Article Text

Download PDFPDF

Original research
Complex immune microenvironment of chordoma: a road map for future treatment
  1. Hua-Qing Niu1,
  2. Bo-Yv Zheng2,
  3. Ming-Xiang Zou3 and
  4. Bo-Wen Zheng3,4
  1. 1Department of Ophthalmology, The Second Affiliated Hospital of Zhengzhou University, Zhengzhou University, Zhengzhou, Henan, China
  2. 2Department of Orthopedics Surgery, General Hospital of the Central Theater Command, Wuhan, China
  3. 3Department of Spine Surgery, The First Affiliated Hospital, Hengyang Medical School, University of South China, Hengyang, China
  4. 4Musculoskeletal Tumor Center, Peking University People's Hospital, Peking University, Beijing, China
  1. Correspondence to Dr Bo-Wen Zheng; xcbowen{at}foxmail.com

Abstract

Background Chordoma, a rare bone tumor, presents limited treatment options and patients typically exhibit poor survival outcomes. While immunotherapy has shown promising results in treating various tumors, research on the immune microenvironment of chordomas is still in its early stages. Therefore, understanding how the immune microenvironment of chordomas influences the outcomes of immunotherapy is crucial.

Methods We employed single-cell RNA sequencing (scRNA-seq), bulk RNA-seq, CellChat, gene set variation analysis, as well as calculation of immune features to further dissect the complex immune microenvironment of chordoma.

Results Previous research by van Oost et al argued that compared with other sarcomas, chordomas typically exhibit an immunologically “hot” microenvironment, a conclusion with which we concur based on their research findings. Additionally, the authors suggest that T cell-mediated immunotherapy is feasible for the majority of chordomas. However, we are inclined to categorize them as an immune-excluded phenotype according to the latest classification methods, rather than persisting with the concepts of “cold” and “hot”. Unlike them, we explored immune infiltration scores (IS), T lymphocyte scoring (TLS), and human leucocyte antigen class I (HLA-I) using Bulk RNA-seq data from 126 chordoma patients and found that higher IS, TLS, and higher HLA-I expression were associated with poorer patient prognosis. Additionally, CellChat analysis of scRNA-seq results from six chordoma patients revealed no direct interaction between T cells and tumor cells.

Conclusions These findings suggested that the efficacy of T cell-based immunotherapy may be limited or even ineffective for patients with chordoma.

  • Bone Cancer
  • Patient reported outcome - PRO
  • T cell

Data availability statement

Data are available upon reasonable request. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

WHAT IS ALREADY KNOWN ON THIS TOPIC

  • The overall therapeutic effect of molecular targeted drugs in chordoma patients is often unsatisfactory and the research on the immune microenvironment of chordomas is still in its early stages. Therefore, understanding how the immune microenvironment of chordomas influences the outcomes of immunotherapy is crucial.

WHAT THIS STUDY ADDS

  • We found that higher immune infiltration scores, T lymphocyte scoring, and HLA-I expression were associated with poorer patient prognosis. Additionally, CellChat analysis of single-cell RNA sequencing results revealed no direct interaction between T cells and tumor cells. These findings suggested that the efficacy of T cell-based immunotherapy may be limited or even ineffective for patients with chordoma.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

  • The composition and interaction mechanisms of the immune microenvironment in chordoma are highly complex, leading to intricate prognostic patterns, and the underlying mechanisms warrant further investigation. In the future, a better understanding of the status and regulatory mechanisms of the chordoma microenvironment is essential for guiding the safer and more effective use of immune checkpoint inhibitor treatment strategies.

Background

Chordoma is a rare bone tumor, and even medical professionals engaged in oncology research are relatively unfamiliar with it. These factors have resulted in basic research on chordoma lagging far behind that of more common malignant tumors so its treatment remains challenging.1 Although in recent years, a few molecular targeted drugs have shown promising efficacy in chordoma patients, overall treatment outcomes are often unsatisfactory due to tumor heterogeneity, and their widespread clinical application is limited.2 We have recently taken an interest in the research conducted by van Oost et al.3 We commend the authors for their series of studies, which contribute to guiding the development and application of immunotherapy, ultimately improving clinical outcomes for chordoma patients. However, we have some questions that require further exploration and research to enhance the completeness and scientific validity of the conclusions.

