand 8 (CCL8) secretion, thereby activating immune response (Huang et al., 2024). These findings demonstrate that precision targeting of a subset of TAM receptors/signaling pathways can effectively remodel the TME, suggesting a potential therapeutic targeting strategy.

The T cell landscape in CRC is equally as complex, with specific subsets correlating with differing patient outcomes. On one hand, cytotoxic CD4⁺ T cells are characterized as cells expressing high levels of effector molecules and co-inhibitory molecules, such as programmed cell death protein 1 (PD-1), pointing to their tumor-suppressing functions (Masuda et al., 2022). Additionally, among CD8⁺ T cells, a Tem phenotype is associated with a less dysfunctional transcriptomic state and positive clinical outcomes; similar results are identified within Tregs population, where transcription factor Helios-positive (Helios⁺) Tregs are associated with improved patient outcomes and present in early-stage cancers. On the other hand, granulysin- and integrin αE-positive (GNLY⁺CD103⁺) T cells, found specifically within tumors (and absent in normal colon tissue), demonstrate a loss of cytokine production and show no correlation with positive outcomes in CRC. Furthermore, peripheral Tregs are associated with poor outcomes (independently of stage), express high levels of cyclic ADP ribose hydrolase (CD38) and lymphocyte-activation gene 3 (LAG3), and produce elevated IL-10, suggesting immunosuppressive activity.

Beyond the adaptive immune response, innate immune cells like mast cells (MCs) also play a significant role in CRC signaling. Sc-seq studies have identified that the activated-to-resting MC ratio is significantly altered in CRC, a pattern validated across multiple datasets (Xie et al., 2023). Functional-wise, gene set variation analysis (GSVA) analysis links activated MCs to immunostimulatory TNFα/NFκB signaling and MHC I/II enrichment, while resting MCs are associated with angiogenesis, a process important for tumor growth. Moreover, a high MC gene signature correlates with better survival in TCGA patients, underscoring the biological significance of these findings. When looking at cellular communication, activated MCs exhibit more intercellular interactions (including myeloid, endothelial, epithelial, fibroblast) than resting MCs, and ligand expression for the tumor progression inhibitor KITLG-KIT axis on interacting cells is identified higher for activated MCs. In support of these findings, in vitro experiments show that KITLG-activated mast cells demonstrate reduced proliferation and migration, suggesting that KITLG/KIT signaling may inhibit CRC progression and confirming its computational significance.

Importantly, the functional impact of immune cells in CRC is critically dependent not only on the differing subtypes and signaling employed, but additionally on the tumor’s spatial organization. Such cellular organization is shown to be influenced by the mismatch repair (MMR) status (Feng et al., 2024). Studies have revealed that proficient MMR (pMMR) tumors, which are typically poor immunotherapy responders, form an organized barrier separating tumor cells from stromal and immune regions. Deficient MMR (dMMR) tumors, on the other hand, show disorganized clusters. Further research uncovered proximity-dependent (<250 µm) chemokine ligand 13-positive (CXCL13⁺) T cell–lysosomal-associated membrane protein 3-positive (LAMP3⁺) dendritic cell (DC) interactions via PD-1/PD-L1, a signaling axis utilized by DC to attenuate T cell activation (Peng et al., 2020), occurring at the tumor-stroma boundaries in dMMR and anti-PD1-responding (partial response (PR)/complete response (CR)) tumors, highlighting distance-critical signaling (Table 2). These findings highlight the fundamental importance of not only intercellular communications, but additionally spatial interconnection between tumor cells and immune/stromal cells and the distance at which they are able to effectively signal.

Thus, the immune cell subsets display functional duality, which suggests that therapies tailored to tumor spatial location is an option. Spatial architecture fundamentally determines therapy response, and effective targeting requires precision against specific dysfunctional states and communication pathways within the tumor. In conclusion, the CRC immune landscape can be characterized as a dynamic ecosystem defined by functional duality as well as spatial context. The success of targeted approaches for CRC treatment (like anti-SIRPα) demonstrate that future treatment strategies should be precisely designed. Effective targeting may potentially include targeting of specific dysfunctional cell states (such as SPP1⁺ TAMs, peripheral Tregs), disruption of pro-tumor communication pathways (such as SPP1-CD44, CD74-MIF), and should account for the spatial context of the TME, including MMR status which shapes these cellular interactions. Moreover, a targeting approach based on tumor sidedness could significantly enhance their efficacy.

3.1.3 Mechanisms of stromal cell signaling in CRC progression

While immune checkpoint blockade has become a standard of care for many cancers alongside chemotherapy, the stromal niche of the TME remains a mediator promoting tumor growth, metastasis, and CRC therapy resistance (Zhao et al., 2023). Single-cell and spatial transcriptomics studies are now uncovering how specific stromal cell populations are involved in the aforementioned processes, revealing new biomarkers and therapeutic vulnerabilities (summarized in Table 5).

Table 5

Table 5. Transcriptomic features of stromal cells in overall and sided CRC progression revealed by single-cell sequencing and spatial transcriptomics.

The CAF population is not uniform, but rather consists of functionally divergent subsets that influence patient response to administered therapy. One such example involves the study of post-neoadjuvant chemotherapy (NAC) biopsies (Qin et al., 2023). Two clinically opposing CAF subsets a favorable subset (PI16⁺/SLIT2⁺/ADAM28⁺), enriched in responders (CR/PR), regulating cell differentiation and immunoregulation as well as a tumor-promoting subset (BMP4⁺/FAP⁺/MMP1⁺), enriched in non-responders (NR), upregulating collagens, matrix metalloproteinases (MMPs), and WNT signaling to drive ECM remodeling and progression. Functionally, fibroblast activation protein α-positive (FAP⁺) CAFs promote epithelial-mesenchymal transition (EMT) and therapy resistance via long non-coding RNA MIR4435-2HG, a molecule previously associated with cancer progression including CRC (Ghasemian et al., 2022), while peptidase inhibitor 16-/slit guidance ligand 2-positive (PI16⁺/SLIT2⁺) progenitor CAFs foster immunity and inhibit tumor progression by recruiting and activating T cells. This occurs through mechanisms like stromal derived factor 1 (CXCL12)-mediated recruitment of CD8⁺ Tem over Tregs and decorin (DCN)-enhanced antigen presentation by DCs for spatial T cell immunostimulatory activation. Thus, these findings establish specific CAF populations as biomarkers of clinical outcome and identify novel mechanisms and targets for modulating the TME to overcome therapy resistance.

In support of the aforementioned findings, a particularly aggressive pro-tumor CAF population - the collagen triple helix repeat containing 1-positive (CTHRC1⁺) fibroblasts - was identified: a tumor-exclusive subset whose abundance increases with cancer progression and correlates with poor patient prognosis (Lu et al., 2025). Functionally, this group is characterized by a potent ECM-remodeling signature (e.g., expression of POSTN, MMP11, TGF-β1) and exhibits heightened transcriptional activity in promoting EMT and angiogenesis. In an attempt to arrange the identified cell population/signaling characteristics spatially within the tumor, the authors employed spatial transcriptomics analysis, revealing that CTHRC1⁺ CAFs co-localize with malignant epithelial cells to form a specialized niche characterized by active EMT, high immunosuppression (featuring TREM2⁺/SPP1⁺ macrophages, Tregs, and Th17 cells), and exclusion of CD8⁺ T cells. Within this niche, they exhibit the strongest interactions with tumor cells, primarily through outgoing signals via WNT, Notch, and ECM-adhesion pathways (for example, COL1A1-ITGB1, LAMA1-ITGA3). Functionally, these pathways are known to enhance tumor cell migration and invasion through WNT5A-LRP6/FZD3 and THBS1-ITGAV ligand-receptor signaling pathways. Crucially, in vitro validation confirms that CTHRC1⁺ CAFs secrete WNT5A, which upregulates the mesothelin (MSLN) gene in CRC tumor cells - an effect abolished by WNT5A inhibition or CTHRC1 knockdown. Collectively, these findings establish that CTHRC1⁺ CAFs drive poor outcomes by creating an immunosuppressive, pro-invasive niche and directly promoting tumor aggressiveness via a CTHRC1-WNT5A-MSLN signaling axis.

The negative impact of pro-tumor CAFs is further supported by sc-seq studies revealing the extensive collaboration of CAFs with other cells within the TME. A potent stromal-myeloid network between FAP⁺ fibroblasts and SPP1⁺ macrophages is detected, with both populations linked to advanced stage, MSI-H status, and reduced patient survival (Qi et al., 2022). These cells co-localize in hypoxic tumor regions and are predicted to engage in direct adhesive interactions (e.g., COL1A1-ITGB1) and paracrine signaling via TGF-β, WNT5A-FZD2, and CCL3-CCR5 axes, promoting recruitment and pro-inflammatory activity of SPP1⁺ macrophages. Ultimately, this synergistic FAP⁺/SPP1⁺ interaction promotes an immunosuppressive, pro-tumorigenic niche that is linked to immunotherapy resistance. Thus, these findings highlight the stromal-myeloid crosstalk as a determinant of patient outcomes and a potential therapeutic target.

Such a collaborative cellular interaction network extends beyond CAFs to the tumor vasculature. Tumor endothelial cells (ECs) in CRC undergo transcriptional reprogramming toward a pro-angiogenic phenotype, upregulating VEGFA-VEGFR2 signaling and ECM genes (e.g., COL4A1, SPARC) while downregulating antigen presentation molecules (MHC I/II) (Xie et al., 2024). Among EC subsets, tip cells dominate this pro-angiogenic response, showing high expression of growth factors (PGF, ANGPT2) and receptors (VEGFR, FLT1). Notably, these tip cells not only dominate the angiogenic response, but are also the primary mediators of MHC loss on the tumor vasculature, thereby actively facilitating immune evasion. Mechanistically, their activity is sustained by a VEGFA–VEGFR2–ESM1 positive feedback loop (based on in vitro data): VEGFA - secreted by myeloid and epithelial cells - activates tip cell VEGFR2, promoting ESM1 release that subsequently upregulates VEGFA in tumor cells, sustaining angiogenesis and promoting tumor growth. These findings are further supported by spatial transcriptomics data, demonstrating the presence of tip cells within stromal VEGF signaling niches. Moreover, the clinical relevance of tip cells is underscored by their correlation with poor prognosis and their reduction following successful anti-PD-1 therapy. Hence, these findings highlight them as a promising therapeutic target.

In conclusion, sc-seq and spatial transcriptomic studies reveal that the CRC TME is regulated by functionally diverse stromal cells that critically influence the clinical outcomes of patients. Novel sc-seq findings demonstrate the functional differences in CAFs, with immunogenic subsets (e.g., PI16⁺/SLIT2⁺) promoting anti-tumor immunity and positive therapy response, while pro-tumor subsets (e.g., FAP⁺/CTHRC1⁺) drive ECM remodeling, immunosuppression, and invasion through pathways like WNT and TGF-β. Importantly, pro-tumor CAFs form spatial niches with immunosuppressive SPP1⁺ macrophages and pro-angiogenic endothelial tip cells, collectively promoting an immune-excluded, pro-metastatic TME. The signaling within this stromal network - mediated by ligand-receptor pairs promoting cell migration and invasion such as WNT5A-LRP6 and COL1A1-ITGB1 - correlates with disease progression, immune evasion, and resistance to chemo- and immunotherapy. Thus, targeting these stromal interactions represents a promising strategy for overcoming treatment resistance and tumor progression in CRC.

