The co-inhibitory immune checkpoint proteins B7-H1(PD-L1) and B7-H4 in high grade glioma: From bench to bedside

Highlights

• •
  Review promising biomarkers which can predicting PD-1/PD-L1 blockade immunotherapy efficacy in high grade gliomas.

  • •
    Elaborate the current status of emerging strategies involving PD-1/PD-L1 inhibitors in high grade gliomas:
    Select potential immunotherapy responders by innovative biomarkers.
  • •
    Combination treatment.
  • •
    Neoadjuvant PD-1 therapy.
• •
  Brief overview of B7-H4 as a promising treatment target.

  • •
    Clinical significance.
  • •
    Regulatory mechanism.
  • •
    Developing immunotherapeutic strategies in Pre-clinical glioma models.

Abstract

Co-inhibitory immune checkpoints play a crucial role in tumor progression, and PD-1/PD-L1 inhibitor has been a breakthrough for treating multiple refractory tumors in last decade. Nevertheless, results of several phase III clinical trials of PD-1/PD-L1 inhibitor are unsatisfactory in high grade gliomas recently. This article reviews the promising biomarkers which can predict the efficacy of PD-1/PD-L1 blockade immunotherapy and current status of emerging strategies involving PD-1/PD-L1 inhibitors, especially the combination treatment and neoadjuvant PD-1 therapy in gliomas. In addition, B7-H4, one of the most promising immune checkpoints, is also briefly reviewed here for its clinical significance, regulatory mechanism and developing immunotherapeutic strategies in pre-clinical glioma models.

Graphical abstract

Introduction

The B7-CD28 superfamily, as vital immune checkpoint proteins in regulating T-cell activation and tolerance, play a significant role in the immune evasion during cancer initiation and progression [1]. Anti-tumor therapy targeting immune checkpoint proteins like PD-1 and CTLA-4 have achieved tremendous success in various cancer types, like melanoma and NSCLC among these years [2]. However, this enthusiasm about eliminating tumor was muted in gliomas. As the most common type of primary brain tumors which are characterized by high morbidity and mortality rates, numerous attempts have been made to develop immune checkpoint inhibitors (ICIs) for glioma treatment. But regrettably, most of these clinical trials were culminated in negative outcomes [[3], [4], [5]].

To address this issue, there are currently two main strategies. One is to predefine the patients who may response to current available immunotherapy, with one feasible method being to screen potential responders through several biomarkers. The other is continued in searching for effective treatment model, like combination regimen, adjuvant treatment or therapy which targets innovative therapeutic targets besides PD-1 or CTLA-4.

In this review, we will demonstrate how researchers try to eliminate glioma through these two strategies and the role B7-CD28 family played in glioma. We also attempt to analyze the underlying causes of unsuccessful outcomes observed in clinical trials of ICIs for glioma, as well as discern potential avenues of future research.

Co-inhibitory immune checkpoint protein superfamily in glioma

Co-stimulation/inhibitory signal is important to regulate the T-cell immune response. B7-CD28 superfamily are viewed as the most important protein molecules in T-cell regulation, with different components lead to different downstream effect. The B7–1/B7–2-CD 28/CTLA-4 pathway is the most well-known regulation proteins in regulating T-cell activity [1]. Besides these, new B7 and CD28 molecules have recently been discovered and new pathways have successively been delineated [[6], [7], [8]]. (detailed in Fig. 1)

Fig. 1.

The CD28 family presents immunoglobulin superfamily members that exhibit a sole immunoglobulin V-like domain. CD28 and CTLA-4 are characterized by a MYPPPY motif, which is crucial for binding B7–1 and B7–2. ICOS, on the other hand, possesses an FDPPPF motif, hence it binds to ICOSL, but fails to interact with B7–1 and B7–2. PD-1, acts as a receptor for PD-L1 and PD-L2, and may additionally bind other receptors on T cells that are yet to be identified, as indicated by the dotted arrows and the question mark. B7 family members are also immunoglobulin superfamily members, with both immunoglobulin-V-like and immunoglobulin-C-like domains. While one type of human B7-H3 displays one V-like and one C-like domain, the most prevalent form is a four-immunoglobulin extracellular domain featuring IgV–IgC–IgV–IgC, as showcased. (This figure is modified from Sharpe AH et al. [7]).

B7-H1, more well-known by the name PD-L1, now is the most widely used immunotherapy target. B7-H1 can negatively regulate T-cell functions by engagement with PD-1. Our previous study found B7-H1 expression in the cytoplasm and/or at cell surface of glioma cells [9]. Interestingly, B7-H1 was upregulated at the growing edge of glioblastoma multiforme (GBM). This research also find B7-H1 was significantly higher in high-grade gliomas than in low-grade ones and a negative correlation between B7-H1 expression with tumor-infiltrating CD8⁺ T cells. Several studies have shown that B7-H1 is upregulated in glioma cells and correlates with poor prognosis and low immune infiltration [[10], [11], [12]]. These findings indicating that B7-H1 can be a novel tumor marker and target for human glioma therapy.

