A comprehensive review on the current applications and future perspectives of radioembolization in endovascular neurosurgery

Abstract

In recent years, there has been advancement in the management of primary and secondary brain tumors within a multidisciplinary framework. However, substantial challenges persist in optimizing treatment strategies, enhancing survival rates, and improving patient prognosis and quality of life. These challenges stem from the complex interplay of biological processes that drive tumor pathogenesis and encompass factors such as tumor heterogeneity and patient variability together with surgical access and resectability in certain brain areas. Current therapeutic modalities – surgical resection, radiotherapy, and systemic pharmacotherapy – all have their own inherent limitations. However, promising alternative treatments are seen in emerging techniques such as endovascular radiosurgery, specifically ones that use intra-arterial delivery of radioactive Yttrium-90 (⁹⁰Y) microspheres. This approach is currently used in hepatocellular carcinoma (HCC) treatment, which allows for the precise and targeted delivery of radiation to the tumor while minimizing systemic toxicity. Advances in imaging modalities such as MRI and single-photon emission computerized tomography (SPECT)/CT facilitate accurate dosimetry planning and ensure optimal therapeutic outcomes. This review provides the first comprehensive synthesis of the rationale, technical considerations, and translational potential of 90Y-based endovascular radiosurgery in neuro-oncology. While intra-arterial therapies and radiotherapy are established in other fields, their convergence for intracranial applications—particularly in bypassing the blood-brain barrier (BBB) and achieving localized brachytherapy—remains in its early preclinical and clinical stages. We highlight specific central nervous system (CNS) use cases, preclinical findings, and procedural adaptations needed for this modality’s advancement in brain tumor care.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Yttrium-90 (⁹⁰Y) radioembolization is a well-established treatment for hepatocellular carcinoma, and its potential application in brain tumors has gained increasing interest.

WHAT THIS STUDY ADDS

• This study summarizes preclinical and early clinical findings that support the feasibility and safety of intra-arterial ⁹⁰Y delivery for brain tumors, including glioblastoma and meningioma.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• These insights lay the groundwork for future clinical trials and may influence the adoption of image-guided intra-arterial radiotherapy as a novel option for patients with recurrent or inoperable brain tumors.

Introduction

In 2020, primary tumors of the central nervous system (CNS) were diagnosed in more than 300 000 individuals worldwide.¹ Glioblastoma accounts for 50.1% of all malignant brain tumors, while meningioma accounts for 55.4% of all primary benign brain tumors.^{2 3} The management of primary and secondary brain tumors has evolved significantly in recent years. However, there remain substantial challenges in optimizing treatment approaches, increasing survival rates, as well as improving the overall prognosis and patient’s quality of life.

Conventional therapeutic modalities employed for the management of brain tumors encompass surgical resection, radiotherapy, and systemic pharmacotherapy – these represent the cornerstone of current treatment paradigms.^{4 5} However, the tightly regulated blood brain barrier (BBB) restricts the diffusion of therapeutic compounds and impedes their access from the blood supply to the tumor.⁶ While surgical intervention holds promise in eliminating cancerous tissue, the intricacy and technical demands of the procedure vary depending on tumor location and stage. In contrast, non-invasive radiotherapy techniques offer exquisite precision in selectively targeting areas of interest with focused radiation, with limitations of delivering effective dose due to the skull, as well as other important and eloquent structures that need to be spared from the radio-trajectories.

In recent years, advances in medical devices and non-invasive technologies have revolutionized the management of neurovascular diseases such as ischemic stroke, brain aneurysms, and various vascular malformations. The technique of intra-arterial administration of chemotherapeutic drugs and radioparticles has established itself as an effective approach for hepatocellular carcinoma (HCC), offering notable survival advantages to patients.^{7 8} Building on this success, there exists considerable potential for utilizing this treatment modality to achieve more precise and targeted delivery of chemotherapeutic and radiotherapeutic agents specifically to brain tumors. Among various intra-arterial modalities, Yttrium-90 (⁹⁰Y) radioembolization has shown success in hepatocellular carcinoma and is now being explored for use in neuro-oncology as a novel method for targeted intratumoral radiotherapy. The potential for using intra-arterial delivery of radioactive ⁹⁰Y microspheres for primary and secondary brain tumors exists due to its ability to provide targeted radiation directly to the tumor, bypassing the BBB and minimizing side effects. This strategy presents as an innovative avenue for targeted intervention or brachytherapy.