3.2 Single-cell sequencing for analysis of total CRC metastasis mechanisms

Despite extensive research in the field of CRC, distant metastasis remains the leading cause of death and a substantial clinical challenge. Similar to primary CRC tumors, metastases are composed of and signaled by not only cancer cells, but additionally the TME. A key driver of metastasis is the formation of a “pre-metastatic niche” (PMN) (Li et al., 2025). This PMN is characterized by several features, including immunosuppression, inflammation, angiogenesis/vascular permeability, lymphangiogenesis, organogenesis, and metabolic reprogramming. Together, these factors enable the seeding and survival of disseminated tumor cells. Therefore, signaling from the TME is as crucial as signaling from cancer cells in promoting CRC metastasis. Given the importance of this process, we have dedicated a section of this review to compiling how single-cell technologies are helping to elucidate its underlying mechanisms for individual cell types and signaling hubs in tumor cells and the TME.

3.2.1 Mechanisms of tumor cell signaling in CRC metastasis

Upon colonizing the liver, CRC cells transcriptionally reprogram themselves to evade immune detection: liver metastases are characterized by a downregulation of genes responsible for immune-related processes, including neutrophil activation, response to TNF, myeloid leukocyte migration, and granulocyte chemotaxis, suggesting a shift toward reduced immunogenicity (Che et al., 2021). These metastases properties are accompanied by a decrease in the rich cellular crosstalk seen in primary tumors. Rather than utilizing extensive cell-cell communication, metastatic tumor cells are predicted to employ a dominant immune-evading signaling axis through the upregulation of the “do not eat me” ligand CD47 (Wu et al., 2022). This interaction is further confirmed by immunohistochemical (IHC) analysis, which revealed that CD47⁺ tumor cells locate in proximity to SIRPA⁺ macrophages.

Thus, CRC liver metastasis involves a shift in cancer cell behavior toward immune silence and reduced intercellular interactions, necessitating specialized treatment strategies. Furthermore, the CD47-SIRPA axis is identified as a promising therapeutic target for metastatic CRC.

3.2.2 Mechanisms of immune cell signaling in CRC metastasis

Single-cell analyses reveal that liver metastases are structurally distinct, supporting a niche where altered macrophage and T cell populations regulate immune evasion and therapy resistance processes. One of the features of the metastatic niche is the dominance of immunosuppressive macrophage populations. In support of this, liver metastases were shown to be enriched with SPP1⁺ and MRC1⁺ CCL18⁺ macrophages (IHC confirmed), which exhibit a M2-polarized phenotype with high expression of anti-inflammatory and pro-M2 conversion genes (APOE and MARCO) and a metabolic shift toward amino acid metabolism, a process tightly linked to tumor outgrowth, metastasis and therapeutic resistance through tight regulation of immune cell activity (Wu et al., 2022; Yang et al., 2023). These macrophages are reprogrammed by metastatic tumor cells themselves, as detected by cell-cell communication analysis, thus revealing the potential mechanism for the formation of a supportive PMN.

Specialization of TAMs in metastases is further evidenced by the redistribution of macrophage subsets. On one hand, primary CRC tumors harbor MHC-low TAMs with impaired antigen presentation, liver metastases are enriched with THBS1⁺ MHC-low TAMs that express pro-angiogenic factors and genes like THBS1 and MARCO, supporting tumor progression (Che et al., 2021). A constant across both sites is the presence of lipid-associated macrophages (LAMs), characterized by increased lipid metabolism, ECM degradation, and complement activation. Their persistence across both primary and metastatic sites suggests a novel location-independent mechanism for the promotion of tumor survival. However, on the other hand, TAMs in the liver communicate less frequently with other cell subtypes when compared to primary tumors, especially with CAFs (Che et al., 2021). This divergence may reflect the distinct ecological pressures at each site. For instance, primary tumors require a dense, supportive stroma to grow and initiate invasion, a process dependent on close CAF interactions. In contrast, metastatic cells, having already completed the invasion and dissemination process, may evolve toward greater self-reliance or shift their dependency to local liver-specific cell types, with their survival “strategy” primarily being immune evasion rather than stromal construction.

Importantly, the administration of chemotherapy remodels this myeloid landscape, although its impact is complex. In one study, it was identified that NAC reverses the immunosuppressive state in responsive patients by reducing the amount of suppressive macrophages and slowing their metabolism (Wu et al., 2022). In contrast, another study identified that chemotherapy reprograms TAMs away from activated states seen in untreated tumors (including pro-inflammatory IL1B⁺ and immunosuppressive CXCL10⁺ subsets) toward less active, immature, and uniformly immunosuppressive phenotypes (Che et al., 2021).

When looking at T cells populations, they are equally dynamic and responsive to both primary/metastatic location and therapy. Treatment-naïve tumors, according to sc-seq findings, contain diverse CD8⁺ T populations, including effector and exhausted T cells, whereas metastatic sites accumulate primarily dysfunctional or exhausted CD8⁺ T cells (Che et al., 2021). A critical finding is that chemotherapy inhibits the accumulation of dysfunctional CD8⁺ T cells in both primary and metastatic sites, as confirmed by flow cytometry and immunofluorescence analysis, while also reducing Treg abundance in primary tumors (although Treg levels remained comparable in liver metastases). Furthermore, differentiation trajectory analysis confirms that dysfunctional CD8⁺ T cells predominated in untreated tumors, while Tregs were persistent regardless of treatment. Mechanistically, proliferating (MKI67⁺) CD8⁺ T cells and exhausted T cells exhibit rich ligand-receptor interaction profiles, communicating with stromal (CAFs, endothelial cells) and immune cells. These findings suggest that for successful chemotherapy of metastatic CRC, the dysfunctional immune cell niche should be reactivated.

Overall, these findings depict a liver metastatic TME that is spatially organized, immunosuppressive and formed by metastatic cells to ensure their survival. This state is maintained by specifically reprogrammed macrophages (SPP1⁺, MRC1⁺CCL18⁺, THBS1⁺) and additionally LAM populations. The mentioned findings carry significant clinical implications. Firstly, the identified macrophage subpopulations represent as novel therapeutic targets for CRC metastases. Secondly, the composition of the TME, in particular macrophage subtypes and T cell dysfunction, could be looked at as biomarkers for predicting chemotherapy response. Lastly, the immunological difference between primary and metastatic sites suggest the need for location-optimized treatment. In conclusion, the liver metastatic TME in CRC is distinct from the primary tumor site and characterized by a more immunosuppressive state that is regulated by reprogrammed macrophage populations. This TME supports metastasis progression, although chemotherapy can reverse this state in responsive patients.

3.2.3 Mechanisms of stromal cell signaling in CRC metastasis

The identification of CAFs is now shifting from a uniform population to distinct subsets whose prevalence is altered by therapeutic intervention. In support of this, a study identified three major groups of CAFs in CRC tumors - secretory, ECM-remodeling and contractile (Che et al., 2021). Notably, out of the mentioned CAF groups, treatment-naïve tumors include ECM-remodeling CAFs that support ECM organization and the activity of collagen metabolism, supporting tumor ECM, whereas chemotherapy-treated primary and metastatic tumors exhibit a shift towards contractile CAF niche formation, enriched for pathways regulating muscle cell differentiation, T cell activation and tumor proliferation. These activated myofibroblasts further reinforce a treatment-resistant state by activating JAG1/NOTCH pathway in an autocrine and a paracrine manner with endothelial cells, promoting blood vessel development and differentiation. These findings indicate that chemotherapy can drive a CAF switch to a pro-proliferative and therapy-resistant state both in primary and metastatic tumors, highlighting the need for stromal-targeting adjuvants.

Apart from resident fibroblasts in CRC metastases, pathways involving trans-differentiation of tumor cells into CAFs are noted. Comparative analysis of primary tumors and metastases has revealed that the latter are enriched in processes regulating EMT, angiogenesis and TGF-β signaling (Yang et al., 2024). Notably, a specific subtype of tumor cells enriched in tumor metastases are able to undergo EMT and further differentiate into growth-regulated oncogene-α-positive (CXCL1⁺) CAFs, which subsequently mature into secreted frizzled-related protein 2-positive (SFRP2⁺) CAFs. The master regulator of this phenotypic switch is the transcriptional factor basic helix-loop-helix protein 40 (BHLHE40), whose expression is associated with poor patient prognosis. Functional studies confirmed that BHLHE40 drives tumor cell proliferation, migration, invasion and liver metastasis both in vitro and in vivo. This discovery of a tumor cell-to-CAF axis demonstrates a novel mechanism of metastatic adaptation and positions BHLHE40 and its downstream pathways as a target for therapeutic intervention to prevent metastatic seeding.

As mentioned previously, the functions of the TME depend not only on its cellular components, but additionally on the specific intercellular communication networks. A key pro-metastatic axis identified with the help of single-cell technologies involves a chemoresistant, chemokine receptor type 4-positive (CXCR4⁺) and glutathione peroxidase 4-positive (GPX4⁺) tumor cell subpopulation with notable enhanced migratory capacity and ferroptosis resistance, as well as a NOX4⁺ TGFB1⁺ CXCL12⁺ CAF population involved in immune regulatory pathways (Z. Wu et al., 2025). The survival of this niche depends on the chemokine-dependent CXCL12-CXCR4 ligand-receptor signaling axis, where CAF-derived CXCL12 binds to CXCR4 on tumor cells, fostering a tumor-promoting niche that suppresses ferroptosis and promotes survival. Targeting this interdependence, a dual-targeting TME-modulating nanoplatform CCM-FSS&CHM-ABI was developed. This strategy simultaneously induces ferroptosis in tumor cells and reprograms supportive CAFs (composed of ferroptosis inducer sulfasalazine and SN38 prodrugs, and CHM-ABI, a chelation product of azithromycin and baohuoside I). This coordinated approach successfully drives CAFs into a quiescent state, while concurrently inducing ferroptotic cell death in a co-culture model system with tumor cells. This coordinated targeting significantly suppresses primary tumor growth and liver metastases in vivo, demonstrating the potent efficacy of simultaneously targeting both the tumor and its stromal niche.