B7-H4 (also known as B7x, B7S1, VTCN1) belongs to the B7 family as a physiologically potent inhibitor of effector T cells (such as CD8, TH1, TH17), and B7-H4 also promotes the immunosuppressive function of regulatory T cells (Tregs) in animal models [[13], [14], [15], [16]]. In gliomas, B7-H4 is generally expressed on the cell membrane of tumor cells and macrophages, but can also be detected in the cerebrospinal fluid as soluble B7-H4 [17]. Highly expressed B7-H4 is positively correlated to higher malignancy, poor prognosis and less infiltration of immune cells in gliomas [17]. Interestingly, B7-H4 is found to be mutually exclusive expressed with other immune checkpoints such as PD-1/PD-L1, which makes B7-H4 a potential target for glioma immunotherapy after the unsatisfactory result of PD-1/PD-L1 inhibitors [18].

Further research is currently underway on the B7 superfamily of proteins. Studies have revealed several B7 homologues, such as B7-H3, are expressed on non-immune cells including fibroblasts and endothelial cells, indicating new mechanisms for regulating T-cell response in peripheral tissues. Additionally, some B7 homologues have unknown receptors, highlighting the possibility of undiscovered immunoregulatory pathways [6,19,20].

Immunotherapy targeting co-inhibitory molecules

There are currently multiple clinical studies investigating immune therapies targeting co-inhibitory molecules in the field of gliomas and several representative ones are summarized in Table 1 [[3], [4], [5],[21], [22], [23], [24], [25], [26]]. The ICIs targeting the PD-1 and CTLA-4 molecules represent the mainstay of those that have entered clinical application phases. Although researchers regrettably find the success of ICIs in other tumor types has not been replicated in gliomas, there are still some good news. Some positive results can be observed in these studies, such as certain subgroups of patients benefiting in large clinical trials [3,21,22], which suggests a need for effective methods to identify potential immunotherapy responders. New treatment modalities, such as combination therapy and neoadjuvant therapy, which has been proved can alter the tumor microenvironment to facilitate the efficacy of immunotherapy, also worth further explore. Furthermore, intracranial administration and the development of new therapeutic targets are promising directions for future exploration.

Table 1.

Representative clinical trials targeting co-inhibitory molecules in glioma.