Endovascular intra-arterial therapy in neuro-oncology

Endovascular approaches to the treatment of brain tumors were reported as early as the 1950s.^{9 10} However, results of early works reduced enthusiasm due to procedural complexity and neurotoxicity concerns. Infusion of carmustine and cisplatin to treat gliomas back in the 1980s and 1990s demonstrated promising results that required further investigation of the technique, as well as the optimal dose. Although it is not common practice, the usage of endovascular intra-arterial delivery of chemotherapeutic agents for brain tumors, including glioblastoma multiforme (GBM), has emerged as an alternative to systemic chemotherapy because this method was proposed as a means for localized dose escalation while minimizing systemic toxicity.^{11 12} As a result, clinicians can pursue dose escalation strategies to enhance intratumoral drug concentrations, achieving a more localized and effective therapeutic response. This effect has been further enhanced by technological advances in endovascular drug, particle, and device delivery systems, as well as the development of diverse treatment approaches for a wide spectrum of neurovascular diseases, including brain aneurysms, arteriovenous malformations, dural arteriovenous fistulas, stroke, subdural hematomas, and meningiomas. To complement the super-selective catheterization approach, the restrictive permeability of the BBB also required circumvention.

Recent intra-arterial protocols have employed chemotherapeutic agents, often in combination with osmotic BBB disruption using hyperosmotic mannitol to enhance drug delivery. A key example is the work by Neuwelt et al, who demonstrated that intra-arterial delivery of carboplatin with mannitol-induced BBB disruption could achieve up to tenfold increases in brain parenchymal drug concentrations.⁵ Similarly, intra-arterial delivery of melphalan has been used in both pediatric and adult brain tumors, particularly in retinoblastoma and recurrent gliomas.¹³ Intra-arterial delivery of biologics is increasingly feasible. MRI-guided endovascular intra-arterial delivery of bevacizumab in combination with osmotic BBB opening has been studied in a patient with recurrent butterfly glioblastoma (bGBM).¹⁴ In a case study by Zawadzki et al, intra-arterial delivery of bevacizumab following BBB disruption was performed under real-time MRI guidance in a patient with recurrent bGBM.¹⁴ By overlaying 3D digital subtraction angiography (3D DSA) on the MRI, catheterization of the appropriate artery and infusion of bevacizumab was successfully achieved.¹⁴ Post-procedure, the patient had no neurological deterioration, no stroke or hemorrhage, reduced contrast enhancement, less edema, and improved cognitive function.¹⁴ These methods demonstrate the feasibility of infusing large-molecule agents intra-arterially with spatial precision. Clinical trials of intra-arterial delivery of chemotherapeutics for gliomas have reported modest efficacy with manageable toxicity. In a Phase I study by Khan et al., intra-arterial carboplatin with mannitol BBB disruption was safely administered in recurrent GBM patients, with radiographic responses seen in approximately 20% of cases and median progression-free survival of 4.2 months. Other methods include MRI-guided focused ultrasound and inhibition of drug efflux transporters.¹⁵ While clinical outcomes from intra-arterial chemotherapy trials have been variable, these studies validated the technical feasibility and relative safety of delivering therapeutic agents directly to intracranial tumors.

In addition to the intra-arterial delivery of chemotherapeutic agents, embolic agents that are administered through super-selective catheterization of tumor feeding arteries have also been identified as a common practice for patients with meningiomas, especially to supplement surgical resection by reducing intraoperative blood loss and facilitating safe resections.¹⁶ This adjunct procedure was first described by Manelfe in the 1970s.¹⁷ Meningiomas are characteristically hypervascular and derive their blood supply predominantly from branches of external carotid artery, such as the middle meningeal artery.¹⁸ This vascular profile makes them particularly amenable to endovascular access and super-selective catheterization. Recent studies have also suggested that preoperative embolization may improve recurrence-free survival in patients with WHO grade I tumors.¹⁹

The procedural frameworks established in these studies, such as super-selective catheterization, infusion monitoring, osmotic BBB disruption, and post-treatment imaging, form a critical translational foundation for adapting endovascular intra-arterial delivery to radioembolization. These same principles now underpin the adaptation of endovascular intra-arterial delivery to radioembolization, where the objective shifts from vascular occlusion or chemotherapy delivery to the precise deposition of therapeutic radiation within brain tumors. In this way, decades of experience with intra-arterial chemotherapy and embolization provide the translational stepping stone for advancing radioembolization as a novel treatment paradigm in neuro-oncology.

