Skip to main content
Download PDF

Single-cell and spatial transcriptomics reveal SPP1-CD44 signaling drives primary resistance to immune checkpoint inhibitors in RCC

Journal of Translational Medicine volume 22, Article number: 1157 (2024) Cite this article

Abstract

Background

Immune checkpoint inhibitors (ICIs) are a cornerstone therapy for advanced renal cell carcinoma (RCC). However, significant rates of primary resistance hinder their efficacy, and the underlying mechanisms remain poorly understood. This study aims to unravel the tumor-immune interactions and signaling pathways driving primary resistance to ICIs in RCC.

Methods

We integrated single-cell RNA sequencing, spatial transcriptomics, and clinical sample analysis to investigate the tumor microenvironment and intercellular signaling. Advanced computational methods, including cell–cell communication networks, pseudotime trajectories, and gene set enrichment analysis (GSEA), were employed to uncover the underlying resistance mechanisms.

Results

Compared to the sensitive group, the primary resistance group exhibited a significant increase in SPP1-CD44 signaling-mediated interactions between tumor cells and immune cells. These interactions disrupted antigen presentation in immune effector cells and suppressed key chemokine and cytokine pathways, thereby impairing effective immune responses. In contrast, the sensitive group showed more active antigen presentation and cytokine signaling, which facilitated stronger immune responses. Furthermore, the interaction between SPP1-secreting tumor cells and CD44-expressing exhausted CD8 + T cells activated the MAPK signaling pathway within CD8 + Tex cells, exacerbating T cell exhaustion and driving the development of ICI resistance in RCC.

Conclusion

Our findings reveal a potential mechanism by which SPP1-CD44 signaling mediates tumor-immune cell interactions leading to ICI resistance, providing a theoretical basis for targeting and disrupting this signaling to overcome primary resistance in RCC.

Introduction

Renal cell carcinoma (RCC) is the most lethal type of cancer among urinary tract tumors, and clear cell renal cell carcinoma (ccRCC) is the most common histopathologic type [1]. The treatment of advanced RCC remains challenging, especially after the failure of initial targeted therapy [2, 3]. Currently, the standard approach for first-line and refractory advanced RCC is a combination of targeted therapy and immune checkpoint inhibitors (ICIs) therapy [4]. However, ICIs show sensitivity in only 10–30% of patients, and the majority of patients demonstrate primary resistance with no objective response to initial therapy [5, 6]. Primary resistance to ICIs results from pre-existing factors [7, 8], and investigating these key pre-existing factors can offer valuable insights into the underlying mechanisms of primary resistance to ICIs therapy and guide the development of strategies to overcome this resistance.

The occurrence of primary resistance to immunotherapy can be ascribed to a combination of tumor cell-intrinsic factors (genomic or proteomic features) and tumor cell-extrinsic factors, genetic heterogeneity, gut microbiome, and other environmental factors [9, 10]. Research suggests that mutations in the PTEN gene, abnormalities in JAK/STAT and Hippo signaling pathways, as well as low tumor mutation burden (TMB), reduce tumor immunogenicity, which hinders effective T cell recognition and weakens the immune response [11, 12]. Furthermore, MAPK pathway activation limits T cell recruitment and function, facilitating immune escape [13]. Tumor cells evade immune detection by downregulating MHC I molecules and upregulating PD-L1, which impairs the efficacy of ICIs [14]. Within the tumor microenvironment, immune-suppressive cells such as Tregs, MDSCs, and TAMs release cytokines like IL-10 and TGF-β, which suppress effector T cell activity [15]. T cell exhaustion, characterized by high expression of inhibitory receptors like PD-1 and LAG-3, further reduces the effectiveness of ICIs by limiting T cell recovery and function [16]. Despite advancements, the mechanisms underlying primary resistance to ICIs in RCC remain insufficiently understood.

Extrinsic mechanisms of resistance, whether primary or adaptive, involve the interplay between tumor cells and immune cells within the tumor microenvironment (TME) [17]. Single-cell RNA sequencing (scRNA-seq) with its high-resolution capabilities has the potential to identify previously uncharacterized cell subpopulations and elucidate their functional roles in tumor progression and treatment response [18]. Additionally, scRNA-seq offers a unique opportunity to investigate cell-to-cell interactions, unveiling intricate communication networks among diverse cell types within the TME [19]. Moreover, the application of single-cell trajectory analysis enables the robust reconstruction of complex developmental pathways and facilitates the delineation of cellular state transitions based on single-cell transcriptomic data, providing a comprehensive approach to comprehending the dynamic changes occurring within the cellular landscape [20].

In this study, we integrated single-cell sequencing data, TCGA database, spatial transcriptomics, and clinical samples to elucidate the complex interactions between tumor and immune cells within the TME of RCC. We found that SPP1-CD44 signaling-mediated interactions between tumor and immune cells were significantly increased in the primary resistance group, potentially impairing antigen presentation and inactivating critical immune pathways. Moreover, interactions between SPP1-secreting tumor cells and CD44-expressing exhausted CD8 + T cells (CD8 + Tex) activated the MAPK pathway, exacerbating CD8 + T cell exhaustion and promoting primary resistance to ICIs. Our findings reveal that SPP1-CD44 signaling drives tumor-immune interactions contributing to ICIs resistance, providing a theoretical basis for targeting this signaling to overcome primary resistance in RCC.

Materials and methods

Data collection and human specimens

The data for single-cell transcriptome sequencing were retrieved from the Single Cell Portal public database (https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.ccell.2021.02.015) [21] . The dataset consisted of a sample of 5 RCC patients treated for the first time with anti-PD1 therapy. According to RECIST v1.1 criteria, the 2 patients who responded to immunotherapy were in the sensitive group, whereas the 3 patients who did not respond were in the primary resistance group [8]. Bulk RNA data for KIRC patients, including 529 ccRCC patients, were obtained from the TCGA database (https://tcgadata.nci.nih.gov/). Spatial transcriptome data were obtained from the GEO database, and the GSE175540 dataset contains 24 ccRCC patients (https://www.ncbi.nlm.nih.gov/geo). Our clinical samples included two cohorts. The first cohort consisted of 8 paraffin-embedded pathology samples from patients treated with ICIs, with 4 cases in the sensitive group and 4 cases in the primary resistance group. The second cohort included 20 biopsy or surgical samples from patients treated with ICIs, with 10 cases in the sensitive group and 10 cases in the primary resistance group. At least two of our pathologists confirmed the ccRCC pathology type, and the baseline characteristics of the cohorts are shown in Supplementary Table 1 and Supplementary Table 2. The collection of samples used in this study was approved by the Ethics Committee of the Affiliated Cancer Hospital of Xinjiang Medical University (S-2024005).

Single-cell transcriptome sequencing data analysis

We utilized the Seurat (version 4.3.0) [22] package to process the sparse expression matrix data, and initially performed quality control (QC) on the data. Following QC and filtering, cells were retained based on the criteria of having between 200 and 2500 RNA features and a mitochondrial gene percentage below 15%. The analysis of single-cell transcriptome data involves normalization, dimensionality reduction and clustering. The integrated and scaled data undergo principal component analysis (PCA), retaining the top 30 principal components. Following this, t-distributed Stochastic Neighbor Embedding (t-SNE) analysis is conducted using the RunTSNE function, along with FindNeighbors and FindClusters functions with a resolution of 0.7, to determine cell clusters. Through automated and manual annotation, a total of 12 cell types were identified. Subsequently, we performed differential gene expression analysis using the Seurat package's FindMarkers function for each cell type within the primary resistant and sensitive subgroups.

