Levels of peripheral immune blood cells are related to the grade of isocitrate dehydrogenase-mutant oligodendroglioma

Jing Cheng; Yanqin Fan; Gang Deng; Baohui Liu; Junmin Wang; Qianxue Chen

doi:10.4103/glioma.glioma_20_19

ORIGINAL ARTICLE

Year : 2019 | Volume : 2 | Issue : 4 | Page : 174-181

Levels of peripheral immune blood cells are related to the grade of isocitrate dehydrogenase-mutant oligodendroglioma

Jing Cheng¹, Yanqin Fan², Gang Deng¹, Baohui Liu¹, Junmin Wang¹, Qianxue Chen¹
¹ Department of Neurosurgery, Renmin Hospital of Wuhan University, Wuhan, Hubei Province, China
² Division of Nephrology, Renmin Hospital of Wuhan University, Wuhan, Hubei Province, China

Date of Submission	26-Oct-2019
Date of Acceptance	26-Nov-2019
Date of Web Publication	23-Jan-2020

Correspondence Address:
Prof. Qianxue Chen
Department of Neurosurgery, Renmin Hospital of Wuhan University, No. 99 Zhang Zhidong Road, Wuhan 430060, Hubei Province
China

Source of Support: This study was supported by the National Natural Science Foundation of China (No. 81572489, to QC)., Conflict of Interest: None

DOI: 10.4103/glioma.glioma_20_19

Abstract

Background and Aim: The immune response to glioma is significantly impaired because of isocitrate dehydrogenase (IDH) mutations. However, the immune reaction to glioma is poorly understood. Materials and Methods: We selected 38 patients with IDH-mutant oligodendroglioma and divided them into low-grade and high-grade groups. Forty healthy people were selected as a control group. Blood samples were collected from the control group and from glioma group patients on the day before surgery and at 3 and 7 days after surgery, and numbers of immune cells were determined. This study was approved by the Institutional Ethics Committee of the Faculty of Medicine at Renmin Hospital of Wuhan University, China (approval No. 2018K-C017) on June 4, 2018. Results: The percentages of CD3+, CD4+, CD4+/CD8+, and CD3− CD19+ B-lymphocytes, and of CD3− CD16+ CD56+ natural killer cells were significantly lower (P < 0.05), and the percentage of CD4+ CD25+ regulatory cells was significantly increased (P < 0.05) in the glioma group compared with the control group. IDH-mutant oligodendroglioma patients with a higher grade of malignancy had lower levels of immune cells preoperatively and postoperatively (P < 0.05), and the levels of immune cells increased following surgery (P < 0.05). Conclusions: IDH-mutant oligodendroglioma patients with high-grade malignancy have a lower number of immune cells in peripheral blood compared with patients with low-grade malignancy. This finding can be used as an effective indicator to evaluate the malignancy and prognosis of IDH-mutant oligodendroglioma and provides a new avenue for the immunotherapy of gliomas.

Keywords: Immune cells, inflammatory cytokines, malignancy, oligodendroglioma, surgery

How to cite this article:
Cheng J, Fan Y, Deng G, Liu B, Wang J, Chen Q. Levels of peripheral immune blood cells are related to the grade of isocitrate dehydrogenase-mutant oligodendroglioma. Glioma 2019;2:174-81

How to cite this URL:
Cheng J, Fan Y, Deng G, Liu B, Wang J, Chen Q. Levels of peripheral immune blood cells are related to the grade of isocitrate dehydrogenase-mutant oligodendroglioma. Glioma [serial online] 2019 [cited 2023 Oct 2];2:174-81. Available from: http://www.jglioma.com/text.asp?2019/2/4/174/276695

Introduction

Gliomas are the most malignant brain tumors,^[1],[2],[3] and they account for more than 40% of brain tumor incidence. Compared with other tumors, patients with glioma have a poor prognosis and high recurrence rate.^[4],[5] The vast majority of gliomas are anaplastic gliomas or glioblastoma and survival times are 12–15 months and 2–5 years, respectively.^[6],[7] Therefore, effective prevention and treatment of glioma is urgently sought. One of the most significant changes to the 2016 World Health Organization (WHO) classification of gliomas was including genetic characteristics for the diagnosis of diffuse gliomas. The most significant change in the new guidelines was the classification of all diffuse infiltrating gliomas, whether astrocytoma or oligodendroglioma, according to isocitrate dehydrogenase (IDH) gene mutation status.^[8] In the new classification system gliomas with similar prognosis and treatment options are grouped together, making the diagnosis and treatment of gliomas more precise. A recent study found that patients with IDH-mutant glioma associated with the CpG island methylator phenotype have a longer survival time than patients with wild-type IDH glioma. CpG islands are always in an unmethylated state in normal human tissues, and CpG island methylator phenotype is an indicator for tumor specificity. A possible reason for this is that the immune response to glioma is significantly impaired because of the IDH mutation.^[9] Other recent studies have also reported that the incidence and prognosis of glioma is associated with immune dysfunction;^[10] however, this is still poorly understood.

