A systematic review of ongoing clinical trials in optic pathway gliomas

Introduction Optic pathway gliomas (OPGs), also known as Visual Pathway Gliomas, are insidious, debilitating tumours. They are most commonly WHO grade 1 pilocytic astrocytomas and frequently occur in patients with neurofibromatosis type 1. The location of OPGs within the optic pathway typically precludes complete resection or optimal radiation dosing, hence outcomes remain poor compared to many other low-grade gliomas. The aim of this systematic review was to formulate a comprehensive list of all current ongoing clinical trials that are specifically looking at clinical care of OPGs in order to identify trends in current research and provide an overview to guide future research efforts. Methods This systematic review was conducted in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. The Cochrane Controlled Register of Trials (CENTRAL) and ClinicalTrials.gov were searched. Inclusion and exclusion criteria were applied and final results were reviewed. Results 501 clinical trials were identified with the search strategy. All were screened and eligible studies extracted and reviewed. This yielded 36 ongoing clinical trials, 27 of which were pharmacological agents in phase I-III. The remaining trials were a mixture of biological agents, radiation optimisation, diagnostic imaging, surgical intervention, and a social function analysis. Conclusion OPG is a complex multifaceted disease, and advances in care require ongoing research efforts across a spectrum of different research fields. This review provides an update on the current state of research in OPG and summarises ongoing trials.


Introduction
Optic pathway gliomas (OPGs), also known as visual pathway gliomas, are insidious, debilitating tumours that account for 3-5% of all paediatric brain tumours. They are a subtype of lowgrade glioma (LGG), most often World Health Organization (WHO) grade 1 juvenile pilocytic astrocytomas (JPAs), with a smaller proportion being pilomyxoid astrocytomas (PXAs) [1]. They are common in patients with neurofibromatosis type 1 (NF1), with up to 20% developing an OPG at a mean age of 4.5-5 years old [2]. OPGs most commonly arise at the chiasmatic-hypothalamic region but can arise anywhere along the optic pathway [3]. Their intimate relationship to the optic apparatus, hypothalamus, ventricular system and brain parenchyma can result in a variety of clinical sequelae including visual loss, endocrinopathies and hypothalamic dysfunction, developmental/neuropsychological disorders, hydrocephalus and focal neurological deficits. This complexity of symptomatology, and the close relationship to key structures make the treatment of OPG challenging. Unlike other JPAs, the location typically precludes complete surgical resection or optimum radiation dosing without incurring an often-unacceptable neurological cost. Furthermore, key aspects of their behaviour including their natural history and optimal management are incompletely understood. Tumour stabilisation, progression or regression can all occur unpredictably. OPG management is highly individualised. Many undergo a period of observation with serial imaging unless there is progressive growth or visual symptoms. Chemotherapy with a carboplatin/ vincristine 'Packer' regimen is often the first line, although alternatives such as the TPCV (thioguanine, procarbazine, lomustine, vincristine) are sometimes used, except in NF1 patients due to the risk of secondary leukaemia [4]. Radiotherapy is typically avoided in younger children due to risks of toxicity wherever possible [5], although it remains an option in the setting of refractory, progressive disease or where visual loss is occurring despite systemic therapy [6,7]. The role of surgical biopsy and debulking surgery remains controversial [8,9]. OPGs require multidisciplinary care by neurosurgeons, neuro-oncologists, radiation oncologists, endocrinologists, ophthalmologists, pathology, neuropsychology, paediatricians, geneticists and a host of allied health professionals. This multifaceted care underlies the complexity of the disease. Advances in the understanding and treatment of OPG could come from a variety of sources, including novel or repurposed pharmacological agents, emerging biological agents and tumour vaccines, refinement of surgical approaches, diagnostic and therapeutic radiological developments or quality of lifefocussed research [10][11][12][13][14][15][16]. The aim of this systematic review was to formulate a comprehensive list of all current ongoing clinical trials that were specifically looking at a clinical care of OPGs in order to identify trends in current research and provide an overview of the field to guide future research efforts. This review includes trials that are registered, and in any stage of recruitment or analysis, but not yet published. A summary of each of the ongoing clinical trials will be presented to facilitate a rapid review of the field.

