Introduction

Brain metastases (BM) constitute a significant disease burden and have a major impact on patient morbidity and mortality ( ). Recent progress in the preclinical and clinical development of molecular targeted drugs has revealed that some of them have potent therapeutic efficacy for BM. Here the preclinical and clinical knowledge on the role of targeting angiogenesis in BM is discussed. In addition, the potential of sensitizing BM to radiation and increasing blood–brain barrier penetrability to improve the outcome of this devastating disease is examined.

Targeting Angiogenesis for Brain Metastases

Angiogenesis is essential for the development and survival of cancer. The formation of new blood vessels plays an important role in tumor progression, metastasis and resistance to chemoradiation ( ). Since the pioneering work of , there has been intense research into the development of anti-angiogenic cancer therapies in primary tumors and a large number of angiogenesis inhibitors are currently in clinical development.

The impact of targeting angiogenesis in BM is less well studied and conflicting findings have been reported. showed that vascular endothelial growth factor (VEGF) promoted the growth of breast cancer BM in nude mice. Targeting the VEGF receptor with the tyrosine kinase inhibitor PTK787/Z222584 led to a significant decrease in BM growth and vascular density ( ). Transfection of colon and lung adenocarcinoma cells with the antisense-VEGF165 gene significantly reduced the incidence of BM after intracarotid injection ( ). Overexpression of VEGF165 induced significant progression of melanoma BM that was associated with brain vessel co-option rather than sprouting angiogenesis, highlighting VEGF as an important therapeutic target for the treatment of BM ( ). used different human and murine tumor cells lines to study BM in SCID, syngeneic BALB/c and syngeneic C3H/He mice. They showed that 95% of early micrometastases demonstrated vascular co-option with little evidence for isolated neurotropic growth. The vascular basement membrane promoted adhesion and invasion of malignant cells and was sufficient for tumor growth prior to any evidence of angiogenesis. Blockade of β1 integrin in tumor cells prevented adhesion to vascular basement membrane and decreased BM development and growth. Hence, β1 integrin could be a useful therapeutic strategy for BM ( ). In an elegant study using a mouse model with a cranial window, showed that tumor cells, upon extravasation, closely correlated with microvessels, either by direct angiogenesis (lung cancer) or vessel co-option (melanoma). Blockade of VEGF-A by bevacizumab inhibited angiogenesis and resulted in prolonged dormancy of lung cancer but not melanoma ( ). In contrast, targeting VEGFR-2 by ZD6474 inhibited angiogenesis but resulted in sustained BM progression via brain vessel co-option ( ). Magnetic resonance imaging (MRI) demonstrated substantially reduced central blood volumes in large BM, derived from a prostate cancer model, after treatment with the VEGFR inhibitor AZD2171 ( ). A rim of elevated blood volume was retained at the tumor–brain interface while small tumors displayed a static response ( ). These data underlie that the outcome of treatment with anti-angiogenic therapies might also depend on the histological type, which should be taken into consideration before deciding to initiate clinical trials with these agents.

Hemorrhagic episodes were recorded in the early clinical studies conducted in patients with metastatic disease, including BM, after treatment with bevacizumab ( ). In consequence, the use of anti-angiogenic drugs such as bevacizumab had been contraindicated for a long time. Hence, patients with BM have been excluded from participating in clinical trials testing the efficacy of anti-angiogenics, even though the exact risk of intracranial bleeding remained largely uninvestigated ( ). However, recent studies and meta-analyses in over 10 000 patients indicated that VEGF targeting therapies are safe for use in patients with BM, as intracranial bleeding incidence is low (0.8–3.3%) and does not exceed baseline ( ). These reports have prompted the European Medicines Agency to reconsider the decision to contraindicate the administration of bevacizumab and allow its use in patients with non-small cell lung cancer (NSCLC), renal cell carcinoma (RCC), colorectal and breast cancer that develop BM ( ). The new guidelines now indicate that treatment decisions should be made after carefully assessing the potential benefits and risks for individual patients ( ). It is expected that the contraindication will also be removed in the USA in the near future.

As a result of the exclusion of patients with active BM from clinical studies, there are only scarce data regarding the efficacy of anti-angiogenic agents in this tumor entity. In a retrospective analysis, identified six patients with NSCLC and treatment naïve or progressive BM who had previously received bevacizumab. The best central nervous system (CNS) response was partial in two, stable disease in three, and progression in one. Bevacizumab treatment resulted in clinical improvement of patient symptomatology and reduced corticosteroid dependence ( ). In an open-label, expanded access study conducted in 212 patients with BM from RCC, sunitinib treatment resulted in an objective response in 26 (12%) patients, without increased toxicity compared to that in the metastatic RCC population ( ). Similar findings were also reported for sorafenib ( ). In a phase III study, the addition of the anti-angiogenic agent thalidomide to whole brain radiation therapy provided no survival benefit in patients with BM and was associated with high toxicity ( ). Notably, the alterations in blood flow during and after treatment were not monitored in that study.

