Standard treatment for glioblastoma involves the DNA-alkylating agent TMZ, which acts by methylating DNA at the O6-guanine residue site (33). By methylating the DNA, TMZ induces DNA cross-linking and eventually cell apoptosis (33). Methylguanine methyltransferase (MGMT), a DNA repair enzyme, can remove the O6-methylguanine thus conferring resistance to TMZ. A subset of glioblastoma, however, express methylated MGMT which makes the repair enzyme inactive and leads to tumor cell TMZ sensitivity and prolongs patient survival (33). Recent work has suggested that adult gliomas demonstrate differential responses to TMZ depending on time of administration, suggesting a beneficial role for a chronotherapeutic approach to TMZ administration (20, 22, 23). One retrospective study on adult glioblastoma found that patients who received TMZ in the mornings had a 3.6 month longer overall survival than patients who received TMZ in the evenings (20). On further risk stratification, this survival benefit was extended to a 6 month increased overall survival in patients with O6-methylguanine-DNA methyltransferase (MGMT)-methylated glioblastoma, who received TMZ in the morning compared to MGMT-methylated patients treated in the evening (20). Given these findings, a follow-up phase II clinical trial was performed and demonstrated that chronotherapy with TMZ is feasible (21). Although this study found no difference in overall survival or adverse effects between patients treated with TMZ in the morning versus evening, the authors note that their small sample size and heterogenous patient population limit what can be concluded in terms of survival benefit (21). In summary, increased TMZ sensitivity in the AM may be driven by diurnal and differential expression of MGMT (34, 35) and may be a suitable target for larger studies in the future.

Another proposed mechanism of differential efficacy of TMZ in gliomas may relate to direct interaction of TMZ to circadian gene expression. One hypothesis is that TMZ sensitivity is specifically tied to the cyclic expression of BMAL1, the binding partner of CLOCK that helps drive transcription of downstream circadian genes (PERs and CRYs) and an essential part of the circadian clock (36, 37). Investigators used both primary human glioblastoma cells and primary mesenchymal murine glioblastoma astrocytes to show that cells are most sensitive to TMZ when it is administered near the daily peak in BMAL1 expression (22). This temporal effect disappeared after a CRISPR-mediated loss of BMAL1, suggesting that the chronotherapeutic sensitivity of TMZ is dependent on BMAL1 (22). Another study took a bioinformatics approach utilizing a database with information on drug sensitivity and gene expression profiles, demonstrating that higher expression of BMAL1 is significantly correlated to higher TMZ sensitivity (23). These findings link the TMZ chronotherapeutic outcomes to the molecular components of the circadian clock. Our understanding of chronotherapy for TMZ remains limited by the lack of large, prospective randomized control trials and future studies should explore how to best time TMZ administration to maximize benefit (38). One promising approach could involve the use of high-throughput sequencing modalities to delineate transient changes in circadian gene expression, which would allow for more accurate dosing of TMZ therapy. RNA sequencing analysis has previously been used to identify molecular pathways involved in TMZ resistance in glioblastoma and could similarly be used to explore circadian regulation of TMZ sensitivity (39, 40). By employing high-frequency output, clinicians may be able to appropriately time treatment to the tumor circadian rhythms in their patients.