Methods
This systematic review was conducted in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines on April 10th, 2020 [17,18]. Two clinical trial databases, the Cochrane C o n t r o l l e d R e g i s t e r o f T r i a l s (C E N T R A L ) a n d ClinicalTrials.gov, were searched. The search terms used to identify clinical trials were as follows: '(optic OR pathway OR visual OR opg OR chiasm OR midline OR hypothalamic OR hypothalamus OR neurofibromatosis OR NF1 OR pediatric OR paediatric) AND (glioma OR pilocytic OR astrocytoma OR pilomyxoid)'. Trials older than 2010 were excluded. Trial titles and abstracts were reviewed by two of the authors. Titles with no English language translation available, and duplicates, were removed. Exclusion criteria were applied; these included titles that were not related to OPG or to trials that would not include OPG as a subgroup based on their inclusion/exclusion criteria. Already-published studies were excluded; these articles were reviewed to ensure that the publications were final and complete, and not an interim analysis or partial publication, with further publication of trial results expected. Studies that had been withdrawn, abandoned or not updated in 3 years or more were also excluded. For identified trials where there had been no recent clinical trial update (in print or online) for > 3 years, contact was made with the principal investigator of the trial to establish progress and assess if the trial was still running. Two authors (CSH and AB) performed the search strategy independently and reviewed the final papers separately. Any conflicting findings were reviewed (by CSH, AB and WI) and a consensus agreed.

Results
Five hundred one clinical trials were identified with the search strategy described. After initial exclusions, 292 were screened. Thirty-six eligible studies were included in the final analysis. The identification, screening, eligibility and inclusion were conducted according to PRISMA guidelines. The strategy is summarised in Fig. 1.
After application of the PRISMA search strategy, we included 36 clinical trials in our final analysis.
A detailed summary of the stratified clinical trials included in the review is presented in Table 1.
Out of the 36 clinical trials assessed, 27 were pharmacological studies. There was one phase 0 trial, 8 phase 0/I drug trials, 7 mixed phase I and phase II, 9 phase II, and 2 phase III trials. There were 2 trials focussed on biological agents, 2 based on optimising radiation therapy, 2 on diagnostic neuroimaging, one on imaging-guided thermotherapy, one on genetic testing and one on social functioning.

Discussion
Systematically reviewing the ongoing clinical trials is important to update clinicians on potential new therapies, avoid duplication of research and identify research trends and developing areas of study to stimulate new investigation. As might be expected, the main focus of ongoing research studies in OPG is related to pharmacological agents. Twenty-seven out of the 36 identified research trials were drug trials. Such investigations are necessary to establish safety and efficacy of all new drugs and are essential to determine if these agents should be adopted or discarded. Many of the new agents were related to the BRAF/MEK/ERK or RAS pathways. A summary of the pharmacological agents under investigation is presented in Table 2.
There were 5 ongoing trials into the same pharmacological agent selumetinib, and 2 looking at trametinib; this raises a question as to whether a collaborative approach would allow data sharing and ensure a common data element and consensus reporting. The included trials were dominated by North American centres and pharmaceutical companies. We did not identify any trials run from a low-or middle-income country (LMIC) despite reports that up to 80% of all paediatric cancer patients occur in these countries. Potential reasons for the lack of clinical trials in LMIC include a lack of specialised centres, equipment and staff, and a lower priority of cancer for healthcare planning strategies [19][20][21]. There were a small number of trials looking at non-pharmacological diagnostics, therapeutics and assessment social functioning relating to OPG. These are summarised in Table 3.

Pending trials
The authors are aware of a further relevant clinical trial that is yet to start, the LOw Grade Glioma In Children (LOGGIC) study. LOGGIC will compare the effect of standard chemotherapy versus MEK inhibitor (trametinib) as first-line agents in paediatric low-grade glioma. The outcome measures will include a child's quality of life, including visual and neurological function. The LOGGIC trial will be run by the European Society for Paediatric Oncology (SIOPE). As with any emerging field, this is likely to be just one of several trials not yet registered on public databases.