Methods

Patients and tissue samples

In this study, we included 126 patients (from January 2002 to December 2019) for bulk RNA-seq profiling. The demographic and clinicopathological characteristics of the bulk RNA-seq cohort were comparable and described in detail in table 1. Patients who had undergone any tumor-specific treatments (including radiotherapy, chemotherapy, or other targeted therapies) and those with concurrent illnesses on admission were excluded, considering the potential interference of prior tumor-specific interventions and comorbidities with the chordoma microenvironment.1 Clinical and pathological characteristics were obtained from medical records. Six fresh tumor specimens were also collected during surgery for single-cell RNA sequencing, providing detailed patient characteristics in table 2.

Table 1

Clinicopathological features of 126 chordoma patients

Table 2

Baseline characteristic of six patients for single-cell RNA sequencing

Patient follow-up data were updated through regular clinical and radiological examinations (up to June 2021). The primary outcome of interest was overall survival (OS), defined as the time from surgery to death from any cause or until the last follow-up.1 Survival outcomes were reviewed when patients remained alive.

Analysis of cell–cell interactions

To assess the potential interactions between T cells and other cells, we used CellChatDB to investigate cell–cell communication based on ligands, receptors, and their interactions. Significant ligand-receptor pairs were extracted based on permutation tests with a p<0.05. The R packages igraph and ggplot2 were employed for visualization of the results.

Characterization of immune features

ESTIMATE (Estimation of Stromal and Immune cells in Malignant Tumor tissues using Expression data) is an R package that outputs the ESTIMATE immune infiltration score (IS).4 For a given sample, gene expression values are rank-normalized and ranked. The empirical cumulative distribution functions of genes in the feature and the remaining genes are calculated. A statistic is computed by integrating the differences between the empirical cumulative distribution functions, similar to the statistic used in gene set enrichment analysis but based on absolute expression rather than differential expression.

Gene set variation analysis

To investigate the clinical significance of T lymphocyte scoring (TLS) and nature kill cells scoring (NKS), we obtained specific gene sets for T lymphocytes and NK cells from the CellMarker database (http://xteam.xbio.top/CellMarker/).1 Subsequently, we performed gene set variation analysis (GSVA) on bulk RNA-seq data to calculate the characteristic scores of T lymphocytes and NK cells for each individual sample.

Statistical analysis

Continuous data were summarized as mean±SD and analyzed using Student’s t-test or one-way analysis of variance. Pearson correlation test was used to evaluate the relationship between continuous variables. The X-tile software (V.3.6. 1) was used to obtain thresholds for continuous data in survival analysis, with OS as the outcome parameter.1 Specifically, the threshold was defined as the point with the smallest p value in the corresponding log-rank test after adjustment. Kaplan-Meier curves for OS were plotted, and statistical analysis was performed using R V.3.5.1 (R Foundation for Statistical Computing, Vienna, Austria). All tests were two sided, and p values less than 0.05 were considered statistically significant.

Results

Commentary

First, the authors posited that chordomas typically exhibit an immune “hot” microenvironment compared with other sarcomas. Traditional classifications categorize tumors into “cold” and “hot” types based on the infiltration of CD3+ and CD8+ T cells in the tumor center and the invasive margin of the tumor microenvironment, using an immune scoring system.5 However, due to significant heterogeneity in the composition and phenotype of tumor-infiltrating lymphocytes, assessing the immune status of tumors solely based on “cold” and “hot” types is incomplete. According to the latest classification methods, immune cells, especially T cells, predominantly infiltrate the fibrous septa in the stroma of chordomas, thus indicating an immune-excluded phenotype.6 7 Additionally, in the two cohorts included by the authors, some patients had a history of radiation or chemotherapy, which could severely disrupt the tumor immune microenvironment and even affect patient prognosis, rendering the study results less rigorous and accurate.1

Immune features

This study included 126 chordoma patients. The demographic and clinicopathological characteristics of the two cohorts were comparable and described in detail in table 1. In order to further investigate, the immunological characteristics of chordomas and their association with clinical parameters, we used the X-tile software to determine the threshold for survival analysis, we divided the IS and TLS into high and low groups according to the cut-off point.8 We found that higher IS and TLS were associated with poorer prognosis (p=0.0012 and p<0.001, respectively, figure 1A,B).

