Introduction

Chordoma is a rare clinically aggressive neoplasm that arises from the embryological remnants of the notochord along the spinal axis [1,2,3]. Although chordomas tend to grow slowly, patients develop severe symptoms such as cranial nerve palsies and brain stem compression, that make successful surgical treatment difficult, and become resistant to radio-chemotherapy. Recently, irradiation with high-energy particles, such as proton beams (Bragg peak), has proven effective for chordomas [3,4,5,6]. Therefore, palliative debulking followed by radiation therapy is the standard treatment. However, patients experience multiple local recurrences, and > 90% of patients die within 10 years [7].

Some studies have suggested clinical prognostic factors for chordomas. Although larger tumor size, non-total resection, older patient age, Ki-67 > 3% were generally used as adverse prognostic factors [8,9,10,11,12], these parameters did not lead to the development of a new treatment strategy. A better understanding of the molecular and genetic basis of chordoma pathogenesis is required to predict the prognosis and develop new targeted therapies.

Vascular endothelial growth factor (VEGF) is a potent activator of angiogenesis that is known to correlate with the progressive course of central nervous tumors [13, 14]. Although limited reliable data are available, some reports have demonstrated that VEGF plays a role in the development and recurrence of chordomas [15, 16]. Therefore, anti-angiogenic therapy, using bevacizumab and sorafinib, has shown efficacy in regulating the growth of chordomas [17, 18].

Tumor-immune microenvironment is also critical for tumor growth [19]. Tumor-associated macrophages (TAMs) and regulatory T cells (Tregs) are major components of immune-suppressive cells [19]. The immune microenvironment is strongly associated with VEGF-A exposure. VEGF-A suppresses antitumor immunity by inhibiting the maturation of dendritic cells and stimulating the proliferation of Tregs [20, 21]. Furthermore, vascular structure is known to be changed following VEGF-A exposure, leading to a leak of the immune-competent cells into the tumor—microenvironment through damaged blood vessels with increased permeability [22]. Recently, some reports have demonstrated that the clinical course of chordomas might be associated with immune-microenvironment [23]. Zou et al. demonstrated that the density of programmed cell death-1 (PD-1)+ tumor infiltrating lymphocytes (TILs), CD4+ TILs, and CD3+ TILs was significantly associated with local recurrence-free survival and overall survival. In addition, the density of Foxp3+ TILs and CD8+ TILs correlated with overall survival [23]. Immunotherapy was regarded as a potential treatment for chordomas [19, 24, 25]. In a phase II clinical trial, adult patients with unresectable localized chordoma who are scheduled to undergo radiation therapy received the yeast-Brachyury vaccine targeting Brachyury, a transcription factor of the T-box family that is typically expressed in notochord and chordoma [26].

In the present study, molecular and histopathological characterizations regarding angiogenic factors, vascular structure, and tumor-immune microenvironment were examined for skull base chordomas. Novel biomarkers strongly associated with poor prognosis were identified that helped achieve an understanding about the clinical course and develop effective treatment strategies.

Materials and methods

This research was approved by the Institutional Review Board at our institutes (Reference number: 20,050,002). Written consent was obtained from all patients.

Study population

In the present study, 27 primary skull base chordomas and 27 recurrent ones (a total of 54 tumors from 42 patients) from December 1999 to March 2018 were analyzed. Twenty-four out of the 54 tumors were paired primary and recurrent tumors obtained from 12 patients. Other 15 primary tumors and 15 recurrent tumors (not paired samples) were also included in the study. Of the 27 recurrent tumors, only three were radiation-naïve, recurrent tumors. All the patients underwent neurosurgical procedures for mass reduction or diagnostic biopsy.

Immunohistochemical analysis

Histopathological analyses were performed on 3-μm sections of formalin-fixed, paraffin-embedded tissue of 54 skull base chordomas (27 primary and 27 recurrent tumors). Histological characteristics were assessed using hematoxylin and eosin (H&E) staining. All the samples were diagnosed by a neuropathologist.

