Next Article in Journal
Stereotactic Radiosurgery for Patients with Brain Metastases from Sarcomas
Previous Article in Journal
Multimodal Deep Learning for Stage Classification of Head and Neck Cancer Using Masked Autoencoders and Vision Transformers with Attention-Based Fusion
Previous Article in Special Issue
Thyroid Cancer—The Tumor Immune Microenvironment (TIME) over Time and Space
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Cancers
Volume 17
Issue 13
10.3390/cancers17132117
cancers-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Immunosuppressive Tumor Microenvironment of Osteosarcoma

by
Aaron Michael Taylor
1,†,
Jianting Sheng
2,†,
Patrick Kwok Shing Ng
1,
Jeffrey M. Harder
3,
Parveen Kumar
1,
Ju Young Ahn
2,4,
Yuliang Cao
2,
Alissa M. Dzis
1,
Nathaniel L. Jillette
1,
Andrew Goodspeed
5,
Avery Bodlak
5,
Qian Wu
6,
Michael S. Isakoff
6,7,
Joshy George
1,
Jessica D. S. Grassmann
1,
Diane Luo
1,
William F. Flynn
1,
Elise T. Courtois
1,
Paul Robson
1,6,
Masanori Hayashi
5,8,
Alini Trujillo Paolillo
9,
Antonio Sergio Petrilli
9,
Silvia Regina Caminada de Toledo
9,
Fabiola Sara Balarezo
10,
Adam D. Lindsay
10,
Bang Hoang
11,
Stephen T. C. Wong
2,4,12,* and
Ching C. Lau
1,6,7,*add Show full author list remove Hide full author list
1
The Jackson Laboratory for Genomic Medicine, Farmington, CT 06032, USA
2
Systems Medicine and Bioengineering Department, Houston Methodist Neal Cancer Center, Houston, TX 77030, USA
3
The Jackson Laboratory, Bar Harbor, ME 04609, USA
4
Department of Biomedical Engineering, Texas A&M University, College Station, TX 77843, USA
5
University of Colorado Anschutz Medical Campus, Aurora, CO 80045, USA
6
University of Connecticut School of Medicine, Farmington, CT 06032, USA
7
Connecticut Children’s Medical Center, Hartford, CT 06106, USA
8
Children’s Hospital Colorado, Aurora, CO 80045, USA
9
Hospital do GRAACC, São Paulo 04039-001, Brazil
10
Hartford Hospital, Hartford, CT 06106, USA
11
Montefiore Medical Center, Bronx, NY 10467, USA
12
Departments of Radiology, Neuroscience, Pathology and Laboratory Medicine, Weill Cornell Medicine, Cornell University, New York, NY 14853, USA
 
add Show full affiliation list
 
remove Hide full affiliation list
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Cancers 2025, 17(13), 2117; https://doi.org/10.3390/cancers17132117
Submission received: 13 May 2025 / Revised: 15 June 2025 / Accepted: 19 June 2025 / Published: 24 June 2025
(This article belongs to the Special Issue Feature Papers in Section "Tumor Microenvironment")
Browse Figures
Versions Notes

Simple Summary

Treatment outcomes for osteosarcoma, the most common bone tumor in children, have not improved for several decades, despite significant research. Knowledge of the immune environment surrounding the tumor would allow us to better understand the mechanisms of tumor formation, treatment resistance, and metastatic growth, and design new immunotherapies to improve tumor treatment. Our aim was to comprehensively profile the treatment-naïve immune environment of the tumor using single-cell and spatial transcriptome profiling, and examine interactions between the tumor and immune cells in order to inform future studies on treatment strategies for osteosarcoma.

Abstract

Background/Objectives: Osteosarcoma is the most common malignant bone tumor in children, characterized by a high degree of genomic instability, resulting in copy number alterations and genomic rearrangements without disease-defining recurrent mutations. Clinical trials based on molecular characterization have failed to find new effective therapies or improve outcomes over the last 40 years. Methods: To better understand the immune microenvironment of osteosarcoma, we performed single-cell RNA sequencing on six tumor biopsy samples, combined with a previously published cohort of six samples. Additional osteosarcoma samples were profiled using spatial transcriptomics for the validation of discovered subtypes and to add spatial context. Results: Analysis revealed immunosuppressive cells, including myeloid-derived suppressor cells (MDSCs), regulatory and exhausted T cells, and LAMP3+ dendritic cells. Conclusions: Using cell–cell communication modeling, we identified robust interactions between MDSCs and other cells, leading to NF-κB upregulation and an immunosuppressive microenvironment, as well as interactions involving regulatory T cells and osteosarcoma cells that promoted tumor progression and a proangiogenic niche.

1. Introduction

Osteosarcoma (OS) is the most common primary malignant bone tumor affecting children and young adults. Current treatment with chemotherapy and surgical resection results in a long-term overall survival rate of ~70% for patients with localized disease but only 20–30% for patients with metastatic disease [1]. No significant advances in treatment or overall survival rate have been made in the last four decades.
Genomic profiling studies of OS tumors from predominantly pediatric populations have shown high levels of chromosomal structural variation, including massive rearrangements resulting from chromothripsis and hypermutation in localized regions (kataegis) that contribute to significant tumoral heterogeneity [2]. Because of their potential to generate novel rearrangements in protein-coding sequences, the genomic structural abnormalities observed in OS would be expected to express neoantigens that should serve as potent targets for immunotherapy (IT). Aligning with this concept, a study of 48 OS patients demonstrated that PD-L1 expression was found in 25% of primary OS samples and correlated with immune cell infiltration and event-free survival [3]. Anti-PD-L1 and anti-CTLA4 treatment also controlled OS tumor growth in a mouse model [4]. However, the clinical trials using anti-PD-L1 and anti-CTLA4 have not shown any efficacy in OS treatment [5,6], suggesting that additional immunosuppressive or resistance factors specific to human disease may need to be addressed in order to develop effective immune therapies.
The tumor microenvironment (TME) of multiple cancers has been shown to play a critical role in tumor progression and treatment response [7]. Previous studies of the OS TME have revealed the presence of multiple immune cell types, e.g., dendritic cells (DCs), macrophages, neutrophils, and lymphoid cells, with a wide range of cell abundance across patient samples [8,9,10]. However, the role of these tumor-infiltrating immune cells in OS is not fully understood. With the recent advancements in single-cell technologies, several groups have applied single-cell RNA sequencing (scRNA-seq) analysis on limited populations of OS patient tumors, including biopsy, post-treatment surgical resection, and lung metastasis samples [11,12,13,14,15]. Apart from the heterogeneous OS tumor cells, those studies also identified several molecular subtypes of immune cells, including TIGIT+-exhausted T cells, FOXP3+ regulatory T cells, and LAMP3+ mature regulatory DCs [11,12,16]. Taken together with the failure of the IT clinical trials, the complex TME within OS may serve an immunosuppressive function. Thus, a better understanding of the OS TME is urgently needed to improve the IT efficacy for OS patients.
To advance our understanding of the OS TME naïve to chemotherapy-induced biological changes, we performed scRNA-seq analysis on six pre-treatment primary tumor biopsy samples of OS. We combined our data with the published six pre-treatment OS sample data [11] to increase our ability to detect rare cell types or subtypes. We described a number of immune cells that potentially contribute to tumor progression and are consistent with the immunosuppressive nature of the TME using subclustering, differential gene expression, and pathway analysis. In addition, we evaluated the spatial distribution of those cell types of interest by spatial transcriptomics analysis on additional pre-treatment biopsy samples of OS.

2. Materials and Methods

2.1. Sample Collection and Tissue Dissociation

Tumor tissue specimens of six OS patients were collected at Connecticut Children’s Medical Center (CCMC) or Children’s Hospital Colorado (CHCO). All specimen collection and experiments were reviewed and approved by the Institutional Review Board of CCMC or CHCO, respectively. Written informed consent was obtained prior to acquisition of tissue from patients. Diagnosis was confirmed by pathologist assessment. The specimens were stored in MACS tissue storage solution (Miltenyi Biotec, Bergisch Gladbach, Germany). The three CCMC samples were transferred to The Jackson Laboratory for Genomic Medicine for the tissue dissociation process.
For CCMC samples, the tissue specimens were minced and enzymatically dissociated in a DMEM medium supplemented with Collagenase II (250 U/mL, Gibco #17101–015, Waltham, MA, USA) and DNase (1 μg/mL, Stemcell #07900, Cambridge, MA, USA) for up to 3 cycles of 37 °C 15 min dissociation with agitation. After each cycle, undigested tissue pieces were settled down by gravity and supernatant was transferred to a cold Buffer I solution (10% FBS/DMEM medium supplemented by EDTA (2 mM) and 2% BSA (Lampire #7500854, Pipersville, PA, USA). A fresh dissociation solution was applied to the undigested tissue for the next digestion cycle. Cells collected from each cycle were merged, spun down and resuspended with ACK lysis buffer (Gibco #A1049201) to remove red blood cells. After 3 min on ice, a lysis reaction was quenched by adding Buffer I solution. After centrifugation, cells were resuspended in Buffer I solution and strained through a 70 μm cell strainer. Cells were stained with propidium iodide (PI) and Calcein Violet (Thermo Fisher Scientific). PI-negative and Calcein Violet-positive viable cells were sorted out using the FACSAria Fusion system (BD Biosciences, Franklin Lakes, NJ, USA).
For CHCO samples, untreated primary tumor samples from biopsies were minced and viably cryopreserved in 90% FBS, 10% DMSO medium immediately after collection in a Corning CoolCell LX container chilled −1 °C/minute until reaching −80 °C, followed by archival storage at −150 °C. Tissue specimens were thawed at 37 °C, then enzymatically dissociated in a 0.1% DNase and Liberase 400 μg/mL (Roche/Sigma-Aldrich, St. Louis, MO, USA) cocktail. Tumor–digest suspensions were strained through a 70 µm filter membrane and washed. Red blood cells were depleted using the ErythroClear™ Red Blood Cell Depletion Kit (STEMCELL Technologies, Vancouver, BC, Canada). The resultant single-cell suspension was sorted for viable cells using DAPI staining and FACS (Beckman Coulter—MoFlo™ XDP, Brea, CA, USA).

2.2. Single-Cell Capture, Library Preparation and RNA-Seq

For CCMC symbols, cells were washed and resuspended in PBS containing 0.04% BSA. Cells were counted on a Countess II automated cell counter (Thermo Fisher, Waltham, MA, USA), and up to 12,000 cells were loaded per lane on 10× Chromium microfluidic chips (10× Genomics, Pleasanton, CA, USA). Single-cell capture, barcoding, and library preparation were performed using the 10× Chromium X version 3.0 chemistry [17], and according to the manufacturer’s protocol. cDNA and libraries were checked for quality on Agilent 4200 Tapestation and quantified by KAPA qPCR before sequencing using a Novaseq 6000 (Illumina, San Diego, CA, USA) v1.5 cycle flow cell lane at 100,000 reads per cell, with a 28–10−10–90 asymmetric read configuration.
For CHCO samples, the enriched tumor-single-cell suspensions were loaded into the Chromium Next GEM Single Cell 3′ platform (10× Genomics) with a target cell recovery for 4000 cells maximum at the Genomics Shared Resource of the University of Colorado Cancer Center. Following GEM capture and barcoded cDNA library construction, libraries were sequenced at a targeted depth of 50,000 reads per cell using an Illumina NextSeq2000.

