Tumor Immune Microenvironment in Pituitary Neuroendocrine Tumors (PitNETs): Increased M2 Macrophage Infiltration and PD-L1 Expression in PIT1-lineage Subset

doi:10.21203/rs.3.rs-3001593/v1

Download PDF

Research Article

Tumor Immune Microenvironment in Pituitary Neuroendocrine Tumors (PitNETs): Increased M2 Macrophage Infiltration and PD-L1 Expression in PIT1-lineage Subset

https://doi.org/10.21203/rs.3.rs-3001593/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 07 Jul, 2023

Read the published version in Journal of Neuro-Oncology →

You are reading this latest preprint version

Purpose

Tumor immune microenvironment in PitNETs and application of current immunotherapy for refractory PitNETs remains debated. We aim to evaluate the immune landscape in different lineages of PitNETs and determine the potential role of pituitary transcription factors in reshaping the TIME, thus promoting the application of current immunotherapy for aggressive and metastatic PitNETs.

Methods

Immunocyte infiltration and expression patterns of immune checkpoint molecules in different lineages of PitNETs were estimated via in silico analysis and validated using an IHC validation cohort. The correlation between varying immune components with clinicopathological features was assessed in PIT1-lineage PitNETs.

Results

Transcriptome profiles from 210 PitNETs/ 8 normal pituitaries (NPs) and immunohistochemical validations of 77 PitNETs/ 6 NPs revealed a significant increase in M2-macrophage infiltration in PIT1-lineage PitNETs, compared with the TPIT-lineage, SF1-lineage subsets and NPs. While CD68 + macrophage, CD4 + T cells, and CD8 + T cells were not different among them. Increased M2-macrophage infiltration was associated with tumor volume (p < 0.0001, r = 0.57) in PIT1-lineage PitNETs. Meanwhile, differentially expressed immune checkpoint molecules (PD-L1, PD1, and CTLA4) were screened and validated in IHC cohorts. The results showed that PD-L1 was highly expressed in PIT1-lineage subsets, and PD-L1 overexpression showed a positive correlation with tumor volume (p = 0.04, r = 0.29) and cavernous sinus invasion (p < 0.0001) in PIT1-lineage PitNETs.

Conclusion

PIT1-lineage PitNETs exhibit a distinct immune profile with enrichment of M2 macrophage infiltration and PD-L1 expression, which contribute to its clinical aggressiveness. Application of current immune checkpoint inhibitors and M2-targeted immunotherapy might be more beneficial to treat aggressive and metastatic PIT-lineage PitNETs.

PitNETs

PIT1

M2-macrophage

PD-L1

immunotherapy

Pituitary neuroendocrine tumors (PitNETs) are the most common neoplasms of the sellar region and the second most frequent intracranial neoplasms. Clinically apparent PitNETs occur in approximately 1/1100 individuals in the general population [1]. The clinical manifestations of PitNETs are primarily attributed to mass effect and hormone excess, leading to a significant reduction in the quality of life and increased mortality [2]. Although most PitNETs are noninvasive and can be effectively managed through surgery and/or medical treatment, a subset of 25–55% of cases exhibits aggressive characteristics. And in rare instances, PitNETs can elicit extracranial metastasis, which is conventionally defined as pituitary carcinomas and updated as metastatic PitNETs [3; 4]. Management of those aggressive and metastatic PitNETs remains challenging, despite extensive surgical intervention and radio-chemotherapy.

The tumor immune microenvironment (TIME) significantly contributes to the development, progression, recurrence, and drug resistance of various malignancies [5]. Although TIME in PitNETs has been widely investigated, the immune pattern and its implications for the clinicopathological characteristics of PitNETs remain controversial [6]. For instance, the infiltration of tumor-associated macrophages (TAMs) and immune checkpoint molecules expression in different subtypes of PitNETs via conventional hormone classification systems varies significantly from study to study, and the crosstalk between those immunological alterations and biological behaviors has not reached a solid consensus. Despite these controversies, immune checkpoint inhibitors (ICIs) tageting PD1/PD-L1 and CTLA4 pathway have been employed to treat aggressive and metastatic PitNETs [7]. The positive clinical response of ICIs applications in PitNETs and the immunotherapy-induced hypophysitis partially warrant its further application in PitNETs [8; 9]. However, only a few patients have been reported, and some cases elicit progressive disease during ICIs employment [10]. Therefore, an improved understanding of the underlying immunological mechanisms driving aggressiveness might provide insights into the mechanisms of different immunotherapy responses and serve as resources for future targeted therapy in aggressive and metastatic PitNETs.

According to the 2022 WHO classification system, pituitary transcription factors, hormones, and low molecular weight cytokeratin (LMWK) have been applied to classify the PitNETs [3]. The interaction between hormones and LMWK with TIME has been widely investigated. However, the potential role of pituitary transcription factors in regulating TIME has not been well established in PitNETs. In this study, we conducted a comprehensive analysis of the TIME in different lineages of PitNETs according to the updated classification system, to determine the potential role of pituitary transcription factors in reshaping TIME and promoting the application of immunotherapeutic strategies in aggressive and metastatic PitNETs. Additionally, immune microenvironment of normal pituitaries (NPs) was estimated to evaluate the potential off-target side effect on pituitary during clinical employment of current immunotherapy.

