- Email: [email protected]
Optic Nerve Sheath Meningiomas — Non-surgical Treatment
Clinical Oncology (2009) 21: 8e13 doi:10.1016/j.clon.2008.10.010
Original Article
Optic Nerve Sheath Meningiomas d Non-surgical Treatment R. I. Smee, M. Schneider, J. R. Williams Department of Radiation Oncology, The Prince of Wales Cancer Centre, High Street, Randwick, New South Wales, Australia
ABSTRACT: Aims: Optic nerve sheath meningiomas typically present with unilateral visual deterioration. Here, a single centre’s experience with radiotherapy aimed at local control and visual stabilisation is presented. Materials and methods: The meningioma database within the Radiation Oncology Department, Prince of Wales Hospital was audited for patients whose meningiomas took origin from the optic nerve sheath. Excluded from this evaluation was any patient whose meningioma secondarily involved the optic nerve. Where vision was not a consideration, treatment was given by stereotactic radiosurgery for patients with retained vision. The remaining patients were treated by fractionated radiotherapy, predominately via a stereotactic approach. The main end points were: lack of radiological progression of the tumour and maintenance of preradiotherapy vision. Results: There were 15 eligible patients, one patient with neurofibromatosis had bilateral optic nerve involvement; thus, 16 optic nerves were treated. Women (10) outnumbered men (five) and the age range was 7e74 years. One patient progressed outside the volume treated (for a geographical failure) with no infield progression. This patient became blind, was re-treated by stereotactic radiosurgery, had tumour control and vision improved. Thus, for 17 optic nerves (or part thereof) treated, all patients ultimately had local control (100%) with worsening of vision only occurring in one patient. No other late morbidity was present for any patient. Conclusion: Optic nerve sheath meningiomas have high local control rates and preservation of vision with radiotherapy. Smee, R. I. et al. (2009). Clinical Oncology 21, 8—13 ª 2008 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Meningiomas, optic nerve, radiotherapy, stereotactic
Introduction Although meningiomas are reasonably common intracranial tumours, optic nerve sheath meningiomas (ONSM) as a specific subsite are infrequent, representing only 1e2% of all primary sites [1e3]. Visual deterioration is a usual presenting feature, although there may be associated retro-orbital headaches and proptosis [4e7]. They are predominantly unilateral, the presence of bilateral disease usually indicates a diagnosis of neurofibromatosis (NF2). The diagnosis of this condition relies upon a combination of the appropriate clinical features with the radiological evidence of a thickened optic nerve on magnetic resonance imaging (MRI) with fat suppression sequences defining the true dimensions of the tumour [1,2,4,5]. Contrast enhancement may be necessary to define involvement of the chiasm. Treatment in the past has usually involved a surgical procedure, either excision of the involved nerve when the patient has unilateral blindness, or decompression when mass effect features are apparent [1,2,4]. However, in the former situation, compromise of function is expected, and in the latter there is no intention to address the tumour itself merely to address the compression features it is 0936-6555/09/210008þ06 $36.00/0
causing. Radiotherapy has a long history of treating meningiomas [8e10], its role in addressing tumours arising from the optic nerve meningeal sheath has as its dual aims control of tumour growth and preservation of nerve function [11e16]. This paper will address a single centre’s experience of this tumour where increasing emphasis has been to use more sophisticated approaches with adjacent structures defined as organs at risk, and the dose to these structures consequently minimised. These results will be presented in the context of other published series.
Materials and Methods All records stored in the Tumour Registry of the Radiation Oncology Department of the Prince of Wales Cancer Centre were searched for patients with a clinical and/or histological diagnosis of meningiomas. This began initially as a retrospective and then subsequently a prospective meningioma database, the start date being January 1990. This database was then sourced to establish tumours arising from the optic nerve sheath (primary) and differentiated from tumours extending to the optic nerve (secondary). The
ª 2008 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
ONSM d NON-SURGICAL TREATMENT
final patient in this review was treated before February 2004, for a minimum follow-up of 48 months. The information used in this review was obtained from the aboveindicated database, as well as from hospital notes, referral and follow-up letters, and infrequently by contacting the patient or their family. This was an ethics-approved study. The year 1990 was chosen as the commencement point of the study as it marked the beginning of any form of stereotactic radiotherapy (SRT) in this department (and in Australia). Single-dose stereotactic radiosurgery (SRS) was initially the only modality available in a highly conformal sense until 1995, when relocatable head fixation devices became available for fractionated stereotactic treatment (SRT). SRS has as its goal the control of the tumour without attention to function, whereas with the relocatable devices used for fractionated SRT, control of the tumour and preservation of function are the twin aims. SRT approaches have the principle of defining a specific point (the isocentre) within the middle of the tumour to which the radiation beam is directed, and then using specialised shaping devices so that the beam conforms to the specific shape and volume of the tumour in a true threedimensional sense. All patients were treated with 6 Mev photons on a linear accelerator, specifically set up for stereotactic work. Beam shaping for SRS was done via multi-sized (in increments of 0.25 cm diameter) cylindrical cones, or a 4 mm leaf width mini multileaf collimator for SRT. Both devices can be attached via the shadow tray aperture. Where the patient was blind in that eye then tumour control was the goal (SRS) and rigid head fixation was used. However, where the aim was to preserve function, i.e. vision, a relocatable head device was used for fractionated SRT. The radiotherapy planning process used fused MRI/ computed tomography images via the planning software platform. This enabled precise demonstration of tissues receiving any dose both in a qualitative and quantitative sense, with an interactive inverse planning component enabling definition of organ-specific dose