Neovascular Glaucoma After Stereotactic Radiotherapy for ...

2 downloads 0 Views 1MB Size Report
Purpose: Enucleation after stereotactic radiotherapy (SRT) for juxtapapillary choroidal melanoma may be re- quired because of tumor progression (TP) or the ...
Int. J. Radiation Oncology Biol. Phys., Vol. 80, No. 2, pp. 377–384, 2011 Copyright Ó 2011 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/$–see front matter

doi:10.1016/j.ijrobp.2010.04.073

CLINICAL INVESTIGATION

Eye

NEOVASCULAR GLAUCOMA AFTER STEREOTACTIC RADIOTHERAPY FOR JUXTAPAPILLARY CHOROIDAL MELANOMA: HISTOPATHOLOGIC AND DOSIMETRIC FINDINGS BRUNO F. FERNANDES, M.D., PH.D.,* DANIEL WEISBROD, M.D.,* YENI H. YU¨CEL, M.D., PH.D.,yz MATTHEW FOLLWELL, M.D.,x HATEM KREMA, M.D.,* MOSTAFA HEYDARIAN, M.D.,x WEI XU, PH.D.,k DAVID PAYNE, M.D.,x HUGH MCGOWAN, M.D.,* ERNEST R. SIMPSON, M.D.,* NORMAND LAPERRIERE, M.D.,x AND ARJUN SAHGAL, M.D.x Departments of *Ocular Oncology, xRadiation Oncology, and kBio-Statistics, Princess Margaret Hospital, University Health Network, Toronto, Ontario, Canada; and yOphthalmic Pathology Laboratory, Department of Ophthalmology & Visual Sciences, and zDepartment of Laboratory Medicine & Pathobiology, St. Michael’s Hospital, Toronto, Ontario, Canada Purpose: Enucleation after stereotactic radiotherapy (SRT) for juxtapapillary choroidal melanoma may be required because of tumor progression (TP) or the development of intractable radiation-induced neovascular glaucoma (NVG). We compare pathologic changes and dosimetric findings in those eyes enucleated secondary to NVG as opposed to TP to better understand potential mechanisms. Methods and Materials: Patients with juxtapapillary choroidal melanoma treated with SRT (70 Gy in 5 fractions, alternate days over a total of 10 days) at the Princess Margaret Hospital, Toronto, Ontario, Canada, who underwent enucleation between 1998 and 2006 were selected. We correlated dosimetric data based on the patient’s original SRT treatment plan with histopathologic findings in the retina, optic nerve head, and anterior chamber. A dedicated ocular pathologist reviewed each case in a blinded fashion. Results: Ten eyes in ten patients were enucleated after SRT. Six were enucleated secondary to NVG and four secondary to because of TP. Aggressive tumor features such as invasion of the sclera and epithelioid cell type were observed predominantly in the TP group. Retinal damage was more predominant in the NVG group, as were findings of radiation-related retinal vascular changes of fibrinoid necrosis and hyalinization. No conclusive radiationrelated effects were found in the anterior chamber. The maximum point dose and dose to 0.1 cc were lower for the anterior chamber as compared with the dose to the tumor, retina, and optic nerve head. The mean 0.1-cc doses to the retina were 69.4 Gy and 73.5 Gy and to the anterior chamber were 4.9 Gy and 17.3 Gy for the NVG group and tumor progression group, respectively. Conclusions: Our findings suggest that NVG is due to radiation damage to the posterior chamber of the eye rather than primary radiation damage to the anterior segment. Ó 2011 Elsevier Inc. Neovascular glaucoma, Stereotactic radiotherapy, Complications, Choroidal melanoma, Histopathology.

(6–8). The aim of radiation as opposed to enucleation is tumor control with organ preservation; however, radiationinduced complications can still result in enucleation. Radiation-induced neovascular glaucoma (NVG) diagnosed by neovascularization of the iris or angle and elevated intraocular pressure is the main radiation-induced complication requiring enucleation (9). On the basis of our reported experience with SRT for juxtapapillary melanoma (6–8), we present a detailed pathologic analysis of eyes enucleated for NVG or tumor progression and correlate these findings with dose–volume histogram (DVH) data based on relevant contoured anatomy features (optic nerve head, retina, and anterior