2.1. Human tissue sample and clinicopathological data

The protocols for human specimens were approved by the Human Investigation Ethics Committee (Institutional Review Boards) of the First-affiliated Hospital of Sun Yat-Sen University. Resected PitNET samples and corresponding clinicopathological characteristics were obtained from the Department of Neurosurgery and Pituitary Tumor Center in the first-affiliated hospital of Sun Yat-Sen University between 2021 and 2023. A total of 77 PitNET patients who underwent endoscopic endonasal transsphenoidal surgery without previous radiotherapy and/or chemotherapy were included. Pathological diagnosis was based on the WHO 2022 classification of PitNETs. All data were obtained from patients who signed informed consent approving the use of their tissue for research purposes after surgery. Normal pituitaries, but not neoplastic adjacent, were obtained within 10h of death from autopsies with no known endocrine, immune, or malignant disease.

2.2. Immunohistochemistry

Surgically resected PitNET specimens and NPs were fixed in 4% paraformaldehyde for 24h and then embedded in paraffin. Paraffin blocks were cut into 4 µm sections and used for hematoxylin and eosin (HE) or immunohistochemistry (IHC) staining. Protein marker expression of immune cell infiltration, expression of immune checkpoint molecules, and PIT1 transcription factor were detected by SP Rabbit&Mouse HRP DAB Kit (catalog no. CW2069S, CWBIO). Slides were incubated with the following primary antibody: CD68 (catalog no.IR61361-2CN, DAKO), CD163 (catalog no.ZM-0428, ZSGB-BIO), CD4 (catalog no. RMA0620, MXBiotechnologies), CD8 (catalog no. ab4055, Abcam), PD-L1 (catalog no. ab205921, Abcam), PD-1 (catalog no. ZM-0381, ZSGB-BIO), CTLA4 (catalog no. ab237712, Abcam), PIT1 (catalog no.ZM0208, ZSGB-BIO), TPIT (catalog no.ZM-0316, ZSGB-BIO), SF1 (catalog no.ZM-0089, ZSGB-BIO). The estimation of PD-L1 expression followed the guideline of Consensus on the Immunohistochemical Tests in Solid Tumors (2021 version). Stained slides were digitally scanned using Slides Scanning Image System (TEKSQRAY). Images were analyzed using Image J software and quantification of positive expression was measured by two independent investigators (Mei Luo and Rui Tang). Cases with staining artifacts or poor tissue integrity following IHC processing were excluded.

2.3. Reclassification of PitNETs in databases according to the 2022 WHO classification system

Based on the 2022 WHO classification, somatotroph tumors, mammosomatotroph tumors, lactotroph tumors, and thyrotroph tumors were directly classified into PIT1-lineage PitNETs, and corticotroph tumors were directly classified into TPIT-lineage PitNETs, and gonadotroph tumors were directly classified into SF1-lineage PiTNETs, even without a pathological diagnosis of the pituitary transcription factors. Since conventional clinically nonfunctioning pituitary adenomas (CNFPAs) without pathological data were not capable of being classified into an exact lineage of PitNETs, patients diagnosed as CNFPAs were excluded from our study. Transcriptomic data of eligible PitNETs and NPs were downloaded after extensive data screening from Gene Expression Omnibus (GEO) and ArrayExpress database E-MTAB databases. The data were normalized using the “normalizeBetween Array” function with the Limma package to remove the batch effect.

2.4. in silico analysis of immune cell infiltration via CIBERSORT algorithm and differentially expressed immune checkpoint molecules

The 22 subsets of immune cell infiltration status were inferred by CIBERSORTx (https://cibersortx.stanford.edu/), a computational analytical approach developed by Newman et al. to provide an estimation of the relative/absolute abundance immune cell infiltration in TIME from the transcriptomic data in tumor [11]. After CIBERSORTx analysis, the immune cell infiltration patterns of PitNETs and NPs were estimated, and the difference among different lineages of PitNETs and NPs were compared. The differentially expressed immune checkpoint molecules among PIT1-lineage PitNETs, non-PIT1-lineage PitNETs, and NPs were determined using the Limma package in R (Version 3.40.6). The thresholds were log2 (fold change) > 1.5 and adjusted p-value < 0.05. Visualization of differentially expressed genes was conducted using the Limma volcano plots package.

2.5. Statistical analysis

Statistical analyses were performed by IBM SPSS (version 23.0, Armonk, NY IBM Corp.), Prism8 (GraphPad, La Jolla), and R project (Version 3.4.1). Normally distributed quantitative variables were presented as Mean ± SEM, and categorical variables are expressed as numbers (percentage). To calculate statistical significance between groups, a Mann-Whitney U test or an unpaired Student’s t-test was applied for the pairwise group comparisons of the quantitative variables, while One-way analysis of variance and the Kruskal-Wallis test was performed for multiple comparisons. Chi-squared test or Fisher’s exact probability test was applied for the comparison of categorical variables. The Pearson correlation coefficient was applied to identify correlations between pairs of parameters. P values less than 0.05 were considered statistically significant. * indicates 0.01 < p < 0.05; ** indicates 0.001 < p < 0.01; *** indicates 0.0001 < p < 0.001; **** indicates p < 0.0001.