Title	Subject	Phase	Study design	Intervention	Study size	Main results
Phase III trial of chemoradiotherapy with temozolomide plus nivolumab or placebo [3] CheckMate 548 NCT02667587	newly diagnosed GBM with methylated MGMT promoter	III	Triple blinding, multicenter, randomized study	RT + TMZ combined with the immune checkpoint inhibitor nivolumab (NIVO) or placebo (PBO)	716 patients randomized	NIVO added to RT + TMZ did not improve survival in patients with newly diagnosed glioblastoma with methylated or determinate MGMT promoter. mPFS (blinded independent central review) was 10.6 months (95% CI, 8.9–11.8) with NIVO + RT + TMZ vs 10.3 months (95% CI, 9.7–12.5) with PBO + RT + TMZ (HR, 1.1; 95% CI, 0.9–1.3) and mOS was 28.9 months (95% CI, 24.4–31.6) vs 32.1 months (95% CI, 29.4–33.8), respectively (HR, 1.1; 95% CI, 0.9–1.3). No new safety signals were observed.
Radiotherapy combined with nivolumab or temozolomide for newly diagnosed glioblastoma with unmethylated MGMT promoter [4] CheckMate 498 NCT02617589	newly diagnosed GBM with unmethylated MGMT promoter.	III	Open label, multicenter, randomized study	nivolumab (NIVO) + RT compared with TMZ + RT	560 patients were randomized, 280 to each arm.	The study did not find a survival beneficial in NIVO + RT arm when compared with TMZ + RT. TMZ + RT demonstrated a longer mOS than NIVO + RT. Median OS was 13.4 months (95% CI, 12.6 to 14.3) with NIVO + RT and 14.9 months (95% CI, 13.3 to 16.1) with TMZ + RT (hazard ratio [HR], 1.31; 95% CI, 1.09 to 1.58; P = 0.0037). Median PFS was 6.0 months (95% CI, 5.7 to 6.2) with NIVO + RT and 6.2 months (95% CI, 5.9 to 6.7) with TMZ + RT (HR, 1.38; 95% CI, 1.15 to 1.65). No new safety signals were detected with NIVO in this study.
Nivolumab versus bevacizumab [5] CheckMate 143 NCT02017717	Recurrent GBM	III	Open label, multicenter, randomized study	nivolumab vs bevacizumab	369 patients randomized	No difference in outcome between the nivolumab or bevacizumab arm Median OS was 9.8 months with nivolumab and 10.0 months with bevacizumab (NS), and the 12-mo OS rate was 42% in both arms. Median PFS was 1.5 months for nivuolumab and 3.5 months for bevacizumab ORRs were 8% for nivolumab and 23% for bevacizumab No steroid use and MGMT promoter methylation were associated with longer OS in the nivolumab arm versus the bevacizumab arm
Treatment with pembrolizumab in PD-L1 positive recurrent glioblastoma: KEYNOTE-028 trial. [21] NCT02054806	Adult recurrent glioblastoma patients with PD-L1-positive tumors	I	Single arm study	Pembrolizumab	26 patients enrolled	median follow-up of 14 months (range, 2–55 months) overall response rate was 8% (95% CI, 1%−26%). Two partial responses, lasting 8.3 and 22.8 months, occurred. Progression-free survival (median, 2.8 months; 95% CI, 1.9–8.1 months) rate at 6 months was 37.7%, and the overall survival (median, 13.1 months; 95% CI, 8.0–26.6 months) rate at 12 months was 58%. Correlation of therapeutic benefit to level of PD-L1 expression, gene expression profile score, or baseline steroid use could not be established. Pembrolizumab monotherapy demonstrated durable antitumor activity in a subset of patients with manageable toxicity in this small, signal-finding, recurrent glioblastoma cohort. Future studies evaluating rationally designed pembrolizumab combination regimens may improve outcomes in patients with recurrent glioblastoma.
Pembrolizumab monotherapy for MSI-H/dMMR recurrent GBM Keynote-158 study [22]	Adults with MSI-H/dMMR recurrent glioma	II	Open label, multicenter study	Pembrolizumab	21 patients enrolled	Among 21 enrolled patients, all received prior temozolomide, 12 (57%) received prior bevacizumab, and 14 (67%) received ≥ 2 prior lines of therapy. Median time from first dose to data cutoff (January 12, 2022) is 50.0 months; 19 (90%) patients discontinued treatment. 1 patient with GBM had PR (ORR, 4.8% [95%CI, 0.1%‒23.8%]); duration of response was 18.9 months, PFS 29.2 months, and OS 32.7 months. 3 (14%) patients had SD; PFS was 23.2, 14.5, and 3.3 months; OS was 23.2, 15.1, and 9.1 months. Overall, median PFS and OS (95%CI) were 1.4 (1.0‒2.1) months and 5.6 (2.6‒16.2) months. Treatment-related AEs occurred in 7 (33%) patients (grade 3/4, n = 1; no grade 5) and led to discontinuation in 1 (5%) patient.
Randomized Phase II and Biomarker Study of Pembrolizumab plus Bevacizumab versus Pembrolizumab Alone for Patients with Recurrent Glioblastoma [23]. NCT02337491	bevacizumab-naïve patients with recurrent glioblastoma	II	Open label, multicenter, randomized study	Pembrolizumab plus Bevacizumab vs Pembrolizumab Alone	80 patients enrolled	80 patients were randomized to pembrolizumab with bevacizumab (cohort A = 50) or pembrolizumab monotherapy (cohort B = 30) For cohort A, PFS-6 was 26.0% [95% confidence interval (CI), 16.3–41.5], median overall survival (OS) was 8.8 months (95% CI, 7.7–14.2), objective response rate (ORR) was 20%, and median duration of response was 48 weeks. For cohort B, PFS-6 was 6.7% (95% CI, 1.7–25.4), median OS was 10.3 months (95% CI, 8.5–12.5), and ORR was 0%. Tumor immune markers were not associated with OS, but worsened OS correlated with baseline dexamethasone use and increased posttherapy plasma VEGF (cohort A) and mutant IDH1, unmethylated, and increased baseline PlGF and sVEGFR1 levels (cohort B). The NANO scale contributed to overall outcome assessment. Pembrolizumab was ineffective as monotherapy or with bevacizumab for recurrent glioblastoma.
Circulating Immune Cell and Outcome Analysis from the Phase II Study of PD-L1 Blockade with Durvalumab for Newly Diagnosed and Recurrent Glioblastoma [24]. NCT02336165	MGMT unmethylated newly diagnosed patients and recurrent patients	II	Open-label, multicenter, non-randomized study	Durvalumab(PD-L1 blockade)	162 patients enrolled	MGMT unmethylated newly diagnosed patients received radiotherapy plus durvalumab (cohort A; n = 40). Bevacizumab-naïve, recurrent patients received durvalumab alone (cohort B; n = 31) or in combination with standard bevacizumab (cohort B2; n = 33) or low-dose bevacizumab (cohort B3; n = 33). Bevacizumab-refractory patients received durvalumab plus bevacizumab (cohort C; n = 22). With a median follow-up for cohort A of 36.8 months, median OS were 15.1months. Median follow-up for cohorts B and B2 was over 36 months and non-evaluable for cohort B3. Median OS was 6.7 m, 8.7mand 9.3 m respectively. mOS for cohort C is 4.5months. PD-L1 blockade and combination with standard or reduced dose bevacizumab was ineffective in glioblastoma.
Intracerebral administration of CTLA-4 and PD-1 immune checkpoint blocking monoclonal antibodies in patients with recurrent glioblastoma: a phase I clinical trial [25]. NCT03233152	Recurrent GBM	I	Single-arm study	intracerebral (IC) administration of ipilimumab (CTLA-4 blockade) +nivolumab (PD-1 blockade) in combination with intravenous administration of nivolumab	27 patients were enrolled	Median OS is 38 weeks (95% CI: 27 to 49) with a 6-month, 1-year, and 2-year OS-rate of, respectively, 74.1% (95% CI: 57 to 90), 40.7% (95% CI: 22 to 59), and 27% (95% CI: 9 to 44). IC administration of NIVO and IPI following maximal safe resection of rGB was feasible, safe, and associated with encouraging OS.
Neoadjuvant anti-PD-1 immunotherapy promotes a survival benefit with intratumoral and systemic immune responses in recurrent glioblastoma [26]	Recurent GBM		Randomized control study	neoadjuvant and adjuvant therapy with pembrolizumab vs only adjuvant pembrolizumab	35 patients enrolled	Patients who were randomized to receive neoadjuvant pembrolizumab, with continued adjuvant therapy following surgery, had significantly extended overall survival compared to patients that were randomized to receive adjuvant, post-surgical programmed cell death protein 1 (PD-1) blockade alone. Neoadjuvant PD-1 blockade was associated with upregulation of T cell- and interferon-γ-related gene expression, but downregulation of cell-cycle-related gene expression within the tumor, which was not seen in patients that received adjuvant therapy alone.

