Introduction

Glioblastoma (GBM) is the most common primary malignant brain tumor in adults, despite progress in current treatment approaches, prognosis remains poor. Hereditary factors relevant to GBM are not yet fully understood; receptor tyrosine kinase (RTK) signaling pathway is one of three major pathways involved in GBM. According to a retrospective biospecimen study of the Cancer Genome Atlas, mutations in epidermal growth factor receptor (EGFR)—an RTK—were observed in 45% of the samples (Cancer Genome Atlas Research 2008). Dysregulation of EGFR has also been identified in other types of cancer such as non-small-cell lung cancer, renal cancer, and colon cancer (Yewale et al. 2013). In addition to dysregulation of EGFR, aberrations in platelet-derived growth factor receptor α (PDGFRA) and mutations in vascular endothelial growth factor (VEGF)/VEGF receptor (VEGFR) are also frequently found in patients with GBM. According to the retrospective biospecimen study of the Cancer Genome Atlas, 13% of abnormalities were found in PDGFRA. Furthermore, as GBM is a solid tumor highly dependent on angiogenesis, aberrant tumor vascular network and augmented VEGF expression are observed in patients (Reardon et al. 2008).

As small molecule drugs, tyrosine kinase inhibitors (TKIs) constitute a majority of approved agents for several cancer types. Tyrosine kinase inhibitors work by blocking the receptor signaling, eventually inhibiting cell growth, proliferation, differentiation, and angiogenesis (Arora and Scholar 2005). Figure 1 below shows how a TKI regulates the cell cycle and acts as an inhibitor.

Fig. 1
figure 1

Receptor tyrosine kinase signaling pathway. ERK extracellular signal-regulated kinase, HIFα hypoxia inducible factor alpha, MEK mitogen-activated protein kinase, VEGF vascular endothelial growth factor

Full size image

Current treatment of GBM includes surgical resection followed by radiation therapy (RT) and temozolomide (TMZ) administration. The standard of care using TMZ significantly increases the median overall survival for patients with GBM by 2.5 months (median overall survival 12.1 months with RT alone vs. 14.6 months with RT plus TMZ), but the median overall survival remains considerably shorter than that in other types of cancer (Stupp et al. 2005). Tumor recurrence and therapeutic resistance after surgery are the main reasons for poor prognosis. Temozolomide has been a part of standard treatment for patients with newly diagnosed GBM since its first approval in the United States in 1999. However, no other drugs have replaced TMZ for newly diagnosed GBM because most drugs have failed to show efficacy in numerous clinical trials. Therefore, developing new drugs other than TMZ for GBM is crucial considering the limited life expectancy of patients with GBM. Small molecule TKIs are recognized as powerful candidates for treating GBM because they block cell signaling pathways such as EGFR, PDGFR, and VEGF/VEGFR. Moreover, small molecule TKIs are usually more affordable and place less financial burden on patients than biological drugs. Tyrosine kinase inhibitors are also more accessible than biological drugs as most TKIs are available as orally administered drugs, whereas biological drugs are not. The present review discusses small molecule TKIs with expanded indications, clinical trial status of TKIs in patients with GBM, and potential TKI candidates for GBM that have shown remarkable preclinical trial results.

Expanded indications for TKIs

As TK plays various roles in the life cycle of cells, many TKIs are used in multiple types of cancer, and it is common to expand indications for TKIs. Small molecule TKIs with expanded indications are summarized in Table 1.

Table 1 Expanded indications of tyrosine kinase inhibitors (TKIs)
Full size table

Imatinib (known as Gleevec®)—called as the “magic bullet”—is one of the most revolutionary drugs; it was first approved as a drug for treating chronic myeloid leukemia by the FDA in 2001. Chronic myeloid leukemia results from a reciprocal translocation between chromosomes 9 and 22 (“Philadelphia chromosome” or “Philadelphia translocation”). This chromosome is unusually short and contains a fusion gene called BCR-ABL. The advent of imatinib—an inhibitor of BCR-ABL—rapidly and dramatically modified the treatment of chronic myeloid leukemia and led to important changes in anti-cancer management.