3.1. Clinicopathological features in the validation cohort of PitNETs

In the validation cohort of PitNETs, a total of 77 patients were included, of which 39 (51%) were female. The mean age at diagnosis was 46 years (range: 15–77 years). Based on the 2022 WHO classification, 48 (62%) cases were diagnosed as PIT1-lineage PitNETs. Of these cases, 14 (18%) were densely granulated somatotroph tumors, 19 (25%) were sparsely granulated somatotroph tumors, 3 (4%) were densely granulated lactotroph tumors, 5 (6%) were sparsely granulated lactotroph tumors, 4 (5%) were mammosomatotroph tumors, and 3 (4%) were thyrotroph tumors. In addition, 12 (16%) cases were diagnosed as TPIT-lineage PitNETs, which included 3 (4%) densely granulated corticotroph tumors, 6 (8%) sparsely granulated corticotroph tumors, and 3 (4%) Crooke cell tumors. Furthermore, 17 (22%) cases were diagnosed as SF1-lineage gonadotroph tumors. The Ki67 index was available for 77 PitNETs, and 66 (86%) had a Ki67 index lower than 3%, while 11 (14%) had a Ki67 index of 3% or higher. The patient cohort included 1 (1%) microadenoma (diameter < 1cm), 53 (69%) macroadenomas (1cm ≤ diameter < 3cm), and 23 (30%) giant adenomas (diameter ≥ 3cm). The mean tumor volume of the included patients was 6.09 cm3 (range: 0.29–42.79 cm3). In terms of invasion, 43 (56%) patients presented with cavernous sinus (CS) invasion. Among the included cases, 72 (94%) were primary and 5 (6%) were recurrent PitNETs. Table 1 provides a detailed overview of the interested characteristics of the included patients.

Table 1

Overview of clinicopathological features for patients in validation cohort.
Clinicopathological features
Gender (n, male/female)	38/39
Age at diagnosis (years, mean and range)	46 [15–77]
Pathological diagnosis (n, %)
PIT1-lineage PitNETs	48, 62%
Somatotroph tumors (densely granulated)	14, 18%
Somatotroph tumors (sparsely granulated)	19, 25%
Lactotroph tumors (densely granulated)	3, 4%
Lactotroph tumors (sparsely granulated)	5, 6%
Mammosomatotroph tumors	4, 5%
Thyrotroph tumors	3, 4%
TPIT-lineage PitNETs	12, 16%
Corticotroph tumors (densely granulated)	3, 4%
Corticotroph tumors (sparsely granulated)	6, 8%
Crooke cell tumor	3, 4%
SF1-lineage PitNETs	17, 22%
Gonadotroph tumors	17, 22%
Ki67 (n, %)
<3%	66, 86%
≥ 3%	11, 14%
Tumor size
Microadenoma (<1cm)	1, 1%
Macroadenomas (1cm-3cm)	53, 69%
Giant adenomas (≥ 3cm)	23, 30%
Tumor volume (mean and range, cm3)	6.09 [0.29–42.79]
Cavernous Sinus invasion (n, %)
Yes	43, 56%
No	34, 44%
Recurrence (n, %)
Primary	72, 94%
Recurrent	5, 6%

3.2. Heterogeneous immune cell infiltration patterns of PitNETs and NPs via in silico analysis

We obtained a total of 6 eligible datasets from GEO and E-MTAB databases for in silico analysis (GSE132982, GSE120350, GSE46311, GSE93825, GSE147786, and E-MTAB-7768). We extracted the transcriptome gene expression profiles if they were capable of being reclassified according to the 2022 WHO classification of PitNETs. Among these datasets, we identified 116 patients as PIT1-lineage PitNETs (53 somatotroph tumors, 5 mammosomatotroph tumors, 48 lactotroph tumors, and 10 thyrotroph tumors) and 94 patients as non-PIT1-lineage PitNETs (93 corticotroph tumors as TPIT-lineage and 1 gonadotroph tumor as SF1-lineage). 8 NPs were included after screening. Deconvolution analysis of transcriptome profile by CIBERSORTx showed that each lineage of PitNETs has a unique immunological infiltration landscape, as shown in Fig. 1A. As the most important immunocytes affecting the clinical efficacy of immunotherapy in tumors, the relative abundance of T cells and macrophage was highly infiltrated and tended to be affected by pituitary transcription factors. Specifically, CD8 + T cells and M2 macrophages were the most abundant immunocytes in PIT1-lineage PitNETs than non-PIT1 lineages and NPs in Fig. 1B.

3.3. M2 macrophages were highly infiltrated in PIT1-lineage PitNETs

To validate the immunocyte infiltration by deconvolution analysis, macrophage and T cell infiltrations were estimated in IHC cohorts. The results demonstrated that the infiltration of CD68 + macrophages was not different among different lineages of PitNETs and NPs in Fig. 2A. Notably, the infiltration of CD163 + M2-like macrophages was significantly increased in PIT1-lineages, compared with TPIT-lineage (p < 0.01), PIT1-lineage (p < 0.0001), and NPs (p < 0.05) in Fig. 2B. To further evaluate the infiltration status of CD163 + M2-like macrophages in PIT1-lineages PitNETs and confirm the effect of PIT1 pituitary transcription factors on TIME, the subgroup analysis data showed that the increased M2 macrophage infiltration was not related to the hormones (Somatotroph, Lactotroph, Thyrotroph, and Mammosomatotroph tumors) and granulation pattern (densely and sparsely) of PIT1-lineage PitNETs in Fig. S1A. While the infiltration of CD4 + T cells (p = 0.23) and CD8 + T cells (p = 0.30) did not show any difference among different lineages of PitNETs and NPs, retrospectively in Fig. 2C and 2D. Representative IHC images of CD68, CD163, CD4, and CD8 were presented in Fig. 2E.