Materials and Methods

Patients

Thirty-eight patients who underwent surgical resection of glioma between February 2012 and February 2018 were selected as the study participants. They were 20 males and 18 females. To exclude age-related effects of the immune system, we selected middle-aged patients 32–55 years old (average age 41.1 ± 2.7 years). All selected patients were confirmed as IDH1- and IDH2-mutant oligodendroglioma patients by imaging, pathology, and genotyping. Patients with abnormal liver or kidney function, an expected survival period of 3 months or less or with serious diseases or blood coagulation disorders were excluded from the study. The 38 IDH-mutant oligodendroglioma patients included WHO Grade II and Grade III oligodendrogliomas and were divided into a low-grade group (15 Grade II) and a high-grade group (23 Grade III) according to the 2016 WHO grading standard.^[11] All glioma patients underwent total surgical resection. The range of resection was evaluated in strict accordance with preoperative and postoperative contrast-enhanced magnetic resonance imaging. None of the patients underwent preoperative radiotherapy or chemotherapy before the operation. None of the patients we selected had common complications such as lung infections and wound infections. Routine treatments, such as intravenous nutrition and anti-infection medication, were administered after the operation. In the same period, 40 healthy people who were examined in our hospital were selected as the control group. They included 22 males and 18 females and their ages ranged from 32 to 55 years and averaged 41.4 ± 7.3 years. All procedures involving human participants were performed in accordance with the ethical standards of the Institutional Ethics Committee of the Faculty of Medicine at Renmin Hospital of Wuhan University, China (approval No. 2018K-C017 on June 4, 2018) and with the 1964 Helsinki Declaration and its later amendments. Written informed consent was obtained from all participants. [Table 1] summarizes the demographics and clinical characteristics of control and glioma groups.

Table 1: Demographics and clinical characteristics of control and glioma groups

Click here to view

Managing patients

The tumor site was the temporal lobe in 20 cases and the frontal lobe in 18 cases. All patients underwent surgical treatment for the first time. Surgery was performed using the pterional approach. After the surgery, tumor sections were prepared for histological examination and the degree of tumor malignancy was divided into Grade II and III according to 2016 WHO^[11] guidelines for the histology and genotype of glioma. The genotype of all patients was IDH-mutant. [Figure 1] shows the clinical data of one of the patients. The WHO Grade II patients were designated the low-grade group, and the WHO Grade III patients were designated the high-grade group. [Table 2] summarizes the 38 patients with IDH-mutant oligodendroglioma.

Figure 1: Male, 39-year-old, right temporal lobe anaplastic oligodendroglioma Grade III, IDH-mutant, 1p/19q combined deletion. (A and C) Preoperative (A) and postoperative (C) magnetic resonance imaging. The white arrow in A shows the lesion site of the glioma. The white arrow in C shows no significant residue after glioma resection. (B) Intraoperative observation. The white arrow shows the glioma observed during surgery. (D) Pathological examination of tumor tissue under light microscopy. The nuclei are round or ovoid and dark stained. The bud-like blood vessels result in the segmentation of tumor cells into the typical oligodendroglioma morphology. Scale bar: 100 μm

Click here to view

Table 2: Demographics and clinical characteristics of the 38 patients with IDH-mutant oligodendroglioma