2.3. Single-Cell Data Processing and Quality Control

Genomics Cell Ranger 10X (3.0.2) was used for read alignment (version 3.0.0, GRCh38) and count matrix generation (https://www.10xgenomics.com/support/software/cell-ranger/latest, accessed 1 December 2019). Reads from each sample were initially filtered using multiple criteria: during FASTQ generation, reads with more than one mismatch in the 8bp i7 index are excluded. Only reads aligned to annotated transcripts with MAPQ scores greater than 255 are retained. Reads containing bases with Q30 scores below 3 are also excluded.
Following alignment, cell barcodes were filtered against a whitelist of barcodes provided by 10X Genomics (<2 mismatches). Barcodes associated with cells were distinguished from those associated with ambient mRNA using an adaptively computed UMI threshold via Cell Ranger, and a digital counts matrix is generated for each sample. Original single-cell data is available at the Gene Expression Omnibus (GSE299023). Single-cell RNA expression data from an additional 6 samples described in Liu et al. [11] were downloaded from the Gene Expression Omnibus (GSE162454).
For all samples, additional preprocessing was performed using Seurat (version 4.0.5) [18] in R (version 4.1.1) [19]. For each cell, the percentage of reads mapping to mitochondrial and ribosomal genes was calculated, and cells were filtered out according to the following criterion: >20% mtRNA, >50% rRNA, or <500 features expressed. Doublet detection and removal were performed using the R-package DoubletFinder (default parameters) [20]. Read counts for each cell were log-normalized (scale factor = 1e6), and the top 2000 variable features (selection method = “vst”) in the dataset were calculated. Data were scaled and corrected for cell cycle score and percentage of mitochondrial RNA (“CellCycleScoring” and “ScaleData” functions, vars.to.regress = c(“percent.mt”, “S.Score”, “G2M.Score”)) prior to principal component analysis (PCA). Harmony (version 0.1.0) [21] was used to correct the top 50 principal components for patient sex and sample source site prior to downstream clustering analyses.

2.4. Clustering and Cell Type Determination

Primary cell clustering was performed using Seurat’s “FindClusters” function (Harmony-correct PCs = 40, res = 0.25). For primary clustering as well as downstream subclustering, the number of principal components were selected using elbow plots, and cluster resolution was chosen based on cluster stability across multiple resolution ranges and whether clusters were well-defined via the UMAP projection. The “FindAllMarkers” function (only.pos = TRUE, min.pct = 0.25, logfc.threshold = 0.25, test.use = “bimod” [22]) was utilized to detect markers for each cluster, with major cell types annotated using a known set of genes, specifically macrophage/DC (MSR1, C1QC, FOLR2), osteosarcoma (SATB2, IBSP, ALPL), NK/T cell (CD3D, TRBC1, NKG7), osteoclast (ACP5, CTSK, MMP9), monocyte/neutrophil (S100A8, S100A9, FCN1), fibroblast (TAGLN, ACTA2, FAP), endothelial (CLEC14A, PLVAP, VWF), B-cell (MS4A1, CD79A, BANK1), plasma cell (IGHG1, IGLC2, IGHG4), and mast cell (TPSB2, TPSAB1, CPA3), as well as proliferation markers (MKI67, TOP2A, TYMS).
Subsequently, major cell types were further subclustered to detect cell subtypes. We used the “subset” function for major cell clusters, reperformed normalization (“CellCycleScoring” and “ScaleData” functions, vars.to.regress = c(“percent.mt”, “CC.Difference”)), and Harmony correction of PCs. Subclustered cell types include macrophage/DCs (PCs = 30, res = 0.2), osteosarcoma/fibroblasts (PCs = 40, res = 0.15), osteoclasts (PCs = 30, res = 0.1), neutrophils (PCs = 30, res = 0.1), and proliferative macrophages/DCs (PCs = 30, res = 0.3). For the NK/T cell cluster, subtypes were identified using Azimuth [18]. Cells within the NK/T cell cluster were mapped to NK/T cell L1 annotations within the reference proliferating blood mononuclear cell (PBMC) dataset [23], and cells with a confidence score >0.75 were retained for downstream analysis. For all visualized dotplots, scaled expression values (“ScaleData”) are shown.

2.5. Copy Number Analysis

In order to characterize the overall copy number changes between the identified osteosarcoma cell populations, as well as to detect subclones within each major cluster, inferCNV (version 1.10.0) [24] was implemented on a sample-specific basis (cutoff = 0.1, HMM = T, denoise = T, analysis_mode = “subclusters”, hclust_method = “ward.D2”, cluster_by_groups = TRUE). All normal cell populations were used for comparison.

2.6. Cell–Cell Interactions

CCCExplorer (version S2C2, https://github.com/methodistsmab/S2C2, accessed on 1 December 2019) [25] was employed for the comprehensive investigation of cell–cell interactions and their impacts on downstream pathways. The integration of a manually curated ligand–receptor database, transcription factor–target database, experimentally validated pathway database, and crosstalk algorithms facilitated a thorough examination of inter-cellular crosstalk. Gene expression changes in ligand–receptor pairs across selected cell types of interest, including various macrophages, T cells, NK cells, dendritic cells, fibroblasts, and neutrophils, were explored using the CCCExplorer ligand–receptor database. The potential downstream effects on cell functions were assessed by applying a crosstalk algorithm to networks generated from KEGG, Ingenuity Pathway Analysis (IPA), and a transcription factor–target interaction database. Attention was given to pairs where elevated expression of ligands from sender cells and receptors from recipient cells was detected. Further analysis of these interactions involved the identification of significant pathway enrichments using permutation test (p < 0.05).

2.7. Visium Spatial Gene Expression Analysis

Formalin-fixed, paraffin-embedded (FFPE) samples of two primary OS pre-treatment biopsy tissue samples (Patients A and B) were obtained from CCMC and stored at 4 °C in the dark. Prior to Visium transcriptomics, RNA quality of FFPE samples was determined by DV200 score using Agilent (Santa Clara, CA, USA) TapeStation 4200 High Sensitivity DNA ScreenTape. Tissue blocks with DV200 scores above 50% were used for downstream processing. Briefly, FFPE sections were placed on a 10× Visium FFPE Gene Expression slide, deparaffinized, H&E stained, then imaged in brightfield using a NanoZoomer SQ (Hamamatsu Photonics, Shizuoka, Japan) slide scanner, followed by incubation with human-specific probe sets provided by the manufacturer for subsequent mRNA labeling and library generation per the manufacturer’s protocol (10× Genomics, CG000407). A library concentration was quantified using a TapeStation High Sensitivity DNA ScreenTape (Agilent) and fluorometry (Thermofisher Qubit) and verified via KAPA qPCR. Libraries were pooled for sequencing on an Illumina NovaSeq 6000 200-cycle S4 flow cell using a 28–10−10–90 read configuration, targeting 100,000 read pairs per spot covered by tissue.
Illumina base call files for all libraries were converted to FASTQs using bcl2fastq v2.20.0.422 (Illumina). Whole Visium slide images were uploaded to a local OMERO server. For each capture area of the Visium slide, a rectangular region of interest (ROI) containing just the capture area was drawn on the whole slide image via OMERO.web, and OMETIFF images of each ROI were programmatically generated using the OMERO Python API (OMERO server v5.6.9, omero-py v5.16.0). FASTQ files and associated OMETIFF corresponding to each capture area were aligned to the GRCh38-specific filtered probe set (10× Genomics Human Probeset v1.0.0) using the version 2.1.0 Space Ranger count pipeline (10× Genomics). Spatial transcriptomics data are available at the Gene Expression Omnibus (GSE299025).
Seurat (version 4.3.0) [18] was used for the normalization, filtering, dimensional reduction, and visualization of the resulting gene expression data. Each section was processed independently. Spots with fewer than 1500 UMIs and 500 genes were removed. Normalization was performed using the “SCTransform” function in Seurat, with batch correction was performed using Harmony (version 0.1.1) [21]. Unsupervised clustering was performed in Seurat using Leiden clustering (“FindClusters”) [26]. Cell type fractions within spots were estimated using Robust Cell Type Decomposition (RCTD, spacexr package version 2.1) [27] and the single-cell RNA-seq data as the reference. Sixteen cell types identified above were used, including three major cell classes (osteosarcoma, endothelial, and osteoclast) and 13 immune cell subtypes. The “Immune” fraction shown was the sum of all the individual subtype contributions. Differences in cell type composition between clusters were determined via a one-sided Wilcoxon rank sum test [28] for positive enrichment. Differences in pathway activity were calculated using PROGENy (version 1.14) [29].

3. Results

3.1. Single-Cell Transcriptomic Analysis of Treatment-Naïve Osteosarcoma

The tumor microenvironment (TME) plays a critical role in tumor progression and treatment response [30,31,32]. Chemotherapy, as the standard treatment regimen for OS, may alter the composition of the TME. To identify the TME cellular composition and understand the role of different cell types in tumor progression and treatment response without the influence of treatment, we aimed to investigate the TME of treatment-naïve OS tumors. We carried out the scRNA-seq of six treatment-naïve OS tumors (Supplementary Table S1) using the 10X Genomics Chromium platform. After quality control and doublet removal, we obtained a dataset of 22,035 cells from six OS tumors (Supplementary Table S1). The number of detected UMIs ranged from 694 to 155,389 per cell, with the number of detected genes ranging from 500 to 8709. To increase our power to identify rare cell types or subtypes, we leveraged the published scRNA-seq dataset from another six treatment-naïve tumors (GSE162454). A total of 40,588 additional cells passed our internal quality control filters and were subsequently integrated with our dataset using the Harmony package. A combined dataset with a total of 62,623 cells from 12 treatment-naïve OS tumors was generated. New data generated from our scRNA-seq cohort contributed 35.19% of cells to the combined dataset (Supplementary Tables S1 and S2).
To determine the cellular composition of the tumors, we performed unsupervised clustering via Seurat’s graph-based clustering method. Seventeen major clusters were identified (Figure 1A). The cell type identification of each cluster was made based on the cluster-specific differential gene expression (Supplementary Table S3) and literature-based cell type markers. The expression of each marker gene was shown in Figure 1D. We identified two clusters of macrophages/DCs (MSR1, C1QC, FOLR2), six clusters of osteoblastic tumor cells (SATB2, IBSP, ALPL), one cluster of natural killer (NK) and T cells (CD3D, NKG7, TRBC1), osteoclasts (ACP5, CTSK, MMP9), monocytes/neutrophils (S100A8, S100A9, FCN1), fibroblasts (TAGLN, ACTA2, FAP), B-cells (MS4A1, CD79A, BANK1), plasma cells (IGHG1, IGLC2, IGHG4), mast cells (TPSB2, TPSAB1, CPA3), and endothelial cells (CLEC14A, PLVAP, VWF). The smallest cluster contained only 36 cells (0.06% of total cells) with no known cell type marker identified, and poor total gene and UMI counts compared to the other clusters. It was assigned as a cluster comprising low-quality cells and removed prior to downstream analyses. Except for the low-quality cluster and one osteoblastic tumor cell cluster, each cluster consisted of cells from both published and our new datasets (Figure 1B, Supplementary Figure S1). The proportion of each cell type and distribution of cells in UMAP was similar between the published and new datasets (Figure 1B,C).