Cell–cell communication analysis

We utilized the R package "CellChat" (version 1.6.1) [19] to investigate intercellular communication interactions at the single-cell level. First, gene expression data were processed to identify highly expressed ligands and receptors within specific cell populations. Following this, communication probabilities for all ligand-receptor pairs associated with each signaling pathway were calculated, which enabled us to construct comprehensive cell–cell communication networks. We then compared the total number of interactions and their respective strengths within the inferred communication networks of primary resistance and sensitive groups. For spatial transcriptomics analysis, we followed the CellChat official guidelines (available at CellChat GitHub) to spatially map intercellular communication between cell populations. Additionally, the “ICELLNET” R package was employed to validate cell communication specifically within the primary resistance group, with detailed procedures available in the ICELLNET repository.

Single-cell trajectory analysis

To investigate cellular state transitions, we conducted trajectory analysis using the Monocle R package (version 2.28.0) [23, 24]. This involved selecting and ordering genes from scRNA-seq data, filtering and estimating size factors, followed by dimensionality reduction using the DDRTree algorithm. Visualization of the trajectory was performed through cell state and cell type plots. Distinct developmental trajectories for each cell type were also depicted. Gene expression along the trajectory was examined by visualizing specific genes with a color gradient.

Gene set enrichment analysis (GSEA) analysis

To investigate the signaling pathway status of the target gene set enrichment, we used the “fgsea” R package (version 1.24.0) [25] to perform functional enrichment analysis. We picked “c2.cp.kegg_legacy.v2023.2.Hs.symbols.gmt” as the reference gene set, and followed the activation or inhibition status of pathways enriched by the target gene set in different subgroups. Pathways with an absolute value of the Normalized Enrichment Score (|NES|) > 1 and a p.adjust < 0.01 were defined as significant pathways.

Correlation analysis

We evaluated the exhaustion score of each CD8 + T cell and the MAPK signaling pathway score of exhausted CD8 + T cells using the "AUCell" (version 1.22.0) software package [26]. This step involved quantifying the activity of gene sets associated with T cell exhaustion and the MAPK signaling pathway, providing a functional readout for each cell. Next, we employed the R package "corrplot" (version 0.92) to calculate and visualize the correlations between the CD8 + T cell exhaustion scores and the MAPK signaling pathway scores. Additionally, we calculated the correlation between CD44 gene expression and the MAPK signaling pathway scores in exhausted CD8 + T cells to explore potential regulatory interactions.

Spatial transcriptome analysis

We processed the spatial transcriptomics data from the GSE175540 dataset, specifically from samples P162, P160, P159, P158, P157, and P156, using the Seurat package. First, we performed normalization for each dataset using the transform function to reduce technical noise and detect highly variable features. Then, we visualized the spatial expression patterns of specific genes using the SpatialFeaturePlot function. Subsequently, we conducted PCA on the normalized data and performed clustering using the nearest neighbor method. We applied the UMAP method for dimensionality reduction to facilitate visualization and displayed the clustering and dimensionality reduction results using the DimPlot and SpatialDimPlot functions. Each cell group was annotated based on known marker genes. Additionally, we mapped the marker gene CA9 of ccRCC, its secreted signaling molecule SPP1, the marker gene CD8A of CD8 + T cells, the exhaustion marker gene HAVCR2, and CD44 onto the spatial transcriptomic maps.

Multiplex immunofluorescence (mIF) analysis

The tumor tissue sections employed in this study were procured from the Pathology Department of our institution. The sections were first deparaffinized and underwent antigen retrieval, followed by overnight incubation with primary antibodies at 4 °C. After washing, the sections were incubated with fluorescently labeled secondary antibodies at room temperature for 2 h, and the nuclei were stained with DAPI. Finally, the sections were mounted and observed using a fluorescence or confocal microscope.

Statistical analysis

For this study, all statistical analyses were performed using R version 4.3.1. Two-tailed p-values were used to evaluate statistical significance. We employed various statistical tests to analyze the data. The Wilcoxon rank-sum test and Kruskal–Walli’s rank-sum test, followed by Dunn's test for multiple comparisons, were utilized for non-parametric data analysis. The Spearman correlation analysis was conducted to assess the relationships between variables, and the log-rank test was used for survival analysis. A p-value of less than 0.05 was considered statistically significant, with significance levels denoted as follows: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001, or the specific p-value was reported when applicable.

Results

Single-cell transcriptomics identifies immune microenvironment differences in RCC patients undergoing ICI treatment

To characterize the immune landscape of RCC patients treated with ICI, we reanalyzed a publicly available scRNA-seq dataset [21]. The dataset consisted of a total of eight RCC patients, five of whom first received ICI therapy, of which two were evaluated as having partial remission (PR) and three were evaluated as having stable disease (SD) and disease progression (PD) after drug administration. We categorized the two patients in PR as the sensitive group (Sen) and those with SD and PD as the primary resistance group (Pri.R). We employed t-distributed random neighborhood embedding (tSNE) to arrange cells into 19 clusters, with 9966 cells classified as the Sen group (Sen) and 5160 cells as the Pri.R group (Fig. 1A). Based on the expression levels of established marker genes as reported previously [27,28,29], we classified all cells into 12 distinct cell types (Fig. 1B, C).

Fig. 1
figure 1

Construction of single-cell transcriptome profiles. A T-SNE plot depicting the distribution of 15,126 tumor cells from 5 RCC patients. B T-SNE plot showing the annotated single-cell profiles consisting of 12 cell types. C Bubble plots showing marker genes expressed in the major cell types; the color of the dots reflects the level of expression, and the size of the dots represents the percentage of cells expressing the marker genes in the different cell types. D Table showing the number of different cells in each group and the percentage of the cells in each group. E T-SNE plots demonstrating the distribution of the expression of PDCD1 and its receptor genes. F Bubble plots showing the distribution of the expression of the immune checkpoint LAG3 and its receptor genes in different subgroups. Pri.R, primary resistance group. Sen, sensitive group. Pcc, peritubular capillary cells

Full size image

Subsequently, we observed alterations in the proportion of immune cell infiltration. In the Pri.R, the most prominent cellular infiltrates were CD8 + Tex cells (20%), CD4 + T cells (19.9%), and B cells (13.6%), whereas the Sen exhibited a predominance of CD8 + Tex cells (49.6%), M2 cells (13.8%), and NK cells (12.4%) in terms of cellular infiltration. By comparing the proportions of various cell types in the infiltrates of both groups, we identified CD8 + Tex cells (-29.6%), B cells (+ 13.1%), CD4T cells (+ 9.1%), and plasma cells (+ 8.0%) as the cell types exhibiting the most pronounced changes (Fig. 1D). Reversing the exhaustion of CD8 + Tex cells and restoring their immune function is a key mechanism of ICI therapy targeting PD-1 and PD-L1 [30]. Therefore, we further evaluated the expression of PD1 and its ligands (PD-L1, PDCD1LG2). The results showed that PDCD1 was enriched in the Sen but sparsely distributed in the Pri.R, while PDCD1LG2 and CD274 were expressed at low levels in both groups, particularly in tumor cells (Fig. 1E).