Overall, CRC metastasis is driven by a dynamic TME. Importantly, CAFs are active participants in promoting metastasis and therapy resistance. The stromal TME composition can be influenced by chemotherapy, resulting in a more aggressive and therapy-resistant tumor state due to contractile CAFs promotion of muscle cell differentiation, T cell activation and tumor cell proliferation. Moreover, tumor cells can differentiate into SFRP2⁺ CAFs, thereby promoting tumor growth and metastasis. Cell-cell communication is also of importance in shaping the TME, where metastatic survival depends on specific signaling (through CXCL12-CXCR4) between GPX4⁺ tumor and NOX4⁺ TGFB1⁺ CXCL12⁺ CAFs. Importantly, dual therapeutic targeting of the aforementioned cell-cell interaction successfully suppressed liver metastases. Thus, for metastasis prevention, successful treatment requires targeting not only tumor cells, but additionally the reprogramming of the TME, as well as their interactions.

4 Insights from single-cell sequencing in left- and right-sided colorectal cancer

While studies have given insights into the molecular mechanisms of overall CRC heterogeneity, the distinct cellular ecosystems underlying its most significant clinical classification - left- versus right-sided origin - remain incompletely defined. To elucidate these differences, we have compiled the current sc-seq data on the unique pathogenic mechanisms characterizing left-and right-sided CRC tumors, making an emphasis on cell-cell interactions (Table 2; Figure 1).

Figure 1

Figure 1. Ligand-Receptor interactions in Left- and Right-sided Colorectal Cancer Microenvironment. This figure depicts intercellular crosstalk between tumor, immune, and stromal cells. Ligand-receptor pairs are shown above the arrows, indicating the direction of signaling between cell populations. The pairs are color-coded by their function: red for tumor-promoting, green for tumor-suppressive, and black for pairs with no indicated function (for more details refer to Table 2 and the text). Generally, LCRC is characterized by a more reactive tumor stroma and a growth factor-driven TME, whereas RCRC has a predominantly immune cell-signaling TME. In left-sided colorectal cancer, bidirectional tumor cells-CAFs signaling promotes the formation of inflammatory CAFs and tumor growth respectively. Concurrently, crosstalk between B cells myeloid cells inhibits immune activation. Specific immune-tumor interactions (such as those between tumor cells and B cells or PLTP⁺ macrophages) further contribute to an immunosuppressive microenvironment. In right-sided tumors, crosstalk between tumor cells and fibroblasts/endothelial cells promotes angiogenesis and ECM remodeling. Furthermore, bidirectional signaling between IS^high tumor cells and SPP1⁺ macrophages/CD161⁺CD8⁺ innate-like T cells impairs immune function by reducing the efficiency of antigen presentation and T cell activation, while also altering cytokine production correspondently. Additional immune cell crosstalk, such as between SPP1⁺ macrophages and CD161⁺CD8⁺ innate like T cells, further dampens cytotoxic T cell capability (Created with Biorender.com).

4.1 Single-cell sequencing to elucidate tumor progression in left- and right-sided colorectal cancer

4.1.1 Mechanisms of tumor cell signaling in left- and right-sided colorectal cancer progression

Sc-seq technology has identified that LCRC and RCRC are driven by different core molecular programs (Liu B et al., 2024) (Table 3; Figure 2). The malignant cells in LCRC are frequently defined by a proliferation stemness (PS) program. The latter activates genes linked to proliferation and stem-like properties (for example, UBE2C, MYBL2 and ALDH1A1), and these proliferative tumor cells are often found in highly glycolytic niches within the tumor and are supported by specific immune cells, including PLTP⁺ macrophages, activated Tregs, and LAYN⁺ CD8⁺ cells. In turn, RCRC is characterized by an immune-secretory (IS) program. Here, the tumor cells elevate expression of immune signaling MHC II molecules (CD74, HLA-DRB1, HLA-DPA1) and immune-regulating secretory proteins (REG4, AGR2, AGR3), with AGR2 and REG4 specifically upregulated in RCRC. These tumor cells reside near hypoxic tumor cores and are surrounded by a different set of immune cells such as SPP1⁺ macrophages and CD161⁺ CD8 T cells. Hence, these findings demonstrate that LCRC and RCRC are molecularly distinct diseases driven by differing molecular programs (PS and IS), each associated with unique metabolic states, specific secreted factors/gene expression, and spatially organized, distinct immune microenvironments. This difference has significant implications for understanding tumor biology and developing sidedness-specific therapies.

Figure 2

Figure 2. Comparative Overview of the Tumor Cell Subtypes and the Tumor Microenvironment in Right-Sided vs. Left-Sided Colorectal Cancer. Left-sided colorectal cancer is characterized by MMP7⁺ and LPCAT2⁺ tumor cells that remodel the extracellular matrix and are sensitive to ferroptosis. The immune landscape, namely CD4⁺ T cells, forms an overall immunosuppressive state (although some studies point to the secretion of pro-inflammatory molecules by macrophage M1-like subsets). The stromal compartment is characterized by CAFs that promote tumor cell migration and invasion. Right-sided colorectal cancer LYZ⁺, TFF1⁺, and MUC5AC⁺ tumor cells, in turn, are metabolically active and upregulate stress and chemokine-related pathways. Immune cells, namely CD8⁺ T cells and CD20⁺ B cells, are immunostimulatory (though subsets of SPP1⁺ and LUCAT1⁺ TAMs exhibit immunosuppressive properties). CAFs, in turn, promote endothelial cell proliferation and angiogenesis (Created with Biorender.com).

The tumorigenic signaling pathways employed by tumor cells also differ by tumor sidedness. For instance, in LCRC, researchers have identified a subgroup of cancer cells enriched for genes involved in maintaining of the gut lining (e.g., TFF1, TFF2, AGR3, MUC5AC) (Guo et al., 2022). These cells exhibit enhanced activity in estrogen and ERBB signaling pathways, which helps to explain why LCRC generally responds better to therapies targeting these pathways. Furthermore, LCRC frequently signals through fibroblast-secreted collagen that binds tumor cell α1β1 integrins, as well as through macrophage-secreted TNF that interacts with tumor cell TNF receptors. These findings provide clues to the mechanistic basis for the established greater sensitivity of LCRC to monoclonal antibodies targeting ERBB receptors.

LCRC may additionally harbor tumor cells with features that may, by contrast to previously mentioned findings, inhibit tumor progression. In a study of Cao et al., LCRC lower malignancy score is linked to lysophosphatidylcholine acyltransferase 2-positive (LPCAT2⁺) tumor cells (Cao et al., 2024), with LPCAT2⁺ tumor cells significantly expanding in the inflammatory phase of a colitis-associated colon cancer mouse model. Functionally, LPCAT2 overexpression significantly inhibits cell proliferation and colony formation in vitro, while repressing tumor growth and Ki67 expression in vivo. Conversely, LPCAT2 downregulation enhances proliferative capacity in vitro and stimulates tumor growth and Ki67 in vivo. Furthermore, findings uncovered that LPCAT2 acts as a tumor suppressor through a novel ferroptosis-inducing PRMT1/SLC7A11 signaling axis. This pathway represents a promising therapeutic vulnerability unique to left-sided tumors.

In contrast, RCRC is characterized by tumor cells enriched in metabolic and inflammatory hallmarks, with glycolysis being particularly prominent in mucinous adenocarcinoma (MAC) (Xu et al., 2025). This enrichment points to a strong dependence on glycolytic metabolism in RCRC tumor cells, especially in the MAC subtype, concurrent with the findings of Liu B et al. (2024). The metabolic shift towards aerobic glycolysis (the Warburg effect) in MAC renders it potentially vulnerable to glycolytic inhibitors or other metabolic-targeted therapies. Moreover, the enrichment of inflammatory hallmarks in RCRC cancer cells suggests anti-inflammatory agents could modulate the TME.

Further underscoring their molecular differences, LCRC and RCRC tumor cells exhibit distinct functional specializations, including enhanced protein folding in left-sided tumors and increased structural homeostasis in right-sided tumors (Hu et al., 2025). A dominant feature of RCRC is a heightened hypoxic state, driven by elevated HIF1A expression in specific subclusters. This hypoxic ecosystem fosters the emergence of functionally distinct tumor populations. For example, hypoxia-responsive RCRC subclusters include LYZ⁺ cells that activate metabolic pathways, TFF1⁺ cells engaged in stress response and MUC5AC⁺ cells upregulating chemokines. In contrast, advanced LCRC is characterized by MMP7⁺ cancer cells expressing ECM remodeling genes and possess significantly higher metastatic potential. Other subpopulations, like ATP5MC2⁺ cells enriched in ATP synthesis, are found across early- and middle-stage CRCs regardless of sidedness. These specialized tumor subpopulations actively shape the TME through specific intercellular ligand-receptor interactions. These include immunosuppressive, angiogenic and migratory interactions through specific tumor cell-TME ligand-receptor pairs (for instance, CCL28-CCR10 with B cells, PDGFA-PDGFRA/PDGFRB with CAFs, and PLAU-PLAUR with TAMs). These findings highlight therapeutic opportunities for side-specific CRC treatment, including hypoxia targeting in RCRC, MMP7⁺ cancer cell targeting in advanced LCRC, and the potential disruption of immunosuppressive, angiogenic, and migratory axes.

Overall, research demonstrates that left-sided and right-sided colorectal cancer are fundamentally distinct diseases at the level of tumor cells. LCRC is characterized by a PS program, lower malignancy scores linked to LPCAT2⁺ cells (which inhibit tumor growth and induce ferroptosis), enhanced mucosal healing/protein folding, and tumor cells localized in glycolytic niches interacting with specific immune cells (PLTP⁺ macrophages, Tregs, LAYN⁺ CD8⁺ cells). In contrast, RCRC is dominated by an IS program, exhibits pronounced glycolysis (especially in MAC), heightened inflammatory states, and tumor cells concentrated near hypoxic cores and interacting with distinct immune cells (SPP1⁺ macrophages, CD161⁺ CD8 T cells). Functionally unique subpopulations exist, including metastatic MMP7⁺ cells in advanced LCRC and hypoxia-responsive subclusters in RCRC. These profound differences in molecular drivers, metabolism, tumor microenvironment, and cellular functions highlight the need for distinct, side-specific therapeutic strategies, such as targeting the LPCAT2/ferroptosis axis or MMP7⁺ cells in LCRC, and glycolysis or hypoxia in RCRC.

4.1.2 Mechanisms of immune cell signaling in LCRC and RCRC progression and immunotherapy

Sc-seq has revealed a complex and sometimes contradictory immunological landscape in CRC, where tumor sidedness is a major, although not exclusive, determinant of immune function. The collective findings on immune cells in regards to tumor side differences are compiled in Table 4.

