Brain tumor-targeted drug delivery strategies

Abstract

Despite the application of aggressive surgery, radiotherapy and chemotherapy in clinics, brain tumors are still a difficult health challenge due to their fast development and poor prognosis. Brain tumor-targeted drug delivery systems, which increase drug accumulation in the tumor region and reduce toxicity in normal brain and peripheral tissue, are a promising new approach to brain tumor treatments. Since brain tumors exhibit many distinctive characteristics relative to tumors growing in peripheral tissues, potential targets based on continuously changing vascular characteristics and the microenvironment can be utilized to facilitate effective brain tumor-targeted drug delivery. In this review, we briefly describe the physiological characteristics of brain tumors, including blood–brain/brain tumor barriers, the tumor microenvironment, and tumor stem cells. We also review targeted delivery strategies and introduce a systematic targeted drug delivery strategy to overcome the challenges.

Graphical abstract

In this review, we briefly describe physiological and pathological conditions of brain tumors, including blood–brain/brain tumor barriers, tumor microenvironment and tumor stem cells; we review the corresponding targeting delivery strategies and introduce a systematic targeted drug delivery strategy to overcome current challenges.

1. Introduction

According to the GLOBOCAN 2008, the worldwide cancer incidence of malignant brain tumors is 3.5 per 100,000 people and about 650 people are diagnosed with malignant brain tumors every day¹. Brain tumors severely threaten human health due to their fast development and poor prognosis. Glioma, the most frequent primary brain cancer, accounts for 29% of all primary brain and CNS tumors and 80% of malignant brain tumors². The median overall survival of patients with glioblastoma (GBM) is only 14.6 months after current multimodal treatment–aggressive surgical resection followed by concurrent or sequential radiation and temozolomide chemotherapies³. One reason why poor prognosis and rapid recurrence are associated with this standard therapy is that the infiltrate growth of gliomas makes it difficult for the surgeon to completely remove pathologic or cancer-infiltrated tissues without affecting normal brain functions⁴. Furthermore, the failure is also ascribed to the side effects of radiotherapy and poor outcome of usual chemotherapy. For many years, researchers have endeavored to deliver therapeutic agents to the tumor region effectively and reduce unnecessary drug accumulation in normal brain and peripheral tissues. For brain tumors, active targeted drug delivery systems have attracted extensive attention in recent decades. Since brain tumors possess many distinctive characteristics from peripheral tumors due to their complicated oncogenesis, many factors must be taken into consideration for effective brain tumor-targeted drug delivery, such as the barriers included in the whole process, the tumor microenvironment, and tumor cells. Now various targets have been exploited to achieve the targeting therapy using nanocarriers. Herein we provide a brief review of several possible targeting delivery strategies for brain tumors.

2. Barriers to targeted drug delivery strategies

The oncogenesis of gliomas is complicated, with various barriers preventing drug from reaching the tumor sites. There are three main barriers for brain tumor treatment: the blood–brain barrier (BBB), the blood–brain tumor barrier (BBTB), and a relatively weak EPR effect. Specific brain tumor development stages require corresponding barrier targeting treatment strategies.

2.1. BBB targeting strategies and related drug delivery systems

At the early stage of brain tumor development and at the infiltration growth region of the tumor, the blood–brain barrier remains intact. The blood–brain barrier, which acts as a natural guard to protect the brain from harmful substances in the blood stream while supplying the brain with the necessary nutrients for proper function, is the key challenge for delivering drugs to brain tumor⁵. The BBB is a specialized system of capillary endothelial cells which are partially covered by pericytes and basement membrane, and almost fully surrounded by the end feet of astrocytes, preventing approximately 98% of the small molecules and nearly 100% of large molecules including recombinant proteins and genes from being transported into the brain and reaching the tumor sites6, 7. The BBB strictly limits drug transport into the brain by serving as a physical (tight junctions), metabolic (enzymes) and immunological barrier⁸.

To tackle this challenge, many kinds of active targeting strategies were adopted for developing effective drug delivery systems to the brain. The active targeting systems are mainly divided into absorptive-mediated transcytosis (AMT), transporter-mediated transcytosis, and receptor-mediated endocytosis (RMT)⁸.