Unanswered questions
This systematic review highlights the dominant pharmacologic-centric nature of current OPG clinical trial research. There are several important areas of OPG care that remain clouded in uncertainty and controversy and are not currently being addressed by the ongoing clinical trials we identified in this study. A key outstanding question is   LGGs. Vemurafenib has shown some promise in the treatment for BRAFV600E mutant LGGs [30]. Tak580 Oral pan-Raf kinase inhibitor. A single-patient study demonstrated apparently good tolerance in a patient with refractory LGG ahead of a larger planned clinical trial [31]. Ulixertinib An ERK1/2 inhibitor. Blocking ERK-mediated signal transduction may inhibit ERK-dependent tumour cell proliferation and survival. RAS pathway targeting agents The RAS/MAPK pathway is important in the control of cancer cell growth and proliferation. The NF1 product neurofibromin functions as a negative regulator of RAS activity. RAS overactivity in patients with NF1 drives tumourigenesis. The MEK pathway is hyperactivated in NF1-deficient astrocytes and this drives NF1-associated optic glioma growth. Hence, the rationale for sustained MEK pharmacologic blockade in attenuating NF1-deficient astrocytes and NF1 optic glioma volume and proliferation [32]. The development of agents that target RAS signalling and associated pathways implicated in the pathogenesis of neurofibromas has led to clinical trials of various new pharmacological agents. Selumetinib/trametinib Selective MEK inhibitor. A phase I study assessing selumetinib dose tolerance and pharmacokinetics in NF1-related plexiform neurofibromas was conducted by Dombi et al. who demonstrated benefit in inoperable plexiform neurofibromas without excess toxicity [33]. A further phase II study in relapsed LLG that included OPG showed sustained partial therapeutic response [34].

MEK162 (binimetinib) Orally available inhibitor of MAPK2. Anti-angiogenic agents
Bevacizumab A humanised monoclonal antibody directed against VEGF, this inhibition leads to a reduction in microvascular growth of tumour blood vessels and thus limits the blood supply to tumour tissues. There are several reports of the benefit of anti-VEGF drugs in uncontrolled case series [35][36][37][38]. Lenvatinib Multireceptor TK inhibitor with preferential anti-angiogenic capacity. It inhibits VEGF receptor family 1-3, FGFR family 1-4, PDGFRa, TK receptor and RET. As a result, it may modulate the tumour microenvironment and anti-tumour responses [39]. Thalidomide An anti-angiogenic agent. The mechanism of anti-angiogenic action is not fully known [40]. Its use has been investigated in the care of high-grade glioma [41]. Lenalidomide An analogue of the anti-angiogenic agent thalidomide. It has demonstrated a significant anti-tumour activity in patients with multiple myeloma and myelodysplastic syndrome. Phase I studies in adults and children demonstrated tolerability but was associated with thromboembolic events and myelosuppression [42][43][44]. Antagonistically, the use of lenalidomide has been associated with secondary primary high-grade astrocytoma development in multiple myeloma [45]. Pomalidomide An analogue of the anti-angiogenic agent thalidomide. It failed to demonstrate benefit in a phase II clinical study as monotherapy for children and young adults with recurrent or progressive primary brain tumours [46]. Miscellaneous pharmacological agents Mebendazole Methyl N-[6-(benzoyl)-1H-benzimidazol-2-yl] carbamate. A benzimidazole anti-helminthic used to treat a variety of worm infections and for echinococcosis (hydatid disease). Mebendazole binds to tubulin subunits preventing polymerisation. Tubulin is a key molecule in cell division and has been proposed as an anti-cancer target. Other potential targets of mebendazole include the Hh signalling pathway, and angiogenesis through VEGF inhibition [47]. Repurposing of mebendazole has been suggested in various tumours and it has been previously trialled in preclinical models of glioblastoma with some success [48][49][50]. Entinostat Entinostat is a synthetic benzamide HDAC type 1 and III inhibitor. It is orally bioactive. Its proposed mode of action is to regulate chromatin structure and through histone deacetylation control epigenetic gene transcription. Other mechanisms of effect may include induction of reactive oxygen species leading to mitochondrial damage and inducing cell cycle arrest [51,52]. Pegylated interferon alfa-2b Pegylated interferon α-2b is a member of the interferon family; these are glycoproteins with anti-cytokine effects that have documented immune-modulating and anti-proliferative effects. One phase II study is examining the effect of pegylated interferon alfa-2b in a child with recurrent or refractory/progressive JPA and OPG in children. Studies have shown that interferon alfa-2b is tolerable and may delay progression in DIPG [53]. Abemaciclib Reversible small molecular CDK inhibitor. Some gliomas have been shown to overexpresses cyclin D1, which in turn increases activity of CDK4 and CDK6; these enzymes phosphorylate (and therefore deactivate) retinoblastoma protein.
Deactivating retinoblastoma protein leads to excess cellular growth by loss of cell cycle control at the G1 to S phases. Abemaciclib has demonstrated some effect in slowing growth in preclinical glioblastoma models [54].
regarding the natural history of OPG and how we can predict outcomes. We have a paucity of understanding of the natural history of OPGs, and we lack tools to predict their clinical course and long-term outcomes. Furthermore, it is controversial which outcome measure should be used. Commonly used oncological outcome measures such as overall survival and (radiological) progression-free survival may not be the most appropriate for the study of OPGs; alternatives include visual function, endocrine/ hypothalamic dysfunction and quality-of-life measures. While one trial identified in this review relates to social functioning, this remains a poorly examined area. Hypothalamic dysfunction is a key component of the OPG disease that is often neglected in clinical studies and can be challenging to quantify. The hypothalamic consequences of existing and emerging therapies need examination. Questions remain regarding the optimal therapeutic management. Although general consensus statements exist, there remains uncertainty around the optimal surgical strategy. The timing and nature of the surgical approaches to OPGs have not been subjected to rigorous trial data. Surgical options include biopsy, partial/subtotal resection and radical resection [13]. Furthermore, the optimal timing of chemotherapy/radiotherapy initiation and the influence this has on overall outcomes are not fully known. The focus of current research, including all trials identified in this study, is on recurrent/refractory OPGs rather than new diagnosis. The long-term efficacy of various treatments for newly diagnosed OPGs is unknown. Optic pathway gliomas are typically treated as a single disease. However, it may be important to stratify and subclassify based on factors such as age (infantile versus juvenile) and based on molecular/genetic subtyping. An example of this is stratification into syndromic/NF1-related and non-syndromic/ non-NF1-related OPGs. These patients have different anatomical predilections, their natural histories can vary and the optimal therapies may be different [22][23][24].

