Introduction

The diagnosis and treatment of intraorbital tumours can be challenging for clinicians. Especially when lesions are located in the orbital apex, where the space is tight and even relatively small lesions may compromise the optic nerve function [1,2,3,4,5,6,7,8,9]. In addition to a carefully taken clinical history, imaging technologies such as computerized tomography (CT), magnetic resonance imaging (MRI) and modern biopsy techniques are used to diagnose an intraorbital tumour [10, 11]. Based on imaging characteristics, location, size and histological features of the tumour, a multidisciplinary team of specialists will evaluate what kind of treatment is preferred. In case of a presumed benign lesion, treatment will usually consist of excisional or debulking surgery or radiotherapy [1,2,3,4,5,6,7,8,9]. However, potential complications of each treatment need to be considered. This is more relevant in benign conditions, where the sequelae of treatment might outweigh the benefits of intervention [1,2,3,4,5,6,7,8,9, 12, 13].

Several studies have demonstrated that surgical excision can be effective for orbital tumours. However, there is an increased risk of complications in apical tumours, such as orbital haemorrhage with subsequent vision loss [1,2,3,4,5,6, 13]. Only few studies have examined the effects of orbital fSRT on long-term visual outcomes. Promising results have been described, but no long-term follow-up studies are available [7,8,9, 14]. Moreover, little is known about the results of fSRT for these indications on globe position (enophthalmos, exophthalmos, hypo-/hyperglobus) and eyelid position (retraction), which are of functional and aesthetic importance.

Therefore, the aim of this study is to assess the long-term outcome in ophthalmic parameters, including the visual acuity, visual field, exophthalmometry value, globe position, deviations of RNFL on OCT and radiation retinopathy in patients who underwent fSRT for (presumed) benign intraorbital tumours.

Methods

This retrospective consecutive series consists of 25 patients with (presumed) benign orbital tumours who visited the Rotterdam Eye Hospital in the Netherlands. Patient with distal location of the tumour, defects in visual field and progression of the tumour were treated with fSRT. Patients were treated at the Radiotherapy Department, Erasmus University Medical Center between 2002 and 2018. The patients were treated with fSRT, a total dose of 50.4 to 54 Gy in 28 to 30 fractions of 1.8 Gy was delivered to the tumour in accordance with ICRU guidelines. The medical records were evaluated for age at presentation, gender, best corrected visual acuity (BCVA) and ophthalmic examination (upper eyelid retraction, proptosis, deviations of RNFL on OCT and radiation retinopathy). Radiation retinopathy was defined as presence of blood, exudation or liquid within the retinal layers. Visual fields were assessed using the Humphrey Field Analyzer (HFA) (II-i Series; Carl Zeiss Meditec, Inc., Dublin, CA), the retinal nerve fibre layer (RNFL) and the presence of liquid on retinal layers were measured using optical coherence tomography (OCT, Canon HS-100). Mean deviations (MD) in decibel (dB) were extracted for the analyses and used as the parameter for visual field defects. Proptosis measurements were performed utilizing calibrated prism-type Hertel exophthalmometers (manufactured by Rodenstock and by Marco). Information on size, type and location of the intraorbital tumour was obtained from imaging technologies such as CT and/or MRI. During the follow-up consultations, patients were asked whether they were content with their functional and cosmetic results after treatment (cosmetic satisfaction). Patients diagnosed with radiation retinopathy received anti-VEGF intravitreal injections to treat macular cysts. Patient younger than 18 years and patients with orbital malignancies were excluded from the study.

Statistical analysis

Characteristics of the study population are described. Mean, median, percentages and standard deviations (before and after treatment) of the visual acuity, visual fields, Hertel exophthalmometry values, presence of eyelid retraction and deviations of RNFL on OCT before and after treatment were assessed using SPSS software (version 25.0).