Standard immunohistochemistry was performed. Sections were deparaffinized and rehydrated. The antigen retrieval was performed in citrate buffer (Ki-67, VEGFR1, CD34, Foxp3, CD163, CD3, CD4, and PD-1), or in Tris buffer (pH 9 for VEGF-A, VEGFR2, CD8, and programmed cell death ligand-1 [PD-L1]) using microwave irradiation or autoclave (PD-L1). The sections were blocked for 60 min in 2.5% horse serum (ImmPRESSTM Detection Systems, Vectorlabs). The sections were incubated overnight at 4 °C with anti-Ki-67 antibody (1:200, M7249, DAKO), anti-VEGF-A antibody (1:200, JH121, Merck Millipore), anti-VEGFR1 antibody (1:200, AF321, R&D SYSTEMS), anti-VEGFR2 antibody (1:600, 55B11, Cell Signaling Technology), anti-CD34 antibody (1:100, F1604, Nichirei Biosciences Inc.), anti-Foxp3 antibody (1:100, ab54501, Abcam), anti-CD163 antibody (1:100, ab87099, Abcam), anti-CD3 antibody (1:100, ab5690, Abcam), anti-CD4 antibody (1:200, 1F6, Nichirei Bioscience Inc.), anti-CD8 antibody (1:50, ab17147, Abcam), anti-PD-1 antibody (1:50, NAT105, Abcam), and anti-PD-L1 antibody (1:500, 28-8, Abcam); thereafter, they were incubated with anti-mouse, rabbit, or goat Ig secondary antibody (ImmPRESSTM Detection Systems, Vectorlabs) for 60 min at room temperature. The products were visualized with peroxidase-diaminobenzidine reaction.

The evaluation of immunohistochemistry was assessed as previously described [27,28,29]. The Ki-67 proliferation index was defined as the percentage of cells with positive nuclear Ki-67 immunostaining at high-power fields (HPF: × 20, 0.95 mm2). The positivity of VEGF-A staining in the tumor cytoplasm or stroma was assessed as the following: ++ , diffuse intense staining; + , diffuse faint staining; and − , negative staining. VEGFR1 and VEGFR2 were expressed on vascular endothelial cell and tumor cell. The expressions of VEGFR1 and VEGFR2 in the tumor cells were assessed as percentage of positive tumor cell analyzed from three areas. For the assessment of microvessel density (MVD), the tissue sections were screened in low-power fields (× 4), and the three most vascularized regions (hot spots) were selected for each region. The counting of microvessels was performed on these regions at high-power fields (HPF: × 20, 0.95 mm2). For the assessment of density of Foxp3, CD163, CD3, CD4, CD8, and PD-1 (+) cells, the tissue sections were screened using each immunohistochemistry at the low-power fields (× 4), and three hot spots were selected. The positive cells were counted in these areas at HPF (× 40, 0.47 mm2). The expression of PD-L1 in tumor cells was assessed as percentage of positive tumor cells analyzed from three areas. All the expressions were determined based on the consensus of four authors with blinded clinical information (YM, RT, KK, and KO).