Unlike TMZ, chronotherapeutic strategies for radiation therapy treatment of gliomas remain controversial. Broadly, the expression of circadian genes Per1 and Per2 are thought to modulate the efficacy of radiotherapy in the treatment of gliomas (24, 25). Per1 expression levels are found to modulate the transcription of a variety of genes, including p53 target genes and checkpoint components for DNA repair (25). Decreasing PER1 expression also results in decreased expression of the Chk2-P53 signaling pathway and C-Myc, which are integral components of DNA-damage repair and apoptosis, respectively (25). As a result, PER1 is seen to play an important role in regulating the DNA damage response and subsequent apoptosis caused by radiation (25). Zhu et al. demonstrated that downregulating PER1 in human glioblastoma cells in vitro using an shRNA lentivirus resulted in minor DNA damage and reduced apoptosis in response to radiation compared to controls (25). Similarly, Zhanfeng et al. found this same positive correlation in vivo, showing that high expression of PER2 in glioma murine tissue was associated with an increased sensitivity to radiation. Specifically, there were higher levels of apoptosis and lower levels of proliferation when radiation was delivered at peak PER2 levels versus trough PER2 levels (24). This suggests there is a benefit of optimizing timing of radiotherapy based on circadian cycling (24). However, the survival benefits seen in vitro in U343 glioma cells and in vivo in glioma-bearing rats were not found in humans. In a retrospective study of 109 patients with high-grade gliomas, Sapienza et al. found no significant difference in progression-free survival or overall survival between patients treated with radiotherapy in the morning versus the afternoon (26). This may be because: i) other mechanisms overpower a modest survival benefit of radiation timing in humans, ii) that the time of radiation effect on tumor cells is much longer than what would occur within a circadian cycle, iii) there are other confounders such as concordant steroid administration that may reset the circadian clock, or iv) that there is not a measurable effect in humans. Despite this finding, the outcomes and toxicity-related benefits of chronoradiotherapy have been shown in other cancers (41–43). For example, a retrospective review on patients with rectal cancer found that patients who received radiotherapy after 12:00 pm had improved response rates compared to patients who were primarily treated before 12:00 pm (43). Similar benefits to chronoradiotherapy were shown for prostate cancer (42) and for bone metastases (41). These conflicting results suggest the need for further exploration into the potential benefits of chronotherapeutic approaches to radiation therapy in malignant brain tumors. As mentioned above, modern sequencing technologies may prove useful in elucidating the role for chronotherapeutic approaches to radiotherapy. Past studies have utilized total RNA sequencing to understand mRNA expression changes underlying radiotherapy resistance in glioblastoma (44), thus a similar analysis of circadian gene expression may allow for effective chronotherapeutic radiation therapy.

The development of immunotherapies, such as chimeric antigen receptor (CAR) T-cells, was a promising advancement in treatment for high-grade gliomas, yet survival outcomes remain poor with immunotherapy. Barriers to immunotherapies such as CAR T-cell therapy in gliomas include off-target effects, poor infiltration into the tumor, and an immunosuppressive tumor microenvironment (45). CAR T-cell response can be enhanced by the addition of a synthetic agonist for the circadian gene Retinoid-related orphan receptor γ (RORγ) (46, 47). Hu et al. demonstrate that adding RORγ, a circadian regulator of BMAL1 (48), to melanoma tumor-specific T cells in vitro and transferring them to tumor-bearing mice results in reduced tumor growth and improved T cell survival (47). Similarly, in a separate study, they show that adding RORγ during ex vivo expansion of a patient’s immune cells increases the antitumor activity of T-helper 17 (Th17) cells that are engineered with a CAR (46). This antitumor activity is seen to persist long-term, with elevated levels of cytokines detected months after infusion (46). The antitumor activity of RORγ involves increasing production of cytokines such as Interleukin-17 and Granulocyte macrophage colony-stimulating factor, as well as co-stimulatory receptors tumor necrosis factor receptor superfamily member 9 (CD137) and Cluster of Differentiation (CD) 226 (47). RORγ also decreases immunosuppression by attenuating the activity of regulatory T cells (Tregs) and reducing expression of CD39, CD73, Programmed Cell Death Protein 1 (PD-1), and T cell immunoreceptor with Ig and ITIM domains (TIGIT) (47). These results offer promising evidence that manipulating the circadian clock may enhance the efficacy of CAR T-cell treatment. Future research should examine whether these findings extend to high-grade gliomas and whether time of administration affects the efficacy and side effect profile of CAR T-cell therapy.

Several small molecule inhibitors have limited efficacy in the treatment of gliomas, despite targeting key growth pathways. Given their molecular potential, their response may be enhanced through further regulation of the tumor’s circadian rhythm. These small molecular inhibitors target the epidermal growth factor (EGF), and mitogen-activated protein kinase (MAPK) pathways and may have maximal therapeutic effect by understanding the chronoregulation of gliomas.

Gliomas can modify their TME through angiogenesis, mediated through the vascular endothelial growth factor (VEGF) pathway and resulting in the generation of new blood vessels to deliver necessary nutrients for further tumor growth and expansion (64, 65). This process is thought to involve circadian regulation, according to results of a study by Pang et al. (66). In this study, authors utilize gene set enrichment analysis on patient glioblastoma samples from publicly available databases to show that angiogenesis is one of the most enriched pathways in glioblastoma tumors with high BMAL1 expression compared to those with low BMAL1 expression (66). Additionally, they use glioblastoma patient-derived cells to show that angiogenesis is greatly reduced in glioblastoma cells with depletion of either CLOCK or BMAL1 (66). Furthermore, the administration of SR9009, a selective androgen receptor modulator that activates the circadian gene Rev-Erbα, decreases angiogenesis in glioblastoma in vitro and reduces intratumoral blood vessels in vivo in murine glioma models (66). Together, these findings highlight the effect of circadian regulation on angiogenesis.