Biomarkers for PD-1 immunotherapy in glioma

As previously stated, there is an urgent need for effective methods to identify subfraction patients who may benefit from immunotherapy. One feasible method is to screen patients by biomarkers. Multiple biomarkers have been investigated in the field of glioma.

PD-1/PD-L1

The PD-1 blockade therapy has become the most widely used immune checkpoint inhibitors, thus making PD-1 the first candidate molecule to be considered for identifying immunotherapy responders in glioma.

It has been found that subgroups with high expression of PD-1 in multiple tumor types (melanoma, lung cancer etc.) do exhibit better responses [27]. Nonetheless, it remains controversial whether this phenomenon can be extrapolated to glioblastoma. Because glioma is the kind of “cold” tumor with less immunological cells infiltrated, it is hard to calculate PD-1 index in tissue accurately [28]. Currently, there is a lack of robust clinical data to support the use of PD-L1 expression as a reliable biomarker for the effective immunotherapy of glioblastoma. Some researchers suggest that T-cell infiltration and indications of T-cell dysfunction may carry greater significance for an accurate prognosis of the response to immune checkpoint blockade [29]. Generally speaking, more clinical trials are necessary to fully understand the role of PD-1 as a reliable biomarker in characterizing therapeutic outcomes.

Tumor mutational burden (TMB)

Tumor mutational burden (TMB) represents the number of mutations per megabase (Mut/Mb) of the tumor genome. High tumor mutational burden (TMB-H) is a leading candidate biomarker for identifying potential ICIs treatment responder based on a mechanism that with a higher number of mutations, and consequentially an increase in the number of mutant proteins, which has neoepitopes and can worked as neoantigen, it is more likely to trigger a robust T cell response [30]. Moreover, TMB-H can induce elevated expression of immune checkpoint molecules, such as PD-L1, which can be specifically targeted by ICIs, consequently augmenting the treatment efficacy [31].

Several clinical studies have demonstrated that various cancer types with TMB-H exhibit a greater respond rate to ICIs treatment when compared with TMB-L ones, like melanoma, bladder cancer and non-small cell lung cancer [32,33]. In glioma, researchers also observe clinical benefit from PD-1 blockade in subset of patients with TMB-H [34].

On June 16, 2020, the Food and Drug Administration (FDA) granted accelerated approval to pembrolizumab (KEYTRUDA, Merck & Co., Inc.) for the treatment of adult and pediatric patients with unresectable or metastatic tumor mutational burden-high (TMB-H) [≥10 mutations/megabase (mut/Mb)] solid tumors, as determined by the targeted sequencing FoundationOne CDx (F1CDx) assay which profiles the total number of synonymous and non-synonymous mutations across 324 cancer-related genes, that have progressed following prior treatment and who have no satisfactory alternative treatment options [35].

The FDA's approval was based on a prospectively-planned retrospective analysis of 10 cohorts of patients with various previously treated unresectable or metastatic TMB-H solid tumors enrolled in a multicenter, non-randomized, open-label trial, KEYNOTE-158 (detailed in Table 1). In this research, a total of 102 patients (13%) had tumors identified as TMB-H. The ORR for these patients was 29% (95% CI: 21,39), and 6%(95% CI:5–8) in the non-TMB-high group [36].