Immunotherapy agents target and inhibit immune checkpoint pathways, helping the immune system to overcome the escape of immunity caused by the over-expression of checkpoints [31]. For this purpose, we examined the expression levels of common immune checkpoints and their ligands. The results showed that LAG3 and its ligand MHC class II molecules were significantly enriched in the sensitive group (Fig. 1F). It was hypothesized that LAG3 and its ligands might be more suitable ICI therapeutic targets for RCC. Taken together, significant differences in the immune microenvironment exist between the Pri.R and Sen, particularly in the expression patterns of CD8 + Tex cells and immune checkpoints.

Immune cell characteristic gene expression patterns in primary resistance and sensitivity

To dissect the underlying mechanisms driving tumors' primary resistance to immune checkpoint inhibitors, differential gene expression analysis was carried out on pivotal immune cells within both two groups (Fig. 2A). We subsequently extracted the intersection of differential genes in these immune cells to identify genes associated with primary resistance and sensitivity, respectively (Fig. 2B). Analysis revealed that six genes (IGHA1, IGHG1, IGHG3, IGKC, IGLC2 and IGLC3) linked to immunoglobulins exhibited marked upregulation in the Pri.R (Fig. 2C, D). Conversely, in the Sen, four genes, encompassing three related to MHC class II molecule-related genes (HLA-DQA1, HLA-DRB1, HLA-DRB5) and the cytokine IL-32, showed significant upregulation (Fig. 2E, F). Additionally, we validated the expression trends of the aforementioned genes in clinical samples through multiplex immunofluorescence staining, confirming consistent expression patterns (Fig. 2G, H).

Fig. 2
figure 2

Differential gene expression analysis of major immune cells. A Volcano plot showing the differential gene expression analysis of major immune cells. Different colors represent upregulated (red) and downregulated (blue) genes. B Upset plots displaying the intersection of DEGs among various immune cells in the Pri.R and Sen, respectively. C Violin plots showing the expression distribution of genes highly expressed in the Pri.R across major immune cells. D Violin plots showing genes with higher expression levels in the Pri.R compared to the Sen. E Violin plots showing the expression distribution of genes highly expressed in the Sen across major immune cells. F Violin plots showing genes with higher expression levels in the Sen compared to the Pri.R. G, H Multiplex immunofluorescence staining showing the protein level expression of differentially expressed genes in clinical samples from different groups (Pri.R = 4, Sen = 4). I Kaplan–Meier survival curves showing the prognostic significance of differentially expressed genes validated using the TCGA dataset. J, K GO enrichment analysis for molecular functions (MF) of highly expressed genes in the Pri.R (J) and Sen (K) groups

Full size image

We further investigated the association of these genes with clinical prognosis in the TCGA-KIRC cohort (n = 529), finding that a higher expression of the immunoglobulin-related gene correlates with poorer overall survival (OS); conversely, HLA-DRB5 emerged as the sole gene with prognostic significance, where its higher expression is associated with improved OS (Fig. 2I). Importantly, the immunoglobulin-related genes predominantly function in antigen and immunoglobulin receptor binding (Fig. 2J). Moreover, genes associated with MHC class II molecules play roles in key immune activities such as antigen binding, immunoglobulin receptor binding, and MHC class II protein complex binding (Fig. 2K).MHC class I molecules present endogenous antigens primarily to CD8 + cytotoxic T cells, whereas MHC class II molecules present exogenous antigens primarily to CD4 + helper T cells [32]. Together, within the context of primary resistance and sensitivity to immune checkpoint inhibitors, our findings reveal that the immune system exhibits differential responses, highlighting the intricate balance between immune recognition and immune evasion.

Impact of intercellular communication differences on ICI treatment response

The interaction between immune cells and tumor cells in the TME is intricate and influences the anti-tumor response [33, 34]. To investigate the impact of tumor cell and immune cell interactions in TME on ICI treatment, we analyzed intercellular communication using the CellChat package [19]. Within the cellular populations of the Pri.R, we identified 12 intercellular communication signals making the largest contributions to incoming or outgoing signaling (Fig. 3A). In contrast, in the cellular populations of the sensitivity group, we detected 19 predominant signals contributing to either signal inflow or outflow (Fig. 3B). Moreover, four ligand-receptor pairs (L-R) in the Pri.R and seven L-R in the Sen served as the primary mediators of communication between tumor cells and immune cells, respectively (Fig. 3C, D). Notably, TOP2 signaling was pro-carcinogenic MIF and SPP1 signaling in both groups. Meanwhile, the groups exhibited activation of ANGPTL, MK, and VEGF signals, which stimulate tumor development and angiogenesis [35, 36]. Nevertheless, the sensitive group exhibited heightened CXCL, CCL, and IFN-II signaling linked to the activation of immune cells, their recruitment, and the induction of anticancer effects.

Fig. 3
figure 3

Intercellular signaling pathway analysis in Pri.R and Sen groups. A, B Heatmaps showing the outgoing and incoming signaling patterns in the Pri.R group and Sen group across various cell types. The relative strength of each signaling pathway is indicated by color intensity. C, D Dot plot representing the communication probability of key signaling interactions in the Pri.R group and Sen group, with the size of the dots indicating the communication probability and color representing the p-value significance. E Bar graphs comparing the number of inferred interactions and interaction strength between the Sen and Pri.R groups. F Network diagrams illustrating the differential number of interactions (left) and differential interaction strength (right) among various cell types between the Sen and Pri.R groups. The thickness of the lines represents the interaction strength. G Bar graph depicting the information flow of key signaling pathways in the Pri.R and Sen groups. H Chord diagram visualizing the SPP1 signaling pathway network, highlighting interactions between different cell types. I Dot plot showing increased signaling interactions in the Pri.R group, with the size of the dots indicating the level of statistical significance (p-value) and the color representing the communication probability

Full size image

Upon analysis of the quantity and intensity of interaction signals between the cell populations of the two groups, we found that the sensitivity group exhibited higher levels of both strength and number of interactions compared to the Pri.R (Fig. 3E, F). Meanwhile, we observed that 3 signals were specifically activated in the Pri.R, 12 signals were specifically activated in the Sen, and 3 more were activated at different levels in both groups (Fig. 3G). Interestingly, the SPP1 signaling pathway exhibits a higher level of activation in the Pri.R. Tumor cells in the Pri.R reshape the immune environment to create a suppressive state by utilizing the SPP1 signaling pathway to communicate with different immune cells, including CD8 + Teff cells, CD8 + Tex cells, NK cells, B cells, and M2 macrophages (Fig. 3H). Additionally, we found that a relative increase in the SPP1-CD44 ligand-receptor pair causes the significant activation of the SPP1 signal in the Pri.R (Fig. 3I). To ensure the robustness of this conclusion, we used the R package "ICELLNET" [37] to analyze cell communication in the Pri.R. The tumor-centered communication network revealed that tumor cells engage in significant interactions with multiple immune cells, particularly CD8 + Tex cells, through the SPP1-CD44 ligand-receptor pair (Supplementary Fig. 1A, B). Moreover, SPP1-CD44 showed the highest specificity in interactions between tumor cells and CD8 + Tex cells, further supporting the role of this signal in fostering resistance by shaping an immunosuppressive microenvironment in the resistance group (Supplementary Fig. 1C, D).

Overall, both the primary resistance and sensitivity groups displayed signals that stimulate tumor proliferation, migration, and invasion, which are crucial elements in the advancement of tumors. The Pri.R had a higher level of activation of immune suppressive signals, which assisted tumor cells in evading immune identification. In contrast, the sensitivity group exhibited increased activation of immune-activating signals that augment the immune system's capacity to identify and eradicate cancerous cells.

