Introduction

One of the crucial treatment challenges of glioblastoma is the restrictiveness of the blood–brain barrier/blood-tumor barrier (BBB/BTB). Glioma progression has been linked to BTB (blood- disruption, causing increased permeability due to structural changes from angiogenesis, astrocytic end feet displacement, and neuronal death [1]. Specifically in high-grade gliomas, structural abnormalities across the BTB can lead to extravasation of blood contents, including solutes, antibodies, fluorescent markers, and immune cells [2]. However, the BTB demonstrates heterogeneous permeability. Specifically, high-grade gliomas have certain intact sections of BTB, which aid in continued tumor growth, invasion, and limitation of therapeutic drug penetration, leading to disease progression and treatment resistance [3]. To overcome these challenges, it is essential to identify BTB modulators that can transiently increase permeability and help the entry of chemotherapy or directed cytotoxic agents in such restricted areas.

The BBB is one of the main components of the neurovascular unit which includes endothelial cells, pericytes, and astrocytic end feet working harmoniously. In a healthy person, the BBB maintains the central nervous system (CNS) homeostasis by regulating the entry and exit of drugs, molecules, and toxins [4]. The key player of the BBB is endothelial cells, mesodermally derived, modified simple squamous epithelial cells that comprise blood vessel walls and are in the most direct contact with blood that circulates throughout the brain [5]. Mechanisms of BBB permeability are mostly centered around the expression and function of endothelial tight junctions along with membrane transport proteins, such as ATP binding cassette (ABC) transporters. These endothelial cells are held together by bicellular and tricellular junction proteins that aid in communication for paracellular transport of drugs, cells and immunologics. Bicellular junctions include the claudin proteins (claudin-1, claudin-3, claudin-5), zonula occluden (ZO-1, ZO-2) and tricellular junctions include (angulin-1/LSR, tricellulin/MarvelD2). Brain endothelial cells also express exceptionally high levels of nutrient transport proteins such as ABC transporters, mainly localized on the luminal side of the vasculature, to assist with pum** specific cytotoxic chemotherapies out of the cell [6, 7]. Both endothelial cells and malignant glioma cells have been reported to express high levels of ABC transporters, specifically P-glycoprotein (P-gp) (encoded by the—ABCB1 gene), ABCG2 and ABCC4 [8, 9]. Overall, these mechanisms are regulated through various modalities, including direct communication with the CNS and neuroimmune modulators released by nearby cells. The dynamic nature of the BBB allows for permeability adjustment in response to changes in the microenvironment, especially in the presence of malignant glioma cells [10, 11].

Ibrutinib is an FDA-approved B-cell lymphoma/lymphocytic leukemia agent that inhibits BTK (Bruton tyrosine kinase) activation, leading to decreased B-cell receptor signaling and decreased proliferative potential. In cardiac endothelium, ibrutinib inhibits vascular cell adhesion, platelet aggregation, and the associated inflammatory responses that occur during endothelial cell activation [https://boevalab.inf.ethz.ch/FREEC/) was applied to the local realigned and refined alignments over a range of ploidy (ploidy = 2, 3, 4, 5, 6, 7, 8). Based on the ploidy esitmate returned from this tool (output_ploidy = 7), variants were filtered to keep those observed to have a freqeuncy ≥ 14% and indicated by functional annotation to be non-synonomous. Surviving variants were then summarized in table form by variant type stratified by chromosome. Variants prior to filtering were also summarized in table form for select genes (Atrx, Egfr, H3f3a, H3f3b, Idh1, Idh2, Tert, Tp53).

Statistical analyses

Data are presented as mean ± standard error of the mean, unless otherwise noted. The Student t-test (2-tailed), and 1-way analysis of variance with Tukey’s test evaluated statistical significance with GraphPad Prism 9.3 software (GraphPad Software, Inc).

Results

High bruton tyrosine kinase (BTK) expression in grade IV human gliomas

Previous studies have demonstrated the high expression of BTK/BMX in glioblastoma [28, 28, 29]. While our brain endothelial cell viability studies exhibit no effects from ibrutinib, cell–cell interaction is reduced after 2 h, with notable recovery over the next 24 h. However, it was not a complete recovery, which could be due to a lack of systemic washout in this in vitro brain endothelial cell only assay. Specifically, junctional protein expression and down-stream ERK pathway proteins were found to be reduced following ibrutinib therapy, which agrees with previous studies showcasing BTK/BMX inhibition causing disruption of epithelial cell tightness [46, 47, 49]. A reduction of pERK has been shown to lead to reduced expression of tight junction proteins in the epididymis of mice, within the blood-testis barrier as reported by Kim et al. [50] Unfortunately, there is no one method to state whether the BBB is intact or not within these glioma models, thus evaluations of drug permeability and immunohistochemical staining of junctional markers is the best surrogate of such determinations. Combined, this data showcases ibrutinib’s ability to reduce expression of ZO-1 and other tight junctions resulting in associated attenuation of p-ERK pathway activation and ABCB1 inhibition. Previous studies have shown high BTK expression on glioma cells and ibrutinib’s efficacy to hinder lymphoma and other solid tumors progression through inhibition of the BTK/BMX pathway. Yet, no studies of ibrutinib’s effect on brain endothelium have been published, which could demonstrate a more complete picture of this drugs effects on the tumor microenvironment [12]. As such, these findings showcasing ibrutinib’s influence of brain endothelial cell–cell integrity and ABCB1 transporter function have larger implications for various neurologic, vascular and oncologic diseases. In our studies, we chose to explore the ABCB1 substrate doxil, and Zhou et al. found that ibrutinib’s use in human-derived glioma rodent models demonstrated additive cytotoxicity with etoposide (ABCB1 and ABCC1 substrate), which suggests that ibrutinib may target more than just one ABC transporter [16].