3.2. Macrophages and Dendritic Cells

The roles of different immune cells in the TME can be diverse and highly dependent on their phenotypic status. To understand the role of different subtypes of immune cells in OS, we looked further into the phenotypic subtypes of various immune cells. First, we investigate the most abundant cell lineage—myeloid cells. Two major myeloid cell clusters were identified in the present dataset—a predominantly macrophage cluster and a macrophage/DC cluster that expressed proliferative markers. While both clusters have strong expression of macrophage markers (MSR1, C1QC, FOLR2), the proliferative macrophage/DC cluster also has expression of cell proliferation genes (MKI67, TOP2A, TYMS) (Figure 1D). Further unsupervised clustering of the macrophage cluster revealed seven sub-clusters with distinct differentially expressed genes (Supplementary Table S4). We were able to detect classically activated macrophages (NR4A3, IL1B, CCL3/4, APOE, TXNIP), LYVE+ macrophages, interferon-stimulated macrophages (CXCL10, ISG15, IFIT1), angiogenic macrophages (SPP1, ADAM8, VIM, VCAN) and one macrophage cluster expressing T cell marker genes (IL32, NKG7, TRAC) (Figure 2A,B).
Although the role of various phenotypic macrophages in OS is not fully elucidated yet, their contribution to the anti-inflammatory and immunosuppressive TME in OS has been suggested based on literature. The LYVE+ macrophage subpopulation has been found in multiple human tissues as tissue-resident macrophages and has been demonstrated to support angiogenesis in different tissues [33,34,35,36,37]. In cancer, they have been shown to cooperate with mesenchymal cells in the TME and support tumor growth in a mammary adenocarcinoma model [38]. Two subtypes of macrophages with angiogenesis signatures (SPP1+, ADAM8+, and VIM+, or VCAN+, respectively) were also detected. SPP1+ and VCAN+ angiogenic macrophages have been associated with worse clinical outcomes in multiple types of epithelial cancers [39]. The SPP1+ macrophage cluster also expressed HMOX1, which has been shown to play a role in the immunosuppressive program of tumor-associated macrophages (TAM) [40]. The VCAN+ angiogenic macrophages have high expression of epidermal growth factor (EGF) family genes, including EREG and AREG. M2-like TAM-secreted EGF has been shown to promote metastasis in ovarian cancer [41]. Interestingly, this population of macrophages also has higher expression of OLR1, a marker of myeloid-derived suppressor cells (MDSC) in other tumors [42]. Macrophages expressing T cell marker genes have been previously found in inflammatory diseases as well as head and neck squamous cell carcinoma [43,44], however, their role in the OS TME is unclear.
Subclustering of the proliferative macrophage/DC cluster revealed seven subpopulations, including three subclusters of dendritic cells and four macrophage subgroups (Supplementary Table S5). Classic dendritic cells 1 (cDC1s) and cDC2s were identified based on the expression of CLEC9A/IRF8/IDO1 and CD1C/CLEC10A, respectively (Figure 2C,D). We also detected another distinct type of dendritic cells based on high expression of LAMP3, CCR7, and IDO1. The cDC2 population was found to be more abundant than cDC1 and LAMP3+ DCs in the tumors studied. The presence of LAMP3+ dendritic cells (LAMP3+ cDCs) or mature regulatory dendritic cells (mregDCs) has also been reported recently in various cancer types, including osteosarcoma [16,39]. One possible mechanism of mregDCs in immunosuppression in OS TME has been suggested through interaction with regulatory T cells (T-regs) via CD274-PDCD1 and PVR-TIGIT signaling [16].
Among proliferative macrophages, similar macrophage subtypes were found as discussed previously, including CCL3–4+/IL1B+ classically-activated macrophages, SSP1+/LYVE1+ angiogenic macrophages, and macrophages expressing T cell associated genes (IL32+, NKG7+, TRAC+). A population of MMP9+/IL7R+ macrophages was found only within the proliferative macrophage cluster.

3.3. Immunosuppressive Neutrophil/Myeloid-Derived Suppressor Cells (MDSCs)

Neutrophils have been known to promote tumor growth, angiogenesis, metastasis and inhibit anti-cancer T cell activity [45]. To investigate the potential immunosuppressive features of neutrophils in OS, the subclustering of the monocyte/neutrophil cluster was carried out, which revealed four different cell types, including two populations of S100A8/9/12+ neutrophils, and one population of each of CDKN1C+/FCGR3A+ non-classical monocytes and FN1+/SPP1+ monocyte-like cells (Figure 3A,B, Supplementary Table S6). Of note, among the two neutrophil populations, the larger population of neutrophils expressed higher levels of S100A8/9/12, and genes previously identified as markers of myeloid-derived suppressor cells (MDSC), including VCAN, CLEC4E, and CSF3R (Figure 3C) [46]. These findings suggest the presence of immunosuppressive neutrophils or MDSC in OS TME and could be one of the major contributors to immunosuppression in OS. In addition, monocyte-like cells are a possible precursor of TAM and can contribute to the accumulation of MDSCs in cancer [47].

3.4. Regulatory and Exhausted T Cells

T cells are essential for the immune anti-tumor response and the cornerstone of successful immunotherapy [48]. The negative regulation of T cells, leading to hypofunctional or exhausted T cells, is a hallmark of many cancers, and understanding the mechanisms that lead to this exhaustion can provide potential targets for immunotherapy development [49,50]. In order to better understand the T cell and NK cell subtypes present within the TME of OS, we mapped cells of the NK/T cell cluster (C02) to the peripheral blood mononuclear cell (PBMC) reference dataset [23] using Azimuth [18]. We separated populations of NK cells (NKG7, KLRD1, TYROBP, GNLY, FCER1G, PRF1, CD247, KLRF1, CST7 and GZMB), helper CD4+ T cells (IL7R, MAL, LTB, CD4, LDHB, TPT1, TRAC, TMSB10, CD3D and CD3G) and cytotoxic CD8+ T cells (CD8B, CD8A, CD3D, TMSB10, HCST, CD3G, LINC02446, CTSW, CD3E and TRAC) (Figure 4A,B, Supplementary Table S7).
We subdivided these categories further using Azimuth to detect the major subtypes of NK-cell and T-cell populations (Figure 4A). Notably, amongst the CD4+ T cells was a robust population of regulatory T cells (RTKN2, FOXP3, AC133644.2, CD4, IL2RA, TIGIT, CTLA4, FCRL3, LAIR2 and IKZF2) (Figure 4B,C). This population represented over 10% (227/2126) of the detected CD4+ T cells. Since the PBMC reference does not include exhausted T cells, to detect this subpopulation, we performed subclustering of the CD8+ effector memory T cells (CD8TEM), and looked at the expression of traditional exhaustion markers, specifically PDCD1, LAG3, and TOX (Figure 4D,E). We found that the largest subpopulation of CD8TEM cells had higher expression of all three marker genes, indicating the robust presence of the exhausted T cell population. Given that CD8TEM also represents the majority of CD8+ T cells within the OS samples, T cell exhaustion within osteosarcoma appears to be prevalent (1249/2610, 48%). Taken together, the significant population of regulatory and exhausted T cells could be one of the major contributors to the immunosuppressive TME in OS.

3.5. Osteosarcoma Cells

Leveraging the large-scale copy number alterations that serve as the signature of osteosarcoma cells, we utilized inferCNV [24] to detect relevant subclones of OS cells with distinct copy number alterations. We observed several subpopulations with normal arm-level copy number profiles, indicating that copy-neutral fibroblast or mesenchymal cells likely existed within our osteosarcoma cell clusters. In order to remove these, we combined all six OS cell clusters potentially containing normal cells derived from the primary clustering analysis shown in Figure 1, and then subclustered the OS cells as a whole (Supplementary Table S8) prior to sample-specific inferCNV analysis. Subclustering results in eight subclusters, including two subclusters of copy-neutral cells (subclusters 6 and 8) (Supplementary Figure S2). After the removal of copy-neutral cell populations, inferCNV analysis was carried out on an individual patient basis. Between three and eight subclones were observed in patients, with one sample having too-few (n = 2) detected osteosarcoma cells to perform inferCNV. Several arm-level copy number alterations were found recurrently in subclones of OS cells across samples, including CNVs previously reported in osteosarcoma. These CNVs include amplifications of chromosome arms 1p [51], 1q [51,52,53,54,55], 7p [56], 8q [51,54,57,58,59,60,61,62], and 20p [63,64,65], as well as deletions of 6q [66,67].

3.6. Cell–Cell Interactions

Next, we sought to investigate how the various cell types identified in our scRNA-seq analysis interact within the OS TME to determine their possible role in immunosuppression. We utilized CCCExplorer [25] to examine ligand–receptor interactions using a curated database of known interacting pairs. We queried CCCExplorer’s database for interactions based on differentially expressed genes calculated using Seurat’s FindMarkers function (Figure 5, Supplementary Table S9). We looked specifically at signaling interactions, involving T-regs, OS, MDSCs, and macrophages in order to determine how this signaling may contribute, to immunosuppression within the tumor, prioritizing interactions that also had downstream pathway expression support.
We found robust interaction between MDSCs and other cell types, including multiple interactions known to be involved in tumorigenesis and immunosuppression. For example, IL1B was highly expressed in MDSC, and its receptor IL1R1 was highly expressed in osteosarcoma cells (Figure 5B,C). This interaction leads to the downstream upregulation of NF-κB through CHUK upregulation, which is known to be highly expressed in osteosarcoma and is implicated in tumorigenesis, and whose inhibition lead to apoptosis and repressed proliferation and invasiveness in cell line models [68]. MDSCs expressing IL1A/B were also predicted to interact with several macrophage subtypes (angiogenic, activated, LYVE1+) through IL1RAP. IL1RAP expression has been identified on the tumor and stromal cells of multiple tumors [69], promoting an immunosuppressive microenvironment, and its interaction with IL1A/IL1B has also been specifically targeted in a phase 1b clinical trial [70].
An association of regulatory T cell signaling with macrophages was observed. We identified LTB overexpression in regulatory T cells, and it interacts as a ligand with LTBR, which was highly expressed in activated and angiogenic macrophages (Figure 5C, Supplementary Figure S3. Ligand–receptor binding between the two activates the noncanonical NF-κB pathway, which also showed evidence of the upregulation of downstream genes, and affects inflammation in the tumor microenvironment of several cancers [71,72].
We also saw signaling between osteosarcoma cells and macrophage populations. OS cells expressed CSF1, which has been shown to be associated with protumor activity of tumor-associated macrophages in multiple sarcomas, including OS, and is a regulator of proliferation and survival [73,74]. This interacts with CSF1R, highly expressed on activated and angiogenic macrophages, and activating several downstream pathways, including the M2-polarization-associated PI3K/AKT pathway (Figure 5C, Supplementary Figure S4). CSF1 secretion by OS cells results in the polarization of bone-marrow-derived macrophages toward an M2 phenotype, and targeting this interaction with a CSF1R inhibitor suppressed OS xenograft growth and lung metastatic potential [74], and is being explored in multiple cancers [75].
We also looked at interactions between other immune cells in osteosarcoma. We detected an interaction between LYVE1+ macrophages and angiogenic macrophages involving CCL2 and CCR1 (Figure 5C, Supplementary Figure S5). LYVE1-expressing macrophages have been shown to have a role in maintaining a proangiogenic perivascular niche in cancer [38], and CCL2 is the primary chemokine responsible for the recruitment of angiogenic macrophages [76]. AKT3 upregulation downstream of this interaction in angiogenic macrophages may lead to inflammatory and angiogenic responses, as has been shown previously [77], although this pathway was not significantly upregulated in our data.

3.7. Spatial Transcriptomics

Like other solid tumors, osteosarcoma has a complex tissue structure and heterogenous cellular organization. To understand potential immunosuppression from a spatial perspective in the OS TME, we studied the cellular organization and architecture of the tissue using spatial transcriptomics analyses with the 10X Genomics Visium platform on two pre-treatment primary tumor biopsy tissue samples. After quality control filtering, normalization, and data integration using Seurat [18] and Harmony [21], we used Robust Cell Type Decomposition (RCTD) [27] to infer spot-level cell composition and compare the overall cell type frequencies in the samples.
Similar to the scRNA-seq data, these samples primarily consisted of OS cells and immune cells, along with populations of endothelial cells and osteoclasts (Figure 6A). Immune cell abundance varied between the samples, with some subtypes being more prevalent in one sample than the other (Figure 6B). For instance, activated macrophages, angiogenic macrophages, and neutrophils/monocytes were more prevalent in Patient A, whereas CD8+ T cells, CD4+ T cells, and IFN-stimulated macrophages were observed more in Patient B.
To better understand the cellular distribution of various cell types, eight clusters with unique gene expression patterns were identified by unsupervised clustering of spots (Figure 6C,D). Four clusters (0–3) defined the largest area of the tumors and were common to both samples, whereas other five clusters showed tumor-specific localizations (Figure 6C,D). Cell composition contributed to these regional identities based on the enrichment of estimated cell type fractions with each cluster (Figure 6E). Both tumors had comparable regions enriched in MDSCs (cluster 0), OS cells (with fewer immune cells, cluster 1), LYVE+ macrophages (cluster 2), or endothelial cells (cluster 3). Patient A had a region marked by neutrophils/monocytes and angiogenic macrophages (cluster 5), and another enriched in activated macrophages and B cells (cluster 7). In contrast, Patient B had a region enriched in CD8+ T cells, CD4+ T cells, and IFN-stimulated macrophages (cluster 4). These differences in immune cell composition between samples were largely attributable to localized phenomena.
To understand the biological significance and phenotypes of different regions within the samples, we ran signaling pathway activity analysis: PROGENy. Different clusters showed distinguished gene expression patterns associated with cancer-relevant signaling pathways (Figure 6F). Common to both tumors, relatively large areas showed gene expression patterns associated with EGFR (cluster 0) and TGFB (cluster 3) signaling pathway activity. Both pathways have been associated with immunosuppression in cancer and response to immunotherapy [78,79]. Consistent with this, these areas contained immunosuppressive cells like MDSCs (cluster 0) and generally lacked enrichment of anti-tumor cells, like CD8+ T cells. In one sample, we observed a hypoxia-associated gene expression pattern (cluster 5), which also had its own distinct enrichment of immune cell types, including angiogenic macrophages and neutrophils/monocytes clustered with osteoclasts. Finally, the region with the clearest enrichment of CD8+ T cells (cluster 4) lacked enrichment of the EGFR, TGFB, and hypoxia-associated gene expression signatures. Overall, these data, for the first time, showed the spatial distribution of different cell types and cell states, particularly in macrophages, in OS TME. Although limited by sample size, these results also suggest that the TME of OS is largely immunosuppressive and its architecture consists of potentially distinct immunosuppressive milieus.