Immunofluorescence analysis

Immunofluorescence triple staining for CD34, VEGFR1, and Brachyury; CD34, VEGFR2 and Brachyury; or VEGFR1, VEGFR2, and PDGFR-β expression was performed for all sections. Tissue sections were incubated with antibodies against CD34 (1:100 or 1:250, NU-4A1 or sc-18917, mouse polyclonal IgG or rat polyclonal IgG, Abcam or Santa Cruz Biotechnology), Brachyury (1:100, H-210, rabbit polyclonal IgG, Santa Cruz Biotechnology), VEGFR1 (1:100, AF321, goat polyclonal IgG, R&D Systems), VEGFR2 (1:100, EIC, mouse monoclonal IgG, Abcam), and PDGFR-β (1:50, Y92, rabbit monoclonal IgG, Abcam) overnight at 4 °C. Appropriate secondary antibodies, Alexa Fluor 488 donkey anti-mouse IgG, Alexa Fluor 488 donkey anti-rat IgG, Alexa Fluor 568 donkey anti-mouse IgG, Alexa Fluor 568 donkey anti-goat IgG, or Alexa Fluor 647 donkey anti-rabbit IgG were used (1:200). Finally, the sections were mounted with mounting medium for fluorescence with 4′,6′-diamidino-2-phenylindole (DAPI; Santa Cruz Biotechnology). Cell images were captured with fluorescence microscopy (Biorevo BZ-9000, KEYENCE).

Cell culture and immunocytochemistry

The human chordoma cell line JHC7 was obtained from the American Type Culture Collection. JHC7 cells were cultured in T-75 cell culture plastic flasks (Thermo Fisher Scientific) with DMEM (Sigma-Aldrich) containing 10% fetal bovine serum (FBS), 2 mM l-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin. After four days, the medium was removed, and fresh media was added. When the cells were 90% confluent, the culture medium was removed, and 3 mL of 0.25% trypsin–EDTA solution (Thermo Fisher Scientific,) was added to the flask. The cells were resuspended in 10 mL fresh media before diluting into new culture flasks.

Immunocytochemistry was performed to clearly detect VEGFR1 and VEGFR2 expressions on the tumor cells. Chordoma cell line JHC7 was plated on 12-mm cover slips coated with poly-l-lysine-coated chamber slides. The cultured cells were fixed with 4% PFA and permeabilized with 0.5% Triton-X. The cells were stained with the following primary antibodies against Brachyury (1:100, H-210, rabbit polyclonal IgG, Santa Cruz Biotechnology), VEGFR1 (1:100, AF321, goat polyclonal IgG, R&D Systems), VEGFR2 (1:100, EIC, mouse monoclonal IgG, Abcam) overnight at 4 °C. Appropriate secondary antibodies, Alexa Fluor 488 donkey anti-rabbit IgG, Alexa Fluor 568 donkey anti-goat IgG, or Alexa Fluor 647 donkey anti-mouse IgG were used (1:200). Finally, the sections were mounted with mounting medium for fluorescence with 4′,6′-diamidino-2-phenylindole (DAPI; Santa Cruz Biotechnology, Texas, USA). Cell images were captured using fluorescence microscopy (Biorevo BZ-9000, KEYENCE).

The specificity of all procedures was checked using negative controls. For negative controls, sections and cells were incubated with non-immune mouse, rabbit, rat and goat IgG at the same concentration used for each antibody.

Radiographical analysis

The gadolinium (Gd)-enhanced T1-weighted magnetic resonance imaging (MRI) or T2-weighted image was used to evaluate tumor growth. The T2-weighted image was used to evaluate the tumor size in one patient because the patient exhibited an allergic reaction to Gd. The tumor growth rate for one year was quantitatively analyzed using the longest diameter [16, 23, 30, 31]. The growth rate was evaluated starting three months after the initial surgery because tumor sizes tended to change early after surgery. We deemed a growth ≧ 3 mm/year to indicate a progressive course. Slow progressive course (growth rate < 3 mm/year) was defined as SP and rapid progressive course (growth rate ≥ 3 mm/year) as RP. The use of 3 mm/year as a growth rate criterion was determined from the median growth speed in the present study, because no previous studies determined a quantitative growth rate for chordomas. Progression-free survival (PFS) was defined as the time that elapsed between treatment initiation and tumor progression.