2.1.1. Absorptive-mediated transcytosis

Absorptive-mediated transcytosis provides a means for the delivery of drugs across the BBB by cationic proteins or cell-penetrating peptides (CPPs). It is triggered by electrostatic interactions between the positively charged moieties of the proteins and negatively charged membrane surface regions on the brain endothelial cells⁹. Typical cationic bovine serum albumin-conjugated, pegylated nanoparticles (CBSA-NP) were prepared by Lu et al.¹⁰ for brain targeting. They demonstrated that the permeability of CBSA-NP was about 7.76 times higher than that of BSA-NP, which offered the possibility of delivering therapeutic agents to CNS. It was reported that plasmid pORF-hTRAIL (pDNA)-incorporated CBSA-NP (CBSA-NP-hTRAIL) colocalized with glycoproteins in brain and tumor microvasculature and accumulated in tumor cells at 30 min after i.v. administration to C6 glioma bearing nude mice, via absorptive-mediated transcytosis¹¹. Aclarubicin (ACL)-loaded cationic albumin-conjugated pegylated nanoparticles (CBSA-NP-ACL) could significantly prolong the survival of the intracranial glioblastoma-bearing mice¹². Du et al.¹³ adopted another cationic protein, wheat germ agglutinin (WGA) conjugated to the surface of liposomes and also demonstrated enhanced BBB transport.

Furthermore, alternative AMT-type cell-penetrating peptide (CPP)-based delivery systems show great ability in BBB transport. CPPs have been used to overcome the lipophilic barrier of cellular membranes and deliver a large variety of cargoes, including peptide/proteins, DNA/oligonucleotide, antibodies, imaging agents, toxins, and nanodrug carriers such as liposomes and micelles¹⁴. CPPs are heterogeneous in size and sequence and are positively charged. Some share common features such as an amphipathic sequence and the ability to interact with lipid membranes. The CPPs are always derived from natural proteins including the transcription-activating factor Tat, penetratin, and the Syn-B vectors, among which Tat might be the most frequently used¹⁵. Qin et al.¹⁶ covalently conjugated cell-penetrating peptide TAT (AYGRKKRRQRRR) to cholesterol for preparing doxorubicin-loaded liposomes for glioma therapy. The biodistribution in the brain and heart demonstrated higher efficiency of brain delivery and lower cardiotoxicity. The survival time of the glioma-bearing rats treated with TAT-modified liposomes was significant prolonged¹⁶. Moreover, self-assembled polymeric micelles modified with transcriptional activator TAT peptide (TAT-PEG-b-Col) were constructed by Liu et al.¹⁷ to successfully deliver antibiotics across the BBB. These studies show the potential of Tat-modified nanoparticle transport into the brain for the diagnosis and treatment of brain diseases¹⁸.

2.1.2. Transporter mediated transcytosis

Since there are many kinds of transport systems in the cerebral endothelium that provide the brain with the necessary nutrients and endogenous substances, transporter mediated transcytosis takes advantage of these transport systems as a promising brain targeting strategy. Transporter mediated transcytosis is substrate-selective, so only drugs that closely mimic the endogenous substrates will be taken up and transported into the brain⁶.

Glucose transporters (GLUT), which facilitate the transport of glucose from the blood to the brain, have a broad prospective use in brain targeting. Liposomes that incorporated a mannose derivative were able to cross the BBB via the glucose transporter GLUT1 in mouse brain¹⁹. Qin et al.²⁰ synthesized a new glycosyl derivative of cholesterol as a material for preparing novel liposomes to overcome the ineffective delivery of normal drug formulations to brain by targeting the glucose transporters on the BBB. Pharmacokinetic and distribution experiments demonstrated that this novel brain drug delivery system possessed good brain targeting ability. Another important transport system is the choline transporter which binds positively charged quaternary ammonium groups or simple cations²¹. Fenart et al.²² examined the ability of 60 nm nanoparticles coated with quaternary ammonium ligands to cross a cell culture model of the blood–brain barrier (BBB) formed by a coculture of bovine brain capillary endothelial cells and rat astrocytes. The coated nanoparticles were able to cross an in vitro model of the BBB (bovine BCEC). Their passage through the endothelial cell monolayer was three- or four-fold higher than that of uncoated nanoparticles, without any modification of paracellular permeability, suggesting that the enhancement was mediated by the BBB choline transporter.