3.4. Enrichment of M2 macrophage infiltration was correlated with the tumor volume in PIT1-lineage PitNETs

After correlation analysis of M2 macrophage infiltration with clinicopathological analysis in PIT1-lineage PitNETs, the results suggested that there was no significant difference in CD163 + M2-like macrophage infiltration between males and females (Fig. 3A, p = 0.85). And CD163 + M2-like macrophage did not seem to be related to the proliferation and CS invasion of PIT1-lineage PitNETs, including Ki67 index (Fig. 3B, p = 0.08), CS invasion (Fig. 3C, p = 0.33), and tumor recurrence (Fig. 3D, p = 0.11). However, there was a notable increase in CD163 + M2-like macrophages in giant adenomas compared to macroadenomas (Fig. 3E, p < 0.01). Moreover, the infiltration of CD163 + M2-like macrophages was positively correlated with the tumor volume of PIT1-lineage PitNETs (Fig. 3F, p < 0.0001, r = 0.57). This suggested that PIT1-mediated CD163 + M2-like macrophage infiltration might play a vital role in the development and growth of PIT1-lineage PitNETs.

3.5. PD-L1 was highly expressed in PIT1-lineage PitNETs

Differentially expressed immune checkpoint molecules (PD-L1, PD1, and CTLA4) were identified from the transcriptome of PitNETs, revealing that PD-L1 was highly expressed in PIT1-lineage PitNETs, while CTLA4 and PD-1 are not different in Fig. 4A. And the mRNA levels of immune checkpoint molecules were not found to be different between PIT1-lineage PitNETs and NPs in Fig. 4B. To evaluate the protein level of PD-L1, PD1, and CTLA-4 in PitNETs and NPs specimens, data from the IHC validation cohort showed that PD-L1 expression was significantly overexpressed in PIT1-lineage PitNETs compared to TPIT-lineage (p < 0.001), SF1-lineage PitNETs (p < 0.0001), and NPs (p < 0.5) in Fig. 4C. Beside, PD1 and CTLA4 were rarely expressed in PitNETs and NPs, without significant differences (p = 0.78 for PD1 in Fig. 4D, and p = 0.78 for CTLA4 in Fig. 4E). Representative IHC images of PD-L1, PD1, and CTLA4 were presented in Fig. 4F.

3.6. Highly expression of PD-L1 contributed to the tumor growth and CS invasion in PIT1-lineage PitNETs

The correlation analysis revealed that PD-L1 expression was not gender-dependent (p = 0.65) and was not associated with Ki67 index (p = 0.52) or tumor recurrence (0.22) in Fig. 5A-C. However, in Fig. 5D, PD-L1 was significantly overexpressed in patients with CS invasion compared to those without (p < 0.0001). Additionally, the tumor volume was positively correlated with PD-L1 expression in PIT1-lineage PitNETs, as shown in Fig. 5E (p = 0.04, r = 0.29). Representative magnetic resonance imaging (MRI) images of CS invasion and non-CS invasion were presented in Fig. 5F. These findings suggest that the PIT1-mediated overexpression of PD-L1 contributes to tumor growth and CS invasion in PIT1-lineage PitNETs.

In our study, PitNETs were classified according to the 2022 WHO classification by pituitary transcription factors, hormones, and LMWK. Our results determined the increased M2-macrophage infiltration and PD-L1 expression in PIT1-lineage PitNETs, which indicate the potential interaction between PIT1 pituitary transcription factor with TIME in PitNETs. And high infiltration of M2-macrophages and PD-L1 overexpression is associated with tumor growth and/or invasiveness in PIT1-lineage PitNETs. All those findings indicate that current ICIs and M2-targeted immunotherapy should be more beneficial to treat aggressive and metastatic PIT1-lineage than TPIT and SF1-lineage PitNETs.

TIME plays a crucial role in the clinical and biological characteristics of tumors [12]. Although the significance of TIME in the pathological progression of PitNETs is increasingly recognized, there is a lack of consensus regarding the immune patterns and their clinical implications for tumor aggressiveness [13; 14]. Among the immune components in TIME, tumor-associated macrophages (TAMs) are considered one of the most abundant immune cells in PitNETs [15]. It is regulated by various factors and interacts with tumor development, invasiveness, and drug resistance through diverse mechanisms [16]. In recent years, studies focusing on TAMs have provided new insights into their recruitment, polarization, and implications for the tumor growth and aggressiveness of PitNETs. Previous research on TAM infiltration primarily aimed to identify its patterns based on the classification of clinical function or hormone excess in PitNETs [17]. However, the results of these studies have been controversial and inconsistent. Some studies reported high infiltration of TAMs in functioning PitNETs, while others did not observe significant differences. The contribution of TAM infiltration to clinical aggressiveness appears to vary among different subtypes of PitNETs according to different studies [13; 15; 18; 19; 20; 21]. Besides, further studies on the polarization of TAMs have revealed that M2-TAMs are more likely to be associated with clinicopathological features compared to M1-TAMs [16]. Several studies have investigated the role of M2-macrophage polarization and its relationship with the aggressiveness of PitNETs. Zhang et al. reported a positive correlation between M2-TAMs and tumor invasion in PitNETs. They found that both GH3 and MMQ cell lines induced M2 polarization in TAMs through mTORC2 and ERK signaling pathways, and activated TAMs promoted invasion via the CCL17/CCR4/mTORC1 axis [21]. Marques observed increased TAM infiltration and an elevated CD163: HLA-DR ratio in PitNETs, indicating M2-TAM prevalence. M2-TAMs were found to promote neovascularization in PitNETs, potentially driven by higher concentrations of M2-polarizing cytokines, such as IL-4, compared to M1-polarizing cytokines like interferon-γ [19]. Yeung et al. reported higher infiltration of M2-TAMs in functioning PitNETs compared to non-functioning PitNETs, and primary tumors had higher infiltration than recurrent tumors based on in silico analysis [22]. Lin et al. demonstrated that an increased M2/M1 ratio was correlated with proliferation and invasion in non-functioning PitNETs [23]. These studies collectively highlight the role of M2-TAMs infiltration in the aggressiveness of PitNETs, suggesting that M2-TAMs may contribute to tumor invasion and neovascularization. According to the results of previous studies, the infiltration of TAMs in PitNETs varies significantly, particularly in non-functioning PitNETs that include different lineages of PitNETs, such as silent PitNETs, null cells, and gonadotroph tumors. The heterogeneity observed in previous studies can, at least partially, be explained by our study with an extended sample size with accurate pathological diagnosis. In the future, it is crucial to conduct larger studies with well-characterized patient cohorts and accurate pathological diagnoses to overcome the heterogeneity observed in studies about TAM infiltration and its implication for clinicopathological features.

