Elsevier

World Neurosurgery

Volume 127, July 2019, Pages e212-e220
World Neurosurgery

Original Article
Immune Microenvironment of Primary and Recurrent Craniopharyngiomas: A Study of the Differences and Clinical Significance

https://doi.org/10.1016/j.wneu.2019.02.212Get rights and content

Objective

This study explored the differences between the immune microenvironments of primary and recurrent craniopharyngiomas (CPs). In addition, we investigated the relationship between the immune microenvironment and clinical characteristics of CP.

Methods

We collected 52 specimens from 26 patients with CPs. For each patient, specimens for both primary and recurrent CPs were obtained. We performed an immunohistochemical analysis of these specimens to determine the distributions of M2 macrophages, CD8+ T cells, programmed cell death 1 ligand 1 (PD-L1), and Ki67.

Results

In recurrent CP specimens, the distributions of M2 macrophages, Ki67, and PD-L1 increased compared with primary CP specimens (P = 0.019, P = 0.0084, and P = 0.0319, respectively). Moreover, the distributions of M2 macrophages, CD8+ T cells, and PD-L1 in papillary CPs were higher than those observed in adamantinomatous craniopharyngiomas (ACPs) (P = 0.0317, P = 0.0359, and P < 0.0001, respectively). In the adult ACP group, M2 macrophages, CD8+ T cells, and PD-L1 were more abundant/expressed than in the child ACP group (P = 0.0159, P = 0.0215, and P < 0.0088, respectively). A positive correlation was found between M2 macrophages and CD8+ T cells (r = 0.4079; P = 0.0027). Correspondingly, M2 macrophages and CD8+ T cells were both positively correlated with PD-L1 (r = 0.4564; P = 0.0007 and r = 0.3987; P = 0.0034, respectively). The observed high expression of M2 macrophages in primary CPs suggests a shortened time for tumor recurrence (P = 0.0131).

Conclusions

The microenvironment of recurrent CP varies from that of primary CP. The abundance of M2 macrophages in primary CP may indicate a risk of early recurrence. Therefore, it is recommended to increase the frequency of follow-up examinations in these patients.

Introduction

Craniopharyngioma (CP) occurs mainly in the 5- to 14-year and 50- to 74-year age groups, accounting for 2% to 5% of primary intracranial tumors and more than half of saddle area tumors in children.1, 2 The histological subtypes of CP include adamantinomatous craniopharyngioma (ACP) and papillary craniopharyngioma (PCP). ACP may be diagnosed in all ages, whereas PCP invariably occurs in adults and is rarely observed in children. Although CP is a benign tumor, it is adjacent to the pituitary, hypothalamus, optic nerve, and internal carotid artery. In addition, its clinical manifestations are complicated and it is associated with a high rate of relapse. Studies investigating CP involved various aspects, such as molecular pathology, cancer stem cells, patient-derived xenografts, and molecular targeted therapy.3 Using new-generation sequencing, it was confirmed that ACP primarily carries CTNNB1 mutations, whereas PCP mainly carries BRAF mutations. Use of a BRAF inhibitor achieved positive results in the treatment of PCP and it is currently under investigation in clinical trials.4, 5

The proliferation of tumors is based on changes in its molecular biology. More recent studies also indicated that the tumor microenvironment plays a key regulatory role in the progression of intracranial malignant tumors, and that targeted therapy against the components of the microenvironment may be performed to achieve certain clinical effects.6 Nevertheless, studies investigating the immune microenvironment of CP are currently limited. In this study, we collected specimens obtained from patients with CPs. For each patient, specimens for both primary and recurrent CPs were collected. We analyzed the distributions of M2 macrophages (main cells of tumor tissue), CD8+ T cells (immune killer cells), programmed cell death 1 (PD-1)/programmed cell death 1 ligand 1 (PD-L1) (immune checkpoint), and Ki67 (proliferation marker), to understand the differences between the immune microenvironments of primary and recurrent CPs. In addition, we investigated the relationship of immune microenvironmental indicators with clinical characteristics.