2.1.3. Receptor mediated transcytosis

Generally, for absorptive-mediated transcytosis the specificity of this BBB targeting strategy is poor since the cationic proteins or CPPs can bind with any negatively charged cell membrane constituents. Moreover, the potential toxicity and immunogenicity of cationic proteins or CPPs limit their use. The transporter-mediated transcytosis is substrate selective, in that only drugs that closely mimic the endogenous carrier substrates will be taken up and transported into the brain. Receptor-mediated transcytosis is considered one of the most mature strategies for brain targeted drug delivery with the characteristics of high specificity, selectivity and affinity, although the ligand may have an effect on homeostasis and natural ligands may compete with the drug ligand to reduce targeting efficiency. Since many kinds of receptors are expressed on the capillary endothelium of the brain, such as transferrin receptor (TfR), the low density lipoprotein receptor (LDLR), the insulin receptor and nicotinic acetylcholine receptors, targeting ligands, including endogenous ligands and ligands based on phage display or structure-guided design, have been exploited to facilitate receptor-mediated BBB transport of drug delivery systems.

One of the most widely characterized receptor-mediated transcytosis systems for brain targeting is the transferrin receptor (TfR), which is highly expressed on endothelial cells of the BBB²³. A transferrin-conjugated drug delivery system for BBB delivery has been developed. Zhang et al.²⁴ prepared transferrin-modified paclitaxel-loaded polyphosphoester hybrid micells (TPM), and in vitro and in vivo brain-targeting efficiencies were evaluated. It was demonstrated that TPM exhibited stronger anti-glioma activity and the mean survival time of mice bearing intracranial U87 MG glioma was significantly prolonged. However, Tf is likely not an ideal brain delivery ligand since the Tf-modified targeted drug delivery system would have to compete with the natural ligand²⁵. As an alternative, a mouse monoclonal antibody (MAb) against the rat TfR, OX26, has been extensively studied. When OX26 is coupled with the liposomes, transferrin receptor-mediated targeting of daunomycin to the rat brain was achieved by using an immunoliposome-based drug delivery system²⁶.

Another common receptor, the low-density lipoprotein (LDL) receptor-related protein (LRP), has been reported to mediate transport of various ligands conjugated to nanocarriers across the BBB. Aprotinin is a LRP ligand and its BBB transport ability was evaluated using an in vitro model of the BBB and in situ brain perfusion. Its transcytosis across bovine brain capillary endothelial cell monolayers was at least 10-fold greater than that of holo-transferrin²⁷. Angiopep, derived from aprotinin with the Kunitz domains of human proteins, exhibited even higher transcytosis capacity and parenchymal accumulation. Sun et al.²⁸ reported angiopep-2 modified cationic liposomes for the efficient co-delivery of a therapeutic gene with paclitaxel to the brain. After treatment with liposomes, the median survival time of brain tumor-bearing mice was significantly longer than that of other groups, making it a promising drug delivery strategy against glioma.

Nicotinic acetylcholine receptors (nAChRs) are a kind of ligand-gated ion channel that is widely expressed in the brain including the brain capillary endothelial cells²⁹. Since they bind the second loop of the three-finger snake toxin with high affinity and selectivity, nAChRs could be exploited to facilitate BBB crossing and intracranial transport of drug delivery systems. Kumar et al.³⁰ reported that a 29-amino acid peptide derived from rabies virus glycoprotein-RVG29 enabled transvascular siRNA delivery to the brain via nAChRs and provided a safe and noninvasive approach for the delivery of therapeutic agents across the BBB. Therefore, nAChR-mediated brain targeting may be a promising strategy for the intracranial transport of drug delivery systems. It was previously reported that a 16 amino acid peptide CDX, which was derived from the loop II region of the snake neurotoxin candoxin, exhibited high binding affinity to nicotine acetylcholine receptors. It was shown that this peptide enabled brain-targeted drug delivery when conjugated to micelles and significantly prolonged the survival time of intracranial glioblastoma-bearing mice³¹.