Statistical analysis

Chi-squared test was used to evaluate the immunohistopathological expressions of patient sex, and distribution of VEGF-A. Student’s t-test was used to compare the recurrent rate; Ki-67 index; the number of CD3, CD4, CD8, CD34, CD163, or Foxp3 positive cells; and percentage of VEGFR1, VEGFR2 and PD-L1 positive tumor cell. Prognostic predictors were examined for tumor growth rate using a multiple linear regression analysis. Receiver operating characteristic (ROC) curves for postoperative recurrence were used to determine the cut-off points for VEGFR1 and VEGFR2 expressions on tumor cells. To evaluate the association with PFS, cut-offs greater than 46% for VEGFR1 and greater than 22% for VEGFR2 were used to include the highest true positive rates and the lowest false positive rates. Analyses were performed using IBM SPSS statistics (IBM Corp.). A p-value < 0.05 was considered statistically significant.

Results

Patients characteristics

Table 1 summarizes patients characteristics; 12 patients (male: 7, female: 5) in the SP group and 15 patients (male: 5, female: 10) in the RP group were analyzed. The age range of the study population was 17 to 79 y; the mean age of those in the SP group was 46.17 ± 21.6 y, while that of those in the RP group was 44.27 ± 20.2 y. There was no significant difference in general characteristics of the patients of the two groups. There was no significant difference in the Ki-67 index of the groups (SP: 4.23%, RP: 5.28%, p = 0.29).

Table 1 Patient characteristics
Full size table

VEGFR1 and VEGFR2 expressions on tumor cells

VEGFR1 and VEGFR2 expressions were detected on human chordoma tumor cells (JHC7) expressing Brachyury. Positive membranous staining of VEGFR1 and VEGFR2, with some faint cytoplasmic staining, was detected on tumor cells. VEGFR1 was more strongly stained on the tumor cells than VEGFR2 (Fig. 1a). Similarly, VEGFR1 and VEGFR2 expressions were clearly detected on endothelial cells with CD34 staining as well as tumor cells in human chordoma specimens (Fig. 1b). Vasculatures stained with VEGFR1 and VEGFR2 exhibited slight or negative PDGFR-β expression of a pericyte marker (Fig. 1b).

Fig. 1
figure 1

VEGFR1 and VEGFR2 expressions on tumor and vascular endothelial cells in chordoma. a Immunofluorescence triple staining for Brachyury, VEGFR1, and VEGFR2 on a human chordoma cell line (JHC7) (original magnification, × 20; magnification bar, 100 μm) (NC, negative control). b Immunofluorescence triple staining for CD34, VEGFR1, and Brachyury; CD34, VEGFR2 and Brachyury; or VEGFR1, VEGFR2 and PDGFR-β on vascular endothelial cells and tumor cells in human chordoma tissue specimens (original magnification, × 20; magnification bar, 100 μm) (V, vascular endothelial cells)

Full size image

The relationship between the percentage of VEGFR1- or VEGFR2-positive tumor cells and tumor growth (mm/year) was evaluated. Patients with a high percentage of VEGFR1- or VEGFR2-positive tumor cells showed higher tumor growth and a linear relationship (VEGFR1, p = 0.00029; VEGFR2, p = 0.0014) (Fig. 2a). PFS of the patients with a higher percentage of VEGFR1- or VEGFR2-positive tumor cells were significantly shorter than that in those with a lower percentage of positive tumor cells (VEGFR1: p = 3.21E-05, VEGFR2: p = 0.037) (Fig. 2b). The percentage of VEGFR1-positive tumor cells in recurrent cases was significantly higher than that in primary cases (33% vs. 53%, p = 0.0053) (Fig. 2c). The percentage of VEGFR2-positive tumor cells in the recurrent cases tended to be higher than that in the primary cases (26% vs. 37%, p = 0.107) (Fig. 2c). These findings were confirmed by comparing the paired initial and recurrent tumors from 12 patients.