Click here to view

Analysis of leukocyte subsets by flow cytometry

Immunofluorescence staining

Venous blood (2 mL) was collected 1 day before surgery and 3 and 7 days after surgery from the glioma group and at equivalent time points from the control group. Six fluorescence-labeled antibodies from SET IMK Plus kit and CD4+ CD25+ Treg kit (CD45-FITC/CD14-PE; IgG1-FITC/IgG2a-PE; CD3-FITC/CD19-PE; CD3-FITC/CD16+56-PE; CD4-FITC/CD8-PE; CD4-FITC/CD25-PE;) for six leukocyte markers were individually mixed with 20 μL of Simul SET IMK Plus kit reagent (Becton Dickinson, Franklin Lakes, NJ, USA) and 100 μL of whole blood. Samples were then incubated for 20 min without light at room temperature. Two milliliters of red blood cell lysate was then added to each sample, mixed and incubated for 10 min at room temperature in the dark. Samples were then centrifuged for 10 min (156.5 ×g). The supernatant was discarded, the cells washed in phosphate buffer, and then resuspended in 300 μL of phosphate buffer saline.

Flow cytometry analysis

FACSCalibur tricolor calibration microspheres (Becton Dickinson) were initially used to calibrate the flow cytometer. After starting up, we first added one drop of quality control microsphere into 1 mL of sheath fluid (phosphate buffer saline). The upper sample is tested after calibration. Cells were then sorted at an excitation wavelength of 488 nm using Simulset software (Becton Dickinson and Company, Franklin Lakes, NJ, USA) and the percentage of cells in each leukocyte subset was determined.

Analysis of CD4+ CD25+ Treg cells

One hundred microliters of whole blood was added to one tube containing 20 μL each of IgG1-FITC and IgG2a-PE (Beckman Coulter, Pasadena, CA, USA) and to another tube containing 20 μL each of CD4-FITC and CD25-PE (Beckman Coulter). Samples were mixed and incubated at room temperature in the dark for 20 min. Then, 2 mL of red blood cell lysate was added to each sample, and incubation continued at room temperature in the dark for 10 min to eliminate interference from red blood cells. Samples were centrifuged for 10 min (156.5 ×g) and the supernatant discarded. Cells were washed with phosphate buffer saline and resuspended in 500 μL phosphate buffer saline. Cells were then analyzed by flow cytometry.

Statistical analysis

SPSS 20 statistical software (IBM, Armonk, NY, USA) was used for statistical analysis. Data are expressed as the mean ± standard deviation. Student's t-test was used for comparison between two groups. Fisher's exact test was used to test population-specific differences in patient demographics. P < 0.05 was considered statistically significant.

Results

IDH- mutant oligodendroglioma patients have lower levels of immune cells compared with healthy subjects

For all patients, we performed detailed immunological monitoring before surgery. As shown in [Figure 2], the percentages of peripheral blood CD3+ (57.2%), CD4+ (36.4%), CD4+/CD8+ (0.9%), and CD3− CD19+ B-lymphocytes (12.2%), and of CD3− CD16+ CD56+ natural killer (NK) cells (16.8%) were significantly reduced in the preoperative glioma group compared with the control group (P < 0.05). These findings indicate that the immune system of IDH-mutant oligodendrogliomas patients was suppressed before surgery. CD4+ CD25+ Treg cells are a group of lymphocytes that negatively regulate immune responses. The percentage of CD4+ CD25+ Treg cells (7.9%) was significantly higher in the preoperative glioma group than in the control group (P < 0.05). This indicated that IDH-mutant oligodendroglioma patients were immunosuppressed before the surgery.

Figure 2: Preoperative immune cell subsets in patients with glioma and healthy subjects. Peripheral blood immune cells were collected from healthy people and glioma patients before surgery. The percentage of cells in immune cell subsets was calculated by flow cytometry. (A) Lymphocyte subsets: CD3+, CD4+, and CD4+/CD8+. (B) Immune cells: CD3− CD19+ B-lymphocytes, CD3− CD16+ CD56+ natural killer cells, and CD4+ CD25+ regulatory (Treg) cells. Data are expressed as the mean ± standard deviation (control group n = 40, glioma group n = 38). *P < 0.05, **P < 0.001, vs. control group;^##P < 0.001, vs. CD3+;^&P < 0.05,^&&P < 0.001, vs. CD3-CD19+ B-lymphocytes (Student's t-test)

Click here to view

Grade III IDH - mutant oligodendroglioma patients have lower levels of immune cells before surgery compared with Grade II patients

We then tested the relationship between peripheral blood immune cell populations and IDH-mutant oligodendroglioma grade. Glioma grades were pathologically determined after tumor resection as Grade II (low grade) or Grade III (high-grade) according to the 2016 WHO classification. As shown in [Figure 3], the percentages of CD3+ (54.3%), CD4+ (33.4%), CD4+/CD8+ (0.7%), and CD3− CD19+ B-lymphocytes (12.4%), and of CD3− CD16+ CD56+ NK cells (18.8%) were significantly decreased (P < 0.05), and the percentage of CD4+ CD25+ Treg cells (9.8%) was significantly increased (P < 0.05) in the peripheral blood of preoperative high-grade glioma patients compared with low-grade patients. These result indicated that the higher degree of glioma malignancy was associated with a lower level of immune cells.