The myeloid compartment demonstrates cellular and functional divergence in regards to tumor side. In LCRC, macrophages are frequently polarized toward an M1-like, pro-inflammatory state, exhibiting high expression of pro-inflammatory molecules and functional enrichment for inflammatory pathways, transcriptional regulation, and B-cell receptor signaling (Hu et al., 2025; Ji et al., 2024). Conversely, RCRC is dominated by M2-like, anti-inflammatory macrophages, which are defined by osteopontin (SPP1) expression and primarily interact with other leukocytes (Hu et al., 2025). These SPP1⁺ macrophages in RCRC upregulate genes that regulate hypoxia, angiogenesis, and EMT, and their high infiltration is linked to a significantly shorter 5-year overall survival (OS) (Ji et al., 2024). However, these side-specific differences are nuanced by contradictory data. Firstly, the study of Hu et al. demonstrates that the accumulation of SPP1⁺ macrophages during malignant progression is noted as a common feature in both LCRC and RCRC, suggesting their role is not entirely specific to the right side (Hu et al., 2025). Moreover, another study characterizes RCRC to exhibit a higher proportion of monocytes in comparison to LCRC (Shang et al., 2024), with these cells activating pathways involved in leukocyte proliferation, antigen presentation, and MHC binding - functions critical for pro-inflammatory macrophage activity. A key driver in this population is the lung cancer associated transcript 1-positive (LUCAT1⁺) monocyte/macrophage subset, which sustains its own activation and contributes to the inflammatory TME by overexpressing the pro-tumorigenic CD74-CD44 axis (Shang et al., 2024; Shi et al., 2006).

Furthermore, the T cell compartment also demonstrates cellular and functional divergence based on tumor side. In LCRC this compartment is often indicative of a suppressed/dysfunctional state: LCRC T cells are predominantly naïve/undifferentiated and enriched with CD4⁺ T cells, with immune cells interacting mainly amongst themselves (Liu D et al., 2024). Furthermore, LCRC T cells display enrichment in stress-related signaling pathways, suggesting a state of chronic activation that may underly their poor response to immunotherapy (Hu et al., 2025). Conversely, the T cell landscape in RCRC is suggestive of a presumably higher immune activation state than in LCRC tumors. Here, the T cell population is enriched with highly differentiated and activated CD8⁺ T cells and engages in frequent and intense tumor-immune cell interactions. However, the nature of these interactions is complex: for instance, NECTIN2-TIGIT is shown to dampen immune response, whereas PLXNB2-SEMA4D presumably promotes immune cell activation and differentiation (Ch’ng and Kumanogoh, 2010; Deuss et al., 2017; Liu D et al., 2024). Moreover, the presence of exhausted T cells in this context may signify a prior active anti-tumor response, potentially predicting better response to therapy. Importantly, another finding highlights that the functional shift of CD8⁺ T cells from a cytotoxic to an exhausted state is actually a stage-dependent process that occurs regardless of the tumors side, suggesting that this may actually be a universal characteristic of CRC tumor biology irrespective of tumor origin (Hu et al., 2025).

The distribution and functional characteristics of B cells also vary based on the tumor site of origin. LCRC exhibits a near-complete depletion of naïve B cells and is characterized by immunosuppressive inter-immune B cell interactions, including immune cell activation-suppressing CD52-SIGLEC10 axis with myeloid cells (Bandala-Sanchez et al., 2018; Ji et al., 2024). These findings reinforce the pattern of a suppressed microenvironment consistent with observations in the myeloid and T cell niches. Conversely, B cells in RCRC play a more favorable role. Despite the majority of inter-immune communications with myeloid cells occurring in both sides (including pro-tumor pairs like SPP1-CD44, LAMP1-FAM3C, and MERTK-GAS6), CD20⁺ B cells specific to the right side are a favorable prognostic marker and have been demonstrated to be essential for the efficacy of anti-PD-1 therapy in murine models and patient-derived organoids (Ji et al., 2024).

Although tumor origin site shapes the immune landscape according to several studies, some core features are conserved between LCRC and RCRC according to others. For instance, the overall expression of immune checkpoint ligands and receptors (ICLs/ICRs) show minimal variation, indicating that sidedness alone in this case is a poor predictor of response to checkpoint blockade (Liu et al., 2022). Furthermore, the pervasive inhibitory communication networks among myeloid, mast, B, and tumor cells via axes such as immune cell inhibition by tumor cells via CD24-SIGLEC10 and LGALS9-HAVCR2 interactions appear to be a fundamental characteristic of the CRC microenvironment, with minimal influence from sidedness (Barkal et al., 2019; Hill et al., 2025).

Overall, research reveals an immunological divergence between LCRC and RCRC that shapes their distinct TMEs, although discrepancies in its nature exist. On one hand, LCRC is described as immunosuppressive, with the T cell population enriched for naïve/undifferentiated CD4⁺ T cells and stress-related signaling pathways, suggesting that they are in a chronic activation state and, therefore, not effective for immunotherapeutic targeting. This is further supported by a nearly-complete depletion of naïve B cells, as well as enhanced immunosuppressive B cell-myeloid cell interactions (such as CD52-SIGLEC10). On the other hand, studies mention that macrophages in LCRC can exhibit high expression of pro-inflammatory molecules and show functional enrichment for inflammatory pathways [although it should be noted that this state of immune cell activation in LCRC may also suggest preceding of chronic inflammation (Zhao et al., 2021)]. Such differences likely stem from several methodological and biological variables. Firstly, cohort heterogeneity presents a significant challenge. In this case, differences in disease stage, prior treatments and molecular subtypes (such as microsatellite instability status) can significantly hinder site-specific comparisons. Secondly, technical divergences in analytical pipelines, such as variations in clustering resolution and cell type annotation criteria, as well as the principles for significant ligand-receptor interaction identification can lead to the characterization of fundamentally different cell populations/intercellular interactions across studies. Thirdly, spatial heterogeneity within the tumor itself due to sampling from either the tumor core or the invasive margin can capture distinct immunological niches, further complicating direct comparisons.

RCRC, in turn, is characterized as immunostimulatory in several studies, with enhanced leukocyte proliferation, antigen presentation and MHC binding in monocytes, as well as the T cell population being enriched for highly differentiated and activated CD8⁺ T cells. Moreover, RCRC shows frequent and intense immune-tumor cell interactions, including through enhancement of macrophage migration by the SEMA4A-PLXND1 axis or by pro-tumorigenic signaling between immune cells (LUCAT1⁺ monocytes and macrophages) with the help of CD74-CD44, and possess metabolically active and highly dense CD20⁺ B cells correlating with favorable prognosis and shown to be essential for anti-PD-1 therapy. Conversely, signs of immunosuppression have been noted, with macrophages in RCRC enriched for anti-inflammatory genes and primarily interacting with leukocytes. Such discrepancies may be described similarly as to those mentioned previously (both biological and methodological).

Importantly, several immune features show no consistent differences by tumor sidedness. The accumulation of exhausted CD8⁺ T cells and SPP1⁺ macrophages during malignant progression was identified to occur in both LCRC and RCRC in several studies. Likewise, the core functional shift of CD8⁺ T cells from cytotoxicity to exhaustion was shown to be stage-dependent rather than location-dependent. CD4⁺ T cell subcluster enrichment in early/middle stages, along with overall patterns of ICL/ICR expression and broad cell-cell communication networks, show minimal variation by sidedness in some studies. CD20⁺ B cells are strongly linked to improved prognosis and enhanced response to anti-PD-1 therapy in RCRC, but B-cell interactions with myeloid cells occur in both tumor locations.

Thus, these findings suggest that, although not unanimously, tumor sidedness informs, but does not exclusively dictate, immunotherapy strategies. One of the options to further clarify the aforementioned findings is for future studies to adopt standardized processing and analysis protocols, coupled with more refined patient stratification that accounts for the multifaceted nature of the disease. Currently, it can be proposed that LCRC may benefit from interventions that convert chronic inflammation into productive antitumor immunity (for example, enhancing antigen presentation or disrupting Treg suppression). RCRC, on the other hand, is immunostimulatory, and enhancing immune signaling crosstalk, such as SEMA4A-PLXND1 between immune and tumor cells could be advantageous. Importantly, both tumor types harbor macrophages with adverse functions that contribute to both immunostimulation and immunosuppression. Therefore, further research is critical to understand the signaling mechanisms of these distinct macrophage populations.

4.1.3 Mechanisms of stromal cell signaling in LCRC and RCRC progression

A critical emerging theme in CRC biology is that tumor sidedness dictates fibroblast function and stromal remodeling (Ahmad Zawawi and Musa, 2022) (Table 5).

LCRC stroma is defined by heterogeneity presented by a specific pro-tumor CAF subset. A sc-seq study revealed that CAFs express high levels of phosphate regulon transcriptional regulatory protein B (RHOB) to promote cell growth and migration, and a distinct myosin heavy chain 11-positive (MYH11⁺) CAFs population is highlighted (C. Wang et al., 2024). This subset ultimately acts as a regulator of LCRC aggressiveness, promoting tumor invasion through TGF-β signaling, upregulation of pro-tumor genes (RAMP1, MYLK, MYH11) and activation of muscle-invasive pathways. Apart from tumor invasion promotion, these CAFs potentially form an immunosuppressive niche through inhibition of immune cell migratory (myeloid cells) CD74-MIF and SPP1-CD44 ligand-receptor signaling interactions. These processes together drive tumor progression and highlight MYH11⁺ CAFs as a potential therapeutic target in LCRC.

In RCRC, CAFs also promote tumor progression but primarily function as drivers of angiogenesis in addition to invasion (Hu et al., 2025). They express proteins like AGR2 to potentiate VEGF and FGF2 signaling, and REG4 to intensify tumor invasion. This process is driven by transcription factors like E2F1 and FOXF1 which promote rapid tumor growth by activating cell cycle and apoptotic pathways that contribute to neovascularization.

Beyond the location-specific stroma signaling, universal mechanisms of CAF activity have been noted (Hu et al., 2025). Across all CRC subtypes, CAFs tend to engage in pro-tumor functions with pathways like “focal adhesion” and “ECM-receptor interaction” being activated. Moreover, a distinguished Stromal3⁺ fibroblast subset is utilized by tumors for building of the ECM. Starting from the earliest stages, CAFs across the colorectum exhibit WNT signaling - a key pathway for invasion promotion - marked by specific regulation of the transcription factor TCF4.

Thus, comparative analyses underscore the impact of tumor sidedness on CAF function. RCRC CAFs promote angiogenesis (via AGR2) and invasion (via REG4 and transcription factors E2F1/FOXF1), whereas LCRC CAFs are characterized by greater heterogeneity and enrichment of MYH11⁺ CAFs, which promote invasion, immunosuppression (via CD74-MIF/SPP1-CD44), and metastasis. Despite these differences, CAFs across both CRC subtypes are shown to activate ECM-remodeling and WNT pathways, helping the tumor to progress. Overall, in addition to the general molecular signaling pathways activated in CAFs to facilitate the progression of the tumor, the anatomic origin of CRC tumors defines distinct stromal pro-tumor programs, making location-specific CAF subsets and their signaling pathways potential therapeutic targets.

5 Conclusion and perspectives

Recent sc-seq and spatial transcriptomic studies have provided insight into CRC progression by describing it as a complex ecosystem of tumor, immune, and stromal cells. The application of sc-seq and spatial transcriptomic technologies has shed light onto CRC progression, therapy resistance and metastasis mechanisms. This review synthesizes how these methods have shifted our understanding of tumor biology, revealing CRC as a complex, sometimes functionally heterogenous ecosystem combining tumor, immune, and stromal cells. Importantly, this ecosystem is not uniform. Its structure, functions, and response to therapy are dependent on the tumor’s embryological origin - left-sided or right-sided. The research on CRC tumor sidedness signaling differences help to understand CRC biology and therefore design more effective therapeutic strategies.