Some limitations of our studies include investigations of ibrutinib’s effects on the 1) tumor immune microenvironment, 2) long-term disease course with combination chemotherapy and 3) varied in vivo models. Our initial studies detailed the effect of ibrutinib on brain endothelium and tumor cells, as the former studies were novel and have the potential to positively benefit field knowledge related to CNS drug entry and impaired disease progression. While our studies did not explore effects of ibrutinib on the immune cells within the tumor microenvironment, previous studies have demonstrated ibrutinib’s ability to decrease glioma derived pericyte expression, which are cells known to create a formidable BBB but also demonstrate scavenger like functions when necessary [14]. In complement, Li et al., found that ibrutinib reduced CD8 + T cell exhaustion both in the in vitro setting and in BTK deficient mice, specifically by downregulating inhibitory receptors and increasing cytokine production [51]. The presence of T cell inhibitory signaling has been implicated in assistance with disease progression in glioblastomas thus further studies are warranted to further delineate the influence of BTK/BMX inhibition on T-cell infiltration, microglial behaviors and cytokine production within immunocompetent models [52].

Additionally, because S635 glioma cells orthotopically injected in immunocompetent hosts resulted in approximately 20–30 + days of survival with ibrutinib alone vs combination therapy, we were unable to assess long-term effects of ibrutinib therapy on disease course. Using another rat glioma model with less invasive quality and pronged survival without therapy could provide additional studies that help to explore systemic and neurologic sequelae from repeat ibrutinib therapy additive and/or synergistic with varied cytotoxic chemotherapy agents. These studies would assist in understanding more about the long-term effects of transient BBB disruption from ibrutinib as related to survival outcomes and tumor biology.

Other than S635, there only exists 9L, C6 and F98 rat derived glioma cells. We chose not to use 9L as it has previously been shown to be more akin histologically to gliosarcoma. C6 and F98 rat glioma lines, while most closely resemble (histologically and molecularly) glioblastoma, its growth patterns evoke alloimmunity or weak immunogenicity, respectively [53,54,55]. As such, its use has been cautioned for use with any evaluations that could aim to evaluate immunotherapeutics or agents that may influence the immune microenvironment [54]. As such, we opted to use a model with no limitations for evaluation. While the use of rodent models are highly necessary to advance to the neurooncologic field, there still exist restrictions within the suitability of models for translational applications.

Shi et al. reported that high BMX expression in glioma stem cells could be inhibited by ibrutinib resulting in decreased cell proliferation alone or with etoposide chemotherapeutic agent [14]. Comparatively, in this study, we found that ibrutinib has no significant effect on S635 rat glioma cell viability, yet it influenced endothelial junction proteins, inhibited efflux of ABCB1 and impaired migration. Classically, it is a known challenge to extrapolate in vitro dosing to rodent models due to the lack of dynamic fluidity as seen in a living system, thus we relied on published data from varied endothelial and cancer lines for treatment of our endothelial and glioma cells [12, 56, 57]. For additional clinical relevance, we investigated the additive effect of ibrutinib with and without doxil or doxorubicin treatment. Doxorubicin is a well-known substrate of both ABCB1 and ABCG2 and an effective chemotherapy drug that has been shown to hinder cancer proliferation of multiple solid tumors [58, 59]. However, doxorubicin cannot penetrate through the BBB easily and even with the assistance of ibrutinib we found no enhancement in survival or tumor volume with combined therapy (Additional file 2: Fig. 2). Doxil is the pegylated liposomal form of doxorubicin which confers a higher likelihood to cross the BBB, therefore these findings further point to the need to strategically select agents that can be paired with ibrutinib which are ABCB1 or potentially ABCC1 substrates and can impair glioma viability, migration and growth [60]. We evaluated free doxorubicin released from doxil within plasma and brain approximately 2 h after administration and found higher concentrations post ibrutinib. Yet, additional evaluations from 6 to 24 h after administration may aid determing differences in the brain and brain tumor settings regarding drug clearance with ibrutinib [24].

In our current study, we have provided evidence that ibrutinib influences brain endothelial integrity, efflux transport, and tumor progression. Our animal studies provide additional data in support of ibrutinib use to provide sustainable effects on BBB/BTB permeability and glioma cell propagation. Current clinical trials are ongoing to explore the use of ibrutinib with chemoradiation against glioblastoma which will provide further data on the use of this agent for varied treatment types (NCT03535350 and NCT05106296). Continued and novel research is needed to investigate ibrutinib’s efficacy with tumor directed agents, changes to the immune microenvironment and prolonged use for malignant gliomas; all with the aim of increasing CNS drug entry and improving disease survival.