Important lessons have been learned from the preclinical and clinical use of anti-VEGF strategies in the treatment of glioblastoma multiforme (GBM) which could possibly also have applications in BM. Administration of cediranib in mice can alleviate brain edema and prolong survival, despite continuous tumor growth ( ). In humans, it decreases corticosteroid dependence but whether this can improve patient outcome remains to be defined ( ). Additionally, special attention should be paid when assessing the response of BM using MRI as the reduction in contrast media uptake may overestimate regression of cerebral lesions ( ). Several clinical trials testing the efficacy of anti-angiogenic agents, either as monotherapy or in combination with radiation and/or chemotherapy, have been initiated in patients with BM. The results of these studies are awaited with great interest and will provide important insight on the efficacy and safety of anti-angiogenic therapies.

Sensitizing Brain Metastases to Radiation

Radiotherapy, either in the form of whole brain radiotherapy (WBRT) and/or stereotactic radiosurgery (SRS), is the cornerstone of treatment. However, local or intracerebal recurrence after radiotherapy is not uncommon and has a negative impact on neurocognitive function. Hence, it is important to explore further the potential of new targeted therapies that enhance radiation efficacy to achieve long-term control of BM while minimizing side effects.

Over a decade ago, montexafin gadolinium (MGd), an MRI-detectable agent, drew significant interest as a potential radiosensitizing agent for patients with BM treated with radiotherapy as it is taken up in tumors at higher concentrations compared to normal tissues ( ). Although its precise mechanism of action remains unclear, it plays a role in electron scavenging and disrupts cellular metabolism by inhibiting key proteins, such as thioredoxin reductase, leading to generation of reactive oxygen species ( ). In a randomized controlled study, the addition of MGd to WBRT failed to improve survival in 401 patients with BM induced by various histological types ( ). The authors of the study reported a small delay in the time to neurologic progression in the combination group in patients with BM originating from lung cancer ( ). In a multicenter randomized phase III clinical trial, 554 patients with NSCLC received WBRT or WBRT plus MGd and the time to neurologic progression was examined ( ). Again, the incorporation of MGd to WBRT failed to meet its primary endpoint. Interestingly, an older preclinical study had failed to demonstrate a radiosensitizing effect for MGd in various tumor cells or xenograft models, raising questions about the efficacy of MGd as a radiosensitizing agent ( ).

Efaproxiral has also been investigated as a radiosensitizer as it reduces hemoglobin’s oxygen binding affinity, leading to increased oxygen release into tissues ( ). In a phase II clinical study that recruited 57 patients with BM, the combination of efaproxiral with WBRT (10×3 Gy) resulted in a median survival of 6.5 months ( ). Following this, a phase III study in 515 patients with BM was initiated. However, the addition of efaproxiral to WBRT failed to improve survival compared to WBRT alone ( ). Similarly, a confirmatory phase III trial in breast cancer patients did not show any survival benefit for the combination arm ( ). Hence, neither MGd nor efaproxiral improve neurologic outcome or survival in patients with BM treated with WBRT.

In a prospective randomized trial, lonidamine, an indazole carboxylic acid, was given with WBRT in patients with BM ( ). Survival and response rate were unaffected by the presence or absence of lonidamine. Similarly, the addition of the hypoxic cell radiosensitizers metronidazole and misonidazole or the proliferation marker bromodeoxyuridine failed to improve the outcome in patients with BM who received WBRT ( ).

Despite the significant effort in evaluating chemotherapeutics, either alone or in combination with radiotherapy, for BM, level I evidence has shown no survival benefit of the alkylating agent temozolomide, nitrosoureas, carboplatin, or tegafur in this disease ( ).

Vorinostat, a histone deacetylase inhibitor, sensitizes tumor cells to radiation. In a recent preclinical study, showed that vorinostat enhanced the response of MDA-MB-231BR breast cancer BM to radiation and prolonged overall survival in mice. Currently, a phase I clinical trial is investigating the radiosensitizing potential of vorinostat in patients with BM, induced by various primary sites, in combination with WBRT. A further phase I study examines the combination of vorinostat with SRS for BM in patients with non-small cell lung cancer and the results are awaited with interest.