Yttrium-90 radioembolization journey

Hepatocellular carcinoma (HCC) accounts for 90% of primary liver cancers.²⁰ Although patients in early stages are eligible for surgical resection or liver transplant which is the mainstay curative therapy, the majority of patients diagnosed with intermediate or advanced tumors on presentation are unresectable.^21–24 In these cases, endovascular intra-arterial therapy is the next best option because it provides locoregional targeting of the tumor. Transarterial chemoembolization (TACE) is the standard of care for patients in the intermediate stage with unresectable HCC. This strategy utilizes catheters to deliver chemotherapeutic agents to obstruct the preselected hepatic arterial branch that feeds the tumor. Unfortunately, not all intermediate HCC patients will benefit equally from TACE due to the population’s heterogeneity, including variations in tumor burden, liver function, and disease etiology.²⁵ Historically, HCC patients who are in the advanced stage of the disease and are not suitable candidates for chemoembolization have the option of radioembolization with ⁹⁰Y.^21–24 As newer techniques and dosimetry approaches develop, recent data from clinical trials also support the incorporation of ⁹⁰Y for early and intermediate stages of HCC with improved safety and efficacy outcomes.^21–24

Studies that investigated ⁹⁰Y radioembolization and compared it to conventional chemoembolization found comparable if not improved results.²⁶ Common toxicities experienced by both groups include fatigue and fever. The time to progression (TTP) for ⁹⁰Y radioembolization also outperformed chemoembolization, with one study reporting 13.3 months vs 8.4 months (P=0.046), respectively.²⁶ Additionally, time to response was notably shorter with ⁹⁰Y compared with chemoembolization.²⁶ For patients with unresectable liver tumors, the median TTP for ⁹⁰Y is found to be more than 26 months, significantly greater than 6.8 months for chemoembolization.²⁷ The median disease control rate was 86% at 3 months post-treatment, which is higher than the disease control rate of 50% by conventional chemoembolization.²⁸ The median time to new hepatic lesions or recurrence was 7.3 months for chemoembolization and greater than 26 months for ⁹⁰Y.²⁷ Some studies found the overall survival status (OSS) among patients with intermediate-stage disease did not differ between the conventional trans-arterial chemoembolization and radioembolization.^{26 27} Similarly, OSS also did not differ between patients undergoing stereotactic body radiation (SBRT) and radioembolization.²⁹ Overall, these findings suggest that ⁹⁰Y radioembolization offers comparable, and in many cases superior, outcomes to conventional chemoembolization, particularly in terms of TTP and disease control, while maintaining a similar safety profile.

Yttrium-90 microspheres

⁹⁰Y is a radioactive isotope that emits a beta particle and decays to zirconium-90. This isotope has a half-life of 2.6 days or approximately 64.2 hours. The decay energy or the energy emitted by the beta particle is between 0.94 MeV to 2.26 MeV, which can penetrate the liver tissue by an average of 2.5 mm to a maximum of 11 mm when embedded in the hepatic microvasculature.^30–32 This value is almost 20 times lower than the penetration depth delivered by, which suggests less radiation exposure to peripheral tissue.^{33 34} For 1 GBq of radioactivity, ⁹⁰Y emits a radiation dose of 50 Gy to 1 kg of tissue.³⁵ This estimation is based on the average pulmonary mass. To deliver ⁹⁰Y through the artery, these isotopic microspheres are contained in carriers to allow for the proximal delivery to the feeding vessels of the tumor.³⁶ The selective endovascular intra-arterial delivery of these isotopes can increase the radiation dose by 10 times compared with that of external beam radiation therapy (EBRT), where the absorbed dose rarely exceeds 70 Gy.³⁷

⁹⁰Y radioembolization is an established locoregional therapy for HCC, offering favorable oncologic outcomes with a lower toxicity profile.²⁸ There are currently two Food and Drug Administration (FDA) approved ⁹⁰Y microsphere products: resin-based SIR resin-based SIR-Spheres (Sirtex Medical, Sydney, Australia) and glass-based TheraSphere (Boston Scientific Corporation, Marlborough, MA). The resin microspheres range from 20 µm to 60 µm in diameter and consist of an ion-exchange polymer matrix, with ⁹⁰Y adsorbed onto the surface.³⁰ Each resin microsphere emits approximately 50 Bq of beta radiation and is typically delivered in higher quantities to achieve therapeutic dosing.³⁰ In contrast, glass microspheres are slightly smaller, ranging from 20 µm to 30 µm, and encapsulate the ⁹⁰Y isotopes within a non-biodegradable glass matrix, emitting around 2500 Bq per sphere.³⁰ Due to their significantly higher radioactivity per sphere—approximately 50 times that of resin microspheres—fewer glass microspheres are required to achieve an equivalent dose, potentially reducing the embolic burden and mitigating ischemic effects on the surrounding liver tissue.³⁸ However, recent improvements in the specific-activity calibration and personalized dosimetry for the SIR-Spheres resin microsphere technology allow resin treatments to deliver more uniform radiation and efficient tumor coverage, achieving therapeutic effects comparable to glass in many clinical scenarios. Enhanced calibration accuracy (up to a 1.23-fold increase over manufacturer estimates) ensures that the intended dose is more reliably delivered while flow-guided, slow infusion techniques enable resin microspheres to reach distal tumor microvasculature with precision, minimizing non-target embolization.³⁹