Figure 3: Analysis of preoperative peripheral immune cell subsets in patients with different grades of glioma. Peripheral blood immune cells were collected preoperatively, and the percentage of cells in each immune cell subset was calculated by flow cytometry. (A) CD3+. (B) CD4+. (C) CD4+/CD8+. (D) CD3− CD19+ B-lymphocytes. (E) CD3− CD16+ CD56+ natural killer cells. (F) CD4+ CD25+ regulatory (Treg) cells. Data are expressed as the mean ± standard deviation (high-grade group n = 23, low-grade group n = 15).^†P < 0.05,^{† †}P < 0.001, vs. high-grade group (Student's t-test)

Click here to view

Grade III IDH- mutant oligodendroglioma patients have lower levels of immune cells after surgery compared with Grade II patients

We then studied the relationship between immune cell populations and the grade of IDH-mutant glioma malignancy after surgery. As shown in [Figure 4], patients were divided into low-grade and high-grade groups. Peripheral blood was collected 3 and 7 days after the surgery. The percentages of CD3⁺ (3 days: 67.3%, 7 days: 63.4%), CD4+ (3 days: 36.5%, 7 days: 32.4%), CD4+/CD8+ (3 days: 1.7%, 7 days: 1.5%), and CD3− CD19+ B-lymphocytes (3 days: 18.3%, 7 days: 16.4%), and of CD3− CD16+ CD56+ NK cells (3 days: 23.7%, 7 days: 21.2%) were higher in the low-grade group compared with those in the high-grade group, and that the percentages of CD4+ CD25+ Treg cells (3 days: 3.8%, 7 days: 4.1%) were lower in the low-grade group compared with the high-grade group (P < 0.05). These results indicated that after surgery, the higher the degree of glioma malignancy, the lower the level of immune cells. This was consistent with the results before surgery.

Figure 4: Preoperative and postoperative analysis of peripheral immune cell subsets in patients with different grades of glioma. Peripheral blood immune cells from glioma patients were collected on days 3 and 7 after surgery. The percentage of cells in each immune cell subset was calculated by flow cytometry. (A) CD3+. (B) CD4+. (C) CD4+/CD8+. (D) CD3− CD19+ B-lymphocytes. (E) CD3− CD16+ CD56+ natural killer (NK) cells. (F) CD4+ CD25+ regulatory (Treg) cells. Data are expressed as the mean ± standard deviation (high-grade group n = 23. Low-grade group n = 15).^‡P < 0.05,^‡‡P < 0.001, vs. high-grade group;^§P < 0.05,^§§P < 0.001, vs. preoperative (t-test)

Click here to view

The level of immune cells was increased after surgery compared with before surgery in IDH-mutant oligodendroglioma patients

Surgery directly affects patient prognosis. Therefore, we analyzed changes in peripheral immune cell subsets between preoperative and postoperative patients. As shown in [Figure 4], there was a significant difference in peripheral immune cell subgroups between pre- and post-surgery. The percentages of CD3+, CD4+/CD8+, and CD3− CD19+ B-lymphocytes and of CD3− CD16+ CD56+ NK cells were higher before surgery (P < 0.05), and the percentage of CD4+ cells at 3 and 7 days after surgery was lower than before surgery (P > 0.05). The number of CD4+ CD25+ Treg cells was lower after surgery compared with before surgery (P < 0.05). These results indicated that the levels of immune cells increased and that immune function was improved following removal of the tumor.