2.2. BBTB targeting strategies and related drug delivery systems

The blood–brain tumor barrier (BBTB), similar to blood–brain barrier (BBB), is located between brain tumor tissues and microvessels formed by highly specialized endothelial cells (ECs), limiting the paracellular delivery of most hydrophilic molecules to tumor tissue³². With the development of brain tumors, tumor cells begin to invade the surrounding normal brain tissue. Only when the tumor cell clusters grow to a certain volume will BBB be damaged and BBTB be formed. The blood–tumor barrier (BTB) of malignant solid tumors growing in peripheral tissues is typically more permeable than the BBTB of similar malignant solid tumors growing in the brain33, 34. The physiologic upper limit of pore size in the BTB of malignant solid tumor-orthotopic RG-2 rat gliomas microvasculature is approximately 12 nm³⁵.

With the deterioration of brain tumors, angiogenesis and gradual impairment of BBB, BBTB becomes the main obstacle of drug delivery nanosystems. At this stage, tumor neovasculature has formed to support the growth of gliomas. The abnormality of microvessels enhances the permeability of BBTB, whereas the cranial microenvironment and/or the gliomas specificity make malignant gliomas less permeable36, 37. Even though the BBB is compromised in malignant gliomas, the permeability differs from other regions. The infiltrating gliomas, especially around the tumor edge, still utilize the available brain vasculature and the BBB still limits glioma-targeted transport of chemotherapeutic agents³⁷. Hence, the receptors present on the BBB/BBTB provide a chance for gliomas-targeted drug delivery at this stage. The strategies proposed for BBTB targeting are mainly based on receptors high expressed on the tumor such as epidermal growth factor receptors and integrin.

The adhesion receptor integrin, which is overexpressed on tumor neovasculture and glioblastoma U87 cells, was identified as a marker of angiogenic vascular tissue38, 39, 40. Particularly, integrin αvβ3 expression is prominent in malignant glioma but not overexpressed on normal brain cells. As the ligands for integrin, cyclic arginine–glycine–aspartic acid (RGD) peptide and its analogs have been widely investigated for the glioma-targeted drug delivery. Herein, the integrin–RGD interaction still provides a promising BBTB-targeted drug delivery strategy. Our group has designed c(RGDyK)-modified polyethylene glycol–polyethylenimine (PEG–PEI) nanoparticles for intracranial glioblastoma-targeted gene delivery. They exhibited high binding affinity with U87 cells and facilitated targeted gene delivery against intracranial glioblastoma in vivo compared to the PEG–PEI gene carrier without RGD modification. It was demonstrated that the therapeutic efficacy of pORF-hTRAIL of this gene carrier was also enhanced as evidenced by a significantly prolonged survival of intracranial glioblastoma-bearing nude mice. These results demonstrated the therapeutic feasibility of this gene delivery system for brain glioblastoma treatment through mediation of integrin αvβ3⁴¹. Zhan et al.⁴² reported a cyclic RGD peptide conjugated PEG–PLA micelle for the intracranial gliomas targeted chemotherapy, and the median survival time of mice bearing intracranial U87MG tumor xenografts was significantly prolonged after treated by c(RGDyK)–PEG–PLA–PTX micelles, suggesting that RGD motif is efficient for targeted drug delivery to integrin αvβ3-overexpressed glioblastoma.

EGFR, a member of ErbB family of receptors, is a transmembrane tyrosine kinase normally expressed in epithelial, mesenchymal, and neuronal tissues43, 44. Overexpression of the epidermal growth factor receptor (EGFR) on the BBTB makes it a promising target for therapy⁴⁵. EGF and anti-EGFR monoclonal antibody are the commonly used EGFR ligands mediating glioma-targeted therapy. Fondell et al.⁴⁶ adopted epidermal growth factor to target the EGFR and developed a new two-step targeting strategy to transport a recently synthesized daunorubicin derivative into cancer cell nuclei using PEG-stabilized targeting liposomes named “Nuclisome-particles”. Tsutsui et al.⁴⁷ constructed a novel drug delivery system using hybrid bionanocapsules (BNCs) conjugated with anti-human EGFR antibody and confirmed the specific delivery of these BNCs to the brain tumor in the in vivo brain tumor model. Anti-human EGFR antibody-conjugated immunoliposomes also were prepared by Feng et al.⁴⁸ and successfully delivered sodium borocaptate into EGFR-overexpressing gliomas in an animal model of brain tumors.