Dosimetric principles and technical considerations for yttrium-90

The ability to deliver proper dosimetry is of critical importance for determining tumor response, toxicity, and survival outcomes. This process also entails mapping the target artery and locating any potential collaterals that may be predisposed to nontarget radioembolization. The purpose is to minimize radiation exposure to adjacent healthy tissues and reduce ischemic changes from embolic effect. Dosimetric planning for HCC has been achieved at pre-procedure to predict response, toxicity, and survival by using Technetium-99m Marcoaggregated Albumin to simulate ⁹⁰Y.⁴⁰ The two most commonly applied dosimetry models to estimate ⁹⁰Y prescription activity are the single-compartmental model based on the Medical Internal Radiation Dose (MIRD) equation and the multi-compartmental partition model.^41–43 These two models differ in their underlying assumptions: the MIRD model assumes uniform distribution of radioactivity across the entire organ, whereas the partition model assumes non-uniform distribution. The underlying assumption for the MIRD model is its largest limitation because the targeted tissue often contains both tumor and non-tumor tissue; therefore, the radioactivity is heterogeneously distributed.⁴⁴ Alternatively, tumor absorbed dose can be calculated in a post hoc manner using voxel-based dosimetry.⁴⁵ Following segmentation, features of these scans can be extracted using kernels in a process called convolution. By using these features from each image voxel, the absorbed dose for each voxel can be calculated.⁴⁵

Pre-clinical evidence of yttrium-90 radioembolization for brain tumors

Endovascular intra-arterial infusion of cancer therapies to manage brain tumors has been proposed as a solution to minimize concerns surrounding the BBB while reducing systemic toxicity.⁴⁶ By concentrating the drug exposure to a targeted location, endovascular intra-arterial drug delivery is capable of increasing drug responses to brain tumors.⁴⁶ The use of ⁹⁰Y to treat brain tumors in canines was explored by one group.⁴⁷ Clinical presentation post-treatment demonstrated improvement by their 1 month follow-up visit, with no new seizures or neurologic deficits. However, seizure activities were observed at the 2 month, 4 month, and 8 month follow-up visits. The radiological findings also varied between the canines. For instance, reduction in tumor volume was more dramatic in some and less dramatic in others. However, canines that showed a greater volume reduction also presented with more favorable survival time. Conversely, cortical atrophy was not observed in post-treatment imaging for all canines.

Future directions for endovascular 90^Y in neuro-oncology

While intra-arterial radioembolization with ⁹⁰Y microspheres is well-established in HCC, its application to neuro-oncology remains largely theoretical and preclinical. Nonetheless, select CNS tumors exhibit biological and anatomical characteristics that may make them suitable candidates for this targeted therapy (figure 1).

Figure 1. Proposed workflow of intra-arterial infusion of ⁹⁰Y. CBCT, Cone Beam Computed Tomography; DSA, Digital Subtraction Angiography; MRI, Magnetic Resonance Imaging; SPECT, Single Photon Emission Computed Tomography.

GBM is the most common and aggressive primary malignant brain tumor in adults, with a median overall survival of 15 to 18 months despite standard-of-care treatment consisting of surgical resection, radiotherapy, and temozolomide.⁴⁸ Nearly all cases recur, and therapeutic options for recurrent GBM are limited and largely palliative. Intra-arterial radioembolization using ⁹⁰Y microspheres has emerged as a novel investigational strategy, aiming to deliver highly localized radiation to the tumor while sparing adjuvant healthy brain parenchyma and bypassing the BBB. Following promising safety signals in a canine glioblastoma model, the FRONTIER Trial (Feasibility of Radiation via Intra-arterial ⁹⁰Y for Neuro-Oncology: Targeted Internal Embolization Research) launched as the first-in-human feasibility study of this approach in recurrent GBM.³ Preliminary results from this prospective, multi-center study that evaluated the safety and feasibility of TheraSphere ⁹⁰Y microsphere therapy in patients with recurrent glioblastoma were presented at the 2025 Society of Interventional Radiology Annual Scientific Meeting.⁴⁹ Early results from six patients show the treatment is well-tolerated with no significant toxicities, technically feasible with accurate dose delivery, and demonstrates localized tumor control, supporting continued investigation.⁴⁹ From a technical standpoint, identifying dominant feeders requires higher-resolution angiography, as the infiltrative nature of GBM can obscure tumor blush. Chen et al, outlined a streamlined approach for planning intra-arterial embolization of brain tumors by integrating MRI with Cone Beam CT and 3D rotational angiography to map vascular supply and guide catheter placement.⁵⁰ Parenchymal blood volume imaging, combined with MRI, enables visualization of perfusion and estimation of infusion volume for mesenchymal stem cells loaded with Delta-24.⁵⁰ Unfortunately, this technique has not been validated for radioembolization, and it is uncertain that the perfusion map corresponds to the amount of radiation delivered. Nonetheless, this targeted approach may offer a viable salvage option for patients with recurrent or inoperable tumors, particularly when other local therapies are not feasible.