Fig. 2
figure 2

Relationship between VEGFR1 or VEGFR2 expression on tumor cells and the clinical course in chordoma patients. a Relationship between the growth rate (mm/year) and percentage of VEGFR1- or VEGFR2-positive tumor cells. b Relationship between progression-free survival (PFS) and percentage of VEGFR1- or VEGFR2-positive tumor cells. Patients with a high percentage of VEGFR1-positive tumor cells ( ≥ 46%) or VEGFR2-positive tumor cells ( ≥ 22%) show significantly shorter PFS. c Percentage of VEGFR1-positive tumor cells in recurrent cases is significantly higher than that in primary cases (33% vs 53%, p = 0.0053). Percentage of VEGFR2-positive tumor cells in recurrent cases tends to be higher than that in primary cases (26% vs 37%, p = 0.107)

Full size image

The percentage of VEGFR1-positive tumor cells was significantly higher in the RP group than in the SP group (60.11 vs 22.61%, p < 0.01) (Fig. 3a, b). The percentage of VEGFR2-positive tumor cells was also significantly higher in the RP group than in the SP group (45.45 vs 16.41%, p = 0.0067). Moreover, VEGF-A expression and MVD were also significantly higher in the RP group than in the SP group (VEGF-A: p = 0.03; MVD, 8.78 vs 15.81 /3 HPF: p = 0.013) (Fig. 3a, b).

Fig. 3
figure 3

Analyses of VEGF-A, VEGFR1, and VEGFR2 expressions on tumor cells. a VEGFR1, VEGFR2, and VEGF-A expression on tumor cells and CD34 expression in the SP and RP groups (SP, patient with growth rate < 3 mm/year; RP, patients with growth rate ≥ 3 mm/year, original magnification, × 40; magnification bar, 50 μm). b Percentage of VEGFR1- or VEGFR2-positive tumor cells is significantly higher in the RP group than that in the SP group. VEGF-A expression and MVD are also significantly higher in the RP group than in the SP group

Full size image

Tumor-infiltrating immune cells

The number of CD3+ T cells was higher in the RP group than in the SP group (10.0 vs 4.97/3 HPF, p = 0.034) (Fig. 4a, c). The number of CD4 +, CD8+ and PD-1+ cells were not significantly difference between the RP and SP groups (CD4, 14.27 vs. 10.43/3HPF: p = 0.28; CD8, 9.96 vs. 12.72/3HPF: p = 0.30; PD-1,4.38 vs. 1.72/3HPF: p = 0.14) (4a–c). The percentage of PD-L1-positive tumor cells was higher in the RP group than in the SP group (38.47 vs 7.67%, p = 0.00064) (Fig. 4b, c). Immunosuppressive cells were also analyzed, and the numbers of Foxp3+ and CD163+ cells was significantly higher in the RP group than in the SP group (Fig. 5a, b).

Fig. 4
figure 4

Analyses of tumor-infiltrating lymphocytes and immune-checkpoint molecules. a Immunohistochemistry of CD3, CD4, CD8, PD-1, and PD-L1 expressions in the SP and RP groups (original magnification, × 40; magnification bar, 50 μm.). b The number of CD3+ T cells is higher in the RP group than in the SP group (10.0 vs 4.97/3 HPF, p = 0.034). No significant difference is present in the number of CD4+ , CD8+ , and PD-1+ cells between the RP and SP groups (CD4, 14.27 vs. 10.43/3HPF: p = 0.28; CD8, 9.96 vs. 12.72/3HPF: p = 0.30; PD-1,4.38 vs. 1.72 /3HPF: p = 0.14). Percentage of PD-L1-positive tumor cells is higher in the RP group than in the SP group (38.47 vs 7.67%, p = 0.00064)

Full size image
Fig. 5
figure 5

Analyses of immunosuppressive cells and tumor growth. a Immunohistochemistry of Foxp3 and CD163 in the SP and RP groups (original magnification, × 40; magnification bar, 50 μm). b The numbers of Foxp3+ and CD163+ cells are higher in the RP group than that in the SP group

Full size image

Multiple linear regression analysis revealed significant relationships between the tumor growth rate (mm/year) and VEGFR1, VEGFR2, or CD163 expression (Table 2).