Importantly, a major challenge in interpreting and comparing the studies discussed in this review is their discrepant findings, particularly regarding immune signaling mechanisms. Such inconsistencies may arise due to both methodological and biological variables. Key factors include heterogeneity in patient cohorts such as the disease stage, prior treatment administration and the tumor molecular subtypes like the microsatellite instability status. Furthermore, sampling from different tumor regions (e.g., the tumor core versus the margin) can capture distinct cellular niches, complicating direct comparisons across studies. Technical variations in data analysis further contributes to divergent results. Differences in clustering resolution, cell type annotation, and criteria for identifying significant ligand-receptor interactions can lead to the characterization of fundamentally different cell populations and communication networks. This methodological variability underscores the need for future studies to adopt standardized workflows for single-cell data processing and analysis coupled with more refined patient stratification that accounts for the complex nature of the disease. It also highlights the necessity for biological validation of computational findings both in vitro and in vivo. Despite these challenges, the co-signaling between cancer cells and the TME remains a critical area for further research, as new mechanisms are continually being uncovered.

Single-cell analyses have revealed that tumor progression, therapy resistance, and metastasis in CRC arise largely from dynamic cellular crosstalk between tumor cells and the TME. Several mechanisms underpinning these processes can be highlighted in overall CRC. Firstly, cellular plasticity within the primary tumor contributes to divergent functional outcomes. For instance, pro-tumorigenic FAP⁺/SPP1⁺/CTHRC1⁺ CAFs drive immunosuppression and invasion whereas immunogenic PI16⁺/SLIT2⁺ CAFs promote T-cell recruitment and activation. Similarly, macrophages can be C1QC⁺ and immunostimulatory or SPP1⁺/SIRPA⁺ and immunosuppressive (Table 4). Secondly, the spatial organization of cells within the primary tumor mass determines their function. For example, pMMR tumors form organized and immune-excluding barriers, while dMMR tumors exert proximity-dependent interactions (e.g., predicted CXCL13⁺ T cells-LAMP3⁺ DCs interactions via PD-1/PD-L1) that determine immunotherapy response. Third, metabolic and signaling focal points drive cell coordination. NU^high tumor cells, regulated by NME1, act as signaling hubs, interacting with stromal and immune cells via ligands like immune-suppressing MIF or immune cell activating PLA2 (Table 2). Moreover, the pro-tumorigenic KITLG-KIT signaling axis between stromal and immune cell compartments defines an anti-tumor niche, while the VEGFA-VEGFR-ESM1 positive feedback loop in endothelial tip cells promotes angiogenesis and immune evasion (Table 5). In metastases, cancer cells undergo a shift toward immune silence and reduced intercellular interactions, as well as reprogramming of macrophages into immunosuppressive SPP1⁺, MRC1⁺CCL18⁺, and THBS1⁺ phenotypes. NAC further remodels this immune landscape, promoting a transition of TAMs from activated to immunosuppressive states. Lastly, SFRP2⁺ CAFs are demonstrated to promote tumor growth and metastasis. Together, these findings underscore key cellular interactions and specific cell subtypes as potential new therapeutic targets for both overall primary and metastatic CRC.

Importantly, the two major CRC subtypes - LCRC and RCRC - are not solely anatomically distinct tumors but functionally divergent diseases with distinguishable cellular ecosystems (Tables 2–5). This may explain their different clinical behaviors and responses to treatment. LCRC is characterized as a tumor subtype featuring MMP7⁺ matrix remodeling, PS meta-program characterized by expression of genes associated with proliferation and stemness, elevated expression of PHLDA2 linked to lymph node metastasis and minimal MHC I expression in tumor cells, while harboring an LPCAT2⁺ tumor cell population that decreases cell proliferation. The immune cell population, in turn, is characterized as immunosuppressive in a number of studies, with the T cell population enriched for naïve/undifferentiated and exhausted CD4⁺ and CD8⁺ T cells and stress-related signaling pathways, suggesting that they are in a chronic activation state. These findings are further supported by the absence of naïve B cells. On the other hand, other studies mention that macrophages in LCRC can exhibit high expression of pro-inflammatory molecules and show functional enrichment for inflammatory pathways. Additionally, a population of MYH11⁺ CAFs expresses pro-tumor and migration genes. RCRC tumor ecosystem, in turn, is comprised of tumor cells with high MHC I expression, characterized by a IS meta program and increased hypoxia in comparison with LCRC. Furthermore, immune cells are characterized as immunostimulatory in several studies, with T cells present in a recently active state and early stage CD8⁺ T cells, LUCAT1⁺ monocytes expressing genes related to interleukin production, as well as CD20⁺ B cells correlating with a favorable prognosis. Contrarily, sings of immunosuppression have been identified in other studies, with the presence of SPP1⁺ macrophages enriched for anti-inflammatory genes and a CD161⁺ T cell subset correlating with lower OS being detected. These discrepancies are further complicated by another group of studies that demonstrate that several immune features show no consistent differences by tumor sidedness, for example, the accumulation of exhausted CD8⁺ T cells and SPP1⁺ macrophages during malignant progression was identified to occur in both LCRC and RCRC. Lastly, CAFs in RCRC were identified to express angiogenesis- and tumor growth-promoting AGR2 and REG4 genes. These findings point to the complexity of tumorigenic signaling within overall CRC and its subtypes, making it troubling to target their cell-specific signaling pathways.

Cell-cell communication analysis examines how the binding of ligands to receptors influences cellular behavior and fate. This analytical approach, widely applied in sc-seq and spatial transcriptomics data analysis using tools such as CellChat, CellPhoneDB, allows to understand the diverse cellular decisions, including decisions to activate the cell cycle or apoptosis, undergo migration or differentiate along the lineage (Efremova et al., 2020; Jin et al., 2021). In CRC, extensive crosstalk has been uncovered between tumor cells and the TME in overall and sided CRC (Table 2). In overall CRC, a significant portion of the cell-cell crosstalk involves tumor cells communicating with fibroblasts and endothelial cells. Some of these interactions include: 1) integrin-mediated signaling—interactions of collagens (COL1A1, COL4A1) and laminins with integrin receptors (ITGB1, ITGAV) are a universal mechanism for enhancing tumor cell migration, invasion, and adhesion; 2) Wnt and Hedgehog pathways—signaling through IHH-PTCH1 and WNT5A-FZD/LRP participates in the regulation of stromal cell differentiation and creating an immune-restrictive ECM; 3) regulation of angiogenesis—the VEGFA-NRP1/2 axis participates in new blood and lymphatic vessels formation. Moreover, the TME actively suppresses anti-tumor immunity through several ligand–receptor–mediated mechanisms, including: 1) checkpoint inhibition by B cell-T cell interactions via the PD-L1/PD-1 axis; 2) CD47-SIRPA and CD52-SIGLEC10 interactions protecting tumor cells from phagocytosis by macrophages; 3) signals from stromal and tumor cells (such as LGALS9-HAVCR2 and CSF1-CSF1R) promoting pro-tumor, M2-like macrophage polarization; 4) a balance of co-stimulatory (CD28⁻CD80/86) and co-inhibitory (CTLA4-CD80/86) signals regulating the T-cell response.

Moreover, cell-cell communication analysis revealed distinct biological preferences between the left and right sides of CRC (Table 2; Figure 1). LCRC is characterized by a more reactive tumor stroma and a growth factor-driven TME: ligand-receptor interactions are shifted towards a tumor-stromal crosstalk, particularly with fibroblasts. Pathways like PLAU-PLAUR and TGFB1-TGFBR1 drive the formation of inflammatory CAFs, collagen deposition, and direct tumor cell proliferation. Furthermore, a notable feature for LCRC is the recruitment and instruction of immunosuppressive B cells via the chemokine-mediated CCL28-CCR10 axis. In contrast, RCRC has an immune cell-signaling TME, dominated by chemokine signaling (e.g., CCL15-CCR1 and CXCL16-CXCR6) that actively recruits specific immunosuppressive SPP1⁺ macrophages yet PD-L1 interacting CD161⁺ CD8⁺ T cells. These recruited cells then participate in immunosuppressive signaling: SPP1⁺ macrophages dampen T-cell cytotoxicity via SPP1-CD44, while tumor cells directly inhibit T cell function by PD-L1-CD80 and TIGIT-NECTIN2 signaling. Lastly, RCRC utilizes signaling pathways through ligand-receptor pairs like DCN-VEGFR2 and SEMA3B-NRP1/2 to promote metastasis and vascular permeability, differing from the canonical VEGFA-driven angiogenesis. Collectively, these findings highlight that while both overall and sided CRC rely on core signaling pathways for the promotion of tumor growth and invasion, the specific ligand-receptor interactions underpinning these processes are sidedness-dependent. LCRC is driven by a stromal and growth-factor-oriented TME, whereas RCRC is characterized by an inflammatory yet immunosuppressive ecosystem. There distinctions underscore the need for subtype-specific therapeutic strategies. We propose that the enhancement of CRC therapy efficiency potentially lies in targeting the aforementioned cell-cell interactions. Such intervention could (i) overcome therapy resistance, by disrupting protective tumor-TME-signaling networks, and (ii) prevent immune cell therapy evasion, by blocking immunosuppressive ligand-receptor interactions that inhibit anti-tumor immune responses.

In order to translate the aforementioned findings into clinical progress, future research should aim to prioritize the adoption of single-cell multi-omics technologies. While sc-seq captures a cell’s phenotypic/transcriptional state, it provides a limited view of the cells’ molecular networks controlling cell identity and function. Single-cell multi-omics methods which simultaneously capture genomic, epigenomic, transcriptomic and proteomic networks in the same cells are essential to overcome this limitation (Bai and Zhu, 2023). This integrated approach can directly explore key biological questions. For instance, methods like G&T-seq profile the genome variation and transcriptome of cells simultaneously and can be utilized to assess the origin and evolution of tumors from LCRC/RCRC with enhanced sensitivity. Furthermore, the established link between RCRC and the CpG island methylation phenotype remains mechanistically unclear (Lee et al., 2017). Applying methods that co-profile the methylome and transcriptome (such as snmCAT-seq for base-resolution cytosine methylation or scATAC-seq to map chromatin accessibility) can directly connect epigenetic landscapes to the gene expression programs that drive sidedness. Furthermore, because transcript levels often poorly predict protein abundance, methods integrating the transcriptome with proteome (for example, CITE-seq/REAP-seq) are critical to determine how the distinct transcriptional networks of LCRC and RCRC are functionally translated into protein-level effectors, revealing the functional mechanisms of tumor divergence. A limitation of the aforementioned methods is the loss of spatial architecture analysis, which is fundamental for understanding tumor signaling. While the studies compiled in our review have begun to map predicted cell-cell interactions at the RNA level, the next step is to integrate spatial context with multi-omics technologies. Emerging spatially resolved multi-omics technologies, for instance, spatial-ATAC-RNA-seq use deterministic barcoding to co-profile accessible chromatin or histone modifications with the transcriptome on the same tissue section. These methods are poised to uncover new biological insights into region-specific epigenetic priming and gene regulation within the distinct histological landscapes of LCRC and RCRC.