2.3. EPR effect-based strategies and related drug delivery systems

As the brain tumor develops, the enhanced permeability and retention effect (EPR effect) appears even though it is significantly weaker in the cranial microenvironment than in peripheral tumors, which enables nanosystems with suitable particle size to enter the brain tumor via endothelial gaps on the microvessels of brain tumors42, 49, 50. Some of the nanoscale drug delivery systems have been developed to make use of the glioma EPR effect for tumor targeting. Huang et al.⁵¹ have developed a tumor-targeting nanoparticle system which possesses passive tumor target ability via the enhanced permeability and retention effect. This system was able to prolong the survival time of intracranial U87MG tumor-bearing nude mice.

3. Tumor microenvironment

Studies in developmental biology and tumor progression have demonstrated that the extracellular microenvironment is not simply a passive structural element in which cells reside, but is an interactive partner that can be altered by and can alter cellular processes and responses⁵².

3.1. Lowered extracellular pH

Lowered extracellular pH (pHe) is a common characteristic of various solid tumors, including brain tumors, which is caused by the high rate of glycolysis in cancer cells, under both aerobic and anaerobic conditions53, 54. It has been reported that the tumor pHe is around 5.7–7.2 depending upon tumor histology, tumor volume, and tumor location, and is lower than normal tissue55, 56. Low extracellular pH may be an important factor inducing more aggressive cancer phenotypes. During recent decades many approaches have been developed to target various tumors by pHe, including a shielding/deshielding mechanism or a pH-triggered drug release mechanism. Huang and coworkers have developed a tumor targeting nanoparticle system (called dtACCP) that responds to the lowered tumor pHe and unregulated matrix metalloproteinase 2 (MMP2) expression in the tumor microenvironment. The nanoparticles are modified with an activatable cell-penetrating peptide that is quenched by a pH-sensitive masking peptide, linked by a MMP2 substrate. Once the modified nanoparticles reach the glioblastoma microenvironment, the negative charge of the masking peptide will be eliminated due to the lowered pHe, while the linker is cleaved by MMP2, exposing CPP to drive the nanoparticle׳s internalization into the tumor cells51, 57. This shielding/deshielding mechanism makes it feasible to develop a lowered tumor pHe-triggered drug delivery system, increasing target drug accumulation at the tumor site and lowering the cytotoxicity to the normal tissue.

3.2. Angiogenesis

With the deterioration of brain tumors, angiogenesis occurs with new vessels developing from pre-existing ones. Angiogenesis is essential for glioma tumor growth and metastasis. The angiogenesis targeting delivery system is capable of specifically delivering anti-angiogenic therapeutics to the tumor neovasculature while minimizing systematic toxicity. Therefore, receptors highly expressed during tumor angiogenesis, such as epidermal growth factor receptors and integrin, provide tumor-targeted drug delivery strategies as mentioned in the BBTB targeting strategies section. Currently, antiangiogenic therapy has been restricted to targeting endothelial cells. It was shown that endothelial cells–pericytes signaling networks could also contribute to tumor angiogenesis and metastasis, suggesting that pericytes could be another promising complementary target for cancer treatment. The activated pericytes are capable of stabilizing blood vessels, providing EC survival signals and protecting ECs from VEGF withdrawal, leading to pericyte-mediated resistance to antiangiogenic therapies58, 59. Guan et al.⁶⁰ developed a pericyte-targeting drug delivery system, TH10 peptide (TAASGVRSMH)-conjugated nanoparticles loaded with docetaxel (TH10-DTX-NP) that can target the NG2 proteoglycan highly expressed in tumor vascular pericytes. Enhanced antitumor effects were achieved, revealing the potency and significance of targeting tumor vascular pericytes using nano-DDS in antiangiogenic cancer therapy. It was reported that doxorubicin-loaded liposomes modified with peptide ligands were constructed for the combined therapy of targeting both tumor pericytes and ECs, displaying enhanced anti-tumor effects and prolonged survival in human neuroblastoma-bearing mice. These results clearly demonstrate that this combined targeting strategy was more effective for destruction of tumor vasculature than either monotherapy⁶¹.