Figure 1

(A) Kaplan-Meier curves of overall survival of 126 chordoma patients stratified by immune score (IS), higher IS was associated with poorer prognosis (p=0.0012). (B) Kaplan-Meier curves of overall survival of 126 chordoma patients stratified by T lymphocyte scoring (TLS), higher TLS was associated with poorer prognosis (p<0.001). (C) Association between HLA-I expression and IS, HLA-I expression was significantly positively correlated with IS (R2=0.21, p<0.001); (D) Association between HLA-I expression and TLS, HLA-I expression was significantly positively correlated with TLS (R2=0.1, p<0.001, respectively). Kaplan-Meier curves of overall survival of 126 chordoma patients stratified by HLA-I expression, low expression of HLA-I was associated with favorable prognosis (p=0.010). (E) Association between HLA-I expression and surrounding muscle invasion (p=0.021); association between HLA-I expression and tumor grade (p<0.015); association between HLA-I expression and AJCC staging System (p=0.028 (IA+IB + IIA vs IIB+III), p=0.011 (IA+IB + IIA vs IVA+IVB)). (F) Number and interaction weight/strength of interactions between T cells and other cells. AJCC, American Joint Committee on Cancer.

Human leucocyte antigen class I

Furthermore, the authors conducted immunohistochemical staining for human leucocyte antigen class I (HLA-I) to assess its expression pattern. Immunohistochemical quantification is typically influenced by factors such as staining intensity, whereas RNA sequencing not only provides digital expression levels for genes but also offers higher quantitative accuracy. To further understand the relationship between HLA-I expression and IS, TLS, and patient survival prognosis, we used gene expression data for HLA-I from RNA sequencing data of 126 chordoma patients. We found that HLA-I expression was significantly positively correlated with IS and TLS (R2=0.21, p<0.001 and R2=0.1, p<0.001, respectively, figure 1C). The immune surveillance function of the body to clear foreign invaders and T cell therapy using immune checkpoint inhibitors (ICIs) both rely on the cytotoxic activity exerted by T cells, and the prerequisite for T cells to exert effector function is the recognition of HLA-I molecules on the surface of tumor cells.9 Decreased expression of HLA-I suggests significant immune evasion, which may indicate tumor progression and poorer prognosis.10 However, our study found that low expression of HLA-I was associated with favorable prognosis for patients (p=0.010, figure 1D) and better tumor phenotypes of invasiveness such as absence of surrounding muscle invasion, low tumor grade and lower American Joint Committee on Cancer staging (p=0.021, p<0.015, and p=0.028 (IA+IB+IIA vs IIB+III), p=0.011 (IA+IB + IIA vs IVA+IVB), respectively, figure 1E). Furthermore, CellChat analysis was performed on the single-cell sequencing data of six chordoma cases previously reported by our team, revealing no significant interaction between T cells and tumor cells (figure 1F). Taken together, these findings suggest that the efficacy of T cell-based immunotherapy may be limited or even ineffective.

Discussion

The relationship between T lymphocytes and prognosis varies among different tumors.6 Our research has found that a higher degree of T lymphocyte infiltration is associated with a poorer prognosis. Previous studies have confirmed differences in the composition and distribution of T lymphocytes in the tumor microenvironment of chordomas.6 It is now known that high expression of FoxP3+TILs is significantly associated with poorer prognosis in chordomas while high expression of CD8+TILs, although correlated with a better prognosis,6 is associated with higher tumor grades in certain tumors.11 In chordomas, higher pathological grades are significantly associated with poorer prognosis.1 In this study, TLS was scored based on the overall T lymphocyte infiltration, reflecting the overall level of T lymphocyte infiltration. In conclusion, the complex composition and distribution of T lymphocytes in the chordoma microenvironment lead to complex prognostic patterns, the specific mechanisms of which warrant further investigation.

Furthermore, high expression of HLA is generally considered to be associated with a better prognosis as it aids in the immune system’s recognition and attack of tumor cells. Its gene expression data can be used to predict immune subtypes in patients receiving immune checkpoint blockade and to stratify patient survival.12 However, a retrospective analysis across 33 tumor types found that higher tumor immunogenicity mediated by HLA expression may sometimes lead to tumor escape under strong selective pressure, resulting in tumors with the highest immune infiltration not necessarily having the best prognosis and exhibiting significantly higher immune exhaustion.13 Strong immune cytotoxicity may lead to immune depletion, impairing the antitumor efficacy of immune checkpoint therapy.13 Moreover, an investigation of 24 chordoma samples found that all samples expressed at least one HLA-I molecule (HLA-A, HLA-B/C, and β2 chain) by immunohistochemistry, but 21 out of 24 samples showed at least one HLA-I expression-deficient molecular subunit14 and HLA-I class expression defects were associated with poorer OS in head and neck cancers, esophageal cancers, and melanomas.15 Considering these results, high expression of HLA-I class in chordomas may likely indicate a significant presence of HLA-I class expression defects, potentially being one of the reasons for poorer prognosis, which aligns with our research findings. Additionally, NK cells and CD8+T cells (antigen-presenting cells for HLA-I) can be inhibited by TGF-β1 activity, thus potentially contributing to poorer prognosis in chordomas.16 The enhancement of the TGF-β signaling pathway is a major mechanism for tumor progression, invasion, and immune suppression.17 The Smad protein family is key in the TGF-β signaling pathway, with Smad3 being highly expressed in chordoma tissues, promoting chordoma progression,18 and the Smad3 inhibitor SIS3 has been shown to selectively inhibit the phosphorylation of TGF-β/Smad3 in other cancers such as colorectal cancer.19 Hence, when Smad is elevated, inhibition of the TGF-β signaling pathway is necessary, as it has been reported to significantly inhibit chordoma invasiveness.20 Furthermore, TGF-β1 is one of the most studied and important subtypes of the TGF-β family and is related to immune regulation,20 further survival analysis of TGF-β1 expression also found that patients in the low expression group had better prognosis (p<0.001, figure 2A).