Gastrointestinal stromal tumors (GISTs) are known as the most common mesenchymal tumors of the gastrointestinal system with an approximate incidence of 10–15 cases per million worldwide (Soreide et al. 2016). Gastrointestinal stromal tumors were considered life threatening owing to their poor response to chemotherapy and RT, until the discovery of activated KIT (CD117) and PDGFRA mutations in 1998 (Mei et al. 2018). Mutations in both receptors lead to dysregulation of downstream intracellular signaling pathways, eventually leading to tumorigenesis. However, structural homology among ABL kinase, KIT, and PDGFRA accelerated the introduction of imatinib in GIST treatment (Mei et al. 2018). Clinical trials in patients with GIST were encouraging, and in 2002, imatinib was approved by the FDA for the treatment of malignant metastatic and/or unresectable GISTs (Dagher et al. 2002). Today, 85–90% of patients with GIST harboring KIT or PDGFRA mutations benefit from imatinib after or before surgery.

Sunitinib (marketed as Sutent®) is an inhibitor of multiple RTKs including VEGFR type 1 and 2 (FLT1 and FLK1/KDR), PDGFR-α and PDGFR-β and the stem cell factor c-KIT receptor. Thus, it is not surprising that sunitinib was the first to be approved as a treatment of renal cell carcinoma, an extremely angiogenesis-dependent cancer (Rizzo and Porta 2017). Furthermore, sunitinib was also approved by the FDA for the treatment of GIST after disease progression or imatinib intolerance. Sunitinib was the first cancer drug to be simultaneously approved for two different indications in 2006.

In 2011, a multinational, randomized, double-blind, placebo-controlled phase III trial of sunitinib showed impressive results in 171 patients with neuroendocrine pancreatic tumors. The median progression-free survival (PFS) values for the sunitinib and placebo arms were 10.2 months and 5.4 months, respectively. Moreover, progression-free survival, overall survival, and objective response rate for sunitinib were also higher than those for the placebo (Raymond et al. 2011). Based on this clinical study, the FDA approved the expanded indications for sunitinib toward progressive neuroendocrine pancreatic cancer that cannot be removed by surgery or that is metastatic.

Unsuccessful clinical study outcomes with tyrosine kinase inhibitors

Owing to the broad availability of TKIs for multiple types of cancer, numerous clinical trials have been conducted for GBM, but no TKI has been approved for GBM until now. Table 2 summarizes negative clinical trials of TKIs for GBM.

Table 2 Tyrosine kinase inhibitors (TKIs) that showed negative results in clinical trials
Full size table

Details about the clinical trials presented in Table 2 (except the Imatinib phase II study; CST1571BGR03 and the Gefitinib phase II study; 1839IL/0116) were retrieved from clinicalTrials.gov, a web-based resource that provides researchers with easy access to publicly available information. The website is maintained by the National Library of Medicine at the National Institutes of Health, but it does not contain information about all the clinical studies conducted in the United States because not all studies are required by law to be registered. However, the rate of study registration has increased over time as more policies and laws requiring registration have been enacted and as more sponsors and investigators have voluntarily registered their clinical studies. Information about clinical trials of TKIs was filtered by the disease “Glioblastoma (GBM)” first; then, the results were re-filtered with the recruitment status “completed” or “terminated”.

All drugs in Table 2 are TKIs, but they can be classified into specific categories on the basis of their targets. Imatinib inhibits ABL, BCR-ABL, PDGFRA, and c-KIT (Iqbal and Iqbal 2014). Dasatinib is an inhibitor of multiple targets such as BCR-ABL, c-KIT, PDGFR-β, and ephrin receptor (Lassman et al. 2015). Vandetanib also inhibits multiple targets such as EGFR, VEGFR2, RET(rearranged during transfection)-tyrosine kinase, Protein tyrosine kinase 6, TIE2, EPH kinase receptor, and Src kinase receptor (Jeong et al. 2013; Lee et al. 2015).

Among the TKIs presented in Table 2, erlotinib, gefitinib, and afatinib have received extensive attention as EGFR TKIs because EGFR mutation is the most common type of mutation in patients with GBM. Erlotinib and gefitinib are classified as first-generation EGFR TKIs because they reversibly bind to only EGFR (Erb1), whereas afatinib is regarded as a second-generation TKI because it binds irreversibly to all ErbB family receptors. Clinical trials have been conducted with positive prospects in patients with glioblastoma multiforme because the TKIs showed positive clinical results in non-small cell lung cancer (NSCLC), which is also associated with EGFR mutations; however, all of them only showed limited efficacy in GBM. Recently, third-generation EGFR TKIs such as osimertinib (Tagrisso®), YH5448 (Lazertinib), and AZD3759 have been developed for NSCLC treatment. They have been tested only in preclinical studies for now; they will be discussed later.

