CoChat

Pathways of immune exclusion in metastatic osteosarcoma are associated with inferior patient outcomes

Open access

The DOI matches its Crossref record; 51 of the 52 references it cites were found in a registry or an index.

Abstract

Background: Current therapy for osteosarcoma pulmonary metastases (PMs) is ineffective. The mechanisms that prevent successful immunotherapy in osteosarcoma are incompletely understood. We investigated the tumor microenvironment of metastatic osteosarcoma with the goal of harnessing the immune system as a therapeutic strategy. Methods: 66 osteosarcoma tissue specimens were analyzed by immunohistochemistry (IHC) and immune markers were digitally quantified. Tumor-infiltrating lymphocytes (TILs) from 25 specimens were profiled by functional cytometry. Comparative transcriptomic studies of distinct tumor-normal lung ‘PM interface’ and ‘PM interior’ regions from 16 PMs were performed. Clinical follow-up (median 24 months) was available from resection. Results: IHC revealed a statistically significantly higher concentration of TILs expressing immune checkpoint and immunoregulatory molecules in PMs compared with primary bone tumors (including programmed cell death 1 (PD-1), programmed death ligand 1 (PD-L1), lymphocyte-activation gene 3 (LAG-3), T-cell immunoglobulin and mucin domain-containing protein 3 (TIM-3), and indoleamine 2,3-dioxygenase (IDO1). Remarkably, these lymphocytes are excluded at the PM interface compared with PM interior. TILs from PMs exhibited significantly higher amounts of PD-1 and LAG-3 and functional cytokines including interferon-γ (IFNγ) by flow cytometry. Gene expression profiling further confirmed the presence of CD8 and CD4 lymphocytes concentrated at the PM interface, along with upregulation of immunoregulatory molecules and IFNγ-driven genes in the same region. We further discovered a strong alternatively activated macrophage signature throughout the entire PMs along with a polymorphonuclear myeloid-derived suppressor cell signature focused at the PM interface. Expression of PD-L1, LAG-3, and colony-stimulating factor 1 receptor (CSF1R) at the PM interface was associated with significantly worse progression-free survival (PFS), while gene sets indicative of productive T cell immune responses (CD8 T cells, T cell survival, and major histocompatibility complex class 1 expression) were associated with significantly improved PFS. Conclusions: Osteosarcoma PMs exhibit immune exclusion characterized by the accumulation of TILs at the PM interface. These TILs produce effector cytokines, suggesting their capability of activation and recognition of tumor antigens. Our findings suggest cooperative immunosuppressive mechanisms in osteosarcoma PMs including immune checkpoint molecule expression and the presence of immunosuppressive myeloid cells. We identify cellular and molecular signatures that are associated with patient outcomes, which could be exploited for successful immunotherapy.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1136/jitc-2020-001772