Discussion

Currently, glioma treatment mainly relies on surgical resection. However, there is usually no clear demarcation between the glioma and the surrounding healthy brain tissue. This complicates resection of the tumor and the glioma usually recurs near the resection area within a few months after surgery.^[12],[13] To overcome these limitations, new therapeutic methods have emerged through preclinical and clinical studies. Among them, immunotherapy shows promise. A recent study reported that the incidence and prognosis of glioma are associated with immune dysfunction.^[10] Lymphocyte counts are often reduced in patients with poor prognosis. Severe lymphopenia is associated with severe opportunistic infections^[14],[15] and glioblastoma patients with induction therapy, resulting in severe lymphopenia have poorer survival.^[16] In 2016 WHO classification of gliomas, IDH gene mutation was added to the diagnosis of gliomas. Changes were made to the classification of oligodendroglioma, according to genotype. A recent study found that patients with IDH-mutant glioma associated with the CpG island methylator phenotype have a longer survival time compared with wild-type IDH glioma patients because of impaired immune function.^[9] Therefore, in this study, levels of peripheral blood immune cells were monitored in IDH-mutant oligodendroglioma patients before and after tumor resection surgery.

T-lymphocytes, including CD4+ helper T-lymphocytes and CD8+ cells, play a critical role in the immune response to glioma. Activated T-lymphocytes secrete cytokines that activate NK cells and B-lymphocytes, to promote the killing of glioma cells. CD8+ lymphocytes can directly recognize and kill glioma cells. The relatively stable state of CD4+ and CD8+ lymphocytes guarantees systemic immune balance, and disturbance of these two subsets causes immune dysfunction that promotes proliferation and reduces the elimination of glioma cells.^[17] Therefore, we selected T lymphocyte subsets as the main indicators of this study. We showed that the percentages of CD3+, CD4+, CD4+/CD8+, and CD3− CD19+ B-lymphocytes and of CD3− CD16+ CD56+ NK cells were significantly lower in the glioma group compared with the control group. These results indicated that the cellular and humoral immune functions were significantly repressed in glioma patients. Lymphocyte proliferation is repressed in glioma patients allowing the growth and metastasis of glioma cells.^[18] Some studies suggest that immune function is more seriously impaired in glioma patients with a higher grade malignancy. A possible mechanism for this is that glioma cells secrete immunosuppressive factors that reduce CD3+ and CD4+ lymphocyte numbers, which suppresses immune function. In addition, decreased numbers of CD4+ cells leads to down-regulation of induced T-lymphocytes, which weakens the inhibitory effect of T-lymphocytes.^[19] In this study, the percentages of CD3+, CD4+, CD4+/CD8+, and CD3− CD19+ B-lymphocytes and of CD3− CD16+ CD56+ NK cells in peripheral blood of high-grade glioma patients were significantly lower before surgery compared with the low-grade group (P < 0.05). After surgery, the percentages of CD3+, CD4+, CD4+/CD8+, and CD3− CD19+ B-lymphocytes, and of CD3− CD16+ CD56+ NK cells were significantly increased in both low-grade and high-grade gliomas. These results indicate that levels of leukocyte subgroups can reflect the efficacy of the operation, which provides a potential cytological indicator for evaluating operation efficacy.

Native Treg cells are a subpopulation of lymphocytes that develop in the medulla of the thymus.^[20] The body's immune responses to tumors are weak or nonexistent. Treg cells function to prevent excessive damage by the immune process, but they also participate in tumor cells escaping surveillance.^[5] The percentage of CD4+ CD25+ Treg cells in the preoperative glioma group was significantly higher than that in the control group. This indicates that the number of Treg cells affects the occurrence and growth of gliomas. Treg cells can be used as an important detection index in patients with glioma. We also found that the proportion of CD4+ CD25+ Treg cells in peripheral blood was related to the grade of glioma. The level of peripheral blood Treg cells was higher in Grade III patients than in Grade II patients indicating that Treg cell-mediated immune tolerance may be directly involved in the progression of IDH-mutant oligodendroglioma.

Conclusion

The immune function of IDH-mutant oligodendroglioma patients is repressed, and a higher grade of malignancy produces a more serious immunosuppressive effect. We hope that this study will provide new evaluation indicators for the invasiveness and prognosis of IDH-mutant oligodendroglioma and that new ideas for the immunotherapy of gliomas will emerge.

Financial support and sponsorship

This study was supported by the National Natural Science Foundation of China (No. 81572489, to QC).

Institutional review board statement

All procedures of this study were approved by the Institutional Ethics Committee of the Faculty of Medicine at Renmin Hospital of Wuhan University (approval No. 2018K-C017) on June 4, 2018.

Declaration of patient consent

The authors certify that they have obtained the appropriate patient consent forms. In the forms, the patients have given consent for their images and other clinical information to be reported in the journal. The patients understood that their names and initials would not be published and due efforts would be made to conceal their identity.