More specifically, circadian genes regulate the expression of several angiogenic factors. For example, the Pang et al. study suggests that CLOCK and BMAL1 drive expression of the pro-angiogenic factor periostin (POSTN) in endothelial cells (66). They propose a mechanism through which the CLOCK-BMAL1 complex regulates expression of the olfactomedin like 3-hypoxia-inducible factor 1-alpha (HIF-1α) axis, which upregulates POSTN and activates TANK-binding kinase 1 signaling in endothelial cells (66). Similarly, a study by Wang et al. found that expression of BMAL1 positively correlates with microvascular density, as well as angiogenic factors HIF-1α, Angiopoietin 2 (ANG2), and VEGF in gliomas (67). Correspondingly, inhibition of BMAL1 in vitro using primary glioma cells results in decreased expression of HIF-1α and VEGF (67). It is likely that BMAL1 promotes angiogenesis through its modulation of these angiogenic factors. Similar results are observed in other cancer types. For example, overexpression of CLOCK in human colorectal carcinoma cell lines correlates with increased expression of angiogenesis-related genes such as HIF-1α, ARNT and VEGF, with CLOCK knockdown showing the opposite results (68). A study in sarcoma and melanoma showed circadian rhythms in VEGF expression are regulated by Period and Cryptochrome1, and in vivo chronotherapeutic administration of anti-angiogenic agents during early light phase decrease tumor growth to a greater degree than administration during early dark phase (69). Additionally, a study in zebrafish embryos showed that BMAL1 positively regulates VEGF expression while PER2 negatively regulates expression (70). Altogether, these findings elucidate the mechanism underlying circadian regulation of angiogenesis through angiogenic factors such as HIF-1α and VEGF, and they highlight the therapeutic potential of targeting this pathway (Figure 2A).

The epithelial-mesenchymal transition (EMT) allows glioma cells to migrate and invade into surrounding tissue (64) and circadian regulation of EMT may provide a novel target for manipulation of the glioma microenvironment to inhibit the invasive nature of high-grade gliomas. In glioblastoma tissue and in cell culture, CLOCK was found to positively regulate glioblastoma migration by upregulating activity of transcription factor nuclear factor-κB (NF-κB), thus promoting mesenchymal differentiation (18, 71). NF-κB is a ubiquitous transcription factor, and one of its roles in glioblastoma involves inducing expression of mesenchymal transcription factors and mesenchymal proteins such as CD44, vimentin, and N-cadherin (71). Additionally, a study by Yu et al. in glioblastoma cells found that knocking down the nuclear receptor REV-ERBß, a repressor of circadian genes, appears to suppress EMT and metastasis of glioblastoma cells (72). Similarly, circadian regulation of EMT has been shown in other cancers such as colorectal cancer. Specifically, increased expression of CLOCK or BMAL1 in colorectal cancers correlates with increased mesenchymal markers and decreased epithelial markers (73). Conversely, silencing CLOCK or BMAL1 has the expected opposite effect, downregulating mesenchymal markers and increasing epithelial ones (68, 73). In breast cancer cells, reduced expression of the circadian gene PER2 was found to correlate with increased expression of the pro-EMT genes Snail Family Transcriptional Repressor 2 (SLUG), Snail Family Transcriptional Repressor 1 (SNAI1), and Twist-related protein 1 (TWIST1) (74). These results suggest a link between the circadian system and increased EMT in various cancer models (Figure 2B). Future studies should determine whether this mechanism is important in glioblastoma as well.