Despite the relatively optimistic response rate observed in TMB-H subfraction patients of multiple cancer types and the FDA's approval of TMB-H as a pan-tumor biomarker for ICI treatment, there is still a controversy attitude among researchers regarding whether the benefits of ICIs in TMB-H tumors can be generalized to all cancer types, especially to those traditionally considered to be immunologically "cold" tumors [31]. In breast cancer, prostate cancer and glioma, which manifest no correlation between CD8 T-cell levels and neoantigen load, TMB-H tumors fail to achieve an ORR of 20% (ORR = 15.3%, 95% CI 9.2–23.4, P = 0.95) and show a markedly lower ORR than TMB-L tumors with an OR of 0.46 (95% CI 0.24–0.88, P = 0.02) [31].

This finding implies that a lower number of tumor-infiltrating T cells may represent a critical factor for the failure of immunotherapy with checkpoint inhibitors in the treatment of glioma, which is consistent with the opinions of previous researchers [37,38]. In gliomas, Touat et al. find that variability in TILs may stem from differences in the quality rather than the quantity of neoantigens, and the number of TILs may also highly varied in different types of gliomas with individual molecular characteristics [39]. We will delve further into this topic in the next dMMR/MSI-H section.

All these findings underscore the intricacy of utilizing TMB as a biomarker for glioma, particularly in the context of recurrent tumors characterized by a distinct mutational landscape resulting from the MMR deficiency together with temozolomide exposure. Furthermore, the determination of the optimal TMB threshold warrants meticulous deliberation. The utilization of high tumor mutational burden (TMB) with a threshold of ≥10 mutations per megabase (mut/Mb), as approved by the FDA, has engendered ongoing debates in the researcher [40]. Prior efforts to define TMB as a biomarker, however, adopted a higher cut-off value of >20 mut/Mb [41], [42], [43]. Additionally, in the KEYNOTE-158 clinical trial, a lower response rate was observed in patients with a TMB of 10–13 mut/Mb compared to those with a TMB of >13 mut/Mb, indicating that the chosen cut-off point may not be optimal for discerning between responders and non-responders.

A clinical trial is ongoing to evaluate the effectiveness of pembrolizumab in patients with recurrent malignant glioma with a hypermutator phenotype (NCT02658279).

dMMR/MSI-H

Mismatch repair deficiency(dMMR)/microsatellites instability high (MSI-H) is a promising biomarker for treatment with immune checkpoint inhibitors (ICIs) [44]. Among the mutations accumulate in MMR-deficient cells, small insertions and deletions (indels) at homopolymers (microsatellites) can cause frameshift mutations, are crucial for producing ‘high-quality’ neoantigens that are recognized by immune cells and potent immune response [39].

On May 23, 2017, the U.S. Food and Drug Administration granted accelerated approval to pembrolizumab (KEYTRUDA, Merck & Co.) for adult and pediatric patients with unresectable or metastatic, dMMR/MSI-H solid tumors that have progressed following prior treatment and who have no satisfactory alternative treatment options. The approval was based on data from 149 patients with MSI-H/dMMR cancers enrolled across five uncontrolled, multi-cohort, multi-center, single-arm clinical trials [45]. In the cohort which comprised of 21 patients with metastatic or unresectable dMMR/MSI-H glioma, following treatment with pembrolizumab, dosed at 200 mg every three weeks until disease progression or unacceptable toxicity, one patient with GBM showed a partial response (ORR, 4.8% [95%CI, 0.1%‒23.8%]). The duration of response was 18.9 months, PFS 29.2 months, and OS 32.7 months. Three (14%) patients had stable disease, with PFS and OS ranging from 3.3 to 23.2 months and 9.1 to 23.2 months, respectively [22].

This result suggests that a subgroup of patients with dMMR/MSI-H gliomas may benefit from PD-1 blockade, indicating a promising future for the use of dMMR/MSI-H as a biomarker to identify glioma responders to ICIs treatment. However, there are still several challenges to imply this biomarker into clinical practice for gliomas, one of which is the difficulty in standardized biomarker testing. According to FDA, approved testing methods include MMR protein immunohistochemistry, polymerase chain reaction (PCR) for MSI status, and/or next-generation sequencing. Unlike in colorectal cancers, where almost all MMR-deficient cases are MSI-H, MMR deficiency does not necessarily lead to MSI-H in gliomas. Single-cell whole-genome DNA sequencing (scWGS) found that only about half of tumor cells actually have an MSI-H phenotype and is not typically revealed by standard bulk sequencing or clinical MSI assays in MMR-deficient gliomas, which indicates that more of the homopolymer indels are subclones and below the detection limits of bulk sequencing in hypermutated gliomas [39].

