Primary central nervous system tumors: future directions in systemic treatment

Primary central nervous system (CNS) tumors are a heterogeneous group of over 100 malignancies [1]. While they account for only 3% of new cancer cases [2], they are among the tumor entities with the highest number of potential life years lost due to their occurrence in younger patient populations and limited survival rates [3]. Moreover, these tumors often pose a considerable burden on patients’ quality of life, as symptoms such as motoric and cognitive deficits and epileptic seizures frequently interfere with activities of daily living [4, 5].

Central nervous system tumor classification has undergone substantial changes in recent years [1, 6]. Currently, definitions integrate histopathological features and molecular alterations, and new as well as previously uncharacterized tumor entities continue to be identified based on genetic alterations and DNA methylation profiling [7]. Across entities, treatment mainly consists of a multimodal approach including surgery, radiotherapy, and systemic treatment [8, 9]. However, local therapies are frequently limited by proximity to eloquent structures, challenging the maximization of extent of resection and radiotherapy target volumes. On the other hand, many antineoplastic drugs cannot penetrate the blood–brain barrier (BBB), limiting their concentrations in the CNS and thereby antitumoral activity [10].

While the BBB may be disrupted to a variable extent in intra-axial tumors, the abundant expression of molecular efflux transporters such as p‑glycoprotein (P-gp) or breast cancer-related protein (BCRP) represents a further challenge to systemic treatment of brain tumors [10]. Moreover, physical factors such as elevated interstitial pressure in brain tumors might contribute to decreased drug efficacy [11]. Particularly gliomas exhibit a high degree of heterogeneity and biological plasticity in their microenvironment, eventually leading to treatment failure [14, 15]. Further research on innovative approaches is therefore urgently needed to improve outcomes in this patient population of high clinical need. For instance, approaches to overcome the BBB include physical (e.g., focused ultrasound, low-dose radiation, nanoparticles) as well as molecular (e.g., receptor-mediated transcytosis, efflux transporter inhibition) strategies (reviewed in more detail in [12, 13]).

In this short review, we will briefly outline current treatment landscapes and future directions in systemic therapies of the most common CNS tumors in adults, diffuse gliomas and meningiomas.

Meningiomas are the most frequent intracranial tumors. Deriving from arachnoid cap cells, most meningiomas show well-demarcated growth and exhibit benign biological behavior. As they show characteristic radiological features, a watch-and-wait approach may be followed depending on the size and location of the lesion. Otherwise, maximal safe resection generally follows a curative intent [9]. However, some meningiomas show higher recurrence rates, intermediate to malignant biological behavior, and may even metastasize extracranially [34, 35]. In line with this, the 2021 WHO classification defines CNS WHO grade 1 (~80–85%), grade 2 (10–15%), and grade 3 (< 5%) meningiomas. Similar to gliomas, histological diagnosis follows an integrated approach considering histomorphological features and molecular alterations [1]. For instance, homozygous deletions of CDKN2A/B or telomerase reverse transcriptase (TERT) promoter mutations are defining hallmarks for CNS WHO grade 3 tumors even in the absence of high mitotic count or spontaneous necroses. In higher-grade meningiomas and/or after incomplete resection, radiotherapy or stereotactic radiosurgery are frequently applied [9]. In contrast, the value of pharmacotherapy remains to be established and is currently limited to situations where local therapies are exhausted. Cytotoxic treatments such as hydroxyurea, temozolomide, irinotecan, or trabectedin have shown limited activity in clinical trials and case series, and targeted approaches are being evaluated [36]. One promising target is the somatostatin receptor 2a (SSTR2a) given its expression in the majority of meningiomas, which is also harnessed in SSTR-targeted positron-emission tomography (PET) [37, 38]. While data on somatostatin analogs such as octreotide are conflicting, radionuclide therapies targeting SSTR2a unite the specificity of SSTR-binding peptides with the cytotoxic activity of radionuclides. This approach follows the “see what you treat” principle by combining diagnostic PET imaging with therapeutic applications (“theranostics”) [39]. Data from small trials and case series showed disease stabilization, and a randomized controlled trial is recruiting (LUMEN‑1, NCT06326190) [40].

Besides targeted sequencing for tumor classification purposes, comprehensive next-generation sequencing panels are increasingly used in clinical routine. This enables detection of alterations potentially amenable to targeted treatment approaches. For instance, NTRK fusions occur in 1–3% of diffuse gliomas and can be targeted by inhibitors such as larotrectinib and entrectinib, representing an additional treatment option in recurrence [41]. Likewise, BRAF mutations can be detected in a low percentage of gliomas, which can be targeted by combined BRAF/MEK inhibition such as using dabrafenib/trametinib [42]. Also FGFR mutations can be found in a rare subgroup, with available inhibitors such as erdafitinib or pemigatinib. Drugs with tissue-agnostic approval by regulatory authorities can be considered, including larotrectinib and entrectinib (NTRK fusions), pembrolizumab (microsatellite instability, high tumor mutational burden, mismatch repair deficiency), dabrafenib/trametinib (BRAF mutations), selpercatinib (RET fusions), and trastuzumab deruxtecan (HER2 overexpression). However, evidence in brain tumors is limited overall, and such approaches should be followed within clinical trials, in well-annotated registries, and/or after exhaustion of further treatment options. For both glioma and meningioma, the European Association of Neuro-Oncology (EANO) has formulated molecular testing guidelines for targeted therapy selection, summarizing the available evidence and designating ESMO Scale for Clinical Actionability of Molecular Targets (ESCAT) tiers [43, 44].

However, the presence of molecular alterations is frequently an imperfect predictor of treatment response. Particularly in gliomas, intratumoral heterogeneity and epigenetic plasticity counteract the antitumoral activity of drugs that are selected based on a single genetic mutation [45]. Model systems reflecting the biological background of the tumor might deliver valuable information as a basis to individually select the most efficacious drug. Personalized approaches based on “ex vivo” drug screening are therefore being evaluated. In the multicentric Austrian ATTRACT trial, resected tumor tissue is cultivated ex vivo and treated with 28 antineoplastic drugs approved in other malignancies [46]. The results of this drug screening are discussed in a multidisciplinary tumor board, and a personalized treatment recommendation is formulated. The patient can then opt to follow this personalized approach after concurrent radiochemotherapy for MGMT promoter-unmethylated glioblastoma. The trial is paralleled by a comprehensive translational research program to identify further treatment targets and biomarkers in this patient cohort with unmet clinical need.