SPP1-CD44 inter-cellular signaling mediates tumor-immune cell interactions driving drug resistance

To further explore the subpopulations of tumor cells secreting SPP1 signals from the perspective of intra-tumor heterogeneity, we employed subgroup clustering analysis to classify tumor cells into eight distinct clusters (Fig. 4A). Next, we arranged the eight tumor subclusters along a proposed temporal trajectory, revealing two distinct cell fate differentiation trajectories. In particular, we found that clusters C2 and C3 were located at the end of cell fate 1, characterizing the primary drug-resistant state, while clusters C1 and C6 were located at the end of cell fate 2, reflecting the heterogeneity of the cell states encompassing both primary drug-resistant and sensitive (Fig. 4B). Additionally, we observed that SPP1's dynamic expression patterns are different in cell fates 1 and 2 (Fig. 4C). Cell fate 1 showed significantly high expression of SPP1 in the C2 and C3 clusters, while cell fate 2 showed no significant differences in its expression patterns (Fig. 4D). Furthermore, our statistical analysis of the population level of the 8 cell clusters revealed that the expression of SPP1 was significantly higher in the Pri.R than in the Sen (Fig. 4E). These findings emphasize the possibility that SPP1 may play a critical role in primary drug-resistant cells in the C2 and C3 clusters.

Fig. 4
figure 4

SPP1-CD44 inter-cellular signaling mediates tumor-immune cell interactions driving resistance. A t-SNE plot showing the clustering of tumor cells into eight distinct groups based on their gene expression profiles. B Pseudotime analysis depicting two cell fate differentiation trajectories in tumor cells. C Expression of SPP1 along the pseudotime trajectory, highlighting different cell fates and clusters. D Box plots showing SPP1 expression across different cell fate branches. E Box plot comparing SPP1 expression between the Pri.R and Sen groups (Wilcoxon test, p = 0.0081). F Volcano plot displaying DEGs in tumor cells, highlighting upregulated (red) and downregulated (blue) genes. G GSEA analysis showing pathways enriched in DEGs from tumor cells. H Bubble plot showing the expression of SPP1 and CD44 across different cell types and conditions (Pri.R and Sen). I GSEA analysis results for various signaling pathways in three immune cell types, with genes ordered from drug-resistant to sensitive groups. J Multiplex immunofluorescence staining showing the protein level expression of SPP1 and CD44 in clinical samples from the Pri.R and Sen groups (Pri.R = 4, Sen = 4). ES, enrichment score

Full size image

Subsequently, we performed differential gene expression analysis on tumor cells and identified a total of 462 differentially expressed genes (DEGs) (Fig. 4F). To further explore the biological importance of these genes that were expressed differently, we employed GSEA and indicated that the genes that were expressed differently were primarily associated with immunological and viral responses, including conditions such as "autoimmune thyroid disease", "graft-versus-host disease" and "systemic lupus erythematosus". Moreover, the increased presence of specific genes involved in important cellular processes such as ribosomes, endocytosis, and antigen processing and presentation suggest that drug-resistant cells may have distinct cellular biology, including modified cellular interactions and protein synthesis pathways (Fig. 4G).

Furthermore, we observed that relevant immune killer cells (NK cells, CD8 + Tex, and CD8 + Teff cells) in the Pri.R highly expressed the receptor gene of the SPP1 signaling molecule, CD44. This suggests the potential importance of the SPP1-CD44 pair in regulating tumor drug resistance and immune responses (Fig. 4H). Importantly, we found a general inactivation of antigen processing and presentation pathways in the relevant immune killer cells. In addition, chemokine pathways and cytokine-receptor interaction pathways were also shown to be inactivated in CD8 + Teff cells. The altered activity of these pathways implies that immune killer cells fail to efficiently acquire and present antigenic information and thus fail to recognize cancer cells or stimulate a stronger immune response. Meanwhile, we observed significant activation of the MAPK signaling pathway in CD8 + Tex cells in the drug-resistant group, which may accelerate the depletion process of these cells (Fig. 4I). Finally, we confirmed that tumor tissues obtained from patients who are resistant to primary drugs exhibit more pronounced fluorescence signals for SPP1 and CD44 (Fig. 4J). Taken together, we describe that the interaction between SPP1-secreting tumor cells and immune cells expressing CD44 ligand may be a driver of primary resistance.

Impact of MAPK pathway activation on CD8 + T cell exhaustion and resistance in ICIs

The primary objective of ICIs is to restore the cytotoxic capabilities of CD8 + Tex cells, a crucial aspect with significant implications for cancer immunotherapy [38, 39]. Next, our study focuses on examining the role and impact of MAPK signaling pathway activation in CD8 + Tex cells within the Pri.R. The correlation analysis showed that the MAPK pathway score was strongly correlated with the degree of CD8 + T cell exhaustion, suggesting that the MAPK pathway may have a crucial function in regulating the exhaustion of T cells (Fig. 5A). Similarly, the expression level of CD44 in CD8 + Tex cells also showed a strong positive correlation with the MAPK pathway score (Fig. 5B). However, the expression of SPP1 showed a weak but significant correlation with the MAPK pathway score (Fig. 5C).

Fig. 5
figure 5

Impact of MAPK pathway activation on CD8 + T cell exhaustion and resistance in ICIs. A Correlation between the MAPK pathway score and the exhaustion score of CD8 + T cells (R = 0.44, p < 2.2e−16). B Correlation between the MAPK pathway score and CD44 expression in exhausted CD8 + T cells, showing a strong positive correlation (R = 0.61, p < 2.2e−16). C Correlation between the MAPK pathway score and SPP1 expression, showing a weak but significant correlation (R = 0.055, p = 2.3e−05). D t-SNE plot showing the clustering of exhausted CD8 + T cells into 12 distinct clusters. E Pseudotime analysis depicting two differentiation trajectories in exhausted CD8 + T cells. F Expression patterns of MAPK-related genes along the pseudotime trajectory, highlighting increased expression at the end of cell fate 2 corresponding to the drug-resistant state. G Expression patterns of CD44 and MHC class II molecules along the pseudotime trajectory, showing increased CD44 expression and decreased MHC class II molecule expression in the resistant state. H Violin plots comparing the expression levels of key immune effector genes between the Sen and Pri.R groups (p < 0.001). I Immunofluorescence images of Pri.R and Sen stained for CD8, LAG3, pERK, and FOS, with quantification showing higher expression levels of these markers in the Pri.R. Scale bar: 10 μm

Full size image

Subsequently, we grouped CD8 + Tex cells into 12 cell clusters and created pseudo-temporal trajectories for these cellular groupings (Fig. 5D). The cells underwent differentiation along two distinct pathways, resulting in the formation of cell fate 1 and cell fate 2. The end of cell fate 1 represented CD8 + Tex cells in the Sen, while the end of cell fate 2 represented CD8 + Tex cells in the Pri.R (Fig. 5E). After analyzing the changing expression patterns of genes related to MAPK, we discovered a considerable increase in the expression of these genes at the end of cell destiny 2, which corresponds to the drug-resistant state (Fig. 5F). This discovery implies that the activation of the MAPK pathway could be a crucial mechanism in the CD8 + Tex cells of the Pri.R.