3.3. Vasculogenic mimicry

Vasculogenic mimicry (VM) occurs when highly invasive tumor cells form vascular channels for transporting plasma and red blood cells to provide the supply required for tumor growth and metastasis themselves. The channels formed by VM, which can be stained by Periodic Acid-Schiff (PAS), are composed of basement membranes and lined by tumor cells, and no endothelial cells are found on the inner wall⁶². In 1999, Maniotis and colleagues⁶³ found that tissue sections from aggressive human intraocular (uveal) and metastatic cutaneous melanomas generally lacked evidence of significant necrosis and contained patterned networks of interconnected loops of extracellular matrix. Red blood cells were detected within these vascular channel networks and no endothelial cells were identified by microscopy or an immunohistochemical panel of endothelial cell markers. From then on, research on vasculogenic mimicry had attracted much more attention and provoked the further thinking about its molecular regulatory mechanism and its application in tumor therapy64, 65.

With further research, VM has been observed in several malignant tumor types such as breast cancer, liver cancer, glioma, ovarian cancer, melanoma, prostate cancer, and others66, 67, 68, 69. Vasculogenic mimicry was previously reported in human glioblastoma tissues and human glioma cell-line xenografts⁷⁰. Yue and Chen⁷¹ examined 45 astrocytomas (including 15 World Health Organization grade II cases, 15 grade III cases, and 15 grade IV cases) for CD34 endothelial cell markers and periodic acid-Schiff (PAS) staining, among which CD34-negative, PAS-positive patterns of VM were observed in malignant astrocytomas. Hence VM provides new insights for the therapeutic intervention of tumor-associated vasculature.

Until now little has been known about the molecular mechanisms involved in VM. It was reported that aggressive VM-positive tumors exhibited higher expression of the basement membrane extracellular matrix (ECM) component laminin 5γ2, MMP1, MMP2, MMP9 and membrane type-1-matrix/metalloproteinase⁷². Paulis et al.⁷³ considered the VE-cadherin/EphA2/MMP-2 as one of the key VM signaling pathways. Besides, the ischemia and hypoxia effects of the local tumor microenvironment also have an impact on the formation of VM channels⁶⁹. Many researchers involved in VM studies have been trying to treat tumors based on an anti-VM therapy. But almost all VM targeting therapies have been performed in vitro. Hess et al.⁷⁴ adopted general inhibitors of protein tyrosine kinases with transient knockout of EphA2 to abrogate the ability of tumor cells to form tubular structures. Downregulating VE-cadherin, using antibodies against human MMPS and the laminin 5γ2 chain are also possible strategies to inhibit VM⁷⁵. Francescone and colleagues⁷⁵ found that vascular channels of VM in GBM were composed of mural-like tumor cells that strongly express VEGF receptor 2 (Flk-1). Therefore, the identification of Flk-1 as a key factor regulating VM could offer a novel therapeutic target for GBM treatment.

It is necessary to combine both classic anti-angiogenesis and anti-VM in anti-tumor therapies. Tumor cells, which line in the inner surface of VM channels, are highly malignant and have high plasticity. They are directly exposed to blood flow and readily allow tumor cells to migrate and metastasize to other organs. With traditional anti-angiogenesis therapy, the vasculogenic mimicry could also nourish the tumor with nutrients and promote metastasis. As a result, multiple antivascular approaches, including targeting VM and angiogenesis together, may represent the best possible regimen in the treatment of tumors. Since VEGF receptors are highly expressed on angiogenic cells and the VM of GBM, a corresponding ligand-mediated drug delivery system may be utilized for the tumor targeting therapy⁷⁵.

4. Tumor cells

Two alternative models have been put forward to explain how tumors initiate and develop. One is the stochastic model, proposing that tumor cells are heterogeneous, and virtually any of them can function as a tumor-initiating cell. Another one is the hierarchical model, encouraged by increasing experimental data, which hypothesizes that only a small subpopulation of tumor stem cells can induce tumorigenesis76, 77. The normal tumor cells, accounting for most of the tumor tissue, are incapable of tumorigenesis and could be killed by common chemotherapeutic drugs, while cancer stem cells have the ability to self-renew, spending most of their time in the G0 phase of the cell division cycle. Cancer stem cells are likely to share many of the properties of normal stem cells that include a long lifespan, relative quiescence, resistance to drugs and toxins through the expression of several ATP-binding cassette (ABC) transporters, an active DNA-repair capacity, and a resistance to apoptosis⁷⁷. In recent years, brain tumor stem cells have been reportedly identified and isolated, and the concept of cancer stem cells was extended to brain tumors78, 79. The brain tumor stem cells (BTSC) are capable for self-renewal, multipotency, and induction of tumorigenesis. BTSC exhibited remarkable resistance to chemotherapy due to their relative quiescence, active DNA repair regulatory systems and high expression of ABC transporters80, 81. Thus, for brain tumor targeted strategies, cancer stem cell targeting is essential. Recently, several stem cell targeting drug delivery systems have been constructed. Zhang et al.⁸² prepared mitochondrial targeting liposomes showing the strongest efficacy in treating MCF-7 cancer cells and cancer stem cells in vitro and in treating relapsed tumor cells in mice. Nevertheless, brain tumor stem cells targeting nanocarriers are still under research.