Cites

51 of 51 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Immunohistochemistry (IHC) demonstrating high concentration of T cells and programmed cell death 1 (PD-1) in pulmonary metastases (PM), particularly at tumor-normal lung interface. Myeloid cells present throughout. (A) Primary bone tumor showing a low concentration of T cells (CD3+) and programmed death ligand 1 (PD-L1), but a higher concentration of CD163+ myeloid cells. (B) H&E with demarcation of tumor-normal lung interface (center green line) and area quantified as the ‘interface’ (IF, outer green lines). PM demonstrates a higher concentration of immune cells (CD3, CD8, CD163) and PD-1/PD-L1 at IF between lung tissue and the metastasis. (C) Quantification of infiltrating CD3+, CD8+, Foxp3+, CD163+ immune cells and PD-1/PD-L1 in osteosarcoma (B=primary bone tumor, PM=entire pulmonary metastasis, IF=PM interface). Gray box drawn to mean, bars showing SD. NS=p>0.05; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. (D) IHCs depicting colocalization of CD163 myeloid marker and PD-L1 at the IF.
Figure 1 Immunohistochemistry (IHC) demonstrating high concentration of T cells and programmed cell death 1 (PD-1) in pulmonary metastases (PM), particularly at tumor-normal lung interface. Myeloid cells present throughout. (A) Primary bone tumor showing a low concentration of T cells (CD3+) and programmed death ligand 1 (PD-L1), but a higher concentration of CD163+ myeloid cells. (B) H&E with demarcation of tumor-normal lung interface (center green line) and area quantified as the ‘interface’ (IF, outer green lines). PM demonstrates a higher concentration of immune cells (CD3, CD8, CD163) and PD-1/PD-L1 at IF between lung tissue and the metastasis. (C) Quantification of infiltrating CD3+, CD8+, Foxp3+, CD163+ immune cells and PD-1/PD-L1 in osteosarcoma (B=primary bone tumor, PM=entire pulmonary metastasis, IF=PM interface). Gray box drawn to mean, bars showing SD. NS=p>0.05; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. (D) IHCs depicting colocalization of CD163 myeloid marker and PD-L1 at the IF.CC BY-NC · PubMed Central
Figure 2 Immunohistochemistry (IHC) analysis reveals high expression of immunoregulatory molecules concentrated at the tumor-normal lung interface in pulmonary metastases (PM). (A) H&E with demarcation of tumor-normal lung interface (center green line) and area quantified as ‘interface’ (IF, outer green lines). PM demonstrating staining of TIM-3, LAG-3, IDO1, and CSF1R at the interface. (B) Quantification of immunoregulatory molecules in PM and particularly at the interface (PM=entire pulmonary metastasis, IF=PM interface). Line between points matches a specific IF with the entire PM from the same specimen. NS=p>0.05; **p<0.01; ****p<0.0001. CSF1R, colony-stimulating factor 1 receptor; IDO1, indoleamine 2,3-dioxygenase; LAG-3, lymphocyte-activation gene 3; TIM-3, T-cell immunoglobulin and mucin domain-containing protein 3.
Figure 2 Immunohistochemistry (IHC) analysis reveals high expression of immunoregulatory molecules concentrated at the tumor-normal lung interface in pulmonary metastases (PM). (A) H&E with demarcation of tumor-normal lung interface (center green line) and area quantified as ‘interface’ (IF, outer green lines). PM demonstrating staining of TIM-3, LAG-3, IDO1, and CSF1R at the interface. (B) Quantification of immunoregulatory molecules in PM and particularly at the interface (PM=entire pulmonary metastasis, IF=PM interface). Line between points matches a specific IF with the entire PM from the same specimen. NS=p>0.05; **p<0.01; ****p<0.0001. CSF1R, colony-stimulating factor 1 receptor; IDO1, indoleamine 2,3-dioxygenase; LAG-3, lymphocyte-activation gene 3; TIM-3, T-cell immunoglobulin and mucin domain-containing protein 3.CC BY-NC · PubMed Central
Figure 3 Multiparameter flow cytometry (MFC) scatterplots demonstrate that tumor-infiltrating lymphocytes (TILs) from pulmonary metastases (PM) express higher levels of checkpoint molecules, effector cytokines and transcription factors compared with primary bone tumors. (A) Relative proportions of CD4+Foxp3+ (regulatory T cell), CD4+Foxp3-, and CD8+TILs isolated from osteosarcoma primary bone tumors (B) and PM. (B) Expression of checkpoint molecules in TILs. Given the sequential manner in which TILs may express checkpoint molecules as they become progressively exhausted, TIM-3+ and LAG-3+ populations are gated out of PD-1+ positive population. Note: For CD4+Foxp3-PD-1+LAG-3+ cells, p=0.064. (C) Expression of the transcription factors T-bet and eomesodermin (EOMES) in PD-1+TILs. Note: For CD4+Foxp3-PD-1+T-bet+ cells, p=0.073. (D) Expression of effector cytokine IFNγ in PD-1+TILs. NS=p>0.05; *p<0.05; **p<0.01. IFNγ, interferon-γ; LAG-3, lymphocyte-activation gene 3; PD-1, programmed cell death 1; TIM-3, T-cell immunoglobulin and mucin domain-containing protein 3.