Section snippets

Patients

This study retrospectively analyzed a total of 52 tumor specimens of primary and recurrent CPs (26 specimens each) and clinical data of 26 patients from September 2008 to February 2017. Clinical data included demographics, clinical symptoms, imaging, surgical conditions, pathological types, and recurrence time (Table 1, Figures 1 and 2). The patients were treated only with surgical resection and did not receive other treatments such as radiation therapy or intracapsular interventions. Recurrent

Characteristics of Immune Microenvironments in Primary and Recurrent CPs

M2 macrophages were mainly found in the stroma and significantly distributed around the perivascular areas and close to tumor cells. Compared with primary CPs, the abundance of M2 macrophages in recurrent CPs was significantly increased. In contrast, the number of CD8+ T cells in recurrent CPs was not significantly different from that observed in primary CPs. Ki67-positive cells were mainly distributed in basal cells and in whorls of tumor epithelium. Moreover, concerning the expression of

Discussion

The tumor microenvironment is composed of tumor cells, extracellular matrix components, immune cells, and the tumor vascular system. It plays an important role in tumor progression and metastasis of intracranial primary and metastatic tumors.11 Tumor-associated macrophages are derived from monocytes in the blood, which are recruited into tumor tissues by various signals, cytokines of tumor cells, and cytokines of tumor microenvironment, polarizing into M1 and M2 types with different functions

Conclusions

Comparisons between primary and recurrent CPs in the same patients indicated that the immune microenvironment of recurrent CPs was significantly changed. The abundance of M2 macrophages in primary CP may suggest a shortened time for tumor recurrence. Consequently, the frequency of follow-up examinations in these patients should be increased. In recurrent tumors, the microenvironment indicators with increased expressions are correlated with poor prognosis, higher chance of immune escape, and

References (31)

  • J.C. Martinez-Gutierrez et al.

    Diagnosis and management of craniopharyngiomas in the era of genomics and targeted therapy

    Neurosurg Focus

    (2016)
  • P.K. Brastianos et al.

    Dramatic response of BRAF V600E mutant papillary craniopharyngioma to targeted therapy

    J Natl Cancer Inst

    (2016)
  • E.R. Parra et al.

    Validation of multiplex immunofluorescence panels using multispectral microscopy for immune-profiling of formalin-fixed and paraffin-embedded human tumor tissues

    Sci Rep

    (2017)
  • W. Zhu et al.

    A reformed surgical treatment modality for children with giant cystic craniopharyngioma

    Child’s Nerv Syst

    (2017)
  • R. Fahlbusch et al.

    Surgical treatment of craniopharyngiomas: experience with 168 patients

    J Neurosurg

    (1999)
  • Cited by (9)

    • Autophagy and immune microenvironment in craniopharyngioma and ameloblastoma

      2021, Experimental and Molecular Pathology
      Citation Excerpt :

      Their immune cell populations are less well described. A previous study in 26 patients showed the absence of PD-L1 expression in immune cells (Lin et al., 2019) similar to our series. The authors of this study also counted the presence of CD206+ macrophages as a marker of M2 macrophages and CD8+ T cells; macrophages were more numerous than CD8+ T cells like in our study; however, a direct comparison is difficult since the counting method is different than ours.

    • Creative and Innovative Methods and Techniques for the Challenges in the Management of Adult Craniopharyngioma

      2020, World Neurosurgery
      Citation Excerpt :

      According to a recent report on IHC, immunofluorescence and RNA in situ hybridization of PD-L1 in craniopharyngioma specimens,103 all the PCPs tested (n = 18) showed expression of PD-L1 in at least some of the tumor cells (average, 35% ± 22% of cells).103 There is also a recent report that recurrent craniopharyngiomas harbor more PD-L1–expressing cells than do primary craniopharyngiomas.104 At the time of writing, there has been no clinical trial to test the efficacy of the immune checkpoint inhibitor for craniopharyngioma.

    View all citing articles on Scopus

    Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

    View full text