4. Discussion

The suboptimal results of recent immunotherapy clinical trials suggested that OS has a complex immune TME. To improve the efficacy of treatments, particularly immunotherapy, a better understanding of the OS immune landscape is urgently needed. To address this, we performed scRNA-seq analysis of six treatment-naïve OS samples and combined these data with published scRNA-seq data of six additional treatment naïve OS samples to generate a comprehensive primary scRNA-seq dataset for OS. We aimed to better understand the potential immunosuppressive TME in OS by focusing on immune cells identified in OS TME.
With the combined dataset, we identified similar immune cell types reported in previous OS scRNA-seq studies [11,12,13,14,15], including immunosuppressive-related cell types, e.g., CD4+ regulatory T cells and CD8+-exhausted T cells. These regulatory and exhausted T cells play well-established roles in the microenvironment of multiple tumors, including osteosarcoma [11,12,16]. T-cell exhaustion is a hallmark of immune evasion in many cancers, and overcoming this phenomenon in solid tumors is essential for the efficacy of emerging treatments such as CAR-T therapy [80,81].
Notably, although previous studies showed the presence of neutrophils in OS, we identified two subtypes of neutrophils including a major subtype expressing previously identified myeloid-derived suppressor cell (MDSC) markers. MDSCs contribute to the pathological immune response in cancer and may be potential targets for immunotherapy development [82]. In addition, Zhou et al. [12] noted an increased presence of neutrophils in primary tumors as compared to recurrent tumors using a chemotherapy-treated cohort. Our treatment-naïve cohort of primary tumors also showed a robust population of neutrophils, and given that the majority were immunosuppressive, this likely represents an important axis of immune evasion particularly for primary osteosarcoma and warrants further investigation.
We also found a population of LAMP3+ cDCs within our cohort, consistent with previous findings in osteosarcoma. These cDCs interact with regulatory T cells to elicit their immunosuppressive functions [16,39]. Liu et al. [16] highlighted the interaction between LAMP3+ mregDCs and regulatory T cells through the interaction of CD274 and PVR with PDCD1 and TIGIT, respectively, specifically within tumors. This underscores the importance of mregDCs in mitigating anti-tumor immunity in osteosarcoma. Our findings confirm the common presence of mregDCs in osteosarcoma, emphasizing the importance of further study of this immunosuppressive cell type.
The presence of immunosuppressive and angiogenic TAMs in osteosarcoma was also found in our dataset. Building on the previous analyses [11,12], we examined M1–M2 macrophage polarity amongst our macrophage subclusters using the markers from those studies, but found no evidence of MKI67+ tissue resident macrophages (Supplementary Figure S6). We did not observe strong evidence of M1 and M2 macrophage polarity in our dataset which aligns with the argument that macrophages have more complex phenotype than M1 and M2 polarity. In our analysis, distinct phenotypic macrophages were found, including classic activated macrophages, angiogenic macrophages, and interferon-stimulated macrophages, based on distinct markers for each subcluster. Additionally, subclustering revealed a distinct population of LYVE1+ angiogenic macrophages, previously found in several cancer subtypes [33,34,35,36,37]. Although the exact role of each macrophage subtype in OS TME remains unclear, as the most abundant immune cell types with distinct phenotypes, macrophages are attractive targets for future immunotherapy development.
Liu et al. previously [11] identified four major subpopulations of osteoclast: mature osteoclasts (CTSK, ACP5, MMP9), proliferative progenitor osteoclasts (CD74, CD14, HLA-DRA, MKI67, CTSKlower, ACP5lower, MMP9lower), and non/hypofunctional osteoclasts with low expression of such markers. While not a focus of this study, our subclustering of osteoclasts revealed a population of progenitor osteoclasts positive for progenitor markers CD74, CD14, HLA-DRA and MKI67, as well as the proliferation marker TOP2A, matching the previous results (Supplementary Figure S7). The other four populations of osteoclasts seemed to closely match the more mature osteoclasts subtypes with few distinctive differentially expressed genes, and there were no strong indications of separate hypofunctional or nonfunctional osteoclast subtypes. Further analysis of the mature osteoclast subtypes using larger datasets may reveal unique phenotypes.
Using CCCExplorer and scRNA-seq data from various cell types, we identified robust interactions between MDSCs and other cells, including macrophages through IL1RAP, leading to NF-κB upregulation and an immunosuppressive microenvironment. Regulatory T cells signaled to macrophages via LTB and LTBR, activating the noncanonical NF-κB pathway. Osteosarcoma cells polarized macrophages to an M2 phenotype via CSF1 and CSF1R, and interactions between LYVE1+ and angiogenic macrophages through CCL2 and CCR1 promoted a proangiogenic niche and inflammatory responses.
Our findings highlight several interactions that have been previously targeted for therapeutic development in osteosarcoma and other cancers. As discussed previously, we found that IL1B was highly expressed in MDSC, and its receptor IL1R1 was highly expressed in osteosarcoma cells, along with downstream upregulation of target pathways. IL1-targeting therapies have been tested in animal and clinical trials for multiple cancer types [83,84,85], including non-small-cell lung cancer, breast cancer, colon cancer, and melanoma, but have not yet been tested for osteosarcoma. Additionally, CSF1, expressed on osteosarcoma cells, interacts with CSF1R, which is highly expressed in the macrophages. This interaction has been previously identified in osteosarcoma, and targeting this axis in cell lines and patient-derived xenografts using the CSF1R inhibitor pexidartinib inhibited tumor and metastatic growth [86]. Although previous studies testing single immunotherapy agents against PD−1 have been unsuccessful [6,87], therapeutics targeting several identified axes of immune suppression may prove fruitful. For example, based on a detected interaction between tumor-associated mesenchymal stem cells and osteosarcoma, a recent study tested the effects of the anti-IL6 receptor antibody, tocilizumab, on a xenograft model of osteosarcoma, and showed a reduction in metastatic potential [88].
We further examined spatial transcriptomics on two pre-treatment biopsy samples to verify the presence of these immunosuppressive cell types and examine how they colocalize within osteosarcoma. We found distinctions between those samples in their immune cell populations, which broadly indicate variations in interferon stimulation and macrophage subtypes. Osteosarcoma cells, MDSCs, and LYVE1+ macrophages were comparable between the two samples, indicating commonalities between tumors. However, the importance of the different immune cell population frequencies between these two tumors, which also showed differences in pathway expression related to growth factor expression, highlight the need for more spatial transcriptomics data to explore cellular heterogeneity in osteosarcoma and its impact on treatment and survival outcomes. Additionally, previous studies using the deconvolution of bulk osteosarcoma tumor versus normal tissue have identified proportional differences between cell types [89] and differential gene expression signatures [90,91]. Future spatial and single-cell studies of osteosarcoma would benefit from the use of normal tissue for comparison in order to better differentiate aberrant gene expression and cell type differences in osteosarcoma.

5. Conclusions

Overall, our findings suggest a broad network of pathological immunosuppression within osteosarcoma. While further orthogonal validation and functional studies are necessary to confirm these findings, it would be prudent to explore treatment strategies that address the observed immunosuppressive signaling within osteosarcoma.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/cancers17132117/s1. Supplementary Figure S1: Major cell type proportions of each sample. Supplementary Figure S2: inferCNV plots for each sample. Supplementary Figure S3: Ligand-receptor interactions and downstream pathway expression originating from regulatory T cells. Supplementary Figure S4: Ligand-receptor interactions and downstream pathway expression originating from osteosarcoma cells. Supplementary Figure S5: Ligand-receptor interactions and downstream pathway expression originating from LYVE1+ macrophage cells. Supplementary Figure S6: M1/M2/tissue-resident macrophage (TRM) subtype markers used by Liu et al. (11) and Zhou et al. (12) and plotted across our identified osteoclast subclusters. Supplementary Figure S7: Osteoclast subtype markers used by Liu et al. (11) plotted across our identified osteoclast subclusters. Supplementary Table S1: Patient and sample characteristics. Supplementary Table S2: Frequency table of the contribution to the primary clusters by (a) sample, (b) patient sex, and (c) cohort. (d) Major cluster membership of each cell. Supplementary Table S3: Full gene list showing markers for each primary cluster. Supplementary Table S4: (a) Frequency table of the contribution to the macrophage subclusters by sample. (b) Macrophage subcluster membership of each cell. (c) Cell markers of each macrophage subcluster. Supplementary Table S5: (a) Frequency table of the contribution to the proliferative macrophage/DC subclusters by sample. (b) Proliferative macrophage/DC subcluster membership of each cell. (c) Cell markers of each proliferative macrophage/DC subcluster. Supplementary Table S6: (a) Frequency table of the contribution to the monocyte/neutrophil subclusters by sample. (b) Monocyte/neutrophil subcluster membership of each cell. (c) Cell markers of each monocyte/neutrophil subcluster. Supplementary Table S7: (a) Azimuth output showing the predicted cell types/subtypes of the NK/T cluster cells after mapping to the PBMC data. Cells that mapped to NK or T cell clusters by Azimuth with a confidence score > 0.75 were retained for downstream analysis (b) Frequency table of the contribution to the NK/T cell L2 subtypes by sample. Supplementary Table S8: (a) Frequency table of the contribution to the osteosarcoma/normal fibroblast subclusters by sample. (b) Osteosarcoma/normal fibroblast subcluster membership of each cell. (c) Cell markers of each osteosarcoma/normal fibroblast subcluster. Supplementary Table S9: CCCExplorer output table, showing significant ligand/receptor interactions between selected cell types.

Author Contributions

Conceptualization, A.M.T., P.K.S.N., J.M.H., P.K., A.G., M.S.I., P.R., M.H. and C.C.L.; Data curation, A.M.T., P.K.S.N., J.M.H., Q.W., M.S.I., M.H., S.R.C.d.T., A.D.L., B.H. and C.C.L.; Formal analysis, A.M.T., J.S., J.M.H., P.K., J.Y.A. and Y.C.; Funding acquisition, E.T.C., P.R., M.H., S.T.C.W. and C.C.L.; Investigation, P.K.S.N., A.M.D., N.L.J., A.G., A.B., J.D.S.G., D.L., W.F.F., E.T.C. and A.T.P.; Methodology, A.M.T., J.S., P.K.S.N., P.K., J.Y.A., Y.C., J.G., J.D.S.G., W.F.F., E.T.C., P.R., M.H., S.T.C.W. and C.C.L.; Project administration, P.K.S.N., W.F.F., E.T.C., P.R., M.H., S.T.C.W. and C.C.L.; Resources, P.K.S.N., A.M.D., N.L.J., A.G., A.B., Q.W., M.S.I., M.H., A.S.P., S.R.C.d.T., F.S.B., A.D.L., B.H. and C.C.L.; Software, A.M.T., J.S., J.M.H., P.K., J.Y.A., Y.C., A.G., J.G., S.T.C.W. and C.C.L.; Supervision, P.K.S.N., M.S.I., J.G., E.T.C., M.H., S.T.C.W. and C.C.L.; Validation, A.M.T., J.S., P.K.S.N., J.M.H., J.Y.A., S.T.C.W. and C.C.L.; Visualization, A.M.T., J.S., P.K.S.N., J.M.H., P.K., J.Y.A., Y.C. and C.C.L.; Writing—original draft, A.M.T., J.S., P.K.S.N., J.M.H., J.Y.A. and C.C.L.; Writing—review and editing, A.M.T., J.S., P.K.S.N., J.M.H., J.G., E.T.C., P.R., M.H., S.T.C.W. and C.C.L. All authors have read and agreed to the published version of the manuscript.