constraints for stereotactic intensity-modulated radiotherapy (IMRT). The planning target volume was the radiologically abnormal area on MRI (contrasted T1, axial and coronal) plus a variable margin, minimum 2 mm. Because treatment was delivered stereotactically there was rigid head fixation, with no organ movement. Daily depth helmet checks were carried out to establish that the relocatable head ring was being set to the same position plus weekly port-films were used for dose verification. This is in the context of many, many thousands of fractions delivered since 1995. Ongoing evaluation of these parameters indicates negligible inter-treatment variation. Regular audit of these ensures stability of the treatment parameters, not requiring any alteration in the treatment technique. The dose used related to the purpose of treatment. SRS was used in three patients who each had had previous resection, one of whom was already blind. The first patient was treated in 1993, receiving 20 Gy to the 100% isodose curve surrounding the tumour (higher doses were thought necessary to control meningiomas in the early 1990s). In addition to this, SRS was implemented as re-treatment of
9
a patient who had an out-of-field failure after initial SRT treatment and received 12 Gy to the 90% isodose (see case 2). For the SRT patients, fractions of 1.8e2 Gy were used to a total of 50.4 Gy where visual preservation was the aim of treatment. The current approach of stereotactic IMRT for this type of patient requires appointment times of about 30e40 min and thus can be fitted into daily busy linear accelerator scheduling. Table 1 outlines the details of radiotherapy treatment received by each patient. The end points used in this study were two-fold: tumour control and vision preservation (and, if possible, improvement). MRI follow-up was the definitive means of evaluating progress with the tumour. Because a number of patients were from interstate, reliance was placed on the MRI report for the evaluation of tumour response. Volume assessment during follow-up was therefore not possible, and the evaluation was thus qualitative, relying on the MRI report in which there is a comment regarding comparison with previous scans. Change was designated as occurring, increase or decrease, if there was a 25% variation in tumour dimension. The general pattern of MRI evaluation (with specific reliance on a fat suppressed T2 sequence) is a follow-up scan at 6 months after treatment, then 12 monthly for 2e3 years, then second yearly, provided no other event occurs. Visual assessment was done on the basis of the ophthalmologist’s report, noting the patient’s comment, only reporting simply as better, worse or the same, given the variable reporting in follow-up letters. Given the small number of patients treated, no statistical evaluation was carried out; all end points are recorded as crude events.
Results During the specified time frame (June 1990eFebruary 2004), with a median follow-up of 86.4 months (range 5.5e157 months), there were 15 patients with ONSM who were treated, and 11 who were seen for consultation only. One patient had NF2 with bilateral biopsy-proven ONSM, one patient had different parts of the same optic nerve treated twice. Thus, for the 15 patients, 17 optic nerves could be evaluated for treatment and outcome. As is typical for meningiomas, there was a predominance of women (10) vs men (five), with an age range of 7e74 years; the 7 year old, a male, had NF2. All patients presented with visual impairment, only one patient was blind in the ipsilateral eye at presentation, one patient became blind in the ipsilateral eye after treatment. Visual deterioration was apparent over many months in four of the 15 patients before referral, with their medical practitioner/ophthalmologist being aware of this. Unfortunately for this retrospective study, formal ophthalmological assessment at the time of initial assessment was not always available for patients. Proptosis was apparent at presentation in five patients; in three patients, ipsilateral orbital/forehead headaches were noted. Treatment was tolerated well, with anterior pin site bruising for a variable time frame of a number of days for
Gender Radiotherapy details Type Dose (Gy) Fractions
CRT, conventional radiotherapy; SRS, stereotactic radiosurgery; FSRT, fractionated stereotactic radiotherapy; MMLC, mini multileaf collimator; IMRT, intensity-modulated radiotherapy. *Recurred outside treated volume, additional tumour treated by SRS. yBoth optic nerves treated.
FSRT/MMLC 50.40 28 FSRT 50.40 28 FSRT 50.40 28 FSRT 50.00 28 SRS 20.00 1 CRT 45.00 25 CRT 55.00 31
SRS 20.00 1
F F F M
F
CRT 56.00 31
SRS 20.00 1
FSRT 50.40 28
F F F F M
M
FSRT/MMLC 50.40 28
FSRT/MMLC 50.40 28
FSRT/MMLC/IMRT 50.68 28
M F M
F
14 13* 12 11 10 9 8 7 6 5 4 3 2 1 Patient
Table 1 e Radiotherapy treatment details
FSRT/MMLC/IMRT 50.05 28
CLINICAL ONCOLOGY
15y
10
patients having rigid head fixation for SRS, and intermittent mouth discomfort for patients having dental plate-based relocatable head ring SRT. Most patients experienced a variable degree of fatigue of short duration after treatment. There was no other reported late morbidity. Because the primary end point was tumour control, this was evaluated in the context of control or progression within the treated volume. An out-of-field failure does not represent a failure within the volume treated, rather inadequate assessment of the volume required for treatment. Hence, outcomes are reported in terms of treated tumours rather than number of optic nerves. Thus, 17 optic nerves had radiologically stable tumours (this representing the number of distinctly different tumours treated), with 15 optic nerves having stable to slightly improved outcome in terms of vision. Two case histories detail the different planning aspects of treating these types of tumour.
Case 1 A 7-year-old boy with known NF2 had bilateral ONSM as his presenting feature. Both nerves had been surgically decompressed to try to preserve vision. Both tumours were radiologically progressing in a 6 month period before referral. Thus, the target volume was both optic nerves and the connecting chiasm, a U-shaped lesion. The defined dose critical structures were the lens, lacrimal glands, retina, temporal and frontal lobes, pituitary and brain stem. The hypothalamus could not be spared. Stereotactic IMRT was used with doseevolume histograms indicating good coverage of the ‘U’ and separation by dose of the adjacent normal tissues (Fig. 1a,b). A mean dose of 5005 cGy in 28 fractions was delivered. Five years after treatment there has been no radiological progression with stable vision and the patient maintaining good academic performance at school.