Conflicts of interests

There are no conflicts of interest.

References

1.	Komori T. The 2016 WHO classification of tumours of the central nervous system: The major points of revision. Neurol Med Chir (Tokyo) 2017;57:301-11.
2.	Kleihues P, Sobin LH. World Health Organization classification of tumors. Cancer 2000;88:2887.
3.	Asklund T, Malmström A, Björ O, Blomquist E, Henriksson R. Considerable improvement in survival for patients aged 60-84 years with high grade malignant gliomas – Data from the Swedish brain tumour population-based registry. Acta Oncol 2013;52:1041-3.
4.	Arai H, Ikota H, Sugawara K, Nobusawa S, Hirato J, Nakazato Y. Nestin expression in brain tumors: Its utility for pathological diagnosis and correlation with the prognosis of high-grade gliomas. Brain Tumor Pathol 2012;29:160-7.
5.	Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 2007;114:97-109.
6.	Wen PY, Kesari S. Malignant gliomas in adults. N Engl J Med 2008;359:492-507.
7.	Louis DN. Molecular pathology of malignant gliomas. Annu Rev Pathol 2006;1:97-117.
8.	Louis DN, Perry A, Reifenberger G, von Deimling A, Figarella-Branger D, Cavenee WK, et al. The 2016 World Health Organization classification of tumors of the central nervous system: A summary. Acta Neuropathol 2016;131:803-20.
9.	Amankulor NM, Kim Y, Arora S, Kargl J, Szulzewsky F, Hanke M, et al. Mutant IDH1 regulates the tumor-Associated immune system in gliomas. Genes Dev 2017;31:774-86.
10.	Yovino S, Kleinberg L, Grossman SA, Narayanan M, Ford E. The etiology of treatment-related lymphopenia in patients with malignant gliomas: Modeling radiation dose to circulating lymphocytes explains clinical observations and suggests methods of modifying the impact of radiation on immune cells. Cancer Invest 2013;31:140-4.
11.	Wen PY, Huse JT. 2016 World Health Organization classification of central nervous system tumors. Continuum (Minneap Minn) 2017;23:1531-47.
12.	Pan IW, Ferguson SD, Lam S. Patient and treatment factors associated with survival among adult glioblastoma patients: A USA population-based study from 2000-2010. J Clin Neurosci 2015;22:1575-81.
13.	Stummer W, Reulen HJ, Meinel T, Pichlmeier U, Schumacher W, Tonn JC, et al. Extent of resection and survival in glioblastoma multiforme: Identification of and adjustment for bias. Neurosurgery 2008;62:564-76.
14.	Mahindra AK, Grossman SA. Pneumocystis carinii pneumonia in HIV negative patients with primary brain tumors. J Neurooncol 2003;63:263-70.
15.	Meije Y, Lizasoain M, García-Reyne A, Martínez P, Rodríguez V, López-Medrano F, et al. Emergence of cytomegalovirus disease in patients receiving temozolomide: Report of two cases and literature review. Clin Infect Dis 2010;50:e73-6.
16.	Grossman SA, Ye X, Lesser G, Sloan A, Carraway H, Desideri S, et al. Immunosuppression in patients with high-grade gliomas treated with radiation and temozolomide. Clin Cancer Res 2011;17:5473-80.
17.	Wu ZB, Qiu C, Zhang AL, Cai L, Lin SJ, Yao Y, et al. Glioma-associated antigen HEATR1 induces functional cytotoxic T lymphocytes in patients with glioma. J Immunol Res 2014;2014:131494.
18.	Fang KM, Yang CS, Lin TC, Chan TC, Tzeng SF. Induced interleukin-33 expression enhances the tumorigenic activity of rat glioma cells. Neuro Oncol 2014;16:552-66.
19.	Miyauchi JT, Chen D, Choi M, Nissen JC, Shroyer KR, Djordevic S, et al. Ablation of neuropilin 1 from glioma-associated microglia and macrophages slows tumor progression. Oncotarget 2016;7:9801-14.
20.	Derbinski J, Schulte A, Kyewski B, Klein L. Pillars article: Promiscuous gene expression in medullary thymic epithelial cells mirrors the peripheral self. Nat Immunol 2001;2:1032-39. Erratum in: J Immunol 2016;196:2915-22.