Another important issue worth attention is the impact of the temozolomide(TMZ) use on subsequent immunotherapy. TMZ has been shown to induce DNA mutations and generate specific mutational signatures, including signature 11 [46]. Pre-clinical data also demonstrate that exposing MMR-deficient glioma cells to TMZ can induce hypermutation with signature 11 and lead to temozolomide resistance [39]. In previous MMR proficient colorectal cancer patients, inactivation of MMR genes, specifically MSH6 mutation, can be pharmacological accomplished through the use of temozolomide [47]. Considering the widespread use of TMZ for glioma treatment, sequential ICIs therapy in glioma patients who have undergone TMZ treatment may be a promising direction for treatment in the future. This treatment strategy may benefit a greater number of patients since the proportion of dMMR/MSI-H in primary gliomas is relatively low, while the dMMR/MSI-H rate increases in recurrent GBM as a result of the impact of temozlomide.

Combination treatment of PD-1 blockage immunotherapy in gliomas

The use of combination therapy in the treatment of glioblastoma is gaining attention due to the disappointing results of several large-scale phase III clinical trials for ICIs monotherapy [3], [4], [5]. Immune checkpoint inhibitors (ICIs) can be paired with non-immune-based therapies such as radiotherapy and chemotherapy, or with other immunotherapies to maximize clinical benefit.

Radiotherapy and chemotherapy can lead to tumor cell death and the releasement of tumor antigens, thus stimulating endogenous immune responses against the tumor. This phenomenon is known as “immunogenic cell death” [48]. Research also show that tumor cells exposed to sublethal doses of radiation or certain chemotherapeutic agents are more susceptible to T-cell-mediated killing. This process is known as “immunogenic modulation” [49,50]. Small-molecule-targeted therapeutics has been shown can alter the immune depression microenvironment in glioma [51,52]. Through these immunogenic consequences, ranging from immunogenic cell death to immunogenic modulation, anticancer therapy can maximize the clinical benefit for patients receiving combination therapies [53,54].

The effectiveness of combined immunotherapies is contingent upon an adequate quantity and longevity of cytotoxic T lymphocytes that respond to tumor antigens. Presently, three predominant strategies for combined therapy are under investigation, which include the administration of two or more immune checkpoint inhibitors, the concomitant use of PD-1 blockade with agents that reduce the T cell activation threshold, and the use of agents targeting metabolic pathways in T cells [2].

Combination with a tumor vaccine regimen is another potential approach to convert “cold tumors” like glioblastoma that do not respond to PD-L1 inhibition due to a lack of infiltrating T cells, into “hot tumors” which are suitable for PD-L1/PD-1 blockade treatments [54].

Neoadjuvant PD-1 therapy in gliomas

The use of neoadjuvant therapy of PD-1 provides promise direction for the treatment of GBM. PD-1 blockade in neoadjuvant settings help in systemic priming and expansion of exhausted T cells within the GBM microenvironment [55], followed by the removal of an immune suppressive tumor microenvironment through surgical resection.

Neoadjuvant PD-1 blockade can enhance anti-tumor immune responses through at least the following mechanisms [26,56]. First, it can release PD-1/PD-L1 checkpoint, enabling modulation of the T cell receptor clonotypes with systemic activation and clonal selection of tumor-specific T cells. By contrast, in adjuvant-only settings, surgery occurs before checkpoint blockade release. Because of the reduced residual antigenic burden, TCR modulation is less robust and fewer tumor-specific T cells are activated. With fewer tumor-specific T cells, the remaining tumor cells are able to proliferate at a more rapid pace. Furthermore, such T cell activation in turn upregulates interferon-γ-related signaling, and consequently induces downregulation of cell cycle-related gene expression within tumor cells, enabling a therapeutic window and resulting in a survival benefit. After surgery, and with continued adjuvant anti-PD-1 monoclonal antibody administration, tumor-specific T cells continue to eliminate residual tumor cells and begin transitioning toward a T memory phenotype.

Neoadjuvant therapy has already led to several inspiring results in GBM patients, underscoring the importance of exploring this approach further. Cloughesy et al. [26] clinical trials have found that neoadjuvant administration of pembrolizumab in combination with adjuvant therapy following surgery significantly extends overall survival and progression-free survival compared to patients who received adjuvant therapy alone. Neoadjuvant PD-1 blockade is also associated with upregulation of T cell and interferon-γ-related gene expression, and downregulation of cell cycle-related gene expression within the tumor, which is not observed in patients that received adjuvant therapy alone. These data were concordant with the study by Schalper et al. and Melero et al. [56]. In addition, Lee et al. [69] found neoadjuvant PD-1 blockade in recurrent GBM patient could induce systemic T cell activation, upregulated genes expression related to DC trafficking and also increased proportion of IFN-γ activated, crosspresenting DC subsets in the tumor microenvironment. Although this study failed to overcome the immunosuppressive tumor environment dominated by microglia and macrophages, it provided us a chance to find the interaction between PD-1 inhibitors, DCs, T cells, and IFN-γ. Therefore, considering the tumor microenvironment remodeling in recurrent GBM treated with aPD-1 mAb neoadjuvant therapy, our ongoing clinical trial (NCT04888611) aims to investigate the clinical feasibility and efficacy of the A2B5+ glioma stem cell like antigen-loaded dendritic cell vaccination (GSC-DCV) combined with aPD-1 mAbs neoadjuvant therapy in recurrent GBM (IDH1/2 -), as well as the underlying mechanisms involved (Fig. 2).