The findings summarized in this review may help guide future clinical trial design by moving toward more targeted approaches that account for the complex ecosystem of LCRC and RCRC. The development of future trials aiming to stratify patients based on integrated “TME subtypes” that incorporate tumor sidedness with specific cellular and molecular features may hold promise for increasing the survival of CRC patients. For instance, for LCRC patients whose tumors are characterized by a reactive stroma enriched in MYH11⁺ CAFs and prominent TGF-β signaling, a potential combination therapy could pair immune checkpoint inhibition with TGF-β pathway inhibitors or specific anti-stromal agents. Moreover, for RCRC patients with an SPP1⁺ macrophage-rich niche and evidence of CD161⁺ T cell-mediated immunosuppression, trial designs could combine immune checkpoint inhibition with agents targeting the SPP1-CD44 axis or other macrophage-reprogramming drugs. Implementing such a precision approach will require the simultaneous development of auxiliary diagnostic assays to reliably identify these TME subtypes in patient biopsies, and may potentially be derived from spatial transcriptomics and multi-omics data. Overall, the combined utilization of specific TME features with corresponding combination therapies hold promise for increasing the likelihood of clinical success and overcoming of therapy resistance in left- and right-sided CRC.

In summary, single-cell technologies have given insight into overall and sided CRC cellular structure. These findings allow moving beyond treatments based solely on the location of the tumor, but instead focus on designing therapies that target the entire network of cells within the tumor. This shift is crucial for overcoming tumor progression, therapy resistance and metastasis in CRC.

Author contributions

MY: Visualization, Writing – original draft, Conceptualization, Writing – review and editing. DM: Supervision, Conceptualization, Writing – review and editing, Visualization. AT: Resources, Conceptualization, Supervision, Writing – review and editing.

Funding

The authors declare that financial support was received for the research and/or publication of this article. This study was performed within the framework of the Basic Research Program at Higher School of Economics University.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The authors declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.

Generative AI statement

The authors declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Abdel Hamid, M., Pammer, L. M., Oberparleiter, S., Günther, M., Amann, A., Gruber, R. A., et al. (2025). Multidimensional differences of right- and left-sided colorectal cancer and their impact on targeted therapies. Npj Precis. Oncol. 9 (1), 116. doi:10.1038/s41698-025-00892-y

PubMed Abstract | CrossRef Full Text | Google Scholar

Ahmad Zawawi, S. S., and Musa, M. (2022). Dynamic Co-Evolution of cancer cells and cancer-associated fibroblasts: role in Right- and left-sided Colon cancer progression and its clinical relevance. Biology 11 (7), 1014. doi:10.3390/biology11071014

PubMed Abstract | CrossRef Full Text | Google Scholar

Armingol, E., Officer, A., Harismendy, O., and Lewis, N. E. (2021). Deciphering cell–cell interactions and communication from gene expression. Nat. Rev. Genet. 22 (2), 71–88. doi:10.1038/s41576-020-00292-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Armingol, E., Baghdassarian, H. M., and Lewis, N. E. (2024). The diversification of methods for studying cell–cell interactions and communication. Nat. Rev. Genet. 25 (6), 381–400. doi:10.1038/s41576-023-00685-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Bai, D., and Zhu, C. (2023). Single-cell technologies for multimodal omics measurements. Front. Syst. Biol. 3, 1155990. doi:10.3389/fsysb.2023.1155990

PubMed Abstract | CrossRef Full Text | Google Scholar

Bandala-Sanchez, E., Bediaga, G., Goddard-Borger, E. D., Ngui, K., Naselli, G., Stone, N. L., et al. (2018). CD52 glycan binds the proinflammatory B box of HMGB1 to engage the Siglec-10 receptor and suppress human T cell function. Proc. Natl. Acad. Sci. 115 (30), 7783–7788. doi:10.1073/pnas.1722056115

CrossRef Full Text | Google Scholar

Barkal, A. A., Brewer, R. E., Markovic, M., Kowarsky, M., Barkal, S. A., Zaro, B. W., et al. (2019). CD24 signalling through macrophage Siglec-10 is a target for cancer immunotherapy. Nature 572 (7769), 392–396. doi:10.1038/s41586-019-1456-0

PubMed Abstract | CrossRef Full Text | Google Scholar

Berntsson, J., Nodin, B., Eberhard, J., Micke, P., and Jirström, K. (2016). Prognostic impact of tumour-infiltrating B cells and plasma cells in colorectal cancer. Int. J. Cancer 139 (5), 1129–1139. doi:10.1002/ijc.30138

PubMed Abstract | CrossRef Full Text | Google Scholar

Beumer, J., and Clevers, H. (2021). Cell fate specification and differentiation in the adult mammalian intestine. Nat. Rev. Mol. Cell Biol. 22 (1), 39–53. doi:10.1038/s41580-020-0278-0

PubMed Abstract | CrossRef Full Text | Google Scholar

Biller, L. H., and Schrag, D. (2021). Diagnosis and treatment of metastatic colorectal cancer: a review. JAMA 325 (7), 669–685. doi:10.1001/jama.2021.0106

PubMed Abstract | CrossRef Full Text | Google Scholar

Blank, A., Roberts, D. E., Dawson, H., Zlobec, I., and Lugli, A. (2018). Tumor heterogeneity in primary colorectal cancer and corresponding metastases. Does the apple fall far from the tree? Front. Med. 5, 234. doi:10.3389/fmed.2018.00234

PubMed Abstract | CrossRef Full Text | Google Scholar

Bray, F., Laversanne, M., Sung, H., Ferlay, J., Siegel, R. L., Soerjomataram, I., et al. (2024). Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA A Cancer J. Clin. 74 (3), 229–263. doi:10.3322/caac.21834

PubMed Abstract | CrossRef Full Text | Google Scholar

Browaeys, R., Saelens, W., and Saeys, Y. (2020). NicheNet: modeling intercellular communication by linking ligands to target genes. Nat. Methods 17 (2), 159–162. doi:10.1038/s41592-019-0667-5

PubMed Abstract | CrossRef Full Text | Google Scholar

Bufill, J. A. (1990). Colorectal cancer: evidence for distinct genetic categories based on proximal or distal tumor location. Ann. Intern. Med. 113 (10), 779–788. doi:10.7326/0003-4819-113-10-779

PubMed Abstract | CrossRef Full Text | Google Scholar

Cao, N., Zhang, F., Yin, J., Zhang, J., Bian, X., Zheng, G., et al. (2024). LPCAT2 inhibits colorectal cancer progression via the PRMT1/SLC7A11 axis. Oncogene 43 (22), 1714–1725. doi:10.1038/s41388-024-02996-4

PubMed Abstract | CrossRef Full Text | Google Scholar

Cervantes, A., Adam, R., Roselló, S., Arnold, D., Normanno, N., Taïeb, J., et al. (2023). Metastatic colorectal cancer: ESMO clinical practice guideline for diagnosis, treatment and follow-up. Ann. Oncol. 34 (1), 10–32. doi:10.1016/j.annonc.2022.10.003

PubMed Abstract | CrossRef Full Text | Google Scholar

Che, L.-H., Liu, J.-W., Huo, J.-P., Luo, R., Xu, R.-M., He, C., et al. (2021). A single-cell atlas of liver metastases of colorectal cancer reveals reprogramming of the tumor microenvironment in response to preoperative chemotherapy. Cell Discov. 7 (1), 80. doi:10.1038/s41421-021-00312-y

PubMed Abstract | CrossRef Full Text | Google Scholar

Chen, K., Collins, G., Wang, H., and Toh, J. W. T. (2021). Pathological features and prognostication in colorectal cancer. Curr. Oncol. 28 (6), 5356–5383. doi:10.3390/curroncol28060447

PubMed Abstract | CrossRef Full Text | Google Scholar

Ch’ng, E. S., and Kumanogoh, A. (2010). Roles of Sema4D and Plexin-B1 in tumor progression. Mol. Cancer 9 (1), 251. doi:10.1186/1476-4598-9-251

PubMed Abstract | CrossRef Full Text | Google Scholar

Ciardiello, F., Ciardiello, D., Martini, G., Napolitano, S., Tabernero, J., and Cervantes, A. (2022). Clinical management of metastatic colorectal cancer in the era of precision medicine. CA A Cancer J. Clin. 72 (4), 372–401. doi:10.3322/caac.21728

PubMed Abstract | CrossRef Full Text | Google Scholar

Colombani, J., and Andersen, D. S. (2020). The drosophila gut: a gatekeeper and coordinator of organism fitness and physiology. WIREs Dev. Biol. 9 (6), e378. doi:10.1002/wdev.378

PubMed Abstract | CrossRef Full Text | Google Scholar

Deuss, F. A., Gully, B. S., Rossjohn, J., and Berry, R. (2017). Recognition of nectin-2 by the natural killer cell receptor T cell immunoglobulin and ITIM domain (TIGIT). J. Biol. Chem. 292 (27), 11413–11422. doi:10.1074/jbc.M117.786483

PubMed Abstract | CrossRef Full Text | Google Scholar

Domanska, D., Majid, U., Karlsen, V. T., Merok, M. A., Beitnes, A.-C. R., Yaqub, S., et al. (2022). Single-cell transcriptomic analysis of human colonic macrophages reveals niche-specific subsets. J. Exp. Med. 219 (3), e20211846. doi:10.1084/jem.20211846

PubMed Abstract | CrossRef Full Text | Google Scholar

Efremova, M., Vento-Tormo, M., Teichmann, S. A., and Vento-Tormo, R. (2020). CellPhoneDB: inferring cell–cell communication from combined expression of multi-subunit ligand–receptor complexes. Nat. Protoc. 15 (4), 1484–1506. doi:10.1038/s41596-020-0292-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Feng, Y., Ma, W., Zang, Y., Guo, Y., Li, Y., Zhang, Y., et al. (2024). Spatially organized tumor-stroma boundary determines the efficacy of immunotherapy in colorectal cancer patients. Nat. Commun. 15 (1), 10259. doi:10.1038/s41467-024-54710-3

PubMed Abstract | CrossRef Full Text | Google Scholar

Fox, D. T., and Spradling, A. C. (2009). The drosophila hindgut lacks constitutively active adult stem cells but proliferates in response to tissue damage. Cell Stem Cell 5 (3), 290–297. doi:10.1016/j.stem.2009.06.003

PubMed Abstract | CrossRef Full Text | Google Scholar

Fridman, W. H., Miller, I., Sautès-Fridman, C., and Byrne, A. T. (2020). Therapeutic targeting of the colorectal tumor stroma. Gastroenterology 158 (2), 303–321. doi:10.1053/j.gastro.2019.09.045