Causes of negative clinical trial results are not yet clearly known, but several hypotheses are highly plausible. Drug concentrations in the sub-therapeutic range in the target organ—the brain—is the primary hypothesis. Altered EGFR signaling pathway is another plausible assumption. Another hypothesis is that the extensively heterogeneous cell types in GBM lead to limited efficacy with a single TKI. Lastly, a few TKIs present undesirable adverse effects during the studies. Detailed descriptions ensue.

Why are TKIs ineffective in GBM?

Restricted delivery of targeted agents across the blood brain barrier (BBB) is a critical cause of poor clinical trial results. Tyrosine kinase inhibitors must penetrate the brain; however, the BBB is a substantial obstacle to overcome. Most TKIs such as imatinib (Raymond et al. 2008), dasatinib (Schiff and Sarkaria 2015), vandetanib (Lee et al. 2015), and nintedanib (Muhic et al. 2013) are reported to be suitable substrates for P-glycoproteins, which are highly expressed in the capillary endothelial cells constituting BBB. They are also known to be substrates for breast cancer resistance protein (BCRP). P-glycoproteins and BCRP are both active efflux transporters that block absorption at the apical membrane of the BBB, and they pump drugs out of the cells.

The CYP3A4 enzyme-induced antiepileptic drug (EIAID) regimen frequently used for patients with GBM is reported to result in poor clinical trial outcomes. Drugs such as carbamazepine, phenytoin, and phenobarbital are enzyme-induced anticonvulsants drugs (EIACDs), which are often used for treating epilepsy in patients with GBM (Englot et al. 2016). As these EIACDs are inducers of cytochrome P450 isoenzyme CYP3A4, several TKIs mainly metabolized by CYP3A4 are significantly affected by EIACDs. Imatinib (Reardon et al. 2005), AEE788 (Reardon et al. 2012), gefitinib (Uhm et al. 2011), and lapatinib plus pazopanib (Iwamoto et al. 2010) are negatively affected by co-administration of EIACDs. It is plausible that substantially decreased plasma exposure of TKIs results in reduced plasma drug concentrations in the brain.

Adaptation to altered signaling pathways is another crucial reason for the negative clinical trial outcomes. Tyrosine kinase inhibitors selectively inhibit single or multiple receptor kinase targets; however, GBM cells may compensate by activating numerous TKs. Moreover, treatment with a single agent was not sufficient to completely block the cell signaling (Stommel et al. 2007). According to Muhic et al. (2013), glioma cells rely on other redundant pathways to support angiogenesis, as observed in several previous studies utilizing small molecule dual-VEGFR/PDGFR inhibitors (sunitinib, sorafenib, vatalanib, cediranib, and pazopanib). Thus, in future studies, it would be more effective to use TKIs in combination with agents that target downstream signaling components or reciprocal signaling components.

Unsuccessful clinical trials may also be attributed to the genetic heterogeneity of GBM. Glioblastoma cell growth is less dependent on a single oncogene (Brown et al. 2016); it shows intratumoral heterogeneity at the genetic and cellular level. Amplification of EGFR, MET and PDGFR are common in GBMs and these abnormalities differentially activate within GBM cell subpopulations (Snuderl et al. 2011). Therefore, further research combining appropriate targeted therapies using various TKIs is needed.

A few severe adverse effects were observed in the studies, which were not negligible. In a clinical study with AEE788, 28% of patients discontinued treatment because of adverse effects such as grade 3/4 aspartate aminotransferase or alanine aminotransferase elevation and bilirubin elevation (Reardon et al. 2012). Eventually, this study was prematurely terminated by the sponsor. Clinical research with dovitinib was discontinued in all patients owing to the unacceptable toxicities—including lipid abnormalities, AST/ALT elevation and thrombocytopenia—observed in 33% of participants (Sharma et al. 2019). A trial with erlotinib showed an unacceptably high number of deaths prompting early closure of the study. Four deaths occurred during the study, and three deaths were definitely related to the treatment: two patients developed refractory bone marrow aplasia and one developed pneumocystis carinii pneumonia, as confirmed by bronchoscopy (Peereboom et al. 2010).