Meningiomas are the most common primary intracranial tumors, with the majority being benign WHO grade I lesions. While gross total resection is often curative, anatomical constraints – particularly in skull base or parasagittal locations – can limit the extent of surgical resection.⁵¹ Additionally, even low-grade tumors can recur, and higher-grade meningiomas (WHO grade II and III) are associated with more aggressive behavior and recurrence rates despite multimodal treatment.⁵² Intra-arterial radioembolization with ⁹⁰Y offers a compelling strategy for these cases. Unlike gliomas, meningiomas often have well-circumscribed angiographic features, allowing for easier identification of dominant feeders and clearer treatment boundaries. From a radiological standpoint, meningiomas are generally radiosensitive, with favorable response rates observed following stereotactic radiosurgery or fractionated radiotherapy. Results from the two Phase II prospective trials, RTOG 0539 and EORTC 22042, showed that adjuvant fractionated stereotactic radiotherapy following gross total resection for patients with recurrent WHO grade 1 or newly diagnosed WHO grade 2 tumors demonstrated favorable overall survival (OS) rates of approximately 96% over and progression-free survival (PFS) of more than 70% over a 3 year period.⁵³ For higher risk meningiomas, classified as new or recurrent WHO grade 3 or recurrent WHO grade 2, the RTOG trial also demonstrated a PFS of 58.8%, local control of 68.9%, and an OS of 78.6% after 3 years.⁵³ Fractionated stereotactic radiotherapy is frequently indicated following surgical resection of higher grade meningiomas due to their unpredictable behavior and heterogeneity, which can lead to considerable recurrence. Beyond their biological radiosensitivity, meningiomas also present a unique clinical opportunity for intra-arterial ⁹⁰Y radioembolization, especially in patients who are not surgical candidates due to tumor location, medical comorbidities, or prior treatment history. These external modalities often require high cumulative doses that may be contraindicated in previously irradiated patients or in tumors adjacent to radiosensitive structures like the optic apparatus or brainstem.^{54 55} Moreover, patients with recurrent or residual meningiomas often have limited options, as additional EBRT may be contraindicated because of cumulative dose constraints or proximity to radiosensitive structures. In this context, intra-arterial ⁹⁰Y radioembolization offers a focal, image-guided alternative to EBRT, enabling high-dose, localized intravascular radiation delivery with rapid dose falloff beyond the target zone. This approach minimizes collateral damage to adjacent brain parenchyma while potentially improving local control in refractory or inoperable tumors. The combination of well-defined vascular anatomy, accessibility to endovascular techniques, and favorable radiobiologic profile makes meningiomas, particularly recurrent, residual, or surgically inoperable lesions, strong candidates for early clinical translation of intra-arterial radioembolization in neuro-oncology. ⁹⁰Y radioembolization could offer a focal radiation boost with rapid dose falloff beyond the target zone, limiting exposure to adjacent tissue.

Future studies are needed to define optimal dosimetry for intracranial applications, but the combination of well-defined vascular supply, inherent radiosensitivity, and anatomical accessibility make recurrent or residual meningiomas a strong candidate for translational evaluation of endovascular radioembolization.

Conclusion

Brain tumor management is challenging due to complex tumor biology and limitations of surgery, radiotherapy, and systemic therapy. Intra-arterial delivery of ⁹⁰Y microspheres—a technique proven in HCC—offers a targeted, low-toxicity alternative. Enhanced by advanced imaging, this endovascular radiosurgery approach may overcome current treatment barriers and improve outcomes through greater precision and reduced morbidity.

Data availability statement

Data sharing not applicable as no datasets generated and/or analyzed for this study.