Figure 3 Multiparameter flow cytometry (MFC) scatterplots demonstrate that tumor-infiltrating lymphocytes (TILs) from pulmonary metastases (PM) express higher levels of checkpoint molecules, effector cytokines and transcription factors compared with primary bone tumors. (A) Relative proportions of CD4+Foxp3+ (regulatory T cell), CD4+Foxp3-, and CD8+TILs isolated from osteosarcoma primary bone tumors (B) and PM. (B) Expression of checkpoint molecules in TILs. Given the sequential manner in which TILs may express checkpoint molecules as they become progressively exhausted, TIM-3+ and LAG-3+ populations are gated out of PD-1+ positive population. Note: For CD4+Foxp3-PD-1+LAG-3+ cells, p=0.064. (C) Expression of the transcription factors T-bet and eomesodermin (EOMES) in PD-1+TILs. Note: For CD4+Foxp3-PD-1+T-bet+ cells, p=0.073. (D) Expression of effector cytokine IFNγ in PD-1+TILs. NS=p>0.05; *p<0.05; **p<0.01. IFNγ, interferon-γ; LAG-3, lymphocyte-activation gene 3; PD-1, programmed cell death 1; TIM-3, T-cell immunoglobulin and mucin domain-containing protein 3.CC BY-NC · PubMed Central
Figure 4 Activated/exhausted lymphocyte, interferon-γ (IFNγ)-responsive and immune checkpoint molecule signatures upregulated at the pulmonary metastasis (PM) interface, but strong myeloid signature throughout the entire tumor. (A) Heatmap representing coefficients of analysis via CIBERSORT shows high CD8 T cell (p=0.0371) and CD4 memory resting T cell (p=0.002) signatures at the PM interface region. There is a slightly stronger M1 macrophage signature (statistically not significant, p=0.547) present at the interface compared with the tumor interior; however, both regions have a much stronger M2 macrophage signature overall. See online supplemental table 2 for raw data and full statistics. (B) Heatmap displaying significant genes that contribute to leading-edge of core enrichment subset via gene set enrichment analysis (GSEA) demonstrating higher expression of immune regulatory molecules at the interface compared with the tumor interior. (C) Heatmap depicting upregulated IFNγ-responsive genes present at the interface by GSEA. (D) Heatmap illustrating coefficients of xCell analysis shows higher expression of markers of cytotoxicity and activation, as well as multiple checkpoint molecules, at the PM interface, with evidence that they are being contributed chiefly by T cells. Intensity represents xCell coefficient, which corresponds to the amount that a particular region (PM interior or PM interface) or cell population (T cells, B cells, or myeloid cells) contributes to the expression of a specific gene. CIBERSORT deconvolution with the leukocyte gene signature matrix LM2221 and xCell22 were used to analyze immune and other cell composition of our data. For GSEA (B and C), genes displayed contribute to the leading-edge core enrichment subset. Expression levels were converted into heatmaps and colors quantitatively correspond to fold changes. FDR, GSEA false discovery rate q value. See online supplemental table 3 for raw rank metric scores for all genes included in GSEA and online supplemental table 4 for xCell coefficients. CIBERSORT, Cell-type identification by estimating relative subsets of RNA transcripts.
Figure 4 Activated/exhausted lymphocyte, interferon-γ (IFNγ)-responsive and immune checkpoint molecule signatures upregulated at the pulmonary metastasis (PM) interface, but strong myeloid signature throughout the entire tumor. (A) Heatmap representing coefficients of analysis via CIBERSORT shows high CD8 T cell (p=0.0371) and CD4 memory resting T cell (p=0.002) signatures at the PM interface region. There is a slightly stronger M1 macrophage signature (statistically not significant, p=0.547) present at the interface compared with the tumor interior; however, both regions have a much stronger M2 macrophage signature overall. See online supplemental table 2 for raw data and full statistics. (B) Heatmap displaying significant genes that contribute to leading-edge of core enrichment subset via gene set enrichment analysis (GSEA) demonstrating higher expression of immune regulatory molecules at the interface compared with the tumor interior. (C) Heatmap depicting upregulated IFNγ-responsive genes present at the interface by GSEA. (D) Heatmap illustrating coefficients of xCell analysis shows higher expression of markers of cytotoxicity and activation, as well as multiple checkpoint molecules, at the PM interface, with evidence that they are being contributed chiefly by T cells. Intensity represents xCell coefficient, which corresponds to the amount that a particular region (PM interior or PM interface) or cell population (T cells, B cells, or myeloid cells) contributes to the expression of a specific gene. CIBERSORT deconvolution with the leukocyte gene signature matrix LM2221 and xCell22 were used to analyze immune and other cell composition of our data. For GSEA (B and C), genes displayed contribute to the leading-edge core enrichment subset. Expression levels were converted into heatmaps and colors quantitatively correspond to fold changes. FDR, GSEA false discovery rate q value. See online supplemental table 3 for raw rank metric scores for all genes included in GSEA and online supplemental table 4 for xCell coefficients. CIBERSORT, Cell-type identification by estimating relative subsets of RNA transcripts.CC BY-NC · PubMed Central