PubMed Abstract | CrossRef Full Text | Google Scholar

Gallo, G., Vescio, G., De Paola, G., and Sammarco, G. (2021). Therapeutic targets and tumor microenvironment in colorectal cancer. J. Clin. Med. 10 (11), 2295. doi:10.3390/jcm10112295

PubMed Abstract | CrossRef Full Text | Google Scholar

Gemoll, T., Strohkamp, S., Schillo, K., Thorns, C., and Habermann, J. K. (2015). MALDI-imaging reveals thymosin beta-4 as an independent prognostic marker for colorectal cancer. Oncotarget 6 (41), 43869–43880. doi:10.18632/oncotarget.6103

PubMed Abstract | CrossRef Full Text | Google Scholar

Ghasemian, M., Rajabibazl, M., Sahebi, U., Sadeghi, S., Maleki, R., Hashemnia, V., et al. (2022). Long non-coding RNA MIR4435-2HG: a key molecule in progression of cancer and non-cancerous disorders. Cancer Cell Int. 22 (1), 215. doi:10.1186/s12935-022-02633-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Guo, L., Wang, C., Qiu, X., Pu, X., and Chang, P. (2020). Colorectal cancer immune infiltrates: significance in patient prognosis and immunotherapeutic efficacy. Front. Immunol. 11, 1052. doi:10.3389/fimmu.2020.01052

PubMed Abstract | CrossRef Full Text | Google Scholar

Guo, W., Zhang, C., Wang, X., Dou, D., Chen, D., and Li, J. (2022). Resolving the difference between left-sided and right-sided colorectal cancer by single-cell sequencing. JCI Insight 7 (1), e152616. doi:10.1172/jci.insight.152616

PubMed Abstract | CrossRef Full Text | Google Scholar

Hill, C. N., Maita, G., Cabrolier, C., Aros, C., Vega-Letter, A. M., Gonzalez, P., et al. (2025). Galectin-9 and Tim-3 in gastric cancer: a checkpoint axis driving T cell exhaustion and Treg-mediated immunosuppression independently of anti-PD-1 blockade. Front. Immunol. 16, 1600792. doi:10.3389/fimmu.2025.1600792

PubMed Abstract | CrossRef Full Text | Google Scholar

Hu, C., Huang, X., Chen, J., Liang, W., Yang, K., Jiang, H., et al. (2025). Dissecting the cellular reprogramming and tumor microenvironment in left- and right-sided Colorectal cancer by single cell RNA sequencing. Transl. Res. 276, 22–37. doi:10.1016/j.trsl.2024.12.002

PubMed Abstract | CrossRef Full Text | Google Scholar

Huang, C., Wang, X., Wang, Y., Feng, Y., Wang, X., Chen, S., et al. (2024). Sirpα on tumor-associated myeloid cells restrains antitumor immunity in colorectal cancer independent of its interaction with CD47. Nat. Cancer 5 (3), 500–516. doi:10.1038/s43018-023-00691-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Ji, L., Fu, G., Huang, M., Kao, X., Zhu, J., Dai, Z., et al. (2024). scRNA-seq of colorectal cancer shows regional immune atlas with the function of CD20+ B cells. Cancer Lett. 584, 216664. doi:10.1016/j.canlet.2024.216664

PubMed Abstract | CrossRef Full Text | Google Scholar

Jin, S., Guerrero-Juarez, C. F., Zhang, L., Chang, I., Ramos, R., Kuan, C.-H., et al. (2021). Inference and analysis of cell-cell communication using CellChat. Nat. Commun. 12 (1), 1088. doi:10.1038/s41467-021-21246-9

PubMed Abstract | CrossRef Full Text | Google Scholar

Lee, M. S., Menter, D. G., and Kopetz, S. (2017). Right versus left Colon cancer biology: integrating the consensus molecular subtypes. J. Natl. Compr. Cancer Netw. 15 (3), 411–419. doi:10.6004/jnccn.2017.0038

PubMed Abstract | CrossRef Full Text | Google Scholar

Leonard, N. A., Reidy, E., Thompson, K., McDermott, E., Peerani, E., Tomas Bort, E., et al. (2021). Stromal cells promote matrix deposition, remodelling and an immunosuppressive tumour microenvironment in a 3D model of Colon cancer. Cancers 13 (23), 5998. doi:10.3390/cancers13235998

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, J., Chen, D., and Shen, M. (2022a). Tumor microenvironment shapes colorectal cancer progression, metastasis, and treatment responses. Front. Med. 9, 869010. doi:10.3389/fmed.2022.869010

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, J., Fu, Y., Zhang, K., and Li, Y. (2022b). Integration of bulk and single-cell RNA-seq data to construct a prognostic model of membrane tension-related genes for Colon cancer. Vaccines 10 (9), 1562. doi:10.3390/vaccines10091562

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, Y., Wang, H., Mao, D., Che, X., Chen, Y., and Liu, Y. (2025). Understanding pre-metastatic niche formation: implications for colorectal cancer liver metastasis. J. Transl. Med. 23 (1), 340. doi:10.1186/s12967-025-06328-2

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, H., and Dilger, P. (2025). Different strategies for cancer treatment: targeting cancer cells or their neighbors? Chin. J. Cancer Res. 37 (2), 289–292. doi:10.21147/j.issn.1000-9604.2025.02.12

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, B., Li, S., Cheng, Y., Song, P., Xu, M., Li, Z., et al. (2024). Distinctive multicellular immunosuppressive hubs confer different intervention strategies for left- and right-sided colon cancers. Cell Rep. Med. 5 (6), 101589. doi:10.1016/j.xcrm.2024.101589

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, D., Li, C., Deng, Z., Luo, N., Li, W., Hu, W., et al. (2024). Multi-omics analysis reveals the landscape of tumor microenvironments in left-sided and right-sided colon cancer. Front. Med. 11, 1403171. doi:10.3389/fmed.2024.1403171

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, X., Xu, X., Wu, Z., Shan, Q., Wang, Z., Wu, Z., et al. (2022). Integrated single-cell RNA-seq analysis identifies immune heterogeneity associated with KRAS/TP53 mutation status and tumor-sideness in colorectal cancers. Front. Immunol. 13, 961350. doi:10.3389/fimmu.2022.961350

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu J, J., Li, H., Wang, L., Wang, S., and Tang, Q. (2024). Spatial transcriptome and single-cell reveal the role of nucleotide metabolism in colorectal cancer progression and tumor microenvironment. J. Transl. Med. 22 (1), 702. doi:10.1186/s12967-024-05495-y

PubMed Abstract | CrossRef Full Text | Google Scholar

Lu, Y., Chen, Y., Wang, Z., Shen, H., Xu, L., Huang, C., et al. (2025). Single-cell and spatial transcriptome profiling reveal CTHRC1+ fibroblasts promote EMT through WNT5A signaling in colorectal cancer. J. Transl. Med. 23 (1), 282. doi:10.1186/s12967-025-06236-5

PubMed Abstract | CrossRef Full Text | Google Scholar

Ma, F., Zhang, S., Song, L., Wang, B., Wei, L., and Zhang, F. (2021). Applications and analytical tools of cell communication based on ligand-receptor interactions at single cell level. Cell and Biosci. 11 (1), 121. doi:10.1186/s13578-021-00635-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Ma, B., Ueda, H., Okamoto, K., Bando, M., Fujimoto, S., Okada, Y., et al. (2022). TIMP1 promotes cell proliferation and invasion capability of right-sided colon cancers via the FAK/Akt signaling pathway. Cancer Sci. 113 (12), 4244–4257. doi:10.1111/cas.15567

PubMed Abstract | CrossRef Full Text | Google Scholar

Masuda, K., Kornberg, A., Miller, J., Lin, S., Suek, N., Botella, T., et al. (2022). Multiplexed single-cell analysis reveals prognostic and nonprognostic T cell types in human colorectal cancer. JCI Insight 7 (7), e154646. doi:10.1172/jci.insight.154646

PubMed Abstract | CrossRef Full Text | Google Scholar

Meyiah, A., Khan, F. I., Alfaki, D. A., Murshed, K., Raza, A., and Elkord, E. (2025). The colorectal cancer microenvironment: preclinical progress in identifying targets for cancer therapy. Transl. Oncol. 53, 102307. doi:10.1016/j.tranon.2025.102307

PubMed Abstract | CrossRef Full Text | Google Scholar

Micchelli, C. A., and Perrimon, N. (2006). Evidence that stem cells reside in the adult Drosophila midgut epithelium. Nature 439 (7075), 475–479. doi:10.1038/nature04371

PubMed Abstract | CrossRef Full Text | Google Scholar

Nemolato, S., Restivo, A., Cabras, T., Coni, P., Zorcolo, L., Orrù, G., et al. (2012). Thymosin β-4 in colorectal cancer is localized predominantly at the invasion front in tumor cells undergoing epithelial mesenchymal transition. Cancer Biol. and Ther. 13 (4), 191–197. doi:10.4161/cbt.13.4.18691

PubMed Abstract | CrossRef Full Text | Google Scholar

O’Brien, S. A., Orf, J., Skrzypczynska, K. M., Tan, H., Kim, J., DeVoss, J., et al. (2021). Activity of tumor-associated macrophage depletion by CSF1R blockade is highly dependent on the tumor model and timing of treatment. Cancer Immunol. Immunother. 70 (8), 2401–2410. doi:10.1007/s00262-021-02861-3

PubMed Abstract | CrossRef Full Text | Google Scholar

Peng, Q., Qiu, X., Zhang, Z., Zhang, S., Zhang, Y., Liang, Y., et al. (2020). PD-L1 on dendritic cells attenuates T cell activation and regulates response to immune checkpoint blockade. Nat. Commun. 11 (1), 4835. doi:10.1038/s41467-020-18570-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Qi, J., Sun, H., Zhang, Y., Wang, Z., Xun, Z., Li, Z., et al. (2022). Single-cell and spatial analysis reveal interaction of FAP+ fibroblasts and SPP1+ macrophages in colorectal cancer. Nat. Commun. 13 (1), 1742. doi:10.1038/s41467-022-29366-6

PubMed Abstract | CrossRef Full Text | Google Scholar

Qin, P., Chen, H., Wang, Y., Huang, L., Huang, K., Xiao, G., et al. (2023). Cancer-associated fibroblasts undergoing neoadjuvant chemotherapy suppress rectal cancer revealed by single-cell and spatial transcriptomics. Cell Rep. Med. 4 (10), 101231. doi:10.1016/j.xcrm.2023.101231

PubMed Abstract | CrossRef Full Text | Google Scholar

Rao, V. S., Gu, Q., Tzschentke, S., Lin, K., Ganig, N., Thepkaysone, M.-L., et al. (2022). Extravesicular TIMP-1 is a non-invasive independent prognostic marker and potential therapeutic target in colorectal liver metastases. Oncogene 41 (12), 1809–1820. doi:10.1038/s41388-022-02218-9