Table 2 Multivariate regression analyses for tumor growth rate (mm/year)
Full size table

There were no statistically significant differences in any parameters regarding angiogenesis or immune-related markers between the radiation-naïve tumors and the post-radiation tumors.

Discussion

VEGF promotes endothelial cell proliferation, and tumor growth. VEGF expression correlated with the prognosis in central nervous tumors, such as malignant gliomas [32]. For example, VEGF-A is strongly expressed in glioblastomas, with the expression being approximately 10 times higher than that in grade II gliomas [33]. Some reports demonstrated that VEGF plays a role in the development, and recurrence of chordoma through upregulated HIF-1α expression under hypoxia [15, 16, 18, 34]. HIF-1α directly regulates the VEGF expression [35]. In the present study, VEGF-A and MVD were associated with the growth rate of chordomas. Angiogenesis played a key role in chordoma growth, which was consistent with the previous study.

VEGF-VEGFRs signaling plays an important role in the development of notochord, which may be related to high expression of chordoma cells [36, 37]. This study showed that both VEGFR1 and VEGFR2 were co-expressed on tumor cells and vascular endothelial cells. It has been reported that not only VEGFR1, but also VEGFR2 was expressed on tumor cells and vascular endothelial cells in various types of tumors [38,39,40,41,42]. VEGFR1 and VEGFR2 expressions on tumor cells were observed in 25-50% of the patients with lung, breast, cervix, larynx, pancreas, and ovarian cancers, as well as melanoma [39,40,41,42]. Further, the expression of VEGFR1 or VEGFR2 on tumor cells was different in various types of tumors. In chordoma, the percentage of VEGFR1- or VEGFR2-positive tumor cells was higher than that in other malignant tumors, suggesting an autocrine regulatory process in tumor growth [41,42,43,44,45,46]. In particular, VEGFR2 is known to play an important role in tumor stem cells proliferation. Blocking VEGFR2 reduced the size of the tumor stem cell pool and stem cell self-renewal potential [47]. In the present study, VEGFR1 and VEGFR2 expression on tumor cells was strongly associated to chordoma growth.

Furthermore, tumor-immune microenvironment was related to the progression of chordoma in the present study. Tumor-infiltrating immune cells, including Tregs, TAMs, and cytotoxic T lymphocytes, are known to be critical determinants of cancer outcome. VEGF-A promotes the immunosuppressive microenvironment [48]. Thus far, few studies have examined the relationship between chordomas and immune microenvironment [49]. Recently, Zou et al. [23, 31] demonstrated that immune cell infiltrates, such as Tregs, were observed in human chordoma specimens, which was associated with poor prognosis. The numbers of Tregs and TAMs were increased in the RP group compared with the SP group in the present study. The role of TAMs, known as immunoregulatory (M2) macrophages [50], has never examined for chordomas. Polarization of TAMs towards an M1 or M2 phenotype depends on the cytokine milieu and the local microenvironment [51]. Arginase I, interleukin (IL)-10, transforming growth factor (TGF)-β and prostaglandin E2 (PGE2) are produced by TAMs, resulting in the suppression of cytotoxic T cell [51]. Therefore, TAMs promote tumor growth, and correlate to tumor malignancy [51]. VEGF-A promotes the migration and differentiation of TAMs from immature myeloid cell [52], which was consistent with the present results.