Case 2 A 55-year-old women presented with a radiologically evident ONM confined to the intra-orbital optic nerve. The planning target volume was thus the radiologically abnormal nerve plus a margin covering the MRI normal component of the optic nerve within the optic canal plus 5 mm of intracranial optic nerve, for a proximal margin of 15 mm. Given that there was poor vision retained, SRT was used. A mean dose of 5040 cGy in 28 fractions was delivered. There was initial visual improvement; however, 12 months after SRT the patient presented with a 3 month history of ipsilateral blindness. MRI indicated a globular mass arising from the residual intracranial optic nerve, adjacent to the chiasm, that had not been treated. This lesion was then treated by SRS with a dose of 1200 cGy to the 90% isodose covering this lesion with no further radiological progression, with some visual improvement to be able to discern light and dark, and shapes. Six years later there has been no manifestation of further recurrence.
ONSM d NON-SURGICAL TREATMENT
11
Fig. 1 e (a) Three-dimensional display of tumour volume and adjacent dose critical structures. (b) Doseevolume histogram representing uniformity of tumour coverage with separation of volume structures between optic nerve sheath meningiomas and low dose normal tissues.
Discussion Visual loss is the main presenting feature for patients diagnosed with an ONSM [1,2,4,17]. However, it is important to differentiate between tumours that arise from the optic nerve sheath (primary) and tumours that arise adjacent to the optic nerve and by direct extension spread along the optic nerve and can cause visual impairment (secondary) [2]. Examples of the latter include meningiomas arising on the cavernous sinus, sphenoid ridge and olfactory groove. This distinction was documented as early as 1949 [6]. All ONSM have the usual female predominance in incidence and are typically unilateral [1,2,5]. The only exception is where the ONSM arises on the background of NF2, where it is usually bilateral. The extent of involvement of the optic nerve can be difficult to determine even with MRI [4]. Involvement of the contralateral side in the absence of NF2 is infrequent, although Dutton [1] reported that 38% of his patients had this extent of disease if the meningioma had spread to involve the nerve sheath within the optic canal. Given the failure pattern in one patient in this series, treating the whole length of the optic nerve would seem more appropriate. If there is intracranial extension, then treating part of the chiasm should also be considered.
The appropriate consideration then is what should be the recommended ‘treatment’ [5]. Although Kennerdell et al. [17] reported that 10% of surgically treated patients had visual improvement after treatment, this was in the context of a decompressive procedure; long-term followup was not reported for this 10% of patients nor indeed even a median follow-up. Basso et al. [2] reported that surgery is not usually successful in preserving vision, with Schick et al. [18] in agreement. Any attempts at appropriate excision will invariably lead to visual deterioration if not blindness [3]. In this circumstance, treatment is usually delayed under medical supervision until the patient is near blind, as in four patients in this series. As with meningiomas at other sites, radiotherapy has a high likelihood of tumour control and preservation of function [7,10]. Andrews et al. [7] recorded 30 patients with 33 optic nerves treated by SRT with a median followup of 89 weeks; tumour control was evident in 100%, with a consistent dose of 50.4 Gy being used. This high tumour control rate is not always seen. Turbin et al. [4] reported eight of 16 patients having surgery and radiotherapy who ultimately progressed where a range of doses was used. Basso et al. [2] noted three of 14 patients progressing with no details regarding the doses or fields used. In this series
12
CLINICAL ONCOLOGY
there was progression outside the treatment volume in only one of 18 ONSM treated. Preservation of function (vision) and, if possible, improvement are the other aspects of considering radiotherapy [11,14]. Turbin et al. [4], in a predominantly surgical paper, indicated that patients receiving radiotherapy alone had the best likelihood of a favourable visual outcome. Baumert et al. [19] reported 95% ‘visual control’ (21/22) with visual improvement in 16 patients. Andrews et al. [7] noted visual deterioration in only two patients; thus, 31 of 33 ONSM had stable to improved vision. Others report similar findings [9,15,16,18,19]. In this series, only one patient (with one optic nerve) had permanent visual deterioration, potentially a treatment-related event. Complications of treatment are also important to consider. Turbin et al. [4] reported that six of 18 patients (33%) had radiotherapy-related complications with retinopathy/vascular occlusion in four, iritis in one and temporal lobe atrophy in one. Richards et al. [13] also reported radiologically detected cerebral change in one of four patients. Andrews et al. [7] noted optic neuritis in one patient. Parsons et al. [20] documented the extent of visual injury after radiotherapy, reporting a 0% incidence of optic nerve injury in 106 optic nerves where total doses of less than 59 Gy were used with fractionation sizes of less than 1.9 Gy. For this population, one patient became blind in the ipsilateral eye after treatment with no radiological progression; thus probably a treatment-related event. The dose delivered to that optic nerve was no greater than that given to other patients. In terms of other effects, shortterm fatigue was frequent, and reversible, but not universal, with no other reported late morbidity. In terms of recommendations for treatment, it is easy to begin by stating that ONSM is not a surgically managed disease unless surgery is used to decompress where a mass effect is evident. Observing a patient with the aim of operating when the patient ultimately becomes blind is also inappropriate. Radiotherapy, in most modern