PubMed Abstract | CrossRef Full Text | Google Scholar

Shang, Z., Xi, S., Lai, Y., and Cheng, H. (2024). Single-cell transcriptomics and Mendelian randomization reveal LUCAT1’s role in right-sided colorectal cancer risk. Front. Genet. 15, 1357704. doi:10.3389/fgene.2024.1357704

PubMed Abstract | CrossRef Full Text | Google Scholar

Shao, X., Lu, X., Liao, J., Chen, H., and Fan, X. (2020). New avenues for systematically inferring cell-cell communication: through single-cell transcriptomics data. Protein and Cell 11 (12), 866–880. doi:10.1007/s13238-020-00727-5

PubMed Abstract | CrossRef Full Text | Google Scholar

Shi, X., Leng, L., Wang, T., Wang, W., Du, X., Li, J., et al. (2006). CD44 is the signaling component of the macrophage migration inhibitory Factor-CD74 receptor complex. Immunity 25 (4), 595–606. doi:10.1016/j.immuni.2006.08.020

PubMed Abstract | CrossRef Full Text | Google Scholar

Siegel, R. L., Kratzer, T. B., Giaquinto, A. N., Sung, H., and Jemal, A. (2025). Cancer statistics, 2025. CA A Cancer J. Clin. 75 (1), 10–45. doi:10.3322/caac.21871

PubMed Abstract | CrossRef Full Text | Google Scholar

Song, G., Xu, S., Zhang, H., Wang, Y., Xiao, C., Jiang, T., et al. (2016). TIMP1 is a prognostic marker for the progression and metastasis of colon cancer through FAK-PI3K/AKT and MAPK pathway. J. Exp. and Clin. Cancer Res. 35 (1), 148. doi:10.1186/s13046-016-0427-7

PubMed Abstract | CrossRef Full Text | Google Scholar

Sphyris, N., Hodder, M. C., and Sansom, O. J. (2021). Subversion of niche-signalling pathways in colorectal cancer: what makes and breaks the intestinal stem cell. Cancers 13 (5), 1000. doi:10.3390/cancers13051000

PubMed Abstract | CrossRef Full Text | Google Scholar

Sun, F., Li, H., Sun, D., Fu, S., Gu, L., Shao, X., et al. (2025). Single-cell omics: experimental workflow, data analyses and applications. Sci. China Life Sci. 68 (1), 5–102. doi:10.1007/s11427-023-2561-0

PubMed Abstract | CrossRef Full Text | Google Scholar

Tang, X., Huang, Y., Lei, J., Luo, H., and Zhu, X. (2019). The single-cell sequencing: new developments and medical applications. Cell and Biosci. 9 (1), 53. doi:10.1186/s13578-019-0314-y

PubMed Abstract | CrossRef Full Text | Google Scholar

Tauriello, D. V. F., and Batlle, E. (2016). Targeting the microenvironment in advanced colorectal cancer. Trends Cancer 2 (9), 495–504. doi:10.1016/j.trecan.2016.08.001

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, R., Li, J., Zhou, X., Mao, Y., Wang, W., Gao, S., et al. (2022). Single-cell genomic and transcriptomic landscapes of primary and metastatic colorectal cancer tumors. Genome Med. 14 (1), 93. doi:10.1186/s13073-022-01093-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, C., Zhao, Y., Zhang, S., Du, M., He, G., Tan, S., et al. (2024). Single-cell RNA sequencing reveals the heterogeneity of MYH11+ tumour-associated fibroblasts between left-sided and right-sided colorectal cancer. J. Cell. Mol. Med. 28 (18), e70102. doi:10.1111/jcmm.70102

PubMed Abstract | CrossRef Full Text | Google Scholar

Wen, J., Wang, S., Guo, R., and Liu, D. (2023). CSF1R inhibitors are emerging immunotherapeutic drugs for cancer treatment. Eur. J. Med. Chem. 245, 114884. doi:10.1016/j.ejmech.2022.114884

PubMed Abstract | CrossRef Full Text | Google Scholar

Wu, Y., Yang, S., Ma, J., Chen, Z., Song, G., Rao, D., et al. (2022). Spatiotemporal immune landscape of colorectal cancer liver metastasis at single-cell level. Cancer Discov. 12 (1), 134–153. doi:10.1158/2159-8290.CD-21-0316

PubMed Abstract | CrossRef Full Text | Google Scholar

Wu, Z., Lin, X., Ying, Y., Fan, G., Shi, J., Zheng, X., et al. (2025). A dual-targeting strategy to inhibit colorectal cancer liver metastasis via tumor cell ferroptosis and cancer-associated fibroblast reprogramming. Bioact. Mater. 52, 73–91. doi:10.1016/j.bioactmat.2025.05.025

PubMed Abstract | CrossRef Full Text | Google Scholar

Xiao, J., Yu, X., Meng, F., Zhang, Y., Zhou, W., Ren, Y., et al. (2024). Integrating spatial and single-cell transcriptomics reveals tumor heterogeneity and intercellular networks in colorectal cancer. Cell Death and Dis. 15 (5), 326. doi:10.1038/s41419-024-06598-6

PubMed Abstract | CrossRef Full Text | Google Scholar

Xie, Z., Niu, L., Zheng, G., Du, K., Dai, S., Li, R., et al. (2023). Single-cell analysis unveils activation of mast cells in colorectal cancer microenvironment. Cell and Biosci. 13 (1), 217. doi:10.1186/s13578-023-01144-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Xie, Z., Niu, L., Du, K., Chen, L., Zheng, G., Dai, S., et al. (2024). Endothelial cell heterogeneity in colorectal cancer: tip cells drive angiogenesis. Cell. Mol. Life Sci. 81 (1), 365. doi:10.1007/s00018-024-05411-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Xu, Y., Fu, G., Chen, X., He, R., Fang, Q., Tan, X., et al. (2025). Integrated single-nuclear RNA sequencing analysis reveals distinct characteristics of mucinous adenocarcinoma in right-sided colon cancer. Int. J. Biol. Macromol. 309, 142744. doi:10.1016/j.ijbiomac.2025.142744

PubMed Abstract | CrossRef Full Text | Google Scholar

Yang, L., Chu, Z., Liu, M., Zou, Q., Li, J., Liu, Q., et al. (2023). Amino acid metabolism in immune cells: essential regulators of the effector functions, and promising opportunities to enhance cancer immunotherapy. J. Hematol. and Oncol. 16 (1), 59. doi:10.1186/s13045-023-01453-1

PubMed Abstract | CrossRef Full Text | Google Scholar

Yang, S., Zhang, D., Sun, Q., Nie, H., Zhang, Y., Wang, X., et al. (2024). Single-cell and spatial transcriptome profiling identifies the transcription factor BHLHE40 as a driver of EMT in metastatic colorectal cancer. Cancer Res. 84 (13), 2202–2217. doi:10.1158/0008-5472.CAN-23-3264

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, L., Li, Z., Skrzypczynska, K. M., Fang, Q., Zhang, W., O’Brien, S. A., et al. (2020). Single-cell analyses inform mechanisms of myeloid-targeted therapies in Colon cancer. Cell 181 (2), 442–459.e29. doi:10.1016/j.cell.2020.03.048

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhao, H., Wu, L., Yan, G., Chen, Y., Zhou, M., Wu, Y., et al. (2021). Inflammation and tumor progression: signaling pathways and targeted intervention. Signal Transduct. Target. Ther. 6 (1), 263. doi:10.1038/s41392-021-00658-5

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhao, Y., Shen, M., Wu, L., Yang, H., Yao, Y., Yang, Q., et al. (2023). Stromal cells in the tumor microenvironment: accomplices of tumor progression? Cell Death and Dis. 14 (9), 587. doi:10.1038/s41419-023-06110-6

PubMed Abstract | CrossRef Full Text | Google Scholar

Glossary

5-FU 5-fluorouracil

BAFF B cell-activating factor

BHLHE40 basic helix-loop-helix protein 40

BTK Bruton tyrosine kinase

C1QC C-chain polypeptide of serum complement subcomponent C1q

CAFs cancer-associated fibroblasts

CCL8 C-C motif chemokine ligand 8

CD103 integrin αE

CD38 cyclic ADP ribose hydrolase

CR complete response

CRC colorectal cancer

CSF1R colony stimulating factor 1 receptor

CTHRC1 collagen triple helix repeat containing 1

CXCL1 growth-regulated oncogene-α-positive

CXCL12 stromal derived factor 1

CXCL13 chemokine ligand 13

CXCR4 chemokine receptor type 4

DC dendritic cell

DCN decorin

dMMR deficient MMR

ECM extracellular matrix

ECs endothelial cells

EMT epithelial-mesenchymal transition

F4/80 adhesion G protein-coupled receptor

FAP fibroblast activation protein α

GNLY granulysin

GPX4 glutathione peroxidase 4

GSEA gene set enrichment analysis

GSVA gene set variation analysis

HDAC histone deacetylase

ICL immune checkpoint ligand

ICR immune checkpoint receptor

IHC immunohistochemical

IS meta-program immune secretory meta-program

IS immune secretory

ITAMs immunoreceptor tyrosine-based activation motifs

KO knockout

LAG3 Lymphocyte-Activation Gene 3

LAMP3 lysosomal-associated membrane protein 3

LAMs Lipid-associated macrophages

LCRC left-sided colorectal cancer

LPCAT2 lysophosphatidylcholine acyltransferase 2

LpMs lamina propria macrophages

LYVE1 lymphatic vessel endothelial hyaluronan receptor 1-positive

MAC mucinous adenocarcinoma

mCRC metastatic colorectal cancer

MCs mast cells

MIF macrophage migration inhibitory factor

MMPs matrix metalloproteinases

MMR mismatch repair

MMs muscularis macrophages

MSLN mesothelin

MTRGs membrane-tension related genes

MYH11 myosin heavy chain 11

NAC neoadjuvant chemotherapy

NME1 nucleoside diphosphate kinase A

NR non-responders

NU^highcells high nucleotide metabolism cells

OS overall survival

PD-1 programmed cell death protein 1

PHLDA2 pleckstrin homology like domain family A member 2

PI16 peptidase inhibitor 16

PLA2 secreted phospholipase A2

pMMR proficient MMR

PMN pre-metastatic niche

PPAR peroxisome proliferator-activated receptor

PR partial response

PS meta-program proliferation stemness meta-program

PS proliferation stemness

RCRC right-sided colorectal cancer

RHOB phosphate regulon transcriptional regulatory protein B

Sc-seq single-cell sequencing

SD disease stabilization

SFRP2 frizzled-related protein 2

SIRPA signal regulatory protein α

SLIT2 slit guidance ligand 2

SmMs submucosal macrophages

SPP1 osteopontin

SYK spleen tyrosine kinase

TAMs tumor-associated macrophages

Tfh T follicular helper

TFRC transferrin receptor

TIMP1 tissue inhibitor of metalloproteinase 1

TME tumor microenvironment

TMSB4X thymosin beta-4