Funding

This work was funded by CapoStrong Fund and the FAPESP (São Paulo Research Foundation) grant number 19/18670–9. Shared services were partially supported in part by the JAX Cancer Center (P30 CA034196). The research was also supported by The Jackson Laboratory CATch program, the Strawbridge Cancer Therapeutic Fund, the T. T. & W.F. Chao Foundation, and the John S Dunn Research Foundation. Additional support was provided by the National Institutes of Health P30CA06934 funded Bioinformatics and Biostatistics Shared Resource core facility (RRID:SCR_021984) and Flow Cytometry shared resources (RRID:SCR_022035) at the University of Colorado Cancer Center, and NIH U01CA253553 at Houston Methodist Cancer Center. This work was further supported by the St. Baldrick’s Foundation Scholar Award (M.H.) and the National Pediatric Cancer Foundation (M.H.).

Institutional Review Board Statement

For samples gathered from Connecticut Children’s Medical Center and processed at The Jackson Laboratory, the study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Connecticut Children’s Medical Center (CCMC−19–135, approval date 1 October 2024). For samples gathered at University of Colorado Anschutz Medical Campus/Children’s Hospital Colorado, ethical review and approval were waived for this study, as it was determined to be using only secondary data and is thus exempt.

Informed Consent Statement

For patients at Connecticut Children’s Medical Center, informed consent was obtained from all subjects involved in the study. For samples gathered from patients at Children’s Hospital Colorado, this is not applicable as the study was determined to not involve human subjects.

Data Availability Statement

Internal single-cell RNA expression and Visium spatial transcriptomics data are available at the Gene Expression Omnibus (GSE299023, GSE299025). Single-cell RNA expression data from the external cohort of six samples was downloaded from the Gene Expression Omnibus (GSE162454). Correspondence should be directed to Dr. Ching C. Lau and Dr. Stephen T.C. Wong.

Acknowledgments

We gratefully acknowledge the contribution of the Single Cell Biology service, the Genome Technologies service, the Histology Services, the Flow Cytometry Service, and cyberinfrastructure high-performance computing resources at The Jackson Laboratory for expert assistance with the work described herein.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Preprint Statement

This article has been submitted as a preprint: Taylor, A. M., et al. Immunosuppressive tumor microenvironment of osteosarcoma. Preprint at https://www.biorxiv.org/content/10.1101/2023.11.01.565008v2 (accessed on 3 November 2023).

Abbreviations

The following abbreviations are used in this manuscript:
MDSCsMyeloid-derived suppressor cells
OSOsteosarcoma
ITImmunotherapy
DCDendritic cells
TMETumor microenvironment
scRNA-seqSingle-cell RNA sequencing
CCMCConnecticut Children’s Medical Center
CHCOChildren’s Hospital Colorado
PIPropidium iodide
PCAPrincipal component analysis
PBMCProliferating blood mononuclear cell
IPAIngenuity Pathway Analysis
FFPEFormalin-fixed, paraffin-embedded
ROIRegion of interest
RCTDRobust Cell Type Decomposition
mregDCsMature regulatory dendritic cells
T-regsRegulatory T cells
N-CMNonclassical monocyte
CD8TEMCD8+ effector memory T cells