As our data showed, M2-TAMs infiltration was significantly enriched in PIT1-lineage PitNETs, which indicates the potential regulatory role of PIT1 for TAMs. Currently, there is limited research on the immunoregulatory significance of PIT1 in TIME. Previous studies have suggested that PIT1 plays a role in promoting metastasis in breast cancer and gastric carcinoma through the PIT-1-CXCL12-CXCR4 axis [24; 25]. A study by A. Martínez-Ordoñez et al. demonstrated that PIT1 induces metabolic reprogramming via the regulation of lactate dehydrogenase A (LDHA) in the progression of human breast tumor cells [26]. In the context of PitNETs, our study is the first to quantitate immune infiltration and immune checkpoint molecule expression according to the classification by pituitary transcription factors. Given PIT1's regulatory role in both the immune system and endocrinology, it is crucial to elucidate its specific immunoregulatory role in PitNETs. Understanding the impact of PIT1 on the immune landscape of PitNETs can provide valuable insights into the interplay between pituitary transcription factors and the immune system in TIME, contributing to a better understanding of PitNET pathogenesis and potential targeted immunotherapeutic strategies.

The immune checkpoint molecules have been extensively investigated as important regulators of TIME [27]. In the context of PitNETs, high levels of PD-L1 expression have been detected and associated with tumor aggressiveness and as a potential predictor of responsiveness to immune checkpoint inhibitors (ICIs). Studies have reported variable levels of PD-L1 expression and its implications for clinicopathological features among different subtypes of PitNETs [28; 29]. Mei et al. found increased PD-L1 expression in somatotroph and lactotroph tumors compared to null cell tumors and gonadotroph tumors. They also observed higher mRNA levels of PD-L1 in primary PitNETs compared to recurrent PitNETs [13; 30]. Another study by P. F. Wang reported frequent expression of PD-L1 in functioning pituitary adenomas, with a positive correlation with aggressive behavior. Interestingly, they found that PD-L1 expression was associated with PRL and GH immunostaining and a higher Ki67 index [31]. M. P. Salomon et al. demonstrated high levels of PD-L1 expression in PitNETs, with somatotroph tumors showing higher expression compared to corticotroph tumors and non-functioning PitNETs [29]. Microarray data by J. Turchini showed preferential expression of PD-L1 (SP263) in PIT1-lineage PitNETs, while neoplastic adjacent normal pituitary gland tissue exhibited abundant PD-L1 expression [32]. According to the updated 2022 WHO classification system for PitNETs, PitNETs with higher PD-L1 expression are attributed to the PIT1-lineage. The findings of our study are consistent with the majority of previous results regarding PD-L1 expression and its implications for aggressiveness in PitNETs. The increased PD-L1 expression observed in PIT1-lineage PitNETs aligns with previous studies that support the potential use of immune checkpoint inhibitors as a therapeutic approach in this subtype of PitNETs.