CD133, a 120 kDa cell-surface protein, is considered as a marker of normal human neuronal precursors and could be used for the enrichment of tumor stem-like cells from brain tumors83, 84. Sun and colleagues⁸⁵ obtained a specific mouse CD133 binding peptide named LS7 (LQNAPRS) by phage-displayed peptide library technology, which could be used to target cancer stem cells. Nestin, a 200–240 kDa intermediate filament protein, is also the most commonly used marker for the isolation and study of brain tumor stem cells. Beck et al.⁸⁶ identified an effective Nestin targeted peptide (AQYLNPS) through the phage display technology, providing a promising opportunity to design a BTSC-targeted drug delivery system. BTSC and neuronal stem cells share many characteristics including these two markers. Up to now, research on BTSC-targeted drug delivery system is only in its infancy, and specific targets are still needed for the BTSC targeted drug delivery systems.

5. Conclusions

In actual research practice, several targeting strategies have been combined to achieve better therapeutic effects. One common dual targeting drug delivery system was the combination of trans-BBB targeting and brain tumor cell targeting in two ways. The first was dual-targeting moiety modification such as transferrin (Tf) and wheat germ agglutinin (WGA)⁸⁷, and p-aminophenyl-α-d-mannopyranoside(MAN) and transferrin (Tf)⁸⁸; the second one is a single-targeting moiety that targets both BBB and tumor cells, such as angiopep-2 targeting to LRP overexpressed on both BBB and glioma cells89, 90. Another dual targeting drug delivery system was combining trans-BBTB targeting with brain tumor cell targeting. For instance, a single targeting moiety (RGD) was utilized to target both BBTB and tumor cells⁴². Anti-angiogenesis and anti-tumor combination treatments gained much attention because of the potential for dual inhibition of both tumor proliferation and tumor invasion. Gao et al. modified nanoparticles with interleukin-13 peptide, targeting GBM cells and RGD targeting the neovasculature to construct a neovasculature and tumor cell dual targeting delivery, which demonstrated a clear anti-tumor effect in vitro and in vivo⁸³.

Taken together, although current brain tumor-targeted drug delivery systems have achieved significant advances in various stages of brain tumors, there are still limitations of the current drug therapy. As a consequence of these challenges, a series of overall targeting drug delivery systems called “systematic targeted drug delivery” or “whole-process targeting” (WPT) were introduced to try to make the best of every potential target for more effective treatment of brain tumors⁴⁸. The “systematic targeted drug delivery” or “whole-process targeting” is recognized as an effective brain tumor targeted drug delivery strategy which considers all the possible targets including barriers, microenvironment and cancer cells. At the early stage of a brain tumor or around the tumor edge of the infiltrating glioma after BBTB formation, the tumor is nourished by the normal brain capillary network. Nanocarriers modified with ligands targeting the BBB and tumor cells are needed. With the development of the tumor, tumor neovasculature has formed to support the growth of gliomas. The BBTB and tumor cells are the main target goals. Along with the deterioration of the brain tumor, the EPR effect in brain tumor tissues appears and the nano-drug delivery vesicles enter the tumor tissue via passive targeting. Furthermore, vasculogenic mimicry, lowered extracellular pH and tumor stem cells provide possible complementary targets which should be taken into consideration for better therapy of the brain tumors (Fig. 1).

Figure 1.

Brain tumor “systematic targeted drug delivery”.

Research on “systematic targeted drug delivery” is a promising approach to the treatment of brain tumors. It will help to fill in the gaps that remain in fully understanding the relationship between the physiological and pathological conditions of brain tumors, providing a basis for various targeted delivery systems. We believed that systematic targeted drug delivery strategies for brain tumors that consider characteristics of brain tumor growth will provide promising ways to treat brain tumor in the future.