series, is reported to have high tumour control figures with now prolonged follow-up. If visual outcome is not a consideration, then treatment can be delivered by SRS with a designated dose of 12e14 Gy to an appropriate isodose. However, where visual stabilisation/improvement is the goal, fractionated radiotherapy is delivered with a dose of 50.4e54 Gy with appropriate fractions of 1.8e1.9 Gy. Although higher doses have been used in some series reporting on the outcomes of treating mengiomas by radiotherapy, there is little evidence to indicate a better outcome in terms of local control than is reported here [21]. Current literature recommendations are for this dose range. The only exception where a higher dose may be required is for meningiomas displaying a more aggressive clinical or histological pattern [21]. The volume of nerve to be treated is critical. Patterns of failure are always the proximal not distal component of the nerve, with no surgical or radiotherapy series reporting failure distal to the globe. Therefore, significant components of the retina do not need to be included, reducing the risk of retinal deficit. None of the surgical series has recorded the
necessity to enucleate the globe because of intra-ocular extension, where the only structure resected is the optic nerve. Proximally the whole length of the nerve needs to be treated, and if there is intracranial extension then part of the chiasm also needs to be treated. In the context of NF2 with bilateral nerve involvement, the authors’ recommendation is to treat both nerves and the whole chiasm. Delivering this treatment stereotactically makes a complex treatment simple and enables the dose to adjacent critical structures to be minimised. ONSM are infrequent, and there is now surgical acceptance that treatment aimed at tumour control and visual stabilisation should be instituted sooner rather than later. SRT presents an ideal clinical circumstance, minimising significant dose to adjacent dose-sensitive structures [13,15,16,19,22]. Author for correspondence: R. Smee, Department of Radiation Oncology, The Prince of Wales Cancer Centre, High Street, Randwick 2031, NSW, Australia; E-mail: [email protected] Received 13 June 2008; received in revised form 14 October 2008; accepted 23 October 2008
References 1 Dutton JJ. Optic nerve sheath meningiomas. Surv Ophthalmol 1992;37:167e183. 2 Basso A, Carrizo A, Kreutel A, Tomecek F. Primary intraorbital meningiomas. In: Schmidek HH, editor. Meningiomas and their surgical management. Philadelphia: WB Saunders; 1991. p. 311e323. 3 Turbin RE, Pokorny K. Diagnosis and treatment of orbital optic nerve sheath meningioma. Cancer Control 2004;11(5):334e341. 4 Turbin RE, Thompson CR, Kennerdell JS, Cockerham KP, Kupersmith MJ. A long-term visual outcome comparison in patients with optic nerve sheath meningioma managed with observation, surgery, radiotherapy, or surgery and radiotherapy. Ophthalmology 2000;109:890e900. 5 Carrasco JR, Penne RB. Optic nerve sheath meningiomas and advanced treatment options. Curr Opin Ophthalmol 2004;15: 406e410. 6 Craig W, Gogela L. Intraorbital meningiomas: a clinicopathologic study. Am J Ophthalmol 1949;32:1663. 7 Andrews DW, Faroozan R, Yang BP, et al. Fractionated stereotactic radiotherapy for the treatment of optic nerve sheath meningiomas: preliminary observations of 33 optic nerves in 30 patients with historical comparison to observation with or without prior surgery. Neurosurgery 2002;51:890e904. 8 Debus J, Wuendrich M, Pirzkall A, et al. High efficacy of fractionated stereotactic radiotherapy of large base-of-skull meningiomas: long-term results. J Clin Oncol 2001;19:3547e3553. 9 Selch M, Ahn E, Laskari A, et al. Stereotactic radiotherapy for treatment of cavernous sinus meningiomas. Int J Radiat Oncol Biol Phys 2004;59:101e111. 10 Pirzkall A, Debus J, Haering P, et al. Intensity modulated radiotherapy (IMRT) for recurrent, residual, or untreated skullbase meningiomas: preliminary clinical experience. Int J Radiat Oncol Biol Phys 2003;55(2):362e372. 11 Lui JK, Forman S, Hershewe GL, et al. Optic nerve sheath meningiomas: visual improvement after stereotactic radiotherapy. Neurosurgery 2002;50:950e957.
ONSM d NON-SURGICAL TREATMENT 12 Narayan S, Cornblath WT, Sandler HM, et al. Preliminary visual outcomes after three-dimensional conformal radiation therapy for optic nerve sheath meningioma. Int J Radiat Oncol Biol Phys 2003;56:537e543. 13 Richards JC, Roden D, Harper CS. Management of sightthreatening optic nerve sheath meningioma with fractionated stereotactic radiotherapy. Clin Experiment Ophthalmol 2005; 33(2):137e141. 14 Melian E, Jay WM. Primary radiotherapy for optic nerve sheath meningioma. Semin Ophthalmol 2004;19(3e4):130e140. 15 Becker G, Jeremic B, Pitz S, et al. Stereotactic fractionated radiotherapy in patients with optic nerve sheath meningioma. Int J Radiat Oncol Biol Phys 2002;54(5):1422e1429. 16 Pitz S, Becker G, Schiefer U, et al. Stereotactic fractionated irradiation of optic nerve sheath meningioma: a new treatment alternative. Br J Ophthalmol 2002;86(11):1265e1268. 17 Kennerdell JS, Maroon JC, Malton M, et al. The management of optic nerve sheath meningiomas. Am J Ophthalmol 1988;106:450e457.
13
18 Schick U, Dott U, Hassler W. Surgical management of meningiomas involving the optic nerve sheath. J Neurosurg 2004;101: 951e959. 19 Baumert BG, Villa S, Studer G, et al. Early improvements in vision after fractionated stereotactic radiotherapy for primary optic nerve sheath meningioma. Radiother Oncol 2004;72(2): 169e174. 20 Parsons JT, Bova FJ, Fitzgerald CR, et al. Radiation optic neuropathy after megavoltage external-beam irradiation: analysis of time-dose factors. Int J Radiat Oncol Biol Phys 1994;30: 755e763. 21 Bulsara KR, Al-Mefty O, Shrieve DC, Angtuaco EJ. Meningiomas. In: Berger MS, Prados MD, editors. Textbook of neurooncology. Philadelphia: Elsevier Saunders; 2005. p. 335e345. 22 Moyer PD, Golnik KC, Breneman J. Treatment of optic nerve sheath meningioma with three-dimensional conformal radiation. Am J Ophthalmol 2000;129:694e696.
Original Article
Optic Nerve Sheath Meningiomas d Non-surgical Treatment R. I. Smee, M. Schneider, J. R. Williams Department of Radiation Oncology, The Prince of Wales Cancer Centre, High Street, Randwick, New South Wales, Australia
ABSTRACT: Aims: Optic nerve sheath meningiomas typically present with unilateral visual deterioration. Here, a single centre’s experience with radiotherapy aimed at local control and visual stabilisation is presented. Materials and methods: The meningioma database within the Radiation Oncology Department, Prince of Wales Hospital was audited for patients whose meningiomas took origin from the optic nerve sheath. Excluded from this evaluation was any patient whose meningioma secondarily involved the optic nerve. Where vision was not a consideration, treatment was given by stereotactic radiosurgery for patients with retained vision. The remaining patients were treated by fractionated radiotherapy, predominately via a stereotactic approach. The main end points were: lack of radiological progression of the tumour and maintenance of preradiotherapy vision. Results: There were 15 eligible patients, one patient with neurofibromatosis had bilateral optic nerve involvement; thus, 16 optic nerves were treated. Women (10) outnumbered men (five) and the age range was 7e74 years. One patient progressed outside the volume treated (for a geographical failure) with no infield progression. This patient became blind, was re-treated by stereotactic radiosurgery, had tumour control and vision improved. Thus, for 17 optic nerves (or part thereof) treated, all patients ultimately had local control (100%) with worsening of vision only occurring in one patient. No other late morbidity was present for any patient. Conclusion: Optic nerve sheath meningiomas have high local control rates and preservation of vision with radiotherapy. Smee, R. I. et al. (2009). Clinical Oncology 21, 8—13 ª 2008 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Meningiomas, optic nerve, radiotherapy, stereotactic
Introduction Although meningiomas are reasonably common intracranial tumours, optic nerve sheath meningiomas (ONSM) as a specific subsite are infrequent, representing only 1e2% of all primary sites [1e3]. Visual deterioration is a usual presenting feature, although there may be associated retro-orbital headaches and proptosis [4e7]. They are predominantly unilateral, the presence of bilateral disease usually indicates a diagnosis of neurofibromatosis (NF2). The diagnosis of this condition relies upon a combination of the appropriate clinical features with the radiological evidence of a thickened optic nerve on magnetic resonance imaging (MRI) with fat suppression sequences defining the true dimensions of the tumour [1,2,4,5]. Contrast enhancement may be necessary to define involvement of the chiasm. Treatment in the past has usually involved a surgical procedure, either excision of the involved nerve when the patient has unilateral blindness, or decompression when mass effect features are apparent [1,2,4]. However, in the former situation, compromise of function is expected, and in the latter there is no intention to address the tumour itself merely to address the compression features it is 0936-6555/09/210008þ06 $36.00/0
causing. Radiotherapy has a long history of treating meningiomas [8e10], its role in addressing tumours arising from the optic nerve meningeal sheath has as its dual aims control of tumour growth and preservation of nerve function [11e16]. This paper will address a single centre’s experience of this tumour where increasing emphasis has been to use more sophisticated approaches with adjacent structures defined as organs at risk, and the dose to these structures consequently minimised. These results will be presented in the context of other published series.
Materials and Methods All records stored in the Tumour Registry of the Radiation Oncology Department of the Prince of Wales Cancer Centre were searched for patients with a clinical and/or histological diagnosis of meningiomas. This began initially as a retrospective and then subsequently a prospective meningioma database, the start date being January 1990. This database was then sourced to establish tumours arising from the optic nerve sheath (primary) and differentiated from tumours extending to the optic nerve (secondary). The
ª 2008 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
ONSM d NON-SURGICAL TREATMENT
final patient in this review was treated before February 2004, for a minimum follow-up of 48 months. The information used in this review was obtained from the aboveindicated database, as well as from hospital notes, referral and follow-up letters, and infrequently by contacting the patient or their family. This was an ethics-approved study. The year 1990 was chosen as the commencement point of the study as it marked the beginning of any form of stereotactic radiotherapy (SRT) in this department (and in Australia). Single-dose stereotactic radiosurgery (SRS) was initially the only modality available in a highly conformal sense until 1995, when relocatable head fixation devices became available for fractionated stereotactic treatment (SRT). SRS has as its goal the control of the tumour without attention to function, whereas with the relocatable devices used for fractionated SRT, control of the tumour and preservation of function are the twin aims. SRT approaches have the principle of defining a specific point (the isocentre) within the middle of the tumour to which the radiation beam is directed, and then using specialised shaping devices so that the beam conforms to the specific shape and volume of the tumour in a true threedimensional sense. All patients were treated with 6 Mev photons on a linear accelerator, specifically set up for stereotactic work. Beam shaping for SRS was done via multi-sized (in increments of 0.25 cm diameter) cylindrical cones, or a 4 mm leaf width mini multileaf collimator for SRT. Both devices can be attached via the shadow tray aperture. Where the patient was blind in that eye then tumour control was the goal (SRS) and rigid head fixation was used. However, where the aim was to preserve function, i.e. vision, a relocatable head device was used for fractionated SRT. The radiotherapy planning process used fused MRI/ computed tomography images via the planning software platform. This enabled precise demonstration of tissues receiving any dose both in a qualitative and quantitative sense, with an interactive inverse planning component enabling definition of organ-specific dose constraints for stereotactic intensity-modulated radiotherapy (IMRT). The planning target volume was the radiologically abnormal area on MRI (contrasted T1, axial and coronal) plus a variable margin, minimum 2 mm. Because treatment was delivered stereotactically there was rigid head fixation, with no organ movement. Daily depth helmet checks were carried out to establish that the relocatable head ring was being set to the same position plus weekly port-films were used for dose verification. This is in the context of many, many thousands of fractions delivered since 1995. Ongoing evaluation of these parameters indicates negligible inter-treatment variation. Regular audit of these ensures stability of the treatment parameters, not requiring any alteration in the treatment technique. The dose used related to the purpose of treatment. SRS was used in three patients who each had had previous resection, one of whom was already blind. The first patient was treated in 1993, receiving 20 Gy to the 100% isodose curve surrounding the tumour (higher doses were thought necessary to control meningiomas in the early 1990s). In addition to this, SRS was implemented as re-treatment of
9
a patient who had an out-of-field failure after initial SRT treatment and received 12 Gy to the 90% isodose (see case 2). For the SRT patients, fractions of 1.8e2 Gy were used to a total of 50.4 Gy where visual preservation was the aim of treatment. The current approach of stereotactic IMRT for this type of patient requires appointment times of about 30e40 min and thus can be fitted into daily busy linear accelerator scheduling. Table 1 outlines the details of radiotherapy treatment received by each patient. The end points used in this study were two-fold: tumour control and vision preservation (and, if possible, improvement). MRI follow-up was the definitive means of evaluating progress with the tumour. Because a number of patients were from interstate, reliance was placed on the MRI report for the evaluation of tumour response. Volume assessment during follow-up was therefore not possible, and the evaluation was thus qualitative, relying on the MRI report in which there is a comment regarding comparison with previous scans. Change was designated as occurring, increase or decrease, if there was a 25% variation in tumour dimension. The general pattern of MRI evaluation (with specific reliance on a fat suppressed T2 sequence) is a follow-up scan at 6 months after treatment, then 12 monthly for 2e3 years, then second yearly, provided no other event occurs. Visual assessment was done on the basis of the ophthalmologist’s report, noting the patient’s comment, only reporting simply as better, worse or the same, given the variable reporting in follow-up letters. Given the small number of patients treated, no statistical evaluation was carried out; all end points are recorded as crude events.
Results During the specified time frame (June 1990eFebruary 2004), with a median follow-up of 86.4 months (range 5.5e157 months), there were 15 patients with ONSM who were treated, and 11 who were seen for consultation only. One patient had NF2 with bilateral biopsy-proven ONSM, one patient had different parts of the same optic nerve treated twice. Thus, for the 15 patients, 17 optic nerves could be evaluated for treatment and outcome. As is typical for meningiomas, there was a predominance of women (10) vs men (five), with an age range of 7e74 years; the 7 year old, a male, had NF2. All patients presented with visual impairment, only one patient was blind in the ipsilateral eye at presentation, one patient became blind in the ipsilateral eye after treatment. Visual deterioration was apparent over many months in four of the 15 patients before referral, with their medical practitioner/ophthalmologist being aware of this. Unfortunately for this retrospective study, formal ophthalmological assessment at the time of initial assessment was not always available for patients. Proptosis was apparent at presentation in five patients; in three patients, ipsilateral orbital/forehead headaches were noted. Treatment was tolerated well, with anterior pin site bruising for a variable time frame of a number of days for
Gender Radiotherapy details Type Dose (Gy) Fractions
CRT, conventional radiotherapy; SRS, stereotactic radiosurgery; FSRT, fractionated stereotactic radiotherapy; MMLC, mini multileaf collimator; IMRT, intensity-modulated radiotherapy. *Recurred outside treated volume, additional tumour treated by SRS. yBoth optic nerves treated.
FSRT/MMLC 50.40 28 FSRT 50.40 28 FSRT 50.40 28 FSRT 50.00 28 SRS 20.00 1 CRT 45.00 25 CRT 55.00 31
SRS 20.00 1
F F F M
F
CRT 56.00 31
SRS 20.00 1
FSRT 50.40 28
F F F F M
M
FSRT/MMLC 50.40 28
FSRT/MMLC 50.40 28
FSRT/MMLC/IMRT 50.68 28
M F M
F
14 13* 12 11 10 9 8 7 6 5 4 3 2 1 Patient
Table 1 e Radiotherapy treatment details
FSRT/MMLC/IMRT 50.05 28
CLINICAL ONCOLOGY
15y
10
patients having rigid head fixation for SRS, and intermittent mouth discomfort for patients having dental plate-based relocatable head ring SRT. Most patients experienced a variable degree of fatigue of short duration after treatment. There was no other reported late morbidity. Because the primary end point was tumour control, this was evaluated in the context of control or progression within the treated volume. An out-of-field failure does not represent a failure within the volume treated, rather inadequate assessment of the volume required for treatment. Hence, outcomes are reported in terms of treated tumours rather than number of optic nerves. Thus, 17 optic nerves had radiologically stable tumours (this representing the number of distinctly different tumours treated), with 15 optic nerves having stable to slightly improved outcome in terms of vision. Two case histories detail the different planning aspects of treating these types of tumour.
Case 1 A 7-year-old boy with known NF2 had bilateral ONSM as his presenting feature. Both nerves had been surgically decompressed to try to preserve vision. Both tumours were radiologically progressing in a 6 month period before referral. Thus, the target volume was both optic nerves and the connecting chiasm, a U-shaped lesion. The defined dose critical structures were the lens, lacrimal glands, retina, temporal and frontal lobes, pituitary and brain stem. The hypothalamus could not be spared. Stereotactic IMRT was used with doseevolume histograms indicating good coverage of the ‘U’ and separation by dose of the adjacent normal tissues (Fig. 1a,b). A mean dose of 5005 cGy in 28 fractions was delivered. Five years after treatment there has been no radiological progression with stable vision and the patient maintaining good academic performance at school.
Case 2 A 55-year-old women presented with a radiologically evident ONM confined to the intra-orbital optic nerve. The planning target volume was thus the radiologically abnormal nerve plus a margin covering the MRI normal component of the optic nerve within the optic canal plus 5 mm of intracranial optic nerve, for a proximal margin of 15 mm. Given that there was poor vision retained, SRT was used. A mean dose of 5040 cGy in 28 fractions was delivered. There was initial visual improvement; however, 12 months after SRT the patient presented with a 3 month history of ipsilateral blindness. MRI indicated a globular mass arising from the residual intracranial optic nerve, adjacent to the chiasm, that had not been treated. This lesion was then treated by SRS with a dose of 1200 cGy to the 90% isodose covering this lesion with no further radiological progression, with some visual improvement to be able to discern light and dark, and shapes. Six years later there has been no manifestation of further recurrence.
ONSM d NON-SURGICAL TREATMENT
11
Fig. 1 e (a) Three-dimensional display of tumour volume and adjacent dose critical structures. (b) Doseevolume histogram representing uniformity of tumour coverage with separation of volume structures between optic nerve sheath meningiomas and low dose normal tissues.
Discussion Visual loss is the main presenting feature for patients diagnosed with an ONSM [1,2,4,17]. However, it is important to differentiate between tumours that arise from the optic nerve sheath (primary) and tumours that arise adjacent to the optic nerve and by direct extension spread along the optic nerve and can cause visual impairment (secondary) [2]. Examples of the latter include meningiomas arising on the cavernous sinus, sphenoid ridge and olfactory groove. This distinction was documented as early as 1949 [6]. All ONSM have the usual female predominance in incidence and are typically unilateral [1,2,5]. The only exception is where the ONSM arises on the background of NF2, where it is usually bilateral. The extent of involvement of the optic nerve can be difficult to determine even with MRI [4]. Involvement of the contralateral side in the absence of NF2 is infrequent, although Dutton [1] reported that 38% of his patients had this extent of disease if the meningioma had spread to involve the nerve sheath within the optic canal. Given the failure pattern in one patient in this series, treating the whole length of the optic nerve would seem more appropriate. If there is intracranial extension, then treating part of the chiasm should also be considered.
The appropriate consideration then is what should be the recommended ‘treatment’ [5]. Although Kennerdell et al. [17] reported that 10% of surgically treated patients had visual improvement after treatment, this was in the context of a decompressive procedure; long-term followup was not reported for this 10% of patients nor indeed even a median follow-up. Basso et al. [2] reported that surgery is not usually successful in preserving vision, with Schick et al. [18] in agreement. Any attempts at appropriate excision will invariably lead to visual deterioration if not blindness [3]. In this circumstance, treatment is usually delayed under medical supervision until the patient is near blind, as in four patients in this series. As with meningiomas at other sites, radiotherapy has a high likelihood of tumour control and preservation of function [7,10]. Andrews et al. [7] recorded 30 patients with 33 optic nerves treated by SRT with a median followup of 89 weeks; tumour control was evident in 100%, with a consistent dose of 50.4 Gy being used. This high tumour control rate is not always seen. Turbin et al. [4] reported eight of 16 patients having surgery and radiotherapy who ultimately progressed where a range of doses was used. Basso et al. [2] noted three of 14 patients progressing with no details regarding the doses or fields used. In this series
12
CLINICAL ONCOLOGY
there was progression outside the treatment volume in only one of 18 ONSM treated. Preservation of function (vision) and, if possible, improvement are the other aspects of considering radiotherapy [11,14]. Turbin et al. [4], in a predominantly surgical paper, indicated that patients receiving radiotherapy alone had the best likelihood of a favourable visual outcome. Baumert et al. [19] reported 95% ‘visual control’ (21/22) with visual improvement in 16 patients. Andrews et al. [7] noted visual deterioration in only two patients; thus, 31 of 33 ONSM had stable to improved vision. Others report similar findings [9,15,16,18,19]. In this series, only one patient (with one optic nerve) had permanent visual deterioration, potentially a treatment-related event. Complications of treatment are also important to consider. Turbin et al. [4] reported that six of 18 patients (33%) had radiotherapy-related complications with retinopathy/vascular occlusion in four, iritis in one and temporal lobe atrophy in one. Richards et al. [13] also reported radiologically detected cerebral change in one of four patients. Andrews et al. [7] noted optic neuritis in one patient. Parsons et al. [20] documented the extent of visual injury after radiotherapy, reporting a 0% incidence of optic nerve injury in 106 optic nerves where total doses of less than 59 Gy were used with fractionation sizes of less than 1.9 Gy. For this population, one patient became blind in the ipsilateral eye after treatment with no radiological progression; thus probably a treatment-related event. The dose delivered to that optic nerve was no greater than that given to other patients. In terms of other effects, shortterm fatigue was frequent, and reversible, but not universal, with no other reported late morbidity. In terms of recommendations for treatment, it is easy to begin by stating that ONSM is not a surgically managed disease unless surgery is used to decompress where a mass effect is evident. Observing a patient with the aim of operating when the patient ultimately becomes blind is also inappropriate. Radiotherapy, in most modern series, is reported to have high tumour control figures with now prolonged follow-up. If visual outcome is not a consideration, then treatment can be delivered by SRS with a designated dose of 12e14 Gy to an appropriate isodose. However, where visual stabilisation/improvement is the goal, fractionated radiotherapy is delivered with a dose of 50.4e54 Gy with appropriate fractions of 1.8e1.9 Gy. Although higher doses have been used in some series reporting on the outcomes of treating mengiomas by radiotherapy, there is little evidence to indicate a better outcome in terms of local control than is reported here [21]. Current literature recommendations are for this dose range. The only exception where a higher dose may be required is for meningiomas displaying a more aggressive clinical or histological pattern [21]. The volume of nerve to be treated is critical. Patterns of failure are always the proximal not distal component of the nerve, with no surgical or radiotherapy series reporting failure distal to the globe. Therefore, significant components of the retina do not need to be included, reducing the risk of retinal deficit. None of the surgical series has recorded the
necessity to enucleate the globe because of intra-ocular extension, where the only structure resected is the optic nerve. Proximally the whole length of the nerve needs to be treated, and if there is intracranial extension then part of the chiasm also needs to be treated. In the context of NF2 with bilateral nerve involvement, the authors’ recommendation is to treat both nerves and the whole chiasm. Delivering this treatment stereotactically makes a complex treatment simple and enables the dose to adjacent critical structures to be minimised. ONSM are infrequent, and there is now surgical acceptance that treatment aimed at tumour control and visual stabilisation should be instituted sooner rather than later. SRT presents an ideal clinical circumstance, minimising significant dose to adjacent dose-sensitive structures [13,15,16,19,22]. Author for correspondence: R. Smee, Department of Radiation Oncology, The Prince of Wales Cancer Centre, High Street, Randwick 2031, NSW, Australia; E-mail: [email protected] Received 13 June 2008; received in revised form 14 October 2008; accepted 23 October 2008
References 1 Dutton JJ. Optic nerve sheath meningiomas. Surv Ophthalmol 1992;37:167e183. 2 Basso A, Carrizo A, Kreutel A, Tomecek F. Primary intraorbital meningiomas. In: Schmidek HH, editor. Meningiomas and their surgical management. Philadelphia: WB Saunders; 1991. p. 311e323. 3 Turbin RE, Pokorny K. Diagnosis and treatment of orbital optic nerve sheath meningioma. Cancer Control 2004;11(5):334e341. 4 Turbin RE, Thompson CR, Kennerdell JS, Cockerham KP, Kupersmith MJ. A long-term visual outcome comparison in patients with optic nerve sheath meningioma managed with observation, surgery, radiotherapy, or surgery and radiotherapy. Ophthalmology 2000;109:890e900. 5 Carrasco JR, Penne RB. Optic nerve sheath meningiomas and advanced treatment options. Curr Opin Ophthalmol 2004;15: 406e410. 6 Craig W, Gogela L. Intraorbital meningiomas: a clinicopathologic study. Am J Ophthalmol 1949;32:1663. 7 Andrews DW, Faroozan R, Yang BP, et al. Fractionated stereotactic radiotherapy for the treatment of optic nerve sheath meningiomas: preliminary observations of 33 optic nerves in 30 patients with historical comparison to observation with or without prior surgery. Neurosurgery 2002;51:890e904. 8 Debus J, Wuendrich M, Pirzkall A, et al. High efficacy of fractionated stereotactic radiotherapy of large base-of-skull meningiomas: long-term results. J Clin Oncol 2001;19:3547e3553. 9 Selch M, Ahn E, Laskari A, et al. Stereotactic radiotherapy for treatment of cavernous sinus meningiomas. Int J Radiat Oncol Biol Phys 2004;59:101e111. 10 Pirzkall A, Debus J, Haering P, et al. Intensity modulated radiotherapy (IMRT) for recurrent, residual, or untreated skullbase meningiomas: preliminary clinical experience. Int J Radiat Oncol Biol Phys 2003;55(2):362e372. 11 Lui JK, Forman S, Hershewe GL, et al. Optic nerve sheath meningiomas: visual improvement after stereotactic radiotherapy. Neurosurgery 2002;50:950e957.
ONSM d NON-SURGICAL TREATMENT 12 Narayan S, Cornblath WT, Sandler HM, et al. Preliminary visual outcomes after three-dimensional conformal radiation therapy for optic nerve sheath meningioma. Int J Radiat Oncol Biol Phys 2003;56:537e543. 13 Richards JC, Roden D, Harper CS. Management of sightthreatening optic nerve sheath meningioma with fractionated stereotactic radiotherapy. Clin Experiment Ophthalmol 2005; 33(2):137e141. 14 Melian E, Jay WM. Primary radiotherapy for optic nerve sheath meningioma. Semin Ophthalmol 2004;19(3e4):130e140. 15 Becker G, Jeremic B, Pitz S, et al. Stereotactic fractionated radiotherapy in patients with optic nerve sheath meningioma. Int J Radiat Oncol Biol Phys 2002;54(5):1422e1429. 16 Pitz S, Becker G, Schiefer U, et al. Stereotactic fractionated irradiation of optic nerve sheath meningioma: a new treatment alternative. Br J Ophthalmol 2002;86(11):1265e1268. 17 Kennerdell JS, Maroon JC, Malton M, et al. The management of optic nerve sheath meningiomas. Am J Ophthalmol 1988;106:450e457.
13
18 Schick U, Dott U, Hassler W. Surgical management of meningiomas involving the optic nerve sheath. J Neurosurg 2004;101: 951e959. 19 Baumert BG, Villa S, Studer G, et al. Early improvements in vision after fractionated stereotactic radiotherapy for primary optic nerve sheath meningioma. Radiother Oncol 2004;72(2): 169e174. 20 Parsons JT, Bova FJ, Fitzgerald CR, et al. Radiation optic neuropathy after megavoltage external-beam irradiation: analysis of time-dose factors. Int J Radiat Oncol Biol Phys 1994;30: 755e763. 21 Bulsara KR, Al-Mefty O, Shrieve DC, Angtuaco EJ. Meningiomas. In: Berger MS, Prados MD, editors. Textbook of neurooncology. Philadelphia: Elsevier Saunders; 2005. p. 335e345. 22 Moyer PD, Golnik KC, Breneman J. Treatment of optic nerve sheath meningioma with three-dimensional conformal radiation. Am J Ophthalmol 2000;129:694e696.