ICIs have indeed demonstrated significant efficacy in various types of cancers, such as melanoma, lung cancer, and urothelial carcinoma [33]. And immune-related hypophysitis, which is a common side effect associated with the use of immune checkpoint inhibitors, suggests that patients with PitNETs might be already vulnerable to ICIs therapy [34]. As a result, ICIs have been investigated as a potential therapeutic option for shrinking tumor volume and normalizing hormone excess in aggressive and metastatic PitNETs [7]. In the context of ICIs, reliable predictive biomarkers are crucial for identifying patients who are more likely to respond to treatment. PD-L1 expression has emerged as an important biomarker in predicting the responsiveness to ICIs in several cancer types [35]. This has generated enthusiasm for exploring the expression of PD-L1 and its potential as a predictive marker or direct target for ICIs therapy in PitNETs. The preclinical study conducted by H.R. Kemeny in a mouse model demonstrated the potential for controlling tumor growth and reducing adrenocorticotropic hormone plasma levels in Cushing's disease using anti-PD-L1 ICIs [36]. In the clinical setting, ICIs have been applied in a small number of cases involving 11 aggressive PitNETs and 13 metastatic PitNETs. The overall response rate, which includes complete response, partial response, and stable disease, was reported as 56.25% for corticotroph tumors and 37.50% for lactotroph tumors. It is worth noting that metastatic PitNETs exhibited a higher sensitivity to ICIs than the aggressive subset. Regarding PD-L1 expression, the data suggest that PD-L1 expression is more dominant in those patients with PIT1-lineage. However, it is important to note that the correlation between increased PD-L1 expression and improved clinical outcomes was not consistently observed and cases without PD-L1 expression also showed favorable clinical efficacy, particularly in metastatic PitNETs [10]. It is widely accepted that highly expression PD-L1 expression does not guarantee that the tumor will respond to ICIs, nor the side effect will overweight the benefit. And negative PD-L1 staining in the tumor center should not preclude ICI administration in PitNETs [7]. Other factors, such as mutation burden, MSI, and CD8 T cell infiltration were associated with the responsiveness [37]. Due to the scarcity of data for ICIs employment in PitNETs, the clinical benefit and the risk of side effects in PIT1-lineage need to be further investigated by clinical trials, regardless of our findings of PD-L1 overexpression in PIT1-lineage. Moreover, it is important to cautiously evaluate the predictive value of PD-L1 using appropriate diagnostic antibodies through rigorous clinical trials. A recent study by V. Suteau et al. revealed that only a small proportion of PitNETs (18%) expressed PD-L1 (22C3), and its expression did not show a significant association with the biological characteristics of the tumors. The study also observed variations in PD-L1 staining intensity when different antibodies (such as QR1) were used, and noted that prominent cytoplasmic staining often obscured true membrane staining [38]. Additionally, the choice of cut-off value to define positive expression is crucial in predicting the clinical outcomes of ICIs. Therefore, if ICIs are to be applied in PitNETs, it is important to investigate whether different antibodies, along with their specific cut-off values for PD-L1 expression, are more suitable for accurately predicting the response to ICI treatment. Given the limited clinical data on ICIs applications in PitNETs, further research is warranted to determine the diagnostic value of PD-L1 immunohistochemical staining in the context of ICIs treatments [39].

In our study, we determined the increased M2-macrophage infiltration and PD-L1 expression in PIT1-lineage PitNETs, which ultimately led to its increased tumor growth and invasion of the cavernous sinus. Our findings provide important insights into the potential immunoregulatory role of PIT1 and lay the groundwork for further investigations on its immunomodulatory functions. Current immune checkpoint inhibitors, alone or in combination with M2-targeted therapy may further improve the clinical outcomes of aggressive and metastatic PIT1-lineage PitNETs.

Funding This research was funded by the Human Provincial Natural Science Foundation of China (2021JJ40868) and the National Natural Science Foundation of China (no.82001738).

Competing Interests The authors have no relevant financial or non-financial interests to disclose

Author Contributions All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Mei Luo and Rui Tang. The first draft of the manuscript was written by Mei Luo and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

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

Ethics approval Research involving human participants, their data or biological materials were reviewed and approved by Medical Ethics Committee of the First Affiliated Hospital of Sun Yat-sen University (protocol number [2022] 229-1).

Consent to participant Informed consent was obtained from all individual participants included in the study.