The expression of CD44 gradually increased along the cell fate 2 trajectory and reached its highest point in the resistant state, whereas the expression of MHC class II molecules (HLA-DOA1, HLA-DRB1, HLA-DRB5) exhibited a declining pattern along the same trajectory (Fig. 5G). Furthermore, our findings revealed a significant reduction in the expression of key immune effector genes like granzyme B (GZMB), perforin (PRF1), and -interferon (IFNG) in the resistant group compared to the sensitive group (Fig. 5H). In clinical samples, we observed that the fluorescence intensity of the MAPK pathway markers, pERK and FOS, was significantly higher in exhausted CD8 + T cells within the Pri.R compared to the Sen (Fig. 5I). Taken together, these findings confirm that the MAPK signaling pathway is markedly activated in CD8 + Tex cells in primary resistant tumor tissues.

Spatial transcriptomics reveals SPP1-CD44 mediated interactions between tumor cells and CD8 + Tex cells

To overcome scRNA-seq's limitations in terms of spatial information, we analyzed the spatial transcriptomics data of six ccRCC frozen sections from the GSE175540 dataset [40]. Initially, we observed a significant level of spatial convergence in the areas where the ccRCC marker gene CA9 and the signaling chemical SPP1 are expressed and distributed. In addition, we examined the expression of the marker genes CD8A and HAVCR2 in exhausted CD8 + T cells, as well as the receptor gene CD44 for SPP1(Fig. 6A). These results suggest that tumor cells actively secrete SPP1 in the space where they are located, and CD8 + Tex cells actively express the ligand CD44 in the space where they are located.

Fig. 6
figure 6

Spatial transcriptomics reveals SPP1-CD44 mediated tumor-immune interactions. A Spatial expression of CA9, SPP1, CD44, CD8A, and HAVCR2 in six ccRCC sections. B Spatial mapping in the P157 sample showing the proximity of tumor cells, exhausted CD8 + T cells, and M2 macrophages. C Number and strength of cell–cell interactions mediated by the SPP1 signaling pathway. D SPP1 signaling pathway network illustrating interactions among tumor cells, CD8 + T cells, and M2 macrophages. E Contribution of each ligand-receptor pair, highlighting SPP1-CD44 as dominant. F Violin plots of SPP1 and CD44 expression across cell types. G Network diagram of spatial SPP1 communication, emphasizing strong interactions between tumor cells, exhausted CD8 + T cells, and M2 macrophages

Full size image

We selected P157 samples extensively infiltrated by tumor cells and CD8 + Tex cells among 6 samples for in-depth analysis. We first annotated five cell types in the P157 samples and performed spatial mapping, observing that CD8 + Tex cells and immunosuppressive M2 macrophages were spatially close to tumor cells early (Fig. 6B). This proximity reveals the existence of complex interactions between these cells that may inhibit effective anti-tumor immune responses. Further analyses visualized the number and strength of these cell–cell interactions and revealed that the SPP1 signaling pathway mediates the interaction between tumor cells and CD8 + Tex cells and M2 cells (Fig. 6C, D). Among them, the SPP1-CD44 pairing interaction contributed the most among all ligand-receptor pairs, emphasizing their dominant role in intercellular communication (Fig. 6E). SPP1 was expressed at significant levels in tumor cells, whereas CD44 was expressed in all cell types, with particularly high expression in CD8 + Tex cells and M2 macrophages (Fig. 6F). Finally, we observed the strongest interactions centered on tumor cells, CD8 + Tex cells, and M2-type macrophages in the spatial mapping of SPP1 intercellular communication in this sample, a network diagram reflecting the critical role of the SPP1-CD44 signaling axis in regulating the dynamics of immune cells within tumors (Fig. 6G).

Thus, we further support the importance of the SPP1 signaling pathway in mediating the interaction between tumor cells and CD8 + Tex cells from spatial transcriptomic data. Enhanced immune escape by tumor cells through remodeling of the immune microenvironment and immune cells may be a potential mechanism for ICI treatment resistance.

Discussion

This study integrates single-cell RNA sequencing, bulk RNA sequencing, spatial transcriptomics, and multiplex immunofluorescence to extensively investigate the intercellular interactions between tumor cells and immune cells and their impact on the response to ICIs. Specifically, our research focuses on the interactions between SPP1-secreting tumor cells and CD44-expressing CD8 + Tex cells. These interactions activate the MAPK signaling pathway within the CD8 + Tex cells, exacerbating T cell exhaustion and diminishing cytotoxic functionality, ultimately leading to resistance to ICI therapy.

Pre-existing immune infiltration characteristics, including the degree of CD8 + T cell infiltration and functional status, are associated with the effectiveness of ICIs before treatment initiation [41, 42]. A primary reason for the ineffectiveness of ICI on a tumor is the T cells' failure to recognize the tumor due to the absence of tumor antigens [43]. Our study observed a close association between deficiencies in the antigen presentation pathway and the development of primary resistance, reinforcing the crucial role of this mechanism in immune evasion and resistance development. Furthermore, the low expression of PD-1 ligands in tumor cells implies that the PD-1/PD-L1 pathway may not be the primary driver of immune suppression and ICI resistance in RCC. Instead, the elevated expression of LAG3 and its ligands points to this pathway as a potentially more critical player in immune evasion mechanisms. Thus, targeting PD-1/PD-L1 alone may be insufficient for effective ICI therapy in RCC. A more promising strategy could involve a combination approach targeting LAG3 or other immune checkpoints to enhance therapeutic outcomes.

Currently, several ICI combination therapies for advanced RCC, including cabozantinib-nivolumab, axitinib-avelumab, lenvatinib-pembrolizumab and axitinib-pembrolizumab, have demonstrated improved overall response rates (ORR), progression-free survival (PFS), and OS [44, 45]. VEGF inhibitors enhance immune response and T cell infiltration in the TME by regulating angiogenesis, creating a synergistic effect with ICIs [46]. Dual immune checkpoint combinations, such as PD-1 with LAG-3 or PD-1 with CTLA-4, have shown significant antitumor responses in metastatic melanoma [47]. These mechanisms are associated with reduced Treg cell populations and increased activation and infiltration of CD8 + T effector and Th1-like CD4 + T cells [48]. Despite initial effectiveness, these treatments face limitations in durability, toxicity, and adaptive resistance. Notably, none of these combinations specifically target the SPP1-CD44 signaling pathway. Thus, further exploration of strategies that disrupt SPP1-CD44-mediated immune suppression may help overcome these limitations and improve outcomes.

Intercellular communication between immune cells and tumor cells is tightly controlled by a complex network [49]. Tumor cells contribute to immune evasion and tumor progression by interacting with resident or recruited non-tumor cells in the TME, leading to non-tumor cell reprogramming and immune microenvironment remodeling [50]. Differences in intercellular communication between the sensitivity and primary resistance groups underscore the impact of TME dynamics on ICI efficacy. In the sensitivity group, signaling pathways like CXCL, CCL, and IFN-II are actively involved in immune cell activation, recruitment, and anti-tumor response induction. These signals effectively enhance the immune system’s recognition and elimination of cancer cells, leading to a strong anti-tumor response [51]. Conversely, the primary resistance group shows a pronounced activation of immunosuppressive signals, particularly via the SPP1 pathway, which facilitates tumor immune evasion. SPP1, or osteopontin, is involved in various cellular processes including cell survival, angiogenesis, and immune modulation [52]. In the primary resistance group, the SPP1-CD44 axis intensifies interactions between tumor and immune cells, suggesting that tumor cells leverage this pathway to suppress immune activity and create a supportive environment for tumor growth and survival. Spatial transcriptomic analysis further confirms the proximity of tumor cells to CD8 + Tex cells, providing evidence of their communication through the SPP1-CD44 signaling pathway.