Figure 5 Strong expression of myeloid genes throughout pulmonary metastases (PM), with highest expression of genes related to dendritic cells (DCs) and myeloid-derived suppressor cells (MDSCs) at the tumor interface. (A) Genes associated with colony-stimulating factor 1 receptor (CSF1R) responsiveness plotted as a heatmap are more strongly expressed at the PM interface, but are also expressed throughout the PM interior. (B) By gene set enrichment analysis (GSEA), genes associated with multiple subclasses of antigen-presenting DCs are significantly upregulated at the PM interface (cDC1, conventional type 1 dendritic cell; cDC2, conventional type 2 dendritic cell; pDC, plasmacytoid dendritic cell; moDC, monocyte-derived dendritic cell). (C) Box-and-whisker plots of canonical genes associated with immunosuppressive monocytic myeloid-derived suppressor cells (M-MDSCs) and polymorphonuclear MDSCs (PMN-MDSCs) displaying upregulation at the PM interface. NS=p>0.05; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 (red=upregulated in the PM interior; blue=upregulated at the PM interface). Z score transformed by subtracting mean from each value and dividing by SD. (D) Heatmap shows heightened expression of cytokines, chemokines and endothelin transcripts associated with development, recruitment and maintenance of PMNs and PMN-MDSCs at the PM interface compared with the PM interior. For GSEA (A and B), genes displayed contribute to the leading-edge core enrichment subset. Expression levels were converted into heatmaps and colors quantitatively correspond to fold changes. FDR=GSEA false-discovery rate q-value. See online supplemental table 3 for raw rank metric scores for all genes included in analysis.
Figure 5 Strong expression of myeloid genes throughout pulmonary metastases (PM), with highest expression of genes related to dendritic cells (DCs) and myeloid-derived suppressor cells (MDSCs) at the tumor interface. (A) Genes associated with colony-stimulating factor 1 receptor (CSF1R) responsiveness plotted as a heatmap are more strongly expressed at the PM interface, but are also expressed throughout the PM interior. (B) By gene set enrichment analysis (GSEA), genes associated with multiple subclasses of antigen-presenting DCs are significantly upregulated at the PM interface (cDC1, conventional type 1 dendritic cell; cDC2, conventional type 2 dendritic cell; pDC, plasmacytoid dendritic cell; moDC, monocyte-derived dendritic cell). (C) Box-and-whisker plots of canonical genes associated with immunosuppressive monocytic myeloid-derived suppressor cells (M-MDSCs) and polymorphonuclear MDSCs (PMN-MDSCs) displaying upregulation at the PM interface. NS=p>0.05; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 (red=upregulated in the PM interior; blue=upregulated at the PM interface). Z score transformed by subtracting mean from each value and dividing by SD. (D) Heatmap shows heightened expression of cytokines, chemokines and endothelin transcripts associated with development, recruitment and maintenance of PMNs and PMN-MDSCs at the PM interface compared with the PM interior. For GSEA (A and B), genes displayed contribute to the leading-edge core enrichment subset. Expression levels were converted into heatmaps and colors quantitatively correspond to fold changes. FDR=GSEA false-discovery rate q-value. See online supplemental table 3 for raw rank metric scores for all genes included in analysis.CC BY-NC · PubMed Central
Figure 6 Forest plots demonstrating that markers of the immune tumor microenvironment (TME) at the pulmonary metastasis (PM) interface correlate with progression-free survival (PFS). (A) HRs for immunohistochemistry (IHC) markers at the PM interface as they relate to PFS. For absolute count biomarkers (CD3, CD8, Foxp3, PD-1, CD163, and LAG-3) the unit is per 100 cells, and for percentage biomarkers (PD-L1, CSF1R, TIM-3, and IDO1), the unit is per 1%. (B) HRs for gene sets at the PM interface as they relate to PFS. NS=p>0.05; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. CSF1R, colony-stimulating factor 1 receptor; IDO1, indoleamine 2,3-dioxygenase; IFNγ, interferon-γ; MHC, major histocompatibility complex; PD-L1, programmed death ligand 1; TIM-3, T-cell immunoglobulin and mucin domain-containing protein 3.
Figure 6 Forest plots demonstrating that markers of the immune tumor microenvironment (TME) at the pulmonary metastasis (PM) interface correlate with progression-free survival (PFS). (A) HRs for immunohistochemistry (IHC) markers at the PM interface as they relate to PFS. For absolute count biomarkers (CD3, CD8, Foxp3, PD-1, CD163, and LAG-3) the unit is per 100 cells, and for percentage biomarkers (PD-L1, CSF1R, TIM-3, and IDO1), the unit is per 1%. (B) HRs for gene sets at the PM interface as they relate to PFS. NS=p>0.05; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. CSF1R, colony-stimulating factor 1 receptor; IDO1, indoleamine 2,3-dioxygenase; IFNγ, interferon-γ; MHC, major histocompatibility complex; PD-L1, programmed death ligand 1; TIM-3, T-cell immunoglobulin and mucin domain-containing protein 3.CC BY-NC · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Cited by