Results

Table 1 presents the characteristics of the study population. The mean age of the patients included in the study was 54 years of age (range 40 to 63 years) and the mean follow-up was 104 months (range 48–215 months). Among presumed benign intraorbital tumours included in this study, 80% were optic nerve sheath meningioma and 20% cavernous haemangioma. The mean BCVA at presentation was 0.56 decimal (0.38 SD). The BCVA increased after treatment to 0.63 decimal (0.37 SD) (p = 0.001). Before treatment, the MD of the visual field was −14.98 dB (12.9 SD). After treatment, a significant lower MD of −4.51 dB (10.8 SD) was observed (p = 0.006). A positive trend was observed at each follow-up visit for the BCVA and visual fields (Fig. 1A and Fig. 1B).

Table 1 Characteristics of the study population before and after treatment.
Full size table
Fig. 1: The mean best corrected visual acuity and the mean deviation in visual fields at each follow-up visit.
figure 1

A The mean best corrected visual acuity in decimal before treatment and during the follow-up. B The mean deviation in decibel obtained from the Humphrey Field Analyzer before treatment and during the follow-up.

Full size image

After treatment, five of 18 patients with abnormal RNFL thickness showed improvement (p = 0.002). A significant reduction of Hertel exophthalmometry values was observed after treatment (mean 14.92 mm (7.9 SD) versus 13.79 mm (7.3 SD) before and after treatment, respectively). The percentage of patients with upper eyelid retraction at presentation was 16%. Half of these patients did not show eyelid retraction at the follow-up period after treatment.

No differences in visual acuity, changes in RNFL or eyelid retraction were observed when comparing patients with presumed optic nerve sheath meningioma and cavernous haemangioma before and after treatment (Table 2). Visual field testing showed a lower MD in patient with an optic nerve sheath meningioma as compared to patients with a cavernous haemangioma (p = 0.007). These differences were higher after treatment (p = 0.000). Significant but small differences were observed in Hertel exophthalmometry values between patients with presumed optic nerve sheath meningioma and cavernous haemangioma. Patient with presumed cavernous haemangioma had a significant reduction of exophthalmos after fSRT (Table 2).

Table 2 Differences in BCVA, visual field and ophthalmic examination between patients with presumed optic nerve sheath meningioma and cavernous haemangioma.
Full size table

Of patients with optic nerve sheath meningioma, 15% developed radiation retinopathy, compared to none of the patients with cavernous haemangioma (Table 3). These three patients were all females. Radiation retinopathy was diagnosed after 14.3 months (median) following treatment. One of the patients showed stable retinal changes without need for treatment. This patient received a low dose of radiation at the globes 24.4 and 22.3 Gy as compared to the mean dose of radiation among the patients included in the study (2.3 Gy vs 15.6 Gy, respectively). These patients received a higher dose of radiation in corresponding eyes as compared to the rest of the patients included in the study (24.4 Gy and 22.3 Gy vs 15.6 Gy). One of these patients did not respond to additional treatment and the BCVA and visual field deteriorated.

Table 3 Characteristics of patients that developed radiation retinopathy.
Full size table

Discussion

This study showed significant improvement in visual acuity, visual field, exophthalmos and eyelid retraction among patients with presumed optic nerve sheath meningioma and presumed cavernous haemangioma, treated with stereotactic radiotherapy. Only three of the 25 patients developed radiation retinopathy with significant deterioration of visual acuity and visual field defects only in one patient. All three patients had optic nerve sheath meningiomas.