Ongoing clinical trials in USA and South Korea

Despite the high clinical trial failure rates of TKIs, considerable effort is directed toward expanding the indications of many other TKIs. Global pharmaceutical companies perform several clinical trials of those TKIs for GBM. Current clinical trials in USA involving small molecule TKIs for GBM are summarized in Table 3. Ongoing clinical trials using TKIs in patients with GBM in South Korea are reviewed in Table 4.

Table 3 Ongoing clinical trials of tyrosine kinase inhibitors (TKIs) for glioblastoma (GBM) in the USA
Full size table
Table 4 Ongoing clinical trials of tyrosine kinase inhibitors (TKIs) for glioblastoma (GBM) in South Korea
Full size table

Information about current clinical trials in South Korea is not integrated; therefore, this review was drafted by exploring three separate websites: Korea Clinical trials information center (K-CLIC, https://www.koreaclinicaltrials.org/), Integrated drug safety information system of the Ministry of Food and Drug Safety (https://nedrug.mfds.go.kr/) and Clinical Research Information Service (CRIS, https://cris.nih.go.kr/cris/). Information about clinical trials was first filtered by the keyword “Glioblastoma” or “brain tumor” in the subject, and the information was re-filtered to ensure that the clinical trials included TKIs.

Approximately 23 clinical trials were first screened using the keyword “Glioblastoma” or “Brain tumor”; however, only two clinical trials remained at the end. This is because most of the ongoing clinical trials in South Korea comprise biologicals such as monoclonal antibody and immune cell therapy.

Lenvatinib (known as Lenvima®) was previously mentioned in Table 1 as it expanded indications from differentiated thyroid cancer to advanced renal cell carcinoma and unresectable hepatocellular carcinoma. Lenvatinib acts as a multiple kinase inhibitor against VEGFR1, 2, and 3 as well as FGFR1, 2, 3, and 4 and PDGFR-alpha. Its sponsor Merck Sharp & Dohme Corp. and collaborator Eisai Inc. are currently recruiting worldwide patients with conditions such as GBM, triple negative breast cancer, and ovarian cancer, to determine the safety and efficacy of lenvatinib (Lenvima®) plus pembrolizumab (Keytruda®). The ClinicalTrials.gov identifier number of the trial is NCT03797326, and it is also known as LEAP-005. The estimated enrollment number of patients is 180, and 16 patients will be enrolled in South Korea. The predicted primary completion date of this study is April 11, 2022.

Buparlisib (BKM-120) is an investigational drug and works as a pan-class I phosphoinositide 3-kinase inhibitor. This clinical trial targets patients with recurrent GBM using buparlisib plus carboplatin or lomustine. The estimated enrollment number of patients is 140, and 5 patients will be enrolled in South Korea.

Promising preclinical trials—every dark cloud has a silver lining

Because GBM is a rare disease, there have been more trials to expand the indications of a drug from brain metastases to GBM than trials to test a drug that only targets GBM. According to the Central Brain Tumor Registry of the United States, the incidence rate of GBM is 3.21 per 100,000 people in USA (Ostrom et al. 2018), whereas that in South Korea is 1.26 per 100,000 (Yun et al. 2019). The accurate incidence of brain metastases is unknown because of methodological limitations of population, hospital, and autopsy studies; it is estimated to be approximately 10 per 100,000 (Stelzer 2013), which is considerably higher than the incidence of primary GBM. Lung cancer and breast cancer are the most common sources of brain metastases in men and women, respectively. Approximately 10% of patients with NSCLC have brain metastases at diagnosis, and approximately 25–40% develop brain metastases during the disease (Abdallah and Wong 2018). Moreover, 5–15% of women with breast cancer are estimated to develop brain metastases (Graesslin et al. 2010). Therefore, apart from treating primary GBM, it is also important to treat brain metastases from NSCLC and breast cancer.