References

  1. Meltzer, P.S.; Helman, L.J.; Longo, D.L. New Horizons in the Treatment of Osteosarcoma. N. Engl. J. Med. 2021, 385, 2066–2076. [Google Scholar] [CrossRef] [PubMed]
  2. Chen, X.; Bahrami, A.; Pappo, A.; Easton, J.; Dalton, J.; Hedlund, E.; Ellison, D.; Shurtleff, S.; Wu, G.; Wei, L.; et al. Recurrent Somatic Structural Variations Contribute to Tumorigenesis in Pediatric Osteosarcoma. Cell Rep. 2014, 7, 104–112. [Google Scholar] [CrossRef] [PubMed]
  3. Koirala, P.; Roth, M.E.; Gill, J.; Piperdi, S.; Chinai, J.M.; Geller, D.S.; Hoang, B.H.; Park, A.; Fremed, M.A.; Zang, X.; et al. Immune Infiltration and Pd-L1 Expression in the Tumor Microenvironment Are Prognostic in Osteosarcoma. Sci. Rep. 2016, 6, 30093. [Google Scholar] [CrossRef] [PubMed]
  4. Lussier, D.M.; Johnson, J.L.; Hingorani, P.; Blattman, J.N. Combination Immunotherapy with A-Ctla-4 and A-Pd-L1 Antibody Blockade Prevents Immune Escape and Leads to Complete Control of Metastatic Osteosarcoma. J. Immunother. Cancer 2015, 3, 21. [Google Scholar] [CrossRef]
  5. D’Angelo, S.P.; Mahoney, M.R.; Van Tine, B.A.; Atkins, J.; Milhem, M.M.; Jahagirdar, B.N.; Antonescu, C.R.; Horvath, E.; Tap, W.D.; Schwartz, G.K.; et al. Nivolumab with or without Ipilimumab Treatment for Metastatic Sarcoma (Alliance A091401): Two Open-Label, Non-Comparative, Randomised, Phase 2 Trials. Lancet Oncol. 2018, 19, 416–426. [Google Scholar] [CrossRef]
  6. Tawbi, H.A.; Burgess, M.; Bolejack, V.; Van Tine, B.A.; Schuetze, S.M.; Hu, J.; D’Angelo, S.; Attia, S.; Riedel, R.F.; Priebat, D.A.; et al. Pembrolizumab in Advanced Soft-Tissue Sarcoma and Bone Sarcoma (Sarc028): A Multicentre, Two-Cohort, Single-Arm, Open-Label, Phase 2 Trial. Lancet Oncol. 2017, 18, 1493–1501. [Google Scholar] [CrossRef]
  7. Binnewies, M.; Roberts, E.W.; Kersten, K.; Chan, V.; Fearon, D.F.; Merad, M.; Coussens, L.M.; Gabrilovich, D.I.; Ostrand-Rosenberg, S.; Hedrick, C.C.; et al. Understanding the Tumor Immune Microenvironment (Time) for Effective Therapy. Nat. Med. 2018, 24, 541–550. [Google Scholar] [CrossRef]
  8. Corre, I.; Verrecchia, F.; Crenn, V.; Redini, F.; Trichet, V. The Osteosarcoma Microenvironment: A Complex but Targetable Ecosystem. Cells 2020, 9, 976. [Google Scholar] [CrossRef]
  9. Zhu, T.; Han, J.; Yang, L.; Cai, Z.; Sun, W.; Hua, Y.; Xu, J. Immune Microenvironment in Osteosarcoma: Components, Therapeutic Strategies and Clinical Applications. Front. Immunol. 2022, 13, 907550. [Google Scholar] [CrossRef]
  10. Tian, H.; Cao, J.; Li, B.; Nice, E.C.; Mao, H.; Zhang, Y.; Huang, C. Managing the Immune Microenvironment of Osteosarcoma: The Outlook for Osteosarcoma Treatment. Bone Res. 2023, 11, 11. [Google Scholar] [CrossRef]
  11. Liu, Y.; Feng, W.; Dai, Y.; Bao, M.; Yuan, Z.; He, M.; Qin, Z.; Liao, S.; He, J.; Huang, Q.; et al. Single-Cell Transcriptomics Reveals the Complexity of the Tumor Microenvironment of Treatment-Naive Osteosarcoma. Front. Oncol. 2021, 11, 709210. [Google Scholar] [CrossRef] [PubMed]
  12. Zhou, Y.; Yang, D.; Yang, Q.; Lv, X.; Huang, W.; Zhou, Z.; Wang, Y.; Zhang, Z.; Yuan, T.; Ding, X.; et al. Single-Cell Rna Landscape of Intratumoral Heterogeneity and Immunosuppressive Microenvironment in Advanced Osteosarcoma. Nat. Commun. 2020, 11, 6322. [Google Scholar] [CrossRef] [PubMed]
  13. Zheng, X.; Liu, X.; Zhang, X.; Zhao, Z.; Wu, W.; Yu, S. A Single-Cell and Spatially Resolved Atlas of Human Osteosarcomas. J. Hematol. Oncol. 2024, 17, 71. [Google Scholar] [CrossRef] [PubMed]
  14. Liu, Y.; He, M.; Tang, H.; Xie, T.; Lin, Y.; Liu, S.; Liang, J.; Li, F.; Luo, K.; Yang, M.; et al. Single-Cell and Spatial Transcriptomics Reveal Metastasis Mechanism and Microenvironment Remodeling of Lymph Node in Osteosarcoma. BMC Med. 2024, 22, 200. [Google Scholar] [CrossRef]
  15. He, X.-Y.; Que, L.-Y.; Yang, F.; Feng, Y.; Ren, D.; Song, X. Single-Cell Transcriptional Profiling in Osteosarcoma and the Effect of Neoadjuvant Chemotherapy on the Tumor Microenvironment. J. Bone Oncol. 2024, 46, 100604. [Google Scholar] [CrossRef]
  16. Liu, W.; Hu, H.; Shao, Z.; Lv, X.; Zhang, Z.; Deng, X.; Song, Q.; Han, Y.; Guo, T.; Xiong, L.; et al. Characterizing the Tumor Microenvironment at the Single-Cell Level Reveals a Novel Immune Evasion Mechanism in Osteosarcoma. Bone Res. 2023, 11, 4. [Google Scholar] [CrossRef]
  17. Zheng, G.X.Y.; Terry, J.M.; Belgrader, P.; Ryvkin, P.; Bent, Z.W.; Wilson, R.; Ziraldo, S.B.; Wheeler, T.D.; McDermott, G.P.; Zhu, J.; et al. Massively Parallel Digital Transcriptional Profiling of Single Cells. Nat. Commun. 2017, 8, 14049. [Google Scholar] [CrossRef]
  18. Hao, Y.; Hao, S.; Andersen-Nissen, E.; Mauck, W.M.; Zheng, S.; Butler, A.; Lee, M.J.; Wilk, A.J.; Darby, C.; Zager, M.; et al. Integrated Analysis of Multimodal Single-Cell Data. Cell 2021, 184, 3573–3587.e29. [Google Scholar] [CrossRef]
  19. R Core Team. R: A Language and Environment for Statistical Computing; R Core Team: Vienna, Austria, 2021. [Google Scholar]
  20. McGinnis, C.S.; Murrow, L.M.; Gartner, Z.J. Doubletfinder: Doublet Detection in Single-Cell Rna Sequencing Data Using Artificial Nearest Neighbors. Cell Syst. 2019, 8, 329–337.e4. [Google Scholar] [CrossRef]
  21. Korsunsky, I.; Millard, N.; Fan, J.; Slowikowski, K.; Zhang, F.; Wei, K.; Baglaenko, Y.; Brenner, M.; Loh, P.-R.; Raychaudhuri, S. Fast, Sensitive and Accurate Integration of Single-Cell Data with Harmony. Nat. Methods 2019, 16, 1289–1296. [Google Scholar] [CrossRef]
  22. McDavid, A.; Finak, G.; Chattopadyay, P.K.; Dominguez, M.; Lamoreaux, L.; Ma, S.S.; Roederer, M.; Gottardo, R. Data Exploration, Quality Control and Testing in Single-Cell Qpcr-Based Gene Expression Experiments. Bioinformatics 2012, 29, 461–467. [Google Scholar] [CrossRef] [PubMed]
  23. Stuart, T.; Butler, A.; Hoffman, P.; Hafemeister, C.; Papalexi, E.; Mauck, W.M.; Hao, Y.; Stoeckius, M.; Smibert, P.; Satija, R. Comprehensive Integration of Single-Cell Data. Cell 2019, 177, 1888–1902.e21. [Google Scholar] [CrossRef] [PubMed]
  24. Tickle, T.; Tirosh, I.; Georgescu, C.; Brown, M.; Haas, B. Infercnv of the Trinity Ctat Project; Klarman Cell Observatory, Broad Institute of MIT and Harvard: Cambridge, MA, USA, 2019. [Google Scholar]
  25. Choi, H.; Sheng, J.; Gao, D.; Li, F.; Durrans, A.; Ryu, S.; Lee, S.B.; Narula, N.; Rafii, S.; Elemento, O.; et al. Transcriptome Analysis of Individual Stromal Cell Populations Identifies Stroma-Tumor Crosstalk in Mouse Lung Cancer Model. Cell Rep. 2015, 10, 1187–1201. [Google Scholar] [CrossRef]
  26. Traag, V.A.; Waltman, L.; Van Eck, N.J. From Louvain to Leiden: Guaranteeing Well-Connected Communities. Sci. Rep. 2019, 9, 5233. [Google Scholar] [CrossRef]
  27. Cable, D.M.; Murray, E.; Zou, L.S.; Goeva, A.; Macosko, E.Z.; Chen, F.; Irizarry, R.A. Robust Decomposition of Cell Type Mixtures in Spatial Transcriptomics. Nat. Biotechnol. 2021, 40, 517–526. [Google Scholar] [CrossRef]
  28. Wilcoxon, F. Individual Comparisons of Grouped Data by Ranking Methods. J. Econ. Èntomol. 1946, 39, 269–270. [Google Scholar] [CrossRef]
  29. Schubert, M.; Klinger, B.; Klünemann, M.; Sieber, A.; Uhlitz, F.; Sauer, S.; Garnett, M.J.; Blüthgen, N.; Saez-Rodriguez, J. Perturbation-Response Genes Reveal Signaling Footprints in Cancer Gene Expression. Nat. Commun. 2018, 9, 20. [Google Scholar] [CrossRef]
  30. Wang, Q.; Shao, X.; Zhang, Y.; Zhu, M.; Wang, F.X.C.; Mu, J.; Li, J.; Yao, H.; Chen, K. Role of Tumor Microenvironment in Cancer Progression and Therapeutic Strategy. Cancer Med. 2023, 12, 11149–11165. [Google Scholar] [CrossRef]
  31. Neophytou, C.M.; Panagi, M.; Stylianopoulos, T.; Papageorgis, P. The Role of Tumor Microenvironment in Cancer Metastasis: Molecular Mechanisms and Therapeutic Opportunities. Cancers 2021, 13, 2053. [Google Scholar] [CrossRef]
  32. Baghban, R.; Roshangar, L.; Jahanban-Esfahlan, R.; Seidi, K.; Ebrahimi-Kalan, A.; Jaymand, M.; Kolahian, S.; Javaheri, T.; Zare, P. Tumor Microenvironment Complexity and Therapeutic Implications at a Glance. Cell Commun. Signal. 2020, 18, 59. [Google Scholar] [CrossRef]
  33. Chakarov, S.; Lim, H.Y.; Tan, L.; Lim, S.Y.; See, P.; Lum, J.; Zhang, X.-M.; Foo, S.; Nakamizo, S.; Duan, K.; et al. Two Distinct Interstitial Macrophage Populations Coexist across Tissues in Specific Subtissular Niches. Science 2019, 363, eaau0964. [Google Scholar] [CrossRef] [PubMed]
  34. Lim, H.Y.; Lim, S.Y.; Tan, C.K.; Thiam, C.H.; Goh, C.C.; Carbajo, D.; Chew, S.H.S.; See, P.; Chakarov, S.; Wang, X.N.; et al. Hyaluronan Receptor Lyve-1-Expressing Macrophages Maintain Arterial Tone Through Hyaluronan-Mediated Regulation of Smooth Muscle Cell Collagen. Immunity 2018, 49, 326–341.e7. [Google Scholar] [CrossRef] [PubMed]
  35. Cho, C.H.; Jun Koh, Y.; Han, J.; Sung, H.K.; Jong Lee, H.; Morisada, T.; Schwendener, R.A.; Brekken, R.A.; Kang, G.; Oike, Y.; et al. Angiogenic Role of Lyve-1–Positive Macrophages in Adipose Tissue. Circ. Res. 2007, 100, e47–e57. [Google Scholar] [CrossRef] [PubMed]
  36. Kieu, T.Q.; Tazawa, K.; Kawashima, N.; Noda, S.; Fujii, M.; Nara, K.; Hashimoto, K.; Han, P.; Okiji, T. Kinetics of Lyve-1-Positive M2-Like Macrophages in Developing and Repairing Dental Pulp In Vivo and Their Pro-Angiogenic Activity In Vitro. Sci. Rep. 2022, 12, 5176. [Google Scholar] [CrossRef]
  37. Tan, Y.; Flynn, W.F.; Sivajothi, S.; Luo, D.; Bozal, S.B.; Davé, M.; Luciano, A.A.; Robson, P.; Luciano, D.E.; Courtois, E.T.; et al. Single-Cell Analysis of Endometriosis Reveals a Coordinated Transcriptional Programme Driving Immunotolerance and Angiogenesis Across Eutopic and Ectopic Tissues. Nat. Cell Biol. 2022, 24, 1306–1318. [Google Scholar] [CrossRef]
  38. Opzoomer, J.W.; Anstee, J.E.; Dean, I.; Hill, E.J.; Bouybayoune, I.; Caron, J.; Muliaditan, T.; Gordon, P.; Sosnowska, D.; Nuamah, R. Macrophages Orchestrate the Expansion of a Proangiogenic Perivascular Niche During Cancer Progression. Sci. Adv. 2021, 7, eabg9518. [Google Scholar] [CrossRef]
  39. Cheng, S.; Li, Z.; Gao, R.; Xing, B.; Gao, Y.; Yang, Y.; Qin, S.; Zhang, L.; Ouyang, H.; Du, P.; et al. A Pan-Cancer Single-Cell Transcriptional Atlas of Tumor Infiltrating Myeloid Cells. Cell 2021, 184, 792–809.e23. [Google Scholar] [CrossRef]
  40. Alaluf, E.; Vokaer, B.; Detavernier, A.; Azouz, A.; Splittgerber, M.; Carrette, A.; Boon, L.; Libert, F.; Soares, M.P.; Le Moine, A.; et al. Heme Oxygenase-1 Orchestrates the Immunosuppressive Program of Tumor-Associated Macrophages. J. Clin. Investig. 2020, 5, e133929. [Google Scholar] [CrossRef]
  41. Zeng, X.-Y.; Xie, H.; Yuan, J.; Jiang, X.-Y.; Yong, J.-H.; Zeng, D.; Dou, Y.-Y.; Xiao, S.-S. M2-Like Tumor-Associated Macrophages-Secreted Egf Promotes Epithelial Ovarian Cancer Metastasis Via Activating Egfr-Erk Signaling and Suppressing Lncrna Limt Expression. Cancer Biol. Ther. 2019, 20, 956–966. [Google Scholar] [CrossRef]
  42. Condamine, T.; Dominguez, G.A.; Youn, J.-I.; Kossenkov, A.V.; Mony, S.; Alicea-Torres, K.; Tcyganov, E.; Hashimoto, A.; Nefedova, Y.; Lin, C.; et al. Lectin-Type Oxidized Ldl Receptor-1 Distinguishes Population of Human Polymorphonuclear Myeloid-Derived Suppressor Cells in Cancer Patients. Sci. Immunol. 2016, 1, aaf8943. [Google Scholar] [CrossRef]
  43. Jiang, Y.; Zhang, S.; Tang, L.; Li, R.; Zhai, J.; Luo, S.; Peng, Y.; Chen, X.; Wei, L. Single-Cell Rna Sequencing Reveals Tcr(+) Macrophages in Hpv-Related Head and Neck Squamous Cell Carcinoma. Front. Immunol. 2022, 13, 1030222. [Google Scholar] [CrossRef] [PubMed]
  44. Fuchs, T.; Hahn, M.; Riabov, V.; Yin, S.; Kzhyshkowska, J.; Busch, S.; Püllmann, K.; Beham, A.W.; Neumaier, M.; Kaminski, W.E. A Combinatorial Alphabeta T Cell Receptor Expressed by Macrophages in the Tumor Microenvironment. Immunobiology 2017, 222, 39–44. [Google Scholar] [CrossRef] [PubMed]
  45. Hedrick, C.C.; Malanchi, I. Neutrophils in Cancer: Heterogeneous and Multifaceted. Nat. Rev. Immunol. 2021, 22, 173–187. [Google Scholar] [CrossRef] [PubMed]
  46. Alshetaiwi, H.; Pervolarakis, N.; McIntyre, L.L.; Ma, D.; Nguyen, Q.; Rath, J.A.; Nee, K.; Hernandez, G.; Evans, K.; Torosian, L.; et al. Defining the Emergence of Myeloid-Derived Suppressor Cells in Breast Cancer Using Single-Cell Transcriptomics. Sci. Immunol. 2020, 5, eaay6017. [Google Scholar] [CrossRef]
  47. Mastio, J.; Condamine, T.; Dominguez, G.; Kossenkov, A.V.; Donthireddy, L.; Veglia, F.; Lin, C.; Wang, F.; Fu, S.; Zhou, J.; et al. Identification of Monocyte-Like Precursors of Granulocytes in Cancer as a Mechanism for Accumulation of Pmn-Mdscs. J. Exp. Med. 2019, 216, 2150–2169. [Google Scholar] [CrossRef]
  48. Khalil, D.N.; Smith, E.L.; Brentjens, R.J.; Wolchok, J.D. The Future of Cancer Treatment: Immunomodulation, Cars and Combination Immunotherapy. Nat. Rev. Clin. Oncol. 2016, 13, 273–290. [Google Scholar] [CrossRef]
  49. Ribas, A.; Wolchok, J.D. Cancer Immunotherapy Using Checkpoint Blockade. Science 2018, 359, 1350–1355. [Google Scholar] [CrossRef]
  50. Chow, A.; Perica, K.; Klebanoff, C.A.; Wolchok, J.D. Clinical Implications of T Cell Exhaustion for Cancer Immunotherapy. Nat. Rev. Clin. Oncol. 2020, 19, 775–790. [Google Scholar] [CrossRef]
  51. Zielenska, M.; Bayani, J.; Pandita, A.; Toledo, S.; Marrano, P.; Andrade, J.; Petrilli, A.; Thorner, P.; Sorensen, P.; Squire, J.A. Comparative Genomic Hybridization Analysis Identifies Gains of 1p35 Approximately P36 and Chromosome 19 in Osteosarcoma. Cancer Genet. Cytogenet. 2001, 130, 14–21. [Google Scholar] [CrossRef]
  52. Ozaki, T.; Schaefer, K.; Wai, D.; Buerger, H.; Flege, S.; Lindner, N.; Kevric, M.; Diallo, R.; Bankfalvi, A.; Brinkschmidt, C.; et al. Genetic Imbalances Revealed by Comparative Genomic Hybridization in Osteosarcomas. Int. J. Cancer 2002, 102, 355–365. [Google Scholar] [CrossRef]
  53. Forus, A.; Weghuis, D.O.; Smeets, D.; Fodstad, Ø.; Myklebost, O.; van Kessel, A.G. Comparative Genomic Hybridization Analysis of Human Sarcomas: Ii. Identification of Novel Amplicons at 6p and 17p in Osteosarcomas. Genes Chromosomes Cancer 1995, 14, 15–21. [Google Scholar] [CrossRef] [PubMed]
  54. Tarkkanen, M.; Elomaa, I.; Blomqvist, C.; Kivioja, A.H.; Kellokumpu-Lehtinen, P.; Böhling, T.; Valle, J.; Knuutila, S. DNA Sequence Copy Number Increase at 8q: A Potential New Prognostic Marker in High-Grade Osteosarcoma. Int. J. Cancer 1999, 84, 114–121. [Google Scholar] [CrossRef]
  55. dos Santos Aguiar, S.; Zambaldi, L.D.J.G.; dos Santos, A.M.; Pinto, W., Jr.; Brandalise, S.R. Comparative Genomic Hybridization Analysis of Abnormalities in Chromosome 21 in Childhood Osteosarcoma. Cancer Genet. Cytogenet. 2007, 175, 35–40. [Google Scholar] [CrossRef]
  56. Entz-Werle, N.; Lavaux, T.; Metzger, N.; Stoetzel, C.; Lasthaus, C.; Marec, P.; Kalita, C.; Brugieres, L.; Pacquement, H.; Schmitt, C.; et al. Involvement of Met/Twist/Apc Combination or the Potential Role of Ossification Factors in Pediatric High-Grade Osteosarcoma Oncogenesis. Neoplasia 2007, 9, 678–688. [Google Scholar] [CrossRef]
  57. Kresse, S.H.; Ohnstad, H.O.; Paulsen, E.B.; Bjerkehagen, B.; Szuhai, K.; Serra, M.; Schaefer, K.; Myklebost, O.; Meza-Zepeda, L.A. Lsamp, a Novel Candidate Tumor Suppressor Gene in Human Osteosarcomas, Identified by Array Comparative Genomic Hybridization. Genes Chromosomes Cancer 2009, 48, 679–693. [Google Scholar] [CrossRef]
  58. Pompetti, F.; Rizzo, P.; Simon, R.M.; Freidlin, B.; Mew, D.J.; Pass, H.I.; Picci, P.; Levine, A.S.; Carbone, M. Oncogene Alterations in Primary, Recurrent, and Metastatic Human Bone Tumors. J Cell Biochem. 1996, 63, 37–50. [Google Scholar] [CrossRef]
  59. Squire, J.A.; Pei, J.; Marrano, P.; Beheshti, B.; Bayani, J.; Lim, G.; Moldovan, L.; Zielenska, M. High-Resolution Mapping of Amplifications and Deletions in Pediatric Osteosarcoma by Use of Cgh Analysis of Cdna Microarrays. Genes Chromosomes Cancer 2003, 38, 215–225. [Google Scholar] [CrossRef]
  60. Ladanyi, M.; Park, C.K.; Lewis, R.; Jhanwar, S.C.; Healey, J.H.; Huvos, A.G. Sporadic Amplification of the Myc Gene in Human Osteosarcomas. Diagn. Mol. Pathol. 1993, 2, 163–167. [Google Scholar] [CrossRef]
  61. Maire, G.; Yoshimoto, M.; Chilton-MacNeill, S.; Thorner, P.S.; Zielenska, M.; Squire, J.A. Recurrent Recql4 Imbalance and Increased Gene Expression Levels Are Associated with Structural Chromosomal Instability in Sporadic Osteosarcoma. Neoplasia 2009, 11, 260–268. [Google Scholar] [CrossRef]
  62. Stock, C.; Kager, L.; Fink, F.M.; Gadner, H.; Ambros, P.F. Chromosomal Regions Involved in the Pathogenesis of Osteosarcomas. Genes Chromosomes Cancer 2000, 28, 329–336. [Google Scholar] [CrossRef]
  63. Bridge, J.A.; Nelson, M.; McComb, E.; McGuire, M.H.; Rosenthal, H.; Vergara, G.; Maale, G.E.; Spanier, S.; Neff, J.R. Cytogenetic Findings in 73 Osteosarcoma Specimens and a Review of the Literature. Cancer Genet. Cytogenet. 1997, 95, 74–87. [Google Scholar] [CrossRef] [PubMed]
  64. Mertens, F.; Mandahl, N.; Örndal, C.; Baldetorp, B.; Bauer, H.C.F.; Rydholm, A.; Wiebe, T.; Willén, H.; Åkerman, M.; Heim, S.; et al. Cytogenetic Findings in 33 Osteosarcomas. Int. J. Cancer 1993, 55, 44–50. [Google Scholar] [CrossRef] [PubMed]
  65. Taylor, A.M.; Sun, J.M.; Yu, A.; Voicu, H.; Shen, J.; Barkauskas, D.A.; Triche, T.J.; Gastier-Foster, J.M.; Man, T.-K.; Lau, C.C. Integrated DNA Copy Number and Expression Profiling Identifies Igf1r as a Prognostic Biomarker in Pediatric Osteosarcoma. Int. J. Mol. Sci. 2022, 23, 8036. [Google Scholar] [CrossRef] [PubMed]