The circadian clock plays a critical role in regulating the immune system through modulation of cytokine expression such as proinflammatory cytokine interleukin-1ß (IL-1ß) (75). Inflammation in the context of cancer is often tumor-promoting by supplying growth factors, proangiogenic factors, and enzymes that modify the extracellular matrix (64). Tumor-derived lactate plays several roles in tumor development, including serving as a proinflammatory mediator that increases cytokines such as IL-1ß, and lactate dehydrogenase A (LDHA) is integral in producing lactate (75, 76). Circadian genes CLOCK and BMAL1 are found to regulate expression of IL-1ß and LDHA in gliomas, with suppression of CLOCK and BMAL1 leading to a reduction in LDHA and IL-1ß levels (75). Similarly, knocking down BMAL1 or CLOCK diminishes the IL-1ß-induced increase in lactate and proinflammatory cytokines such as IL-8 and IL-6 (75). Increased expression of each part of this CLOCK-BMAL1- IL-1ß-LDHA feedback loop is correlated with poor prognosis and shorter survival (75). These findings therefore demonstrate an autocrine signaling loop that could be targeted to disrupt the dysregulated inflammation and metabolism that benefits glioma cells (Figure 2C).

Glucocorticoids, such as dexamethasone, are often used to alleviate cerebral edema and inflammation from intracranial tumors (77). Recently, there have been more studies linking chronopharmacology of glucocorticoid administration and the expression of circadian genes in the immune cells (78). For example, a study by Fonken et al. showed that glucocorticoids entrain circadian clock gene expression in microglia by inducing the expression of the Per1 gene (78). This study also showed that microglia demonstrate temporal fluctuations in inflammatory cytokines such as tumor necrosis factor α (TNFα), IL-1β and IL6 (78). These findings show that microglia demonstrate circadian rhythms in inflammatory responses, and these rhythms can be influenced by administration of glucocorticoids (78). Glucocorticoids are widely used among patients with all brain tumors, and these results suggest an important link between steroid administration and circadian-regulated changes in immune cells in the glioma TME. Future studies should examine whether there are differential effects in treatment response, based on what time of day the glucocorticoids are administered.

Recently, circadian markers such as CLOCK and BMAL1 have been tied to immune suppression in glioblastomas by increasing microglia infiltration. High CLOCK expression is positively correlated with an increase in microglia and a decrease in CD8 activated T-cells and dendritic cells (79). In addition, both CLOCK and BMAL1 expression correlate with expression of microglial markers (79) and the infiltrating microglia are biased towards the immunosuppressive (M2) phenotype (79, 80). Overall, by increasing the expression of immunosuppressive microglia in the TME, CLOCK and BMAL1 genes can effectively minimize the response of glioma to immunotherapies.

Because of their interference with circadian recruitment of microglia, these circadian genes can decrease immune suppression in glioma. For example, administration of SR9009, the REV-ERBα agonist that inhibits BMAL1 expression, decreases intratumoral immune-suppressive microglia in GSCs (80). Similarly, depletion of CLOCK or BMAL1 in murine models results in reduced infiltration of microglia and improved overall survival (79). These studies highlight circadian regulation of immune suppression in the glioma TME and demonstrate how this regulation can be targeted therapeutically.

A specific signaling pathway between GSCs and microglia, the CLOCK-olfactomedin-like 3 (OLFML3)-HIF1α-legumain (LMGN)-cluster of differentiation 162 (CD162) axis, has been identified and serves as a potential therapeutic target for glioblastoma (Figure 2D) (80). GSCs utilize CLOCK and BMAL1 to transcriptionally regulate LMGN and OLFML3, two chemokines that promote recruitment of microglia (79, 80). Expression of the circadian gene CLOCK specifically demonstrates a positive correlation with LMGN and OLFML3 in glioblastoma (79, 80). Inhibiting any part of this CLOCK–OLFML3–HIF1α–LGMN axis results in reduced immunosuppressive microglial recruitment and prolonged survival in glioblastoma murine models (80). Similarly, treatment with either SR9009 or anti-CD162 enhances survival in this glioblastoma mouse model in response to anti-programmed cell death protein 1 (PD1) therapy (80). This finding is significant because gliomas are known to upregulate immune checkpoint molecule programmed death-ligand 1 (PD-L1) in order to suppress the immune response and evade immunotherapies (81). Targeting the CLOCK–OLFML3–HIF1α–LGMN–CD162 axis reduces PD-L1 expression and augments anti-PD1 therapy in glioblastoma mice, overall highlighting the therapeutic potential of this pathway (80).