Although EGFR TKIs such as erlotinib (Tarceva®), gefitinib (Iressa®) and afatinib (Gilotrif®) are effective in treating NSCLC, all failed to show efficacy in GBM; however, preclinical data of the third-generation EGFR TKIs in brain metastases are promising.

Tyrosine kinase inhibitors that showed positive preclinical results are summarized in Table 5.

Table 5 Tyrosine kinase inhibitors (TKIs) with preclinical study results
Full size table

Osimertinib, also known as Tagrisso®, is a third-generation EGFR TKI, with sales of $1860 million in 2018. Moreover, it is also anticipated to be AstraZeneca’s largest selling drug in 2019. A preclinical study using EGFR-mutant NSCLC brain metastasis mouse model revealed that osimertinib is highly distributed in severely compromised immunodeficient mouse brain, to a similar extent as in the primary tumor (H1975), showing AUC tissue: plasma ratios of 1.7–2.8. Furthermore, osimertinib was more highly distributed in the mouse brain than gefitinib or afatinib; the brain: plasma Cmax ratio for osimertinib was 3.41, whereas the ratio was only 0.21 and < 0.36 for gefitinib and afatinib, respectively. The unbound brain-to-plasma partition ratio (Kpuu,brain) was 0.39 and 0.02 for osimertinib and gefitinib, respectively, but it could not be determined for afatinib (Ballard et al. 2016).

Another preclinical study of osimertinib was conducted in six different GBM cell lines and mice. It showed that osimertinib significantly inhibited the growth of the six GBM cell lines (U87, U251, U118, LN229, T98G, and LN18) in a dose-dependent manner, with IC50 values ranging from 1.25 to 3.00 µM. Notably, its inhibitory activity on GBM cell growth was 10-fold higher than that of either of the first-generation EGFR inhibitors. Moreover, survival of the GBM-bearing mice was significantly prolonged after treatment with osimertinib (Liu et al. 2019).

Lazertinib (YH25448) is also an oral, mutant-selective, irreversible third-generation EGFR TKI, which received enormous attention as Yuhan corporation licensed it out to Janssen in 2018. A preclinical study using EGFR-mutant brain metastasis mouse model reported that YH25448 inhibited intracranial tumor growth more effectively than osimertinib, at 10 and 25 mg/kg once daily during a 49-day treatment period. Although all mice were dead in the vehicle group at 42 days after tumor implantation, lazertinib-treated mice showed significantly longer survival than osimertinib-treated mice. Moreover, in the H1975-luc brain metastasis xenograft model, YH25448 demonstrated excellent BBB penetration. The Kpuu,brain value of YH25448 was 0.29 which is comparable to that of osimertinib, which was previously reported as 0.39. Moreover, YH25448 is not a substrate for BCRP and only a weak substrate for MDR1, suggesting that YH25448 may be less affected by drug efflux transporters (Yun et al. 2019).

AZD3759, another small molecule EGFR TKI, also presents good preclinical trial results with regard to BBB penetration in mice. A study in 2016 evaluating the efficacy of AZD3759 in lung cancer brain metastasis mouse model reported that the Kpuu,brain and Kpuu,CSF for AZD3759 were 0.65 and 0.42, respectively, which were significantly higher than those for erlotinib (Kpuu,brain, 0.13; Kpuu,CSF, 0.14). It has been reported that a Kpuu,brain of > 0.4 is necessary for good CNS penetration in humans. Moreover, the Kpuu,CSF values for AZD3759 in two human patients were 1.1 and 1.4, which strongly suggests that AZD3759 should be similarly effective both intracranially and extracranially (Yang et al. 2016).

Good BBB penetration does not always guarantee efficacy against GBM; however, considering the number of clinical trials that showed limited BBB infiltration (negative result), these preclinical data seem to be extremely promising.

Conclusions

In this review, we discuss small molecule TKIs that could be used for treating patients with GBM. We also outline plausible hypotheses for the negative clinical trial outcomes of many TKIs in patients with GBM. Although none of the TKIs have been approved for GBM treatment, several clinical and preclinical trials are currently ongoing. Tyrosine kinase inhibitors have the potential to treat GBM if they overcome certain limitations such as BBB penetration, adaptation to altered signaling pathways, and heterogeneity of GBM cells. The first small molecule TKI that can be used for GBM treatment has been anticipated since 1999.