  66. Ohata, N.; Ito, S.; Yoshida, A.; Kunisada, T.; Numoto, K.; Jitsumori, Y.; Kanzaki, H.; Ozaki, T.; Shimizu, K.; Ouchida, M. Highly Frequent Allelic Loss of Chromosome 6q16-23 in Osteosarcoma: Involvement of Cyclin C in Osteosarcoma. Int. J. Mol. Med. 2006, 18, 1153–1158. [Google Scholar] [CrossRef]
  67. Tarkkanen, M.; Karhu, R.; Kallioniemi, A.; Elomaa, I.; Kivioja, A.H.; Nevalainen, J.; Böhling, T.; Karaharju, E.; Hyytinen, E.; Knuutila, S. Gains and Losses of DNA Sequences in Osteosarcomas by Comparative Genomic Hybridization. Cancer Res. 1995, 55, 1334–1338. [Google Scholar]
  68. Tan, B.; Yuan, Z.; Zhang, Q.; Xiqiang, X.; Dong, J. The Nf-Kappab Pathway Is Critically Implicated in the Oncogenic Phenotype of Human Osteosarcoma Cells. J. Appl. Biomed. 2021, 19, 190–201. [Google Scholar] [CrossRef]
  69. Zhang, Y.; Chen, X.; Wang, H.; Gordon-Mitchell, S.; Sahu, S.; Bhagat, T.D.; Choudhary, G.; Aluri, S.; Pradhan, K.; Sahu, P.; et al. Innate Immune Mediator, Interleukin-1 Receptor Accessory Protein (Il1rap), Is Expressed and Pro-Tumorigenic in Pancreatic Cancer. J. Hematol. Oncol. 2022, 15, 70. [Google Scholar] [CrossRef]
  70. Jauhari, S.; Jimeno, A.; Hreno, J.; Bauml, J.M.; Garcia-Ribas, I.; Öhman, M.W.; Rydberg-Millrud, C.; Cohen, R.B. Safety, Tolerability, and Preliminary Efficacy of Nadunolimab, a First-in-Class Monoclonal Antibody against Il1rap, in Combination with Pembrolizumab in Subjects with Solid Tumors. J. Clin. Oncol. 2022, 40, 2527. [Google Scholar] [CrossRef]
  71. Browning, J.L.; Miatkowski, K.; Sizing, I.; Griffiths, D.; Zafari, M.; Benjamin, C.D.; Meier, W.; Mackay, F. Signaling through the Lymphotoxin Beta Receptor Induces the Death of Some Adenocarcinoma Tumor Lines. J. Exp. Med. 1996, 183, 867–878. [Google Scholar] [CrossRef]
  72. Yang, X.; Cheng, H.; Chen, J.; Wang, R.; Saleh, A.; Si, H.; Lee, S.; Guven-Maiorov, E.; Keskin, O.; Gursoy, A. Head and Neck Cancers Promote an Inflammatory Transcriptome through Coactivation of Classic and Alternative Nf-Κb Pathways. Cancer Immunol. Res. 2019, 7, 1760–1774. [Google Scholar] [CrossRef]
  73. Hamilton, J.A. Colony-Stimulating Factors in Inflammation and Autoimmunity. Nat. Rev. Immunol. 2008, 8, 533–544. [Google Scholar] [CrossRef] [PubMed]
  74. Fujiwara, T.; Yakoub, M.A.; Chandler, A.; Christ, A.B.; Yang, G.; Ouerfelli, O.; Rajasekhar, V.K.; Yoshida, A.; Kondo, H.; Hata, T. Csf1/Csf1r Signaling Inhibitor Pexidartinib (Plx3397) Reprograms Tumor-Associated Macrophages and Stimulates T-Cell Infiltration in the Sarcoma Microenvironment. Mol. Cancer Ther. 2021, 20, 1388–1399. [Google Scholar] [CrossRef] [PubMed]
  75. Peyraud, F.; Cousin, S.; Italiano, A. Csf-1r Inhibitor Development: Current Clinical Status. Curr. Oncol. Rep. 2017, 19, 70. [Google Scholar] [CrossRef]
  76. Owen, J.L.; Mohamadzadeh, M. Macrophages and Chemokines as Mediators of Angiogenesis. Front. Physiol. 2013, 4, 159. [Google Scholar] [CrossRef]
  77. Vergadi, E.; Ieronymaki, E.; Lyroni, K.; Vaporidi, K.; Tsatsanis, C. Akt Signaling Pathway in Macrophage Activation and M1/M2 Polarization. J. Immunol. 2017, 198, 1006–1014. [Google Scholar] [CrossRef]
  78. Wang, X.; Semba, T.; Manyam, G.C.; Wang, J.; Shao, S.; Bertucci, F.; Finetti, P.; Krishnamurthy, S.; Phi, L.T.H.; Pearson, T.; et al. Egfr Is a Master Switch between Immunosuppressive and Immunoactive Tumor Microenvironment in Inflammatory Breast Cancer. Sci. Adv. 2022, 8, eabn7983. [Google Scholar] [CrossRef]
  79. Ni, Y.; Soliman, A.; Joehlin-Price, A.; Rose, P.G.; Vlad, A.; Edwards, R.P.; Mahdi, H. High Tgf-Beta Signature Predicts Immunotherapy Resistance in Gynecologic Cancer Patients Treated with Immune Checkpoint Inhibition. NPJ Precis. Oncol. 2021, 5, 101. [Google Scholar] [CrossRef]
  80. Pauken, K.E.; Wherry, E.J. Overcoming T Cell Exhaustion in Infection and Cancer. Trends Immunol. 2015, 36, 265–276. [Google Scholar] [CrossRef]
  81. Gumber, D.; Wang, L.D. Improving Car-T Immunotherapy: Overcoming the Challenges of T Cell Exhaustion. EBioMedicine 2022, 77, 103941. [Google Scholar] [CrossRef]
  82. Wu, Y.; Yi, M.; Niu, M.; Mei, Q.; Wu, K. Myeloid-Derived Suppressor Cells: An Emerging Target for Anticancer Immunotherapy. Mol. Cancer 2022, 21, 184. [Google Scholar] [CrossRef]
  83. Garlanda, C.; Mantovani, A. Interleukin-1 in Tumor Progression, Therapy, and Prevention. Cancer Cell 2021, 39, 1023–1027. [Google Scholar] [CrossRef] [PubMed]
  84. Fang, Z.; Jiang, J.; Zheng, X. Interleukin-1 Receptor Antagonist: An Alternative Therapy for Cancer Treatment. Life Sci. 2023, 335, 122276. [Google Scholar] [CrossRef] [PubMed]
  85. Villatoro, A.; Cuminetti, V.; Bernal, A.; Torroja, C.; Cossío, I.; Benguría, A.; Ferré, M.; Konieczny, J.; Vázquez, E.; Rubio, A.; et al. Endogenous Il-1 Receptor Antagonist Restricts Healthy and Malignant Myeloproliferation. Nat. Commun. 2023, 14, 12. [Google Scholar] [CrossRef] [PubMed]
  86. Smeester, B.A.; Slipek, N.J.; Pomeroy, E.J.; Laoharawee, K.; Osum, S.H.; Larsson, A.T.; Williams, K.B.; Stratton, N.; Yamamoto, K.; Peterson, J.J. Plx3397 Treatment Inhibits Constitutive Csf1r-Induced Oncogenic Erk Signaling, Reduces Tumor Growth, and Metastatic Burden in Osteosarcoma. Bone 2020, 136, 115353. [Google Scholar] [CrossRef]
  87. Boye, K.; Longhi, A.; Guren, T.; Lorenz, S.; Næss, S.; Pierini, M.; Taksdal, I.; Lobmaier, I.; Cesari, M.; Paioli, A. Pembrolizumab in Advanced Osteosarcoma: Results of a Single-Arm, Open-Label, Phase 2 Trial. Cancer Immunol. Immunother. 2021, 70, 2617–2624. [Google Scholar] [CrossRef]
  88. Baglio, S.R.; Lagerweij, T.; Pérez-Lanzón, M.; Ho, X.D.; Léveillé, N.; Melo, S.A.; Cleton-Jansen, A.-M.; Jordanova, E.S.; Roncuzzi, L.; Greco, M. Blocking Tumor-Educated Msc Paracrine Activity Halts Osteosarcoma Progression. Clin. Cancer Res. 2017, 23, 3721–3733. [Google Scholar] [CrossRef]
  89. Poudel, B.H.; Koks, S. The Whole Transcriptome Analysis Using Ffpe and Fresh Tissue Samples Identifies the Molecular Fingerprint of Osteosarcoma. Exp. Biol. Med. 2024, 249, 10161. [Google Scholar] [CrossRef]
  90. Ho, X.D.; Nguyen, H.G.; Trinh, L.H.; Reimann, E.; Prans, E.; Kõks, G.; Maasalu, K.; Le, V.Q.; Nguyen, V.H.; Le, N.T.N.; et al. Analysis of the Expression of Repetitive DNA Elements in Osteosarcoma. Front. Genet. 2017, 8, 193. [Google Scholar] [CrossRef]
  91. Ho, X.D.; Phung, P.; Le, V.Q.; Nguyen, V.H.; Reimann, E.; Prans, E.; Kõks, G.; Maasalu, K.; Le, N.T.; Trinh, L.H.; et al. Whole Transcriptome Analysis Identifies Differentially Regulated Networks between Osteosarcoma and Normal Bone Samples. Exp. Biol. Med. 2017, 242, 1802–1811. [Google Scholar] [CrossRef]
Cancers 17 02117 g001
Figure 1. Primary clustering identifies major cell types in pre-treatment osteosarcoma tumors. (A) UMAP visualization of Harmony-corrected principal components, with cell type clusters separated by color. (B) Cell type proportions of each sample cohort. Macrophages and osteosarcoma were the most abundant cell types in both cohorts. (C) UMAP visualization of Harmony-corrected principal components in each cohort. (D) Expression of selected cell-type-specific markers in major cell clusters. Markers were chosen from the differentially expressed genes in Supplementary Table S3 based on fold-change, cell cluster specificity, and their prior published use as cell type markers.
Figure 1. Primary clustering identifies major cell types in pre-treatment osteosarcoma tumors. (A) UMAP visualization of Harmony-corrected principal components, with cell type clusters separated by color. (B) Cell type proportions of each sample cohort. Macrophages and osteosarcoma were the most abundant cell types in both cohorts. (C) UMAP visualization of Harmony-corrected principal components in each cohort. (D) Expression of selected cell-type-specific markers in major cell clusters. Markers were chosen from the differentially expressed genes in Supplementary Table S3 based on fold-change, cell cluster specificity, and their prior published use as cell type markers.
Cancers 17 02117 g001
Cancers 17 02117 g002
Figure 2. Subclustering of macrophage/DC clusters. (A) UMAP visualization of Harmony-corrected principal components of the macrophage cluster. Cell subtypes separated by color. (B) Expression of selected cell-subtype-specific markers amongst macrophage subclusters. Several subtypes of activated, interferon (IFN)-stimulated, and angiogenic macrophages were identified. (C) UMAP visualization of Harmony-corrected principal components of the proliferative macrophage/DC cluster. Cell subtypes separated by color. (D) Expression of selected cell-subtype-specific markers amongst macrophage/DC subclusters. LYVE+ macrophages were detected, along with populations of cDC1 and cDC2, in addition to a putative LAMP3+ mregDC population.
Figure 2. Subclustering of macrophage/DC clusters. (A) UMAP visualization of Harmony-corrected principal components of the macrophage cluster. Cell subtypes separated by color. (B) Expression of selected cell-subtype-specific markers amongst macrophage subclusters. Several subtypes of activated, interferon (IFN)-stimulated, and angiogenic macrophages were identified. (C) UMAP visualization of Harmony-corrected principal components of the proliferative macrophage/DC cluster. Cell subtypes separated by color. (D) Expression of selected cell-subtype-specific markers amongst macrophage/DC subclusters. LYVE+ macrophages were detected, along with populations of cDC1 and cDC2, in addition to a putative LAMP3+ mregDC population.
Cancers 17 02117 g002
Cancers 17 02117 g003
Figure 3. Identification of MDSCs amongst the neutrophil/monocyte population. (A) UMAP visualization of Harmony-corrected principal components of the major neutrophil/monocyte cluster. Cell subtypes separated by color. (B) Dotplot expression of selected markers amongst neutrophil and monocyte subclusters (“N-CM” = non-classical monocyte). (C) Visualization of MDSC feature expression. Seurat’s “AddModuleScore” function was used to create an S100A gene expression score in the leftmost plot. The expression of other MDSC signature genes (VCAN, CLEC4E and CSF3R) is also shown.
Figure 3. Identification of MDSCs amongst the neutrophil/monocyte population. (A) UMAP visualization of Harmony-corrected principal components of the major neutrophil/monocyte cluster. Cell subtypes separated by color. (B) Dotplot expression of selected markers amongst neutrophil and monocyte subclusters (“N-CM” = non-classical monocyte). (C) Visualization of MDSC feature expression. Seurat’s “AddModuleScore” function was used to create an S100A gene expression score in the leftmost plot. The expression of other MDSC signature genes (VCAN, CLEC4E and CSF3R) is also shown.
Cancers 17 02117 g003
Cancers 17 02117 g004
Figure 4. Identification of NK/T cell subtypes using Azimuth. (A) Identified NK/T cells and their subtypes. Cells identified with >0.75 confidence score were mapped to Azimuth’s “L1” identities (‘Major Cell Cluster’) from the reference “PBMC” dataset. Further annotation using the “L2” cell identities (‘Specific Cell Type’) is also shown, along with a subset of the signature genes that define those cell subtypes. (B) UMAP visualization of Harmony-corrected principal components of identified NK/T cell populations, colored by L1 identities. (C) NK cell, CD4+ T cell, and CD8+ T cell scores were created using Seurat’s “AddModuleScore” function from the Azimuth-defined cell type markers. (D) UMAP visualization of Harmony-corrected principal components of identified CD8 T-effector-memory (CD8TEM) cell subpopulations. (E) Dotplot expression of exhaustion markers amongst CD8TEM subpopulations. Exhaustion markers were overexpressed in the largest subcluster of CD8TEMs.
Figure 4. Identification of NK/T cell subtypes using Azimuth. (A) Identified NK/T cells and their subtypes. Cells identified with >0.75 confidence score were mapped to Azimuth’s “L1” identities (‘Major Cell Cluster’) from the reference “PBMC” dataset. Further annotation using the “L2” cell identities (‘Specific Cell Type’) is also shown, along with a subset of the signature genes that define those cell subtypes. (B) UMAP visualization of Harmony-corrected principal components of identified NK/T cell populations, colored by L1 identities. (C) NK cell, CD4+ T cell, and CD8+ T cell scores were created using Seurat’s “AddModuleScore” function from the Azimuth-defined cell type markers. (D) UMAP visualization of Harmony-corrected principal components of identified CD8 T-effector-memory (CD8TEM) cell subpopulations. (E) Dotplot expression of exhaustion markers amongst CD8TEM subpopulations. Exhaustion markers were overexpressed in the largest subcluster of CD8TEMs.
Cancers 17 02117 g004
Cancers 17 02117 g005
Figure 5. Upregulated ligand–receptor interactions from S2C2, separated by ligand-expressing cell type. (A) (left) A chord diagram from ligand–receptor interaction network between tested cell types. (right) A chord diagram highlighting the interaction originating from myeloid-derived suppressor cells (MDSCs). Width of links are proportional to the number of ligand–receptor interactions identified. (B) Ligand–receptor interactions and downstream pathway expression originating from myeloid-derived suppressor cells (MDSCs). Only identified ligand–receptor interactions with significant enrichment of the downstream pathway are shown. (left) Ligand and receptor expression, annotated by receiver cell type. (right) Boxplot showing the expression of genes in pathways downstream of the ligand–receptor interaction in the receiver cell type, with outliers shown as circles. (C) Summary of the selected interactions between immune cell types and osteosarcoma.
Figure 5. Upregulated ligand–receptor interactions from S2C2, separated by ligand-expressing cell type. (A) (left) A chord diagram from ligand–receptor interaction network between tested cell types. (right) A chord diagram highlighting the interaction originating from myeloid-derived suppressor cells (MDSCs). Width of links are proportional to the number of ligand–receptor interactions identified. (B) Ligand–receptor interactions and downstream pathway expression originating from myeloid-derived suppressor cells (MDSCs). Only identified ligand–receptor interactions with significant enrichment of the downstream pathway are shown. (left) Ligand and receptor expression, annotated by receiver cell type. (right) Boxplot showing the expression of genes in pathways downstream of the ligand–receptor interaction in the receiver cell type, with outliers shown as circles. (C) Summary of the selected interactions between immune cell types and osteosarcoma.
Cancers 17 02117 g005
Cancers 17 02117 g006
Figure 6. Spatial transcriptomics analysis of osteosarcoma. (A) Estimated fractions of major classes of cell types per spot as determined by deconvolution using RCTD and the single cell reference dataset. The immune class was aggregated from values inferred for 12 subtypes of immune cells. Osteosarcoma, endothelial, and osteoclast were calculated directly. (B) Estimated fractions of 13 immune cell subtypes per spot inferred by deconvolution showed variability between patients. (C) H&E images overlayed with spot cluster membership determined by unsupervised clustering of gene expression defining regional patterning. (D) Cluster membership information depicted as a UMAP plot (left) and as a percent of total spots for each patient (right). (E) Heatmap with z-scores based on the median cell type fraction per cluster. (F) Heatmap with z-scores based on the median PROGENy scores per cluster. These scores of relative pathway activity based on weighted gene expression levels were calculated per spot. Asterisks indicate the adjusted significance level for enrichment determined by one-sided Wilcoxon rank sum test. Spots in a single cluster were compared to all other spots. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Figure 6. Spatial transcriptomics analysis of osteosarcoma. (A) Estimated fractions of major classes of cell types per spot as determined by deconvolution using RCTD and the single cell reference dataset. The immune class was aggregated from values inferred for 12 subtypes of immune cells. Osteosarcoma, endothelial, and osteoclast were calculated directly. (B) Estimated fractions of 13 immune cell subtypes per spot inferred by deconvolution showed variability between patients. (C) H&E images overlayed with spot cluster membership determined by unsupervised clustering of gene expression defining regional patterning. (D) Cluster membership information depicted as a UMAP plot (left) and as a percent of total spots for each patient (right). (E) Heatmap with z-scores based on the median cell type fraction per cluster. (F) Heatmap with z-scores based on the median PROGENy scores per cluster. These scores of relative pathway activity based on weighted gene expression levels were calculated per spot. Asterisks indicate the adjusted significance level for enrichment determined by one-sided Wilcoxon rank sum test. Spots in a single cluster were compared to all other spots. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Cancers 17 02117 g006
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Taylor, A.M.; Sheng, J.; Ng, P.K.S.; Harder, J.M.; Kumar, P.; Ahn, J.Y.; Cao, Y.; Dzis, A.M.; Jillette, N.L.; Goodspeed, A.; et al. Immunosuppressive Tumor Microenvironment of Osteosarcoma. Cancers 2025, 17, 2117. https://doi.org/10.3390/cancers17132117