Tregs also have been shown to play an important role in the regulation of immune responses via suppression of proliferation of other T cells present in the tumor—microenvironment [53]. Tregs produce immunosuppressive cytokines such as TGF-β, IL-10, and IL-35. Tregs induce the apoptosis of cytotoxic T cells by the action of granzyme B. Consuming IL-2 by Tregs inhibits effector T cell proliferation, resulting in impaired immune response [54]. Upregulation of VEGF-A promotes the recruitment of Tregs to the tumor microenvironment via over-expression of chemokine–chemokine ligand 28 (CCL28) by tumor cells and induces VEGFR-2-expressing Tregs [55]. Zou et al. [31] demonstrated CD8+ /Foxp3+ ratio to be an independent prognostic factor for overall survival of chordomas. This study also suggests that the presence of Tregs provides important prognostic information for chordomas. VEGF-A exposure increased the permeability of blood vessels with upregulated VEGFR1 and VEGFR2 expressions [22, 29, 56]. Immunocompetent cells can easily leak through the damaged vascular structures in the tumor—microenvironment [29, 56]. Permeability of blood vessels in chordoma was suggested to be high because high expressions of VEGF-A, VEGFR1, and VEGFR2 without strict coverage of PDGFR-β expressing pericytes were observed in tumor specimens.

The present results suggest that anti-angiogenic therapy, such as VEGF-targeting drugs (bevacizumab) can be effective for chordoma. In fact, bevacizumab induced durable stabilization for chordoma patients [57]. Furthermore, VEGFRs-targeting drugs (sorafenib, sunitinib and VEGFRs epitope peptide derived vaccine) may be more effective because chordoma cells expressed VEGFR1 and VEGFR2, and CD4+ CD25+ Foxp3+ Tregs that are known to express VEGFR2 [58, 59]. Sorafenib was used for advanced chordoma in phase II trial, leading to prolonged PFS [18]. However, we must pay attention for high PD-L1 expression in the RP group of chordoma. Previous studies have shown that high PD-L1 expression correlated with worse OS in patients with some central nervous tumors [60]. The relationship between PD-L1 expression and prognosis was controversial for chordoma [61]. Zou et al. [62] showed that the expression of PD-L1 is associated with favorable prognosis of spinal chordoma. In the present study, high PD-L1 expression was shown in the RP group of chordomas, which was related to poor prognosis. The expression of PD-L1 is induced by various proinflammatory molecules, such as interferon (IFN)-γ and VEGF-A, in the tumor—microenvironment [63,64,65]. Therefore, combinational therapy using anti-angiogenic therapy and checkpoint inhibitor can be more effective.

The present study evaluated VEGFR1 and VEGFR2 expression on tumor cells and found the expressions to be significantly correlated with an aggressive clinical course of chordomas. The percentage of VEGFR1- and VEGFR2-positive tumor cells might be helpful for predicting prognosis and for developing new targeted therapies. Furthermore, immunosuppressive cells, such as Tregs and TAMs, and immune checkpoint molecules were also associated with tumor growth. These tumor-microenvironments must be comprehensively regulated for the treatment of chordomas.

The prevalence of chordoma was reported as 0.5–0.8 per 1,000,000 [6, 66]. Reports on clival chordoma are more valuable because most examinations are performed using spinal chordomas [6, 66]. Limitation of the present study was the small number of paired primary and recurrent samples from the same patients. Some patients were initially operated at other hospitals, and came to our hospital at the time of recurrence. Ideally, studies using a larger number of paired samples should be conducted to confirm the findings of this study. Furthermore, the biology of chordomas has not been fully elucidated. To improve the patient clinical course, the tumor biology, including bone destruction, needs to be analyzed. For those analysis, in vitro and in vivo analyses using human chordoma cells are indispensable. Further analyses should be conducted using human chordoma cells to evaluate the relationship between VEGFR1 and VEGFR2 expression on tumor cells and the tumor biology, including bone destruction.

Conclusions

VEGFR1 and VEGFR2 that were strongly expressed on chordoma tumor cells. The percentage of VEGFR1- and VEGFR2-positive tumor cells significantly correlated with an aggressive clinical course. In addition, increased numbers of immunosuppressive cells, including that of Tregs and TAMs, and immune checkpoint molecule were also associated with the growth of chordomas.