In our study was observed a significant improvement of visual acuity. However, the changes in BCVA were small. Most of the patients included in this study had a relatively good visual acuity at the moment of presentation. Visual acuity is a subjective measurement which deteriorates at a late stage of optic nerve compression. Our patients were followed each year and treated immediately if progression on the visual field was observed. This might explain the small changes observed in this study. In contrary of the visual acuity, visual field showed a remarkable improvement with an MD of −14.98 dB and −4.51 dB before and after fSRT. The mechanism explaining these improvements are not clear yet. However, different studies have hypothesised that the compression of the optic nerve can lead to loss of ganglion cells and consequently visual field loss and deterioration of visual acuity [15, 16]. In less profound compressions, a recovery of ganglion cells and subsequently improvement of visual field might occur after decompression methods such as radiation therapy or debulking surgery [15]. In line with our results, other smaller studies have also shown that radiation therapy improves the visual field at patients with optic nerve sheath meningioma and presumed orbital cavernous haemangiomas [9, 17,18,19].

Three patients who developed radiation retinopathy after orbital fSRT had been diagnosed with this condition. We do not have a clear explanation why retinopathy developed only among patients with presumed optic nerve sheath meningioma. In our series, the anterior part of this tumour was situated closer to the eye than the apical cavernous haemangiomas. Therefore, the retina of these patients might have been exposed to higher radiation doses, which may cause radiation necrosis and damage. Previous studies have shown that the total received dose of radiotherapy was also higher among patients with presumed optic nerve sheath meningioma compared to patients with presumed cavernous haemangioma [20, 21]. The reported risk of radiation retinopathy is low, but the risk is increased when the dose exceeds 45 to 55 Gy, as in these patients [20, 21]. However, studies have shown that fSRT uses multiple beams of radiation to accurately focus and shape treatment to the lesion, thereby limiting radiation exposure of adjacent normal tissue [9, 22]. In our study, the dose of radiation received to the globes of the patients who developed retinopathy was very small (2.3 Gy, 24.4 Gy and 22.3 Gy). We did not observe a significantly higher dose of radiation between the globes of patients who did and did not developed retinopathy. From these results, we cannot conclude a direct relation between the radiation dose to the globe and the development of the retinopathy. Therefore, more studies are needed to examine radiation retinopathy among patients treated with fSRT. We cannot rule out that more patients in this series may develop complications as the follow-up period varied between 24 and 215 months, but as we observed that complications mostly occurred within the first 24 months, this is less probable. One study among 26 patients with presumed optic meningioma and treated with fSRT has also reported one patient with radiation retinopathy. However they didn’t report the time of occurrence [23]. Moreover, the development of secondary malignancy or malignant transformation of a benign tumour in patients treated with fSRT was shown to be low and similar to the risk of general population to have a primary tumour [24]. Slightly significant differences on eyelid position were observed before and after treatment. A significant decrease of proptosis was observed after treatment. However, the difference in the measurements were small. We acknowledge that in our study, we had a small group of the patients with presumed cavernous haemangioma. This might have led to a slight deviation of the results on the comparison of clinical ophthalmologic outcomes between the two groups. Further, all included patients had a presumed diagnosis based on clinical and radiographic characteristics. Histopathology was not performed due to localisation and high risk of potential complications from a biopsy [13]. This might have led to a misdiagnosis of the tumour and therefore incorrect management.

However, to our best knowledge this is the first study focussing on visual outcomes following sFRT for presumed optic nerve sheath meningioma and orbital cavernous haemangiomas with a median follow up period of more than 104 months.

Conclusion

Optic nerve sheath meningiomas and orbital cavernous haemangiomas are benign in nature but may compromise the visual function due to their growth. Treatment is necessary if the vision is threatened. This long-term follow-up study shows that fSRT can be a non-invasive treatment modality with few complications and good long-term visual and cosmetic outcomes. Radiation-induced retinopathy was more frequent in tumours that extended towards the globe.

Summary

What was known before

  • Optic nerve sheath meningiomas and orbital cavernous haemangiomas are benign in nature but may compromise the visual function due to their growth.

What this study adds

  • fSRT is an effective treatment modality for this subset of orbital tumours with few complications and good long-term visual and cosmetic outcomes. FSRT is non-invasive and safer than surgery for lesions in the posterior orbit or around the optic nerve.