AMA Style

Taylor AM, Sheng J, Ng PKS, Harder JM, Kumar P, Ahn JY, Cao Y, Dzis AM, Jillette NL, Goodspeed A, et al. Immunosuppressive Tumor Microenvironment of Osteosarcoma. Cancers. 2025; 17(13):2117. https://doi.org/10.3390/cancers17132117

Chicago/Turabian Style

Taylor, Aaron Michael, Jianting Sheng, Patrick Kwok Shing Ng, Jeffrey M. Harder, Parveen Kumar, Ju Young Ahn, Yuliang Cao, Alissa M. Dzis, Nathaniel L. Jillette, Andrew Goodspeed, and et al. 2025. "Immunosuppressive Tumor Microenvironment of Osteosarcoma" Cancers 17, no. 13: 2117. https://doi.org/10.3390/cancers17132117

APA Style

Taylor, A. M., Sheng, J., Ng, P. K. S., Harder, J. M., Kumar, P., Ahn, J. Y., Cao, Y., Dzis, A. M., Jillette, N. L., Goodspeed, A., Bodlak, A., Wu, Q., Isakoff, M. S., George, J., Grassmann, J. D. S., Luo, D., Flynn, W. F., Courtois, E. T., Robson, P., ... Lau, C. C. (2025). Immunosuppressive Tumor Microenvironment of Osteosarcoma. Cancers, 17(13), 2117. https://doi.org/10.3390/cancers17132117

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop