Primary optic nerve sheath meningioma

Keywords::

• arachnoid
• meningioma
• merlin
• neurofibromatosis
• optic nerve

Recent advances in molecular pathobiology and refinements in radiotherapy techniques have accelerated the evolution of the understanding and management of primary optic nerve sheath meningioma, and generated a new interest in this otherwise highly rare orbital tumor. Once considered benign but incurable, primary optic nerve sheath meningioma is now amenable to treatment with fractionated stereotactic radiotherapy, which results in preservation of vision and even in improvement in some selected patients, with very few side effects. Discoveries on signaling pathways that regulate growth and cellular adhesion in CNS glia are promising, and may offer new pharmacological tools.

Primary optic nerve sheath meningioma (pONSM) arises from the meningoendothelial cap cells of arachnoid villi within the dura of the optic nerve anywhere between the globe and the canalicular portion of the nerve [1–3]. A total of 92% of pONSMs develop within the intraorbital part whereas 8% arise from the intracanalicular segment [1]. They comprise 1–2% of all meningiomas, 5–10% of all orbital tumors and one-third of primary optic nerve tumors [1–3]. The mean age at diagnosis is reported to be 40.8 years, with a slight female preponderance [1,2]. However, 4–7% of pONSMs may be encountered in patients younger than 20 years of age and, compared with adults, tumors in this age group tend to be bilateral, grow faster and show intracranial extension more frequently [4]. In addition, pONSMs are more commonly associated with neurofibromatosis type 2 in children and the tumors are usually bilateral and multifocal [3–5].

The clinical presentation of painless progressive loss of vision, optic atrophy and retinochoroidal shunting veins on the optic disc (optociliary shunt vessels), also known as the Hoyt–Spencer triad, is typically suggestive of pONSM [1–3]. However, since any chronic compressive optic nerve lesion may present similarly, this triad of findings is no longer considered pathognomonic. Other ocular findings include lower eyelid edema and chemosis, proptosis, afferent pupillary defect, ocular motility disturbances, color vision deficiencies, disk edema and retinal or choroidal folds [1–3]. Common symptoms include gaze-evoked transient visual obscurations, diplopia, and orbital pain or headaches [2,3]. Visual field testing may reveal peripheral constrictions, central or cecocentral scotomas, and enlarged blind spot [3].

The diagnosis of pONSM is most often based on imaging results. CT scans may demonstrate diffuse tubular, fusiform or globular enlargement of the optic nerve and sometimes the ‘tram track’ sign, which is a hyperintense nerve sheath with central lucency [2,3,6]. Linear calcifications within the sheath may also contribute to the formation of this sign [3,6]. Occasionally, CT scans may show expanded posterior ethmoid or sphenoid sinuses called pneumosinus dilatans, which may be a clue for intracanalicular meningioma [3,6]. On MRI scans, pONSMs appear isointense to extraocular muscles on T1- and T2-weighted images [6]. Currently, gadolinium-enhanced fat suppression T1-weighted sequences best demonstrate the tumor surrounding the optic nerve. Meningiomas express somatostatin receptors, which has been the basis for utilizing indium-111-diethylenetriamine pentaacetic acid-octreotide in imaging pONSMs [7]. Saeed et al. found 100% sensitivity and 97% specificity of this technique in diagnosing pONSM and demonstrated significantly decreased uptake after radiotherapy, which may be used quantifiably in monitoring tumor response during follow-up [7]. However, the use of this somatostatin analogue will probably remain limited to difficult cases owing to its prohibitive cost.

Histopathologically, most pONSMs are of either the meningothelial or transitional meningioma subtypes. Almost all are slow-growing, well-differentiated, benign WHO grade I tumors [8]. However, the exceptionally rare rhabdoid, anaplastic and papillary variants are highly malignant (WHO grade III) [8]. Although the pathogenesis of meningioma has not been fully elucidated, studies on the NF2 gene, which is located on chromosome 22q12.2, continue to provide significant information. It has been shown that mutations in both alleles of the NF2 gene are found in 60% of sporadic meningiomas and in all neurofibromatosis type 2-associated meningiomas [9]. The NF2 gene encodes moesin-ezrin-radixin-like-protein (Merlin), a 69-kDa cytoskeleton–membrane-linking protein that acts as a tumor suppressor by regulating cellular adhesion, contact inhibition and EGF receptor response [9–12]. Several mechanisms of action for Merlin have been proposed. Houshmandi et al. elegantly demonstrated in knockout mice that Merlin negatively regulates glial cell proliferation through binding to erythroblastic leukemia viral oncogene homologue 2 (ErbB2), thus preventing the activation of src tyrosine kinases [13]. Src plays key roles in cellular motility, growth and proliferation through downstream signaling via FAK and paxillin [13]. Striedinger et al. studied the evolutionarily conserved Hippo pathway, which regulates cellular proliferation and apoptosis, in human meningioma samples and found that Merlin loss was associated with increased S-phase entry, cellular proliferation and increased intracellular Yes-associated protein (YAP) levels [11]. The authors concluded that Merlin regulated cellular proliferation in human meningioma cells by suppressing the YAP protein, the key element in the Hippo pathway [11]. An alternative hypothesis came from the study by Chang et al. in which the loss of heterozygosity of NF2 in meningiomas was linked to an increase in proto-oncogene Mdm2 levels, which in turn downregulated the tumor suppressor activity of the p53 pathway [14]. In addition, the same authors demonstrated the crucial role of the p53 codon 72 arginine-to-proline polymorphism in meningiomas. Accordingly, the presence of the Pro72 allele is associated with a higher WHO grade and aggressive meningioma, whereas the Arg72 allele induces apoptosis [14]. These new findings will certainly pave the way for precise targeted therapy in the near future.

The management of pONSMs still generates controversy. Above all, the main objective is to preserve visual function and impede the progression of the tumor. In the past, observation until disfiguring proptosis, intracranial extension or severe loss of vision developed was the standard care. Currently, a more aggressive approach is adopted, as the tumor is slowly progressive and leads to seizures, hemiparesis and legal blindness in most patients if left untreated or treated late.

Surgery enjoys a limited role in the management of pONSMs. Anteriorly located extradural, exophytic and focal tumors can be excised without damaging the optic nerve [15,16]. Tumors that leave no prospect of useful vision and threaten the unaffected eye through intracranial extension can be resected [15,16]. In addition, severe proptosis, corneal exposure and compartment syndrome of the optic nerve may be indications for surgery [16]. Roser et al. reported their surgical experience on 22 patients and found that six had preserved or improved vision [15]. They recommend surgery for patients with rapid decline in visual acuity in order to immediately relieve the compression on the optic nerve, which cannot be achieved with radiotherapy [15].

Current evidence suggests that radiotherapy is the best option for long-term preservation of visual function and local tumor control. In the 1980s, conventional radiotherapy administered as 50–54 Gy in 175–200 cGy per fraction often resulted in improvements in both visual fields and visual acuity outcomes [17,18]. These techniques employed lateral, two parallel or coplanar opposing fields with relatively few side effects, which included dry eyes, radiation retinopathy, cataracts and temporal lobe atrophy [18].

Stereotactic fractionated radiotherapy (SFRT) represents a significant improvement over conventional techniques by offering stereotactic precision and conformality to the shape of the tumor, thus sparing adjacent normal tissues [18]. In 2007, Jeremic and Pitz reviewed the results of seven studies reporting on 89 patients treated with SFRT [18]. Prescribed radiation doses varied between 40 and 59.1 Gy, and the documented improvements in visual acuity varied between 42 and 100% [18]. It emerged from these studies that SFRT given as 50–54 Gy in 1.8 Gy per fraction provides at least 80% stability or improvement in visual function [18]. Another study reported on 32 patients who received a median total dose of 54.9 Gy with 1.8 Gy per fraction five times a week [19]. After a mean follow-up of 4.5 years, the authors found that 38% of patients had improvement in pre-existing clinical symptoms and 97% of patients had stable or improved visual acuity [19]. Interestingly, 19% of patients had reduced tumor volume, whereas in 81% of cases the tumor remained stable based on imaging studies [19]. Richards et al. similarly noted that no change in tumor volume was observed following SFRT [20]. Recently, Saeed et al. published the retrospective data of three orbital centers involving 34 patients [21]. A total of 22 patients received conventional radiotherapy, while 12 received stereotactic fractionated conformal radiotherapy. Stabilization or improvement in visual function was documented in 91% of patients [21]. Importantly, no significant difference was found in terms of visual outcome between conventional radiotherapy and SFRT groups [21].

A remarkable technical advance is the use of frameless image-guided robotic radiosurgery (CyberKnife^®, Accuray Inc., CA, USA), which has the advantages of more available beam trajectories and enhanced dose conformality and homogeneity due to the absence of a fixed rigid headframe [22]. A preliminary study on three patients showed progressive improvements in visual fields and acuity after receiving an 80% prescribed isodose of 20 Gy in four sessions [22]. Full restoration to normal vision was documented in each patient 1 year later.

Conclusion

Primary optic nerve sheath meningioma is a rare orbital tumor, thus designing randomized controlled trials to assess the best treatment option seems unrealistic. However, the existing literature convincingly favors fractionated stereotactic radiotherapy as the most effective modality to stop the progression of the tumor and to stabilize or even to ameliorate the visual function. Technical improvements and refinements with increased precision will further limit the collateral damage caused by radiation, and this will allow the institution of an earlier treatment as soon as the diagnosis is made, even if the patient is asymptomatic or does not have any visual functional defects. Surgery will perhaps enjoy a minor role and will most likely be reserved for advanced cases or for biopsy purposes in atypical tumors.

Another major advance will undoubtedly come from the pathogenetic studies, particularly on patients with NF2. Further information on the role of NF2 protein Merlin and related intracellular pathways, membrane receptors and key molecules will be instrumental in the development of targeted therapies. The ErbB2 inhibitor trastuzumab and Src kinase inhibitor dasatinib are already in use against breast cancer and may have potential therapeutic applications in pONSMs as well [13].

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.