N.A. Tritos, and K.K. Miller, Diagnosis and Management of Pituitary Adenomas: A Review. Jama 329 (2023) 1386-1398.
S. Melmed, U.B. Kaiser, M.B. Lopes, J. Bertherat, L.V. Syro, G. Raverot, M. Reincke, G. Johannsson, A. Beckers, M. Fleseriu, A. Giustina, J.A.H. Wass, and K.K.Y. Ho, Clinical Biology of the Pituitary Adenoma. Endocrine reviews (2022).
S.L. Asa, O. Mete, A. Perry, and R.Y. Osamura, Overview of the 2022 WHO Classification of Pituitary Tumors. Endocrine pathology 33 (2022) 6-26.
P. Burman, O. Casar-Borota, L.G. Perez-Rivas, and O.M. Dekkers, Aggressive pituitary tumors and pituitary carcinomas: from pathology to treatment. The Journal of clinical endocrinology and metabolism (2023).
S. Yenyuwadee, K. Aliazis, Q. Wang, A. Christofides, R. Shah, N. Patsoukis, and V.A. Boussiotis, Immune cellular components and signaling pathways in the tumor microenvironment. Semin Cancer Biol 86 (2022) 187-201.
P. Marques, A.L. Silva, D. López-Presa, C. Faria, and M.J. Bugalho, The microenvironment of pituitary adenomas: biological, clinical and therapeutical implications. Pituitary 25 (2022) 363-382.
M.D. Ilie, A. Vasiljevic, E. Jouanneau, and G. Raverot, Immunotherapy in aggressive pituitary tumors and carcinomas: a systematic review. Endocrine-related cancer 29 (2022) 415-426.
J.J. Wright, A.C. Powers, and D.B. Johnson, Endocrine toxicities of immune checkpoint inhibitors. Nature reviews. Endocrinology (2021).
A. Shalit, P. Sarantis, E. Koustas, E.M. Trifylli, D. Matthaios, and M.V. Karamouzis, Predictive Biomarkers for Immune-Related Endocrinopathies following Immune Checkpoint Inhibitors Treatment. Cancers 15 (2023).
M.D. Ilie, C. Villa, T. Cuny, C. Cortet, G. Assie, B. Baussart, M. Cancel, P. Chanson, B. Decoudier, E. Deluche, A.L. Di Stefano, D. Drui, S. Gaillard, B. Goichot, O. Huillard, A. Joncour, D. Larrieu-Ciron, R. Libe, G. Nars, A. Vasiljevic, and G. Raverot, Real-life efficacy and predictors of response to immunotherapy in pituitary tumors: a cohort study. European journal of endocrinology 187 (2022) 685-696.
A.M. Newman, C.L. Liu, M.R. Green, A.J. Gentles, W. Feng, Y. Xu, C.D. Hoang, M. Diehn, and A.A. Alizadeh, Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 12 (2015) 453-7.
A. Christofides, L. Strauss, A. Yeo, C. Cao, A. Charest, and V.A. Boussiotis, The complex role of tumor-infiltrating macrophages. Nature immunology 23 (2022) 1148-1156.
Y. Mei, W.L. Bi, J. Agolia, C. Hu, A.M. Giantini Larsen, D.M. Meredith, S. Al Abdulmohsen, T. Bale, G.P. Dunn, M. Abedalthagafi, and I.F. Dunn, Immune profiling of pituitary tumors reveals variations in immune infiltration and checkpoint molecule expression. Pituitary (2021).
M.D. Ilie, A. Vasiljevic, P. Bertolino, and G. Raverot, Biological and Therapeutic Implications of the Tumor Microenvironment in Pituitary Adenomas. Endocrine reviews (2022).
P. Marques, S. Barry, E. Carlsen, D. Collier, A. Ronaldson, S. Awad, N. Dorward, J. Grieve, N. Mendoza, S. Muquit, A.B. Grossman, F. Balkwill, and M. Korbonits, Chemokines modulate the tumour microenvironment in pituitary neuroendocrine tumours. Acta neuropathologica communications 7 (2019) 172.
C. Han, S. Lin, X. Lu, L. Xue, and Z.B. Wu, Tumor-Associated Macrophages: New Horizons for Pituitary Adenoma Researches. Frontiers in endocrinology 12 (2021) 785050.
N. Inoshita, and H. Nishioka, The 2017 WHO classification of pituitary adenoma: overview and comments. Brain Tumor Pathol 35 (2018) 51-56.
J.Q. Lu, B. Adam, A.S. Jack, A. Lam, R.W. Broad, and C.L. Chik, Immune Cell Infiltrates in Pituitary Adenomas: More Macrophages in Larger Adenomas and More T Cells in Growth Hormone Adenomas. Endocrine pathology 26 (2015) 263-72.
P. Marques, S. Barry, E. Carlsen, D. Collier, A. Ronaldson, N. Dorward, J. Grieve, N. Mendoza, R. Nair, S. Muquit, A.B. Grossman, and M. Korbonits, The role of the tumour microenvironment in the angiogenesis of pituitary tumours. Endocrine 70 (2020) 593-606.
M. Principe, M. Chanal, M.D. Ilie, A. Ziverec, A. Vasiljevic, E. Jouanneau, A. Hennino, G. Raverot, and P. Bertolino, Immune Landscape of Pituitary Tumors Reveals Association Between Macrophages and Gonadotroph Tumor Invasion. The Journal of clinical endocrinology and metabolism 105 (2020).
A. Zhang, Y. Xu, H. Xu, J. Ren, T. Meng, Y. Ni, Q. Zhu, W.B. Zhang, Y.B. Pan, J. Jin, Y. Bi, Z.B. Wu, S. Lin, and M. Lou, Lactate-induced M2 polarization of tumor-associated macrophages promotes the invasion of pituitary adenoma by secreting CCL17. Theranostics 11 (2021) 3839-3852.
J.T. Yeung, M.D. Vesely, and D.F. Miyagishima, In silico analysis of the immunological landscape of pituitary adenomas. Journal of neuro-oncology 147 (2020) 595-598.
K. Lin, J. Zhang, Y. Lin, Z. Pei, and S. Wang, Metabolic Characteristics and M2 Macrophage Infiltrates in Invasive Nonfunctioning Pituitary Adenomas. Frontiers in endocrinology 13 (2022) 901884.
C. Tang, X. Lei, L. Xiong, Z. Hu, and B. Tang, HMGA1B/2 transcriptionally activated-POU1F1 facilitates gastric carcinoma metastasis via CXCL12/CXCR4 axis-mediated macrophage polarization. Cell Death Dis 12 (2021) 422.
A. Martinez-Ordoñez, S. Seoane, P. Cabezas, N. Eiro, J. Sendon-Lago, M. Macia, T. Garcia-Caballero, L.O. Gonzalez, L. Sanchez, F. Vizoso, and R. Perez-Fernandez, Breast cancer metastasis to liver and lung is facilitated by Pit-1-CXCL12-CXCR4 axis. Oncogene 37 (2018) 1430-1444.
A. Martínez-Ordoñez, S. Seoane, L. Avila, N. Eiro, M. Macía, E. Arias, F. Pereira, T. García-Caballero, N. Gómez-Lado, P. Aguiar, F. Vizoso, and R. Perez-Fernandez, POU1F1 transcription factor induces metabolic reprogramming and breast cancer progression via LDHA regulation. Oncogene 40 (2021) 2725-2740.
C. Sun, R. Mezzadra, and T.N. Schumacher, Regulation and Function of the PD-L1 Checkpoint. Immunity 48 (2018) 434-452.
P. Wang, T. Wang, Y. Yang, C. Yu, N. Liu, and C. Yan, Detection of programmed death ligand 1 protein and CD8+ lymphocyte infiltration in plurihormonal pituitary adenomas: A case report and review of the literatures. Medicine 96 (2017) e9056.
M.P. Salomon, X. Wang, D.M. Marzese, S.C. Hsu, N. Nelson, X. Zhang, C. Matsuba, Y. Takasumi, C. Ballesteros-Merino, B.A. Fox, G. Barkhoudarian, D.F. Kelly, and D.S.B. Hoon, The Epigenomic Landscape of Pituitary Adenomas Reveals Specific Alterations and Differentiates Among Acromegaly, Cushing's Disease and Endocrine-Inactive Subtypes. Clinical cancer research : an official journal of the American Association for Cancer Research 24 (2018) 4126-4136.
Y. Mei, W.L. Bi, N.F. Greenwald, Z. Du, N.Y. Agar, U.B. Kaiser, W.W. Woodmansee, D.A. Reardon, G.J. Freeman, P.E. Fecci, E.R. Laws, Jr., S. Santagata, G.P. Dunn, and I.F. Dunn, Increased expression of programmed death ligand 1 (PD-L1) in human pituitary tumors. Oncotarget 7 (2016) 76565-76576.
P.F. Wang, T.J. Wang, Y.K. Yang, K. Yao, Z. Li, Y.M. Li, and C.X. Yan, The expression profile of PD-L1 and CD8(+) lymphocyte in pituitary adenomas indicating for immunotherapy. Journal of neuro-oncology 139 (2018) 89-95.
J. Turchini, L. Sioson, A. Clarkson, A. Sheen, and A.J. Gill, PD-L1 Is Preferentially Expressed in PIT-1 Positive Pituitary Neuroendocrine Tumours. Endocrine pathology 32 (2021) 408-414.
S.P. Kubli, T. Berger, D.V. Araujo, L.L. Siu, and T.W. Mak, Beyond immune checkpoint blockade: emerging immunological strategies. Nat Rev Drug Discov 20 (2021) 899-919.
T. Maghathe, W.K. Miller, L. Mugge, T.R. Mansour, and J. Schroeder, Immunotherapy and potential molecular targets for the treatment of pituitary adenomas resistant to standard therapy: a critical review of potential therapeutic targets and current developments. Journal of neurosurgical sciences 64 (2020) 71-83.
M. Yi, D. Jiao, H. Xu, Q. Liu, W. Zhao, X. Han, and K. Wu, Biomarkers for predicting efficacy of PD-1/PD-L1 inhibitors. Mol Cancer 17 (2018) 129.
H.R. Kemeny, A.A. Elsamadicy, S.H. Farber, C.D. Champion, S.J. Lorrey, P. Chongsathidkiet, K.I. Woroniecka, X. Cui, S.H. Shen, K.E. Rhodin, V. Tsvankin, J. Everitt, L. Sanchez-Perez, P. Healy, R.E. McLendon, P.J. Codd, I.F. Dunn, and P.E. Fecci, Targeting PD-L1 Initiates Effective Antitumor Immunity in a Murine Model of Cushing Disease. Clinical cancer research : an official journal of the American Association for Cancer Research 26 (2020) 1141-1151.
B. Shum, J. Larkin, and S. Turajlic, Predictive biomarkers for response to immune checkpoint inhibition. Semin Cancer Biol 79 (2022) 4-17.
V. Suteau, A. Collin, P. Menei, P. Rodien, M.C. Rousselet, and C. Briet, Expression of programmed death-ligand 1 (PD-L1) in human pituitary neuroendocrine tumor. Cancer immunology, immunotherapy : CII 69 (2020) 2053-2061.
[Consensus on the immunohistochemical tests of PD-L1 in solid tumors (2021 version)]. Zhonghua Bing Li Xue Za Zhi 50 (2021) 710-718.