Fig. 2.

Schematic diagram of a randomized, controlled, double-blind phase IIb trial using neoadjuvant PD-1 blockade alone or combined with GSC-DCV in patients with recurrent glioblastoma (IDH1/2 -). All patients received preoperative aPD-1 mAbs treatment 14±5 days before surgery. Following surgery, eligible patients were randomized in a 1:1 ratio to receive either a combination treatment of aPD-1 mAbs with GSC-DCV or aPD-1 mAbs with placebo. The above-mentioned postoperative therapy was given every three weeks until disease progression or unacceptable toxicity.

These results suggest that neoadjuvant immunotherapeutic approaches and rational combinations may be beneficial for understanding how immunotherapeutic interventions can influence the tumor microenvironment and improve patient outcomes. Therefore, neoadjuvant administration of PD-1 blockade may represent a more efficacious approach to treating this uniformly lethal brain tumor.

Regulation of B7-CD28 superfamily in cancer cells and immune cells

The ‘two-signal’ model of lymphocyte activation was proposed as an explanation for T-cell activation and tolerance. The study on 'signal 2′, which is commonly attributed to co-stimulation molecules, has gained increasing attention and numerous studies have uncovered the role of the B7-CD28 superfamily in either negatively or positively regulating immune cells, as well as its involvement in cancer initiation and progression. Recently, cutting-edge researches have put their focus towards elucidating the metabolic regulatory functions of PD-1/PD-L1 in immune cells and cancer cells, as well as the glycosylation biology of the B7 superfamily, particularly B7-H4. Additional information can be found in TableS1.

The emerging role of B7-H4 in gliomas

The clinical significance of B7-H4 in gliomas

Contrary to the minimal expression of B7-H4 in normal tissues, more than half of the gliomas are B7-H4 positive and nearly 20% are B7-H4 high expression detected by immunohistochemistry [18]. Increased expression of B7-H4 on glioma cells is associated with higher grade, more malignant molecular features, and poorer prognosis [17,18]. Besides, higher expression of B7-H4 on glioma macrophages and microglia indicates both shorter OS (overall survival) and PFS (progression-free survival) of glioma patients [17]. Moreover, B7-H4 expression is also positively correlated with the stemness of glioma cells [17,57].

The expression pattern of B7-H4 is highly associated with immune phenotype of gliomas. Distinct from PD-1/PD-L1, less immune cell infiltration including tumor infiltrating lymphocytes (TILs) and tumor associated macrophages (TAMs) is found in high expressed B7-H4 glioma tissues [18]. Although expression of other immune checkpoints such as CTLA-4, TIM-3 and LAG-3 are tightly associated with PD-L1, B7-H4 is rarely co-expressed with each of these immune checkpoints and demonstrates a mutually exclusive expression pattern in gliomas. In fact, only 2% of the glioma tissues showed co-expression of high B7-H4 and high PD-L1, which indicates B7-H4 as a promising immunotherapy target of glioma [18].

Mechanism of B7-H4 expression in gliomas

The exact mechanism that regulates B7-H4 expressing merely in normal tissues while broadly in glioma cells remains unclear. In peripheral tumors, upregulation of B7-H4 on tumor cells can be stimulated by hypoxia, immunosuppressive cytokines and several miRNAs [58,59]. Considering the immune distinctiveness of central nervous system, whether these stimuli can induce the expression of B7-H4 on glioma cells needs further investigation. In gliomas, B7-H4 expression is also upregulated on tumor infiltrating macrophages/microglia through JAK-STAT3 pathway induced by IL-6 secreting from glioma-initiating cells [17]. Although the B7-H4 ligands or receptors have not been determined yet, several mechanisms of B7-H4 regulating immune cells in tumor microenvironment have been found in peripheral tumor models. Increased B7-H4 expression in hepatocellular carcinoma can cause the exhaustion of tumor infiltrating CD8⁺T cells through Eomes upregulation, and blockade of B7-H4 can inversely promote the anti-tumor effects of CD8⁺T cells [60]. B7-H4 also promotes CD4⁺T cells converse into immunosuppressive Tregs through AKT-Foxo pathway and further reduces the efficacy of anti-CTLA-4 immunotherapy in multiple syngeneic tumor models [61]. Our recent research also find B7-H4 may induce the early dysfunction of tumor infiltrating CD8⁺T cells in glioma through downregulating AKT-eNOS pathway (Fig. 3). The immunosuppressive effect of B7-H4 on tumor infiltrating T cells is remarkably distinct from PD-L1, which high PD-L1 expression commonly associated with increased infiltrating T cells (so called immunologically ‘hot’ tumors) [18]. These differential patterns of B7-H4 and PD-L1 expression make B7-H4 as a potentially valuable biomarker for various immunotherapeutic strategies.