The identification of primary drug-resistant tumor cells secreting SPP1 emphasized the crucial significance of intra-tumor heterogeneity in cancer progression and treatment resistance. This supports the theory that intra-tumor cell heterogeneity and drug resistance are related and that specific subpopulations may be key targets for more precise therapeutic interventions [53, 54]. Antigen presentation is typically a crucial stage in the immune system's activation of specific T-cell responses [55]. Nevertheless, when the signals for antigen presentation are deactivated in CD8 + Tex cells and other immune effector cells (such as CD8 + Teff cells, NK), these cells become incapable of efficiently identifying and attacking cancer cells that display tumor antigens. Chemokines and cytokines are key signaling molecules that regulate immune cell migration and function [56]. Inactivation of these signaling molecules affects the recruitment and activation of immune cells, especially T cells and NK cells [57], as corroborated by our observations and analyses in drug-resistant TME. Previous studies have shown that activation of the MAPK pathway is closely associated with the development and progression of a variety of tumors [58]. Similarly, SPP1 was previously thought to be proposed to promote tumor growth through activation of the MAPK signaling pathway [59]. Despite this, traditional population-level studies fall short in pinpointing the specific cells where MAPK pathway activation occurs and its precise role. Our study reveals that MAPK signaling activation within CD8 + Tex cells may heighten exhaustion levels and contribute to ICI resistance. This high-resolution analysis of complex cellular interactions offers essential insights and provides a foundation for future in-depth investigations.

Our study has certain limitations. Firstly, the lack of a large-sample clinical validation cohort for ICI treatment in advanced RCC, as not yet extensively practiced in China, could potentially affect the broader applicability and clinical significance of our research findings. Secondly, our observed findings and conclusions have not been substantiated by in vivo and in vitro experimental evidence. Thirdly, the tumor immune microenvironment is dynamic, with immune cell-tumor cell interactions subject to change over time. Since this study relies on static data from a single time point, it does not fully capture the dynamic changes in the immune microenvironment throughout tumor progression.

Conclusion

In summary, our study highlights the role of complex intercellular interactions between tumor cells and immune cells in driving primary resistance to ICIs. We found that interactions between SPP1-secreting tumor cells and CD44-expressing CD8 + Tex cells activate the MAPK pathway within these CD8 + Tex cells, exacerbating exhaustion and reducing cytotoxicity, ultimately leading to ICI resistance in RCC. Future research should validate these findings in larger clinical cohorts and further investigate underlying mechanisms through in vivo and in vitro experiments. Targeting SPP1-CD44 interactions to enhance antigen presentation and immune activation may offer a promising strategy to improve ICI efficacy.

Data availability

On reasonable request, the corresponding author will provide the datasets used and analyzed in this study.

Abbreviations

RCC:

Renal cell carcinoma

ccRCC:

Clear cell renal cell carcinoma

Pri.R:

Primary resistance group

Sen:

Sensitive group

L-R:

Ligand-receptor pairs

SPP1:

Osteopontin

ScRNA-seq:

Single-cell RNA sequencing

CD8 + Tex:

Exhausted CD8 + T cells

ORR:

Overall response rates

PFS:

Progression-free survival

OS:

Overall survival

ICIs:

Immune checkpoint inhibitor

TMB:

Tumor mutation burden

mIF:

Multiplex immunofluorescence

DEGs:

Differentially expressed genes

GSEA:

Gene set enrichment analysis

References

  1. Gross-Goupil M, Kwon TG, Eto M, Ye D, Miyake H, Seo SI, Byun SS, Lee JL, Master V, Jin J, et al. Axitinib versus placebo as an adjuvant treatment of renal cell carcinoma: results from the phase III, randomized ATLAS trial. Ann Oncol. 2018;29(12):2371–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Au L, Hatipoglu E, Robert de Massy M, Litchfield K, Beattie G, Rowan A, Schnidrig D, Thompson R, Byrne F, Horswell S, et al. Determinants of anti-PD-1 response and resistance in clear cell renal cell carcinoma. Cancer Cell. 2021;39(11):1497-1518 e1411.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Braun DA, Hou Y, Bakouny Z, Ficial M, Angelo MS, Forman J, Ross-Macdonald P, Berger AC, Jegede OA, Elagina L, et al. Interplay of somatic alterations and immune infiltration modulates response to PD-1 blockade in advanced clear cell renal cell carcinoma. Nat Med. 2020;26(6):909–18.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Wang Q, Peng H, Qi X, Wu M, Zhao X. Targeted therapies in gynecological cancers: a comprehensive review of clinical evidence. Signal Transduct Target Ther. 2020;5(1):137.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Yao C, Zhang T, Wu T, Brugarolas J. Facts and hopes for immunotherapy in renal cell carcinoma. Clin Cancer Res. 2022;28(23):5013–20.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Dizman N, Lyou Y, Salgia N, Bergerot PG, Hsu J, Enriquez D, Izatt T, Trent JM, Byron S, Pal S. Correlates of clinical benefit from immunotherapy and targeted therapy in metastatic renal cell carcinoma: comprehensive genomic and transcriptomic analysis. J Immunother Cancer. 2020;8(2): e000953.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Dong X, Yang Y, Zhou Y, Bi X, Zhao N, Zhang Z, Li L, Hang Q, Zhang R, Chen D, et al. Glutathione S-transferases P1 protects breast cancer cell from adriamycin-induced cell death through promoting autophagy. Cell Death Differ. 2019;26(10):2086–99.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Cabanos HF, Hata AN. emerging insights into targeted therapy-tolerant persister cells in cancer. Cancers (Basel). 2021;13(11):2666.

    Article  CAS  PubMed  Google Scholar 

  9. Zhou F, Qiao M, Zhou C. The cutting-edge progress of immune-checkpoint blockade in lung cancer. Cell Mol Immunol. 2021;18(2):279–93.

    Article  CAS  PubMed  Google Scholar 

  10. Kalbasi A, Ribas A. Tumour-intrinsic resistance to immune checkpoint blockade. Nat Rev Immunol. 2019;20(1):25–39.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Moroishi T, Hayashi T, Pan WW, Fujita Y, Holt MV, Qin J, Carson DA, Guan KL. The hippo pathway kinases LATS1/2 suppress cancer immunity. Cell. 2016;167(6):1525-1539.e1517.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Lasfar A, Gogas H, Zloza A, Kaufman HL, Kirkwood JM. IFN-λ cancer immunotherapy: new kid on the block. Immunotherapy. 2016;8(8):877–88.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Smith MP, Sanchez-Laorden B, O’Brien K, Brunton H, Ferguson J, Young H, Dhomen N, Flaherty KT, Frederick DT, Cooper ZA, et al. The immune microenvironment confers resistance to MAPK pathway inhibitors through macrophage-derived TNFα. Cancer Discov. 2014;4(10):1214–29.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Babikr F, Wan J, Xu A, Wu Z, Ahmed S, Freywald A, Chibbar R, Wu Y, Moser M, Groot G, et al. Distinct roles but cooperative effect of TLR3/9 agonists and PD-1 blockade in converting the immunotolerant microenvironment of irreversible electroporation-ablated tumors. Cell Mol Immunol. 2021;18(12):2632–47.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Cannarile MA, Weisser M, Jacob W, Jegg AM, Ries CH, Rüttinger D. Colony-stimulating factor 1 receptor (CSF1R) inhibitors in cancer therapy. J Immunother Cancer. 2017;5(1):53.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Haber DA, Settleman J. Cancer: drivers and passengers. Nature. 2007;446(7132):145–6.