Figure 2

(A) Kaplan-Meier curves of overall survival of 126 chordoma patients stratified by TGF-β expression, low TGF-β expression group was associated with favorable prognosis (p<0.001). (B) Association between HLA-I expression and NK Score (NKS), HLA-I expression was significant negative correlated with NKS (R2=−0.16, p=0.038). (C) Kaplan-Meier curves of overall survival of 126 chordoma patients stratified by NKS, high NKS was associated with favorable prognosis (p<0.001).

Additionally, our research findings indicate that low expression of HLA-I is associated with a better prognosis. According to previous reports, low expression or loss of HLA-I may render tumor cells more sensitive to attack by NK cells.21 Reduced expression of HLA-I diminishes the ability of tumor cells to evade recognition by NK cells through antigen presentation, making it easier for NK cells to identify and clear these tumor cells.22 In line with this, we performed GSVA scoring of NK cells based on Bulk RNA-seq data from 126 patients. We found a significant negative correlation between NKS and HLA-I expression levels (R2=−0.16, p=0.038, figure 2B). Furthermore, high NKS predicted a favorable prognosis for patients (p<0.001, figure 2C), further supporting the above conclusion. It has been reported that IL-2 receptor signaling is crucial for the activation and effector functions of human NK cells, and administration of IL-2 along with cetuximab to chordoma patients has shown greater clinical benefits.23 Moreover, interferon-stimulated gene expression has been detected in chordomas, and interferons enhance the proliferation, differentiation, and activation of NK cells, enhancing their cytotoxic effects.24 25 Numerous reports support that oncolytic viruses increase NK cell infiltration, modulate NK cell activity, and enhance antitumor effects.26 TGF-β antagonists also enhance NK cell-mediated antitumor immune responses.16 Taken together, the combination of these approaches may hold more significance.

Although some data suggest that immunotherapy is a promising treatment strategy for chordomas, the results remain controversial, and there are limited data on the efficacy of immune checkpoint blockade in chordomas.27–30 Furthermore, previous studies have indicated that only 13%–45% of solid cancer patients show a favorable response to immunotherapy, suggesting the existence of alternative immune checkpoint pathways or pathways that disrupt immune surveillance.31 Therefore, it is crucial to identify other new immunotherapy targets. Currently, various drugs targeting the PD-1/PD-L1 pathway in chordomas are undergoing clinical trials.32 33 Pembrolizumab and nivolumab, monoclonal antibodies targeting PD-1, have shown significant efficacy in other areas and have become the most widely used drugs in chordoma ICI therapy.34 Reports have also suggested that the tumor molecule HHLA2 is widely expressed in chordoma tissues and its stromal expression with other immune checkpoint molecules such as PD-L1, B7H3, and Galectin-9 correlates with tumor phenotype, microenvironmental immune levels, and clinical outcomes.35 HHLA2 mediates tumor evasion of host immune attack by inhibiting the proliferation and cytokine production of CD4+ and CD8+ T cells, thus promoting disease progression through binding with TMIGD2.36 37 Notably, PD-L1 exhibits widespread coexpression with HHLA2, while its coexpression with other immune checkpoint molecules is limited, indicating independent biological functions of these proteins in chordomas. This suggests that single therapy using PD-1/PD-L1 inhibitors alone may not achieve satisfactory efficacy in patients, and future combination therapies targeting multiple immune checkpoints and signaling pathways may be a promising strategy for treating this disease.35 Additionally, research has shown that dendritic cells and macrophages highly express PD-L1 regulatory factors in the chordoma microenvironment, suggesting that combination blockade of these factors may provide better therapeutic effects.34 M2-type macrophages infiltrate heavily in chordomas, participating in the formation of the chordoma immune suppressive microenvironment, but whether they serve as prognostic factors and therapeutic targets for chordomas remains unclear.17 Overall, ICI therapy for chordomas is mainly limited to PD-1/PD-L1, with relatively few studies on other targets, and these are limited to preclinical stages. The expression and clinical effects of other immune checkpoint molecules, including VISTA, HVEM, ICOSL, and GITRL, should also be evaluated in chordomas. Additionally, most patients do not benefit from current ICI therapy and may experience immune-related adverse events.38 Therefore, a better understanding of the status and regulatory mechanisms of the chordoma microenvironment is crucial for guiding the safer and more effective use of ICI treatment strategies.