Limitations
We found few trials that were specifically focussed on OPGs as a distinct tumour subtype. As OPGs are a subset of LGG, they may be bundled into trials under umbrella terms like 'LGG' or 'glioma' or 'JPA/PXA'. We not only developed a search strategy aimed at capturing trials that had specific reference to OPG (with the search terms; optic/visual/pathway/chiasm/midline/ hypothalamus/hypothalamic) but also included terms to capture trials that may include OPGs without specific reference to them, e.g. by using terms (Neurofibromatosis/NF1/paediatric/paediatric) combined with generic blanket terms (glioma/ astrocytoma/pilocytic/pilomyxoid). Where appropriate, we screened the trial protocols and inclusion/exclusion criteria to determine if OPGs might be included in the trials. As with any systematic review, there is always a risk that our search strategy missed a relevant trial. This risk is higher with a disease like OPG that is a subcategory of a wider disease. In order to try and minimise this danger, we used a reproducible search strategy that conformed to PRISMA guidelines. We elected to include all trials that could recruit OPGs; however, the vast majority are not focussed on OPGs but on LGG in general. We ensured that Various possibilities including modulation of autophagy, cellular metabolism or direct chemotoxic effects [55]. Sirolimus/everolimus First-and second-generation mTOR inhibitors. The mTOR pathway is important in control of cellular growth. These drugs have been trialled in giant cell astrocytomas as part of tuberous sclerosis complex, and everolimus is currently used in some cases that are not amenable to surgical resection [56,57].

Memantine
An NMDAR1 antagonist thought to have anti-tumour and neuroprotective effects. It has been studied in glioblastoma.

Poly-ICLC
A synthetic double-stranded RNA complex. Poly-ICLC is a ligand for toll-like receptor-3 and MDA-5 that can activate immune cells (including dendritic cells and natural killer cells) and therefore act as a stimulating immunomodulating agent to induce tumour cell killing [58]. Cyclophosphamide A chemotherapeutic agent. Cyclophosphamide forms an active metabolite phosphoramide mustard in cells with low levels of ALDH, as is the case in some tumour cells. The phosphoramide mustard forms irreversible DNA interstrand cross-links that trigger tumour cell apoptosis. It has been shown to enhance glioma virotherapy by inhibiting innate immune responses [59][60][61]. 5′Azacytidine An inhibitor of DNA methylation. This may be beneficial in LGGs with hypermethylation as part of their genetic phenotype [62].
BRAF rapidly accelerated fibrosarcoma gene B (B-Raf = protein), CDK cyclin-dependent kinase, DIPG diffuse intrinsic pontine glioma, ERK extracellular signal-regulated kinases, FGFR fibroblast growth factor receptor, HDAC histone deacetylase, Hh hedgehog, LGG low-grade glioma, MAPK selective mitogen-activated protein kinase, MEK selective mitogen-activated protein kinase, mTOR mammalian target of rapamycin, NF1 neurofibromatosis 1, NMDAR N-methyl-D-aspartate receptor, OPG optic pathway glioma, PDGFRa platelet-derived growth factor receptor alpha, RET rearranged during transfection receptor, Poly-ICLC polyinosinic-polycytidylic acid-poly-L-lysine carboxymethylcellulose, TK tyrosine kinase, VEGF vascular endothelial growth factor all included trials would allow OPGs as part of their criteria, but we cannot be sure that they will eventually recruit patients with this pathology in representative numbers, or if they will examine the data adequately to allow subgroup analysis of this pathology.