    Article  CAS  PubMed  Google Scholar 

  17. Sharma P, Hu-Lieskovan S, Wargo JA, Ribas A. Primary, adaptive, and acquired resistance to cancer immunotherapy. Cell. 2017;168(4):707–23.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Westoby J, Artemov P, Hemberg M, Ferguson-Smith A. Obstacles to detecting isoforms using full-length scRNA-seq data. Genome Biol. 2020;21(1):74.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Jin S, Guerrero-Juarez CF, Zhang L, Chang I, Ramos R, Kuan CH, Myung P, Plikus MV, Nie Q. Inference and analysis of cell-cell communication using Cell Chat. Nat Commun. 2021;12(1):1088.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Chen H, Albergante L, Hsu JY, Lareau CA, Lo Bosco G, Guan J, Zhou S, Gorban AN, Bauer DE, Aryee MJ, et al. Single-cell trajectories reconstruction, exploration and mapping of omics data with STREAM. Nat Commun. 2019;10(1):1903.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Bi K, He MX, Bakouny Z, Kanodia A, Napolitano S, Wu J, Grimaldi G, Braun DA, Cuoco MS, Mayorga A, et al. Tumor and immune reprogramming during immunotherapy in advanced renal cell carcinoma. Cancer Cell. 2021;39(5):649-661.e645.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Hao Y, Hao S, Andersen-Nissen E, Mauck WM 3rd, Zheng S, Butler A, Lee MJ, Wilk AJ, Darby C, Zager M, et al. Integrated analysis of multimodal single-cell data. Cell. 2021;184(13):3573-3587.e3529.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Qiu X, Mao Q, Tang Y, Wang L, Chawla R, Pliner HA, Trapnell C. Reversed graph embedding resolves complex single-cell trajectories. Nat Methods. 2017;14(10):979–82.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Trapnell C, Cacchiarelli D, Grimsby J, Pokharel P, Li S, Morse M, Lennon NJ, Livak KJ, Mikkelsen TS, Rinn JL. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat Biotechnol. 2014;32(4):381–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Korotkevich G, Sukhov V, Budin N, Shpak B, Artyomov MN, Sergushichev A Fast gene set enrichment analysis. bioRxiv 2021;060012.

  26. Sun K, Xu R, Ma F, Yang N, Li Y, Sun X, Jin P, Kang W, Jia L, Xiong J, et al. scRNA-seq of gastric tumor shows complex intercellular interaction with an alternative T cell exhaustion trajectory. Nat Commun. 2022;13(1):4943.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Zhang L, Yu X, Zheng L, Zhang Y, Li Y, Fang Q, Gao R, Kang B, Zhang Q, Huang JY, et al. Lineage tracking reveals dynamic relationships of T cells in colorectal cancer. Nature. 2018;564(7735):268–72.

    Article  CAS  PubMed  Google Scholar 

  28. Pritykin Y, van der Veeken J, Pine AR, Zhong Y, Sahin M, Mazutis L, Pe’er D, Rudensky AY, Leslie CS. A unified atlas of CD8 T cell dysfunctional states in cancer and infection. Mol Cell. 2021;81(11):2477–24932410.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Wisdom AJ, Mowery YM, Hong CS, Himes JE, Nabet BY, Qin X, Zhang D, Chen L, Fradin H, Patel R, et al. Single cell analysis reveals distinct immune landscapes in transplant and primary sarcomas that determine response or resistance to immunotherapy. Nat Commun. 2020;11(1):6410.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Hsu CL, Ou DL, Bai LY, Chen CW, Lin L, Huang SF, Cheng AL, Jeng YM, Hsu C. Exploring markers of exhausted CD8 T cells to predict response to immune checkpoint inhibitor therapy for hepatocellular carcinoma. Liver Cancer. 2021;10(4):346–59.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Lakatos E, Williams MJ, Schenck RO, Cross WCH, Househam J, Zapata L, Werner B, Gatenbee C, Robertson-Tessi M, Barnes CP, et al. Evolutionary dynamics of neoantigens in growing tumors. Nat Genet. 2020;52(10):1057–66.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Zhan X, Dravid G, Ye Z, Hammond H, Shamblott M, Gearhart J, Cheng L. Functional antigen-presenting leucocytes derived from human embryonic stem cells in vitro. Lancet. 2004;364(9429):163–71.

    Article  PubMed  Google Scholar 

  33. Lahiri A, Maji A, Potdar PD, Singh N, Parikh P, Bisht B, Mukherjee A, Paul MK. Lung cancer immunotherapy: progress, pitfalls, and promises. Mol Cancer. 2023;22(1):40.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Ho DW, Tsui YM, Chan LK, Sze KM, Zhang X, Cheu JW, Chiu YT, Lee JM, Chan AC, Cheung ET, et al. Single-cell RNA sequencing shows the immunosuppressive landscape and tumor heterogeneity of HBV-associated hepatocellular carcinoma. Nat Commun. 2021;12(1):3684.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Guo Y, Luo F, Xu D. Cardioprotective properties of ANGPTL3 deficiency. J Am Coll Cardiol. 2017;70(16):2098–9.

    Article  PubMed  Google Scholar 

  36. Storkebaum E, Ruiz de Almodovar C, Meens M, Zacchigna S, Mazzone M, Vanhoutte G, Vinckier S, Miskiewicz K, Poesen K, Lambrechts D, et al. Impaired autonomic regulation of resistance arteries in mice with low vascular endothelial growth factor or upon vascular endothelial growth factor trap delivery. Circulation. 2010;122(3):273–81.

    Article  CAS  PubMed  Google Scholar 

  37. Noël F, Massenet-Regad L, Carmi-Levy I, Cappuccio A, Grandclaudon M, Trichot C, Kieffer Y, Mechta-Grigoriou F, Soumelis V. Dissection of intercellular communication using the transcriptome-based framework ICELLNET. Nat Commun. 2021;12(1):1089.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Andreatta M, Corria-Osorio J, Muller S, Cubas R, Coukos G, Carmona SJ. Interpretation of T cell states from single-cell transcriptomics data using reference atlases. Nat Commun. 2021;12(1):2965.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Yang W, Bai Y, Xiong Y, Zhang J, Chen S, Zheng X, Meng X, Li L, Wang J, Xu C, et al. Potentiating the antitumour response of CD8(+) T cells by modulating cholesterol metabolism. Nature. 2016;531(7596):651–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Meylan M, Petitprez F, Becht E, Bougouin A, Pupier G, Calvez A, Giglioli I, Verkarre V, Lacroix G, Verneau J, et al. Tertiary lymphoid structures generate and propagate anti-tumor antibody-producing plasma cells in renal cell cancer. Immunity. 2022;55(3):527-541e525.

    Article  CAS  PubMed  Google Scholar 

  41. Şenbabaoğlu Y, Gejman RS, Winer AG, Liu M, Van Allen EM, de Velasco G, Miao D, Ostrovnaya I, Drill E, Luna A, et al. Tumor immune microenvironment characterization in clear cell renal cell carcinoma identifies prognostic and immunotherapeutically relevant messenger RNA signatures. Genome Biol. 2016;17(1):231.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Kallies A, Zehn D, Utzschneider DT. Precursor exhausted T cells: key to successful immunotherapy? Nat Rev Immunol. 2020;20(2):128–36.