Conclusion

In general, chordomas present complex immune microenvironmental characteristics. While immunotherapy has shown promising results in treating various tumors, research on the immune microenvironment of chordomas is still in its early stages. Therefore, understanding how the immune microenvironment of chordomas influences the outcomes of immunotherapy is crucial. We believe that carefully designed experimental studies in the future will contribute to further understanding of the immune landscape of chordomas.

Data availability statement

Data are available upon reasonable request. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by the Institutional Review Board at The Second Affiliated Hospital of Zhengzhou University, Henan, P.R. China. Reference: No.24-04000. Participants gave informed consent to participate in the study before taking part.

Acknowledgments

HQN and MXZ contributed equally to this work. The data that support the findings of this study are available from the corresponding author BWZ or the first co-author MXZ upon reasonable request. We are grateful for the support of the National Natural Science Foundation of China (82003802 to Dr Tao-Lan Zhang [Department of Pharmacy, The First Affiliated Hospital, Hengyang Medical School, University of South China] and 82002364 to Dr Ming-Xiang Zou [Department of Spine Surgery, The First Affiliated Hospital, Hengyang Medical School, University of South China]), Natural Science Foundation of Hunan Province (2019JJ50542 and 2023JJ50156 to Dr Tao-Lan Zhang [Department of Pharmacy, The First Affiliated Hospital, Hengyang Medical School, University of South China], 2023JJ40596 to Chao Xia [Department of Spine Surgery, The First Affiliated Hospital, Hengyang Medical School, University of South China], 2023JJ40587 to Wei Huang [Department of Epidemiology and Health Statistics, School of Public Health, Central South University / The First Affiliated Hospital, Health Management Center, Hengyang Medical School, University of South China] and 2021JJ40509 to Dr Ming-Xiang Zou [Department of Spine Surgery, The First Affiliated Hospital, Hengyang Medical School, University of South China]), Hunan Provincial Natural Science Foundation for Excellent Youth Scholars (2023JJ20035 to Dr Ming-Xiang Zou [Department of Spine Surgery, The First Affiliated Hospital, Hengyang Medical School, University of South China]), Clinical Research 4310 Program of the First Affiliated Hospital of the University of South China (20224310NHYCG04 to Dr Tao-Lan Zhang [Department of Pharmacy, The First Affiliated Hospital, Hengyang Medical School, University of South China]), Project for Clinical Research of Hunan Provincial Health Commission (20201978 to Dr Tao-Lan Zhang [Department of Pharmacy, The First Affiliated Hospital, Hengyang Medical School, University of South China], 20201962 to Chao Xia [Department of Spine Surgery, The First Affiliated Hospital, Hengyang Medical School, University of South China] and 20201956 to Dr Ming-Xiang Zou [Department of Spine Surgery, The First Affiliated Hospital, Hengyang Medical School, University of South China]), Research Foundation of Education Bureau of Hunan Province (22B0441 to Wei Huang [Department of Epidemiology and Health Statistics, School of Public Health, Central South University / The First Affiliated Hospital, Health Management Center, Hengyang Medical School, University of South China]).

References

Footnotes

  • Collaborators Not applicable.

  • Contributors HQN, BYZ, and BWZ contributed to the conception and design of the study. MXZ and BWZ contributed to drafting and revision of the manuscript. All authors read and approved the final manuscript. BWZ is the guarantor.

  • Funding This work was supported by China Scholarship Council (202106370071 to BWZ), the Science and Technology Innovation Program of Hunan Province (2023RC3172 to MXZ), and the National Natural Science Foundation of China (82002364 to MXZ).

  • Competing interests The authors declare that they have no competing interests

  • Provenance and peer review Not commissioned; externally peer reviewed.