Conclusion
OPG are a debilitating childhood cancer that causes a significant burden of suffering. Our understanding of this disease is limited and we lack effective treatments and clear Table 3 Ongoing non-pharmacological trials related to optic pathway glioma

Object of investigation Explanation
Non-pharmacological therapeutics Ad-RTS-hIL-12 An inducible adenoviral (Ad) vector encoding human pro-inflammatory cytokine interleukin-12 (hIL-12) p70 transgene. This is under the transcriptional control of the RheoSwitch Therapeutic System (RTS) and can be activated by administration of the proprietary diacylhydrazine-based activator ligand veledimex. Activation and transcription of IL-12 is proposed to have immunomodulating and anti-neoplastic activities [63]. B7-H3-specific CAR T cell Chimeric antigen receptor (CAR) T cell therapy has emerged as a potential treatment in a range of cancers. CAR T cells are synthetic molecules composed of antibody binding domains connected to T cell activating domains and co-stimulatory domains. This allows highly specific tumour antigen recognition and subsequent T cell activation. The B7-H3 antigen is a transmembrane protein that is overexpressed in a range of tumours, and so manufacture of CAR T cells targeted to B7-H3 may be effective in a range of solid cancers including glioma [64,65]. Proton radiotherapy There were two studies identified that relate to proton radiotherapy. Protons are high-energy subatomic particles that when accelerated can be used to deliver focussed ionising radiation to a tumour in order to damage its DNA-killing tumour cells and inhibiting growth.
One study is focussed on assessing long-term neurocognitive, neuroendocrine and ototoxicity outcomes. The other is assessing the capacity of proton beam therapy to successfully avoid hippocampal irradiation in LGG. MRI-guided laser heat ablation MRI-guided laser ablation (MLA) is a minimally invasive laser surgery technique that involves a scalp incision and a small burr hole through which a laser probe is inserted into the centre of a brain tumour under MRI guidance. The probe delivers hyperthermic ablation to the tumour which can destroy tumour tissue and also disrupts the tumour blood-brain barrier potentially enhancing the delivery of chemotherapeutic agents [66].

Frameshift peptides
A frameshift mutation is caused by insertion of deletion of a nucleotide sequence into DNA that creates a 'frameshift' as it is not a multiple of 3 (the number of nucleotides in a codon). This frameshift alters the whole DNA sequence and results in an altered gene and subsequent protein product. These 'frameshift peptides' can act as new epitopes that are potential targets for tumour vaccines. The study NCT04212351 proposes to determine if NF1-associated LGGs have a specific frameshift peptide protein profile that could be targeted with a specific vaccine [67]. Imaging diagnostics Fast paediatric imaging with Wave-CAIPI Wave-CAIPI (controlled aliasing in parallel imaging) is an MRI-based technology that is faster than conventional MRI imaging. Parallel imaging works by acquiring a reduced amount of k-space data with an array of receiver coils. Research is underway to see if it can provide equivalent information to the current gold-standard MRI [68,69]. Positron emission tomography 18F-Fluorodeoxyglucose (FDG) positron emission tomography (PET) is a nuclear imaging modality that identifies increased glucose uptake in tissues. FDG-PET can be fused with MRI to improve the identification of high-grade features in gliomas [70]. Psychological assessment Social functioning Children with brain tumours are at risk for a number of late psychological effects, including neurocognitive and social deficits. These can be assessed with psychological questionnaires [16]. management consensus [1,25]. This manuscript details a systematic review of current ongoing trials in OPG. We find that from an initial yield of 501 hits, 36 ongoing trials fulfilled criteria for inclusion. The majority of these are studies or pharmacological agents, mostly phase I or II. OPG is a complex multifaceted disease, and advances in care are likely to require ongoing research efforts across a spectrum of different research fields [26]. Studies investigating surgical interventions and quality of life were notably absent. This review provides an update on the current state of research in OPG and highlights the main agents under investigation. We hope that this updates and stimulates clinicians and research scientists to engage with this important topic.