    Article  CAS  PubMed  Google Scholar 

  43. Gubin MM, Zhang X, Schuster H, Caron E, Ward JP, Noguchi T, Ivanova Y, Hundal J, Arthur CD, Krebber WJ, et al. Checkpoint blockade cancer immunotherapy targets tumour-specific mutant antigens. Nature. 2014;515(7528):577–81.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Motzer RJ, Tannir NM, McDermott DF, Arén Frontera O, Melichar B, Choueiri TK, Plimack ER, Barthélémy P, Porta C, George S, et al. Nivolumab plus ipilimumab versus sunitinib in advanced renal-cell carcinoma. N Engl J Med. 2018;378(14):1277–90.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Braun DA, Bakouny Z, Hirsch L, Flippot R, Van Allen EM, Wu CJ, Choueiri TK. Beyond conventional immune-checkpoint inhibition—novel immunotherapies for renal cell carcinoma. Nat Rev Clin Oncol. 2021;18(4):199–214.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Robert L, Ribas A, Hu-Lieskovan S. Combining targeted therapy with immunotherapy. Can 1+1 equal more than 2? Semin Immunol. 2016;28(1):73–80.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Larkin J, Chiarion-Sileni V, Gonzalez R, Grob JJ, Rutkowski P, Lao CD, Cowey CL, Schadendorf D, Wagstaff J, Dummer R, et al. Five-year survival with combined nivolumab and ipilimumab in advanced melanoma. N Engl J Med. 2019;381(16):1535–46.

    Article  CAS  PubMed  Google Scholar 

  48. Diab A, Tykodi SS, Daniels GA, Maio M, Curti BD, Lewis KD, Jang S, Kalinka E, Puzanov I, Spira AI, et al. Bempegaldesleukin plus nivolumab in first-line metastatic melanoma. J Clin Oncol. 2021;39(26):2914–25.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Graham DB, Xavier RJ. Pathway paradigms revealed from the genetics of inflammatory bowel disease. Nature. 2020;578(7796):527–39.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. de Visser KE, Joyce JA. The evolving tumor microenvironment: from cancer initiation to metastatic outgrowth. Cancer Cell. 2023;41(3):374–403.

    Article  PubMed  Google Scholar 

  51. Mempel TR, Lill JK, Altenburger LM. How chemokines organize the tumour microenvironment. Nat Rev Cancer. 2023;24(1):28–50.

    Article  PubMed  Google Scholar 

  52. Xu C, Sun L, Jiang C, Zhou H, Gu L, Liu Y, Xu Q. SPP1, analyzed by bioinformatics methods, promotes the metastasis in colorectal cancer by activating EMT pathway. Biomed Pharmacother. 2017;91:1167–77.

    Article  CAS  PubMed  Google Scholar 

  53. Tammela T, Sanchez-Rivera FJ, Cetinbas NM, Wu K, Joshi NS, Helenius K, Park Y, Azimi R, Kerper NR, Wesselhoeft RA, et al. A Wnt-producing niche drives proliferative potential and progression in lung adenocarcinoma. Nature. 2017;545(7654):355–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Wang B, He BS, Ruan XL, Zhu J, Hu R, Wang J, Li Y, Yang YH, Liu ML. An integrated microfluidics platform with high-throughput single-cell cloning array and concentration gradient generator for efficient cancer drug effect screening. Mil Med Res. 2022;9(1):51.

    PubMed  PubMed Central  Google Scholar 

  55. Esser-Kahn AP. Kiss-and-tell way to track cell contacts. Nature. 2018;553(7689):414–5.

    Article  CAS  PubMed  Google Scholar 

  56. Sokol CL, Camire RB, Jones MC, Luster AD. The chemokine receptor CCR8 promotes the migration of dendritic cells into the lymph node parenchyma to initiate the allergic immune response. Immunity. 2018;49(3):449-463e446.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Xiong W, He W, Wang T, He S, Xu F, Wang Z, Wang X, Guo H, Ling J, Zhang H, et al. Smad4 deficiency promotes pancreatic cancer immunogenicity by activating the cancer-autonomous DNA-sensing signaling axis. Adv Sci (Weinh). 2022;9(7): e2103029.

    Article  PubMed  Google Scholar 

  58. Xie Z, Peng M, Lu R, Meng X, Liang W, Li Z, Qiu M, Zhang B, Nie G, Xie N, et al. Black phosphorus-based photothermal therapy with aCD47-mediated immune checkpoint blockade for enhanced cancer immunotherapy. Light Sci Appl. 2020;9:161.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  59. Sun BS, Dong QZ, Ye QH, Sun HJ, Jia HL, Zhu XQ, Liu DY, Chen J, Xue Q, Zhou HJ, et al. Lentiviral-mediated miRNA against osteopontin suppresses tumor growth and metastasis of human hepatocellular carcinoma. Hepatology. 2008;48(6):1834–42.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We thank all authors who contributed valuable methods and data and made them public.

Funding

The National Natural Science Foundation of China (No. 82260791). Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2023D01C124). Science and Technology Assistance Program for Xinjiang (2024E02058).

Author information

Author notes

  1. Junfeng Zhang, Qingyan Peng and Jin Fan contributed equally to this work.

Authors and Affiliations

  1. Department of Urology, Xinjiang Medical University Affiliated Cancer Hospital, Urumqi, China

    Junfeng Zhang, Xing Bi, Shuai Yuan, Wei Jiang, Ting Pan & Peng Chen

  2. School of Nursing, Xinjiang Medical University, Urumqi, China

    Qingyan Peng

  3. Cancer Institute, Xinjiang Medical University Affiliated Cancer Hospital, Urumqi, China

    Fuzhong Liu

  4. Department of Urology, The Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, Enshi, Hubei, China

    Hongbo Chen & Kailing Li

  5. Department of Oncology, The Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, Enshi, Hubei, China

    Jin Fan

  6. Department of Pediatric, The Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, Enshi, Hubei, China

    Sihai Tan

Authors
  1. Junfeng Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Qingyan Peng
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Jin Fan
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Fuzhong Liu
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Hongbo Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Xing Bi
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Shuai Yuan
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Wei Jiang
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Ting Pan
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Kailing Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Sihai Tan
    View author publications

    You can also search for this author inPubMed Google Scholar

  12. Peng Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

JZ, QP and JF designed the research and wrote the paper; ST and PC guided the research and ideas; FL and HC helped with R-language technical support; KL, XB and TP helped guide statistical analysis; WJ and SY collected clinical specimens from our hospital. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Sihai Tan or Peng Chen.

Ethics declarations

Ethics approval and consent to participate

The collection of samples used in this study was approved by the Ethics Committee of the Affiliated Cancer Hospital of Xinjiang Medical University (S-2024005). All participants were provided with written informed consent at the time of enrollment.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

12967_2024_6018_MOESM1_ESM.png

Supplementary Material 1: Supplementary Fig. 1. Analysis of cell–cell communication between tumor cells and various immune cells in the Pri.R group using the R package ICELLNET. A. Network illustrating interactions between tumor cells and other cell types. B. Heatmap showing the most significant L-R interactions contributing to the communication network in the Pri.R group. C. Heatmap displaying tumor-specific interactions with other cell types. D. Heatmap showing the specificity of cell–cell interactions mediated by the SPP1-CD44 ligand-receptor pair

Supplementary Material 2.

Supplementary Material 3.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, J., Peng, Q., Fan, J. et al. Single-cell and spatial transcriptomics reveal SPP1-CD44 signaling drives primary resistance to immune checkpoint inhibitors in RCC. J Transl Med 22, 1157 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12967-024-06018-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12967-024-06018-5

Keywords

Journal of Translational Medicine

ISSN: 1479-5876

Contact us