The circadian clock is associated with immune evasion in tumors (82). In particular, a study by Wu et al. found that the circadian clock is associated with an immune evasion phenotype (82). Using RNA-sequencing data to infer level of immune cell infiltration, they demonstrated a positive correlation of core circadian genes with infiltration of Tregs and Mast cells, but a negative correlation with infiltration of Th2 cells, Th1 cells, natural killer T cells, CD8 T cells, CD8 naïve T cells, and CD4 T cells (82). This finding shows the broad impact of the circadian clock on the immune cells that infiltrate the TME. Using bioinformatics approaches, the authors demonstrate that the disrupted circadian clock in cancer contributes to T-cell exhaustion through persistent elevation of inhibitory checkpoints (82). Specifically, there was a positive correlation between all clock genes and immune inhibitory checkpoints PD-L1, PD-L2, cytotoxic T-lymphocyte–associated antigen 4 (CTLA-4) and transforming growth factor beta (TGFB) (Figure 2D) (82). Treatments aimed at targeting the circadian system in gliomas may therefore diminish the ability of these tumors to evade the immune system.

As previously mentioned, MGMT plays a critical role in promoting TMZ-resistance in gliomas by repairing double-stranded DNA breaks (33). Studies by Damato et al. demonstrate a significant 6-month difference in survival between patients treated in the AM and PM, highlighting the role of the circadian system on TMZ sensitivity in MGMT-methylated gliomas (20). Further studies by Marchenay et al. and Martineau-Pivoteau et al. demonstrate circadian rhythms in MGMT protein activity in both human and mouse cells, respectively (34, 35). Marchenay et al. use serial blood samples across a 24-hour period from healthy volunteers to demonstrate circadian rhythms in MGMT activity in circulating mononuclear cells (34). Similarly, Martineau-Pivoteau et al. use liver samples obtained from mice at eight different circadian times to demonstrate circadian rhythms in MGMT activity, with the highest activity occurring during the active period (35). These findings demonstrate circadian regulation of MGMT-methylation, and understanding this cycling may allow for more precise timing of TMZ administration to maximize overall survival.

Mutations in isocitrate dehydrogenase 1 (IDH1) and isocitrate dehydrogenase 2 (IDH2) are associated with improved survival rates among patients with glioblastoma and are positive prognostic predictors of overall survival (83, 84). However, there are few studies examining the role of IDH mutations in tumorigenesis and in modulating the circadian cancer pathway. Although most studies exploring circadian regulation of gliomas do not stratify based on IDH mutation status, De La Cruz Minyety et al. found that PER gene expression predicts survival in high-grade glioma patients independently of IDH mutational status (84). Interestingly, another study of in vitro glioma cells demonstrated that IDH1 mutations are associated with lower expression of BMAL1, CLOCK, PER genes, and CRY genes compared to their wildtype counterparts (85). Together these findings suggest that IDH mutations may regulate circadian gene expression, but do not show clear evidence for circadian regulation of IDH mutational status. Given the paucity of research into this important prognostic factor for glioblastoma, further studies should elucidate whether IDH mutational status could be used as a marker for assessing the effectiveness of chronotherapy.

High-grade gliomas are known to express growth factors and growth factor receptors (86), and several malignant behaviors such as proliferation, invasion, angiogenesis, and decreased apoptosis involve growth factor signaling (87). For example, Insulin Growth Factor 1 (IGF-1) receptor signaling is one mechanism through which GSCs become resistant to radiotherapy (88). Specifically, acute radiation increases expression of IGF1R and secretion of IGF-1, which activates the phosphatidylinositol 3-kinase (PI3K)-Protein kinase B (Akt) pathway to prevent apoptosis and promote survival (88, 89). Recent work by Alonso-Gomez et al, Mazzoccoli et al, and Chaurdhari et al. revealed IGF-1 levels in the liver and serum exhibit circadian rhythms (90–92). Additionally, circadian regulation of IGF-1 has been shown to regulate cancer progression in non-small cell lung cancer cells (90–93). IGF-1 entrains the circadian clock in the liver, highlighting the two-way relationship between circadian-regulated targets and the circadian clock itself (94–96). Similarly, nerve growth factor (NGF) was found to entrain the circadian clock in hamsters, and epidermal growth factor (EGF) is seen to induce clock gene expression in neural stem cells (97–99). Likewise, fibroblast growth factor (FGF) and platelet-derived growth factor (PDGF) induce robust PER1 expression in murine fibroblasts (100). Together, these data show that regulation of growth factors represents another bi-directional link between the circadian clock and cancer biology. This link should be further researched and targeted to potentially decrease radiotherapy resistance in glioblastoma.