See all 124

Metadata

Authors
Ligon, J. A., Choi, W., Cojocaru, G., Fu, W., Hsiue, E. H.-C., Oke, T. F., Siegel, N., Fong, M. H., Ladle, B., Pratilas, C. A., Morris, C. D., Levin, A., Rhee, D. S., Meyer, C. F., Tam, A. J., Blosser, R., Thompson, E. D., Suru, A., McConkey, D., Housseau, F., Anders, R., Pardoll, D. M., & Llosa, N.
Affiliations
Johns Hopkins University (US)
Corresponding author
Llosa, Nicolas
Venue
Journal for ImmunoTherapy of Cancer
Publisher
BMJ
Published
2021-05
Volume
9
Issue
5
Pages
e001772
ISSN
2051-1426 (ISSN-L)
OpenAlex
W3160748480
Cited by
124
References
52
Citations by year
2026: 21 · 2025: 30 · 2024: 27 · 2023: 26 · 2022: 15 · 2021: 5
FWCI
6.64
Type
journal article
Language
en
Open access
Locations
  • Journal for ImmunoTherapy of Cancer (publishedVersion, cc-by-nc)
  • PubMed Central (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Bloomberg Philanthropies
  • St. Baldrick’s foundation
  • National Pediatric Cancer Foundation
  • Giant Food
  • Bloomberg-Kimmel Institute for Immunotherapy
  • BMS II-ON
  • Johns Hopkins Hospital
Topics
  • Health Sciences › Medicine › Oncology › Cancer Immunotherapy and Biomarkers
  • Health Sciences › Medicine › Oncology › Sarcoma Diagnosis and Treatment
  • Life Sciences › Immunology and Microbiology › Immunology › Immunotherapy and Immune Responses
Keywords
immune exclusion, tumor-infiltrating lymphocytes, tumor microenvironment, PD-1, PD-L1, LAG-3, TIM-3, IDO1, interferon-gamma, myeloid-derived suppressor cells, M2 macrophages, CSF1R, progression-free survival, immunotherapy resistance, lung metastasis
MeSH terms
Myeloid-Derived Suppressor Cells, Progression-Free Survival, Immune Checkpoint Proteins, Tumor-Associated Macrophages, Bone Neoplasms, Humans, Immunotherapy, Lung Neoplasms, Lymphocyte Activation, Macrophage Activation, Retrospective Studies, Osteosarcoma, Biomarkers, Tumor, Cytokines, Lymphocytes, Tumor-Infiltrating, Electronic Health Records, Tumor Microenvironment, Transcriptome

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.