No competing interests reported.

s1.jpeg
Figure S1 Subgroup analysis of CD163+ M2-like macrophage infiltration and PD-L1 expression. (A) CD163+ M2-like macrophage infiltration was not different among all subtypes of PIT1-lineage PitNETs. (B) No difference was observed for PD-L1 expression in all subtypes of PIT1-lineage PitNETs.

Download PDF

Journal Publication

published 07 Jul, 2023

Read the published version in Journal of Neuro-Oncology →

Editorial decision: Major revision
13 Jun, 2023
Reviews received at journal
03 Jun, 2023
Reviewers agreed at journal
02 Jun, 2023
Reviewers invited by journal
31 May, 2023
Editor assigned by journal
31 May, 2023
Submission checks completed at journal
31 May, 2023
First submitted to journal
30 May, 2023

You are reading this latest preprint version

Tumor Immune Microenvironment in Pituitary Neuroendocrine Tumors (PitNETs): Increased M2 Macrophage Infiltration and PD-L1 Expression in PIT1-lineage Subset

Status:

Journal Publication

Version 1

Abstract

Purpose

Methods

Results

Conclusion

Figures

1. Introduction

2. Materials and Methods

2.1. Human tissue sample and clinicopathological data

2.2. Immunohistochemistry

2.3. Reclassification of PitNETs in databases according to the 2022 WHO classification system

2.5. Statistical analysis

3. Results

3.1. Clinicopathological features in the validation cohort of PitNETs

3.2. Heterogeneous immune cell infiltration patterns of PitNETs and NPs via in silico analysis

3.3. M2 macrophages were highly infiltrated in PIT1-lineage PitNETs

3.4. Enrichment of M2 macrophage infiltration was correlated with the tumor volume in PIT1-lineage PitNETs

3.5. PD-L1 was highly expressed in PIT1-lineage PitNETs

3.6. Highly expression of PD-L1 contributed to the tumor growth and CS invasion in PIT1-lineage PitNETs

4. Discussion

5. Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1