Fig. 3.

Interaction of B7-H4 between glioma cells, T cells and macrophage/microglia in glioma immune microenvironment. Currently unclear interaction is denoted by ‘?’ sign. B7-H4R, B7-H4 receptor; sB7-H4, soluble B7-H4.

B7-H4 based immunotherapeutic strategies in gliomas

Since the immunosuppressive effects of B7-H4 in tumor microenvironment have been discovered, immunotherapies targeting B7-H4 have achieved impressive success in multiple pre-clinical murine models [61,62]. Representative strategies including monoclonal anti-B7-H4 antibodies, bispecific antibodies and suppression of B7-H4 glycosylation [63], [64], [65]. In xenograft glioma murine model, blocking B7-H4 on glioma cells and macrophages also induces the increased T cell anti-tumor function and reduced glioma growth [17]. Similar to the PD-1/PD-L1 immunotherapies, the main obstacles in application of B7-H4 targeted immunotherapies from murine tumor models into clinical trials are the tremendous difference in the immunity between human and mice [66]. The immunological heterogeneity of glioma microenvironment may also restrict the efficacy of B7-H4 targeted immunotherapies in human gliomas. Therefore, newly developed glioma models including patient-derived xenografts, organoids and customized 3D models have been recently used in glioma immunological investigations [67]. Among these glioma models, we found 3D-bioprinted glioma models are especially suitable for T-cell related research due to their reproducibility, visualization and similarity of mimicking T cell infiltration in gliomas. Moreover, B7-H4 shows biomimicry of immunosuppressive functions in 3D-bioprinted glioma models compared with human glioma tissues and traditional orthotopic murine gliomas (Fig. 4).

Fig. 4.

Representative 3D-bioprinted glioma models in glioma immunological investigations. A-B, Schematic diagram of 3D-bioprinted glioma models involving glioma cells, microglia and T cells. C—H, Visualization of the process of immune cell infiltration in a representative 3D-bioprinted glioma model shows CD3⁺T cells and CD11B⁺microglia infiltrates from peripheral region into tumor region at day5 and uniformly distributed in the whole glioma model at day8. I-J, Both infiltrated CD8⁺T cells (CD8⁺TILs) and CD8⁺CD137⁺T cells (CD8⁺CD137⁺TILs) are remarkably decreased in 3D-bioprinted glioma model with high B7-H4 compared to model with low B7-H4.

Although early clinical trials of anti-B7-H4 antibodies have begun in patients with peripheral tumors, B7-H4 based immunotherapeutic strategies have not been applied in clinical trials in glioma patients. Nevertheless, B7-H4 can be a valuable biomarker to predict efficacy of dendritic cell vaccines treatment in glioblastoma from a phase II clinical trial [68].

Conclusion and future direction

In this article, we briefly review the co-inhibitory immune checkpoint proteins B7-H1(PD-L1) as well as B7-H4 in glioma. To date, the PD-1/PD-L1 blockade immunotherapy has showen promising efficacy in specific gliomas such as patients with dMMR/MSI-H, while PD-1/PD-L1 blockade alone failed to achieve satisfactory outcomes in large sample clinical trials. Concerning immunotherapy in high grade gliomas, emerging strategies involving combined immune-checkpoint treatment and drugs as well as vaccines anti-immunosuppressive microglia/macrophages or brain tumor stem cells, have the potential to remodel suppressive immune microenvironment in glioma patients, however, further investigations especially large sample clinical trials are warranted to confirm the efficacy of these strategies. Meanwhile, newly discovered immune checkpoints including B7-H4 also provide a promising target in glioma immunotherapy in pre-clinical studies and further researches are needed to confirm the feasibility of these newly discovered immune checkpoint blockade immunotherapy in glioma patients.

Funding

This work was funded by grants from National Key R&D Program of China (2022YFC3401600) and CSCO-MSD Cancer Research Program (Y-MSDPU2022–0246).

Institutional review board statement

Not included.

Informed consent statement

Not included.

Data availability statement

Not included.

CRediT authorship contribution statement

Ying Qi: Writing – review & editing, Writing – original draft. Xiaoming Huang: Writing – original draft. Chunxia Ji: Writing – review & editing. Chaojun Wang: Supervision. Yu Yao: Writing – review & editing, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.