INTRODUCTION Juxtapapillary choroidal melanoma poses a challenge to the ocular oncologist, given greater recurrence rates, a higher likelihood of extraocular extension, an overall worse visual prognosis, and a greater risk of treatment-related complications (1, 2). With respect to treatment modalities, brachytherapy with standard plaque radiotherapy has been cautioned against because of the physical difficulty of positioning a plaque close to the optic nerve head where tilting can result in tumor underdosing (3). Alternatively, external beam radiotherapy has been used successfully with protons or charged particles (4), Gamma Knife technology (5), or linear accelerator–based stereotactic radiotherapy (SRT)

Reprint requests to: Arjun Sahgal, M.D., 610 University Ave., Toronto, Ontario, M5G2M9, Canada. Tel: (416) 946-2000; Fax: (416) 946-2227; E-mail: [email protected]

Conflict of interest: none. Received Feb 18, 2010, and in revised form March 18, 2010. Accepted for publication April 29, 2010. 377

I. J. Radiation Oncology d Biology d Physics

378

Volume 80, Number 2, 2011

Fig. 1. Eye with neovascular glaucoma due to stereotactic radiotherapy for juxtapapillary choroidal melanoma. (a) Neovascularization of iris surface at pupillary margin (arrows). (b) Early, (c) mid, and (d) late fluorescein angiography of anterior segment showing hyperfluorescence of iris neovascularization. As the angiogram progresses, blurring of the margins of the neovascular fronds occurred from leakage of dye from the new vessels.

chamber). Even though there have been reports on the histopathologic changes of enucleated eyes after different forms of radiotherapy for uveal melanoma (10–15), to our knowledge, this is the first study to detail pathologic changes due to SRT for juxtapapillary tumors.

ameter away from the treated tumor margin, with or without macular involvement. Radiation optic neuropathy was defined as optic disc pallor or swelling, without pre-existing peripapillary hemorrhages (7). Radiation maculopathy was defined as the presence of ischemic and/or exudative changes involving the macula secondary to irradiation of a tumour not encroaching on the anatomical macula. Clinical photos of these selected complications are provided in Fig. 2.

METHODS AND MATERIALS Patient selection Patients with juxtapapillary choroidal melanoma who underwent enucleation after treatment with SRT at the Ocular Oncology Service, Princess Margaret Hospital (Toronto, Ontario, Canada), between October 1998 and January 2006 were selected. Ethics approval was obtained from the institutional research ethics board. Details of the entire cohort of 64 consecutive patients who have received SRT at our institution including their demographics, SRT technique, and outcomes have been previously published (7). Enucleation was indicated in 10 patients. Six cases were because of NVG and four cases because of tumor progression. The clinical diagnosis of NVG was based on the presence of iris or angle neovascularization (Fig. 1), associated with an increase in intraocular pressure. Clinically, local tumor recurrence was defined as tumor enlargement of at least 30% in thickness or 500 mm of tumor base recorded within two successive post-treatment visits. Other radiation-related complications based on clinical evaluation were also recorded, including radiation cataract, tumor vasculopathy, radiation retinopathy, radiation maculopathy and optic neuropathy. Tumor vasculopathy was defined as the presence of retinal hemorrhage, exudate, or vascular occlusion within one disc diameter of the tumor margin. Radiation retinopathy was defined as the presence of at least two of these findings more than one disc di-

Radiation technique The details of our stereotactic radiation technique and delivery have been described in previous publications (6–8). The total prescribed radiation dose was 70 Gy given in 5 fractions, every alternate day, over 10 days for each patient.

Histopathology After histopathologic processing, two independent ocular pathologists evaluated each case and were masked to all clinical data and the reason for enucleation. Specific pathologic features related to the tumor, optic nerve, retina, and anterior segment of the eye were described. We also specifically examined each specimen for vessel hyalinization and fibrinoid necrosis. Although these are nonspecific features due to chronic vascular stress, they are histopathologic findings associated with radiation effect. The histopathologic findings were unmasked, and patients were divided into two groups depending on the reason for enucleation (tumor progression or NVG) and analyzed.

Dosimetric analysis and statistics The original treatment plans were retrospectively reviewed and dosimetric data obtained. Dose–volume histogram analysis was performed for the contoured anatomic features consisting of the tumor,

Stereotactic radiotherapy and neovascular glaucoma d B. F. FERNANDES et al.

379

Fig. 2. Color fundus photographs of choroidal melanoma before and after stereotactic radiotherapy. (a) Juxtapapillary choroidal melanocytic lesion on initial examination. (b) Increase in base diameter of same lesion 6 months after initial presentation. (c) Same lesion after stereotactic radiotherapy with intraretinal hemorrhage and cotton wool spots overlying tumor (tumor vasculopathy) and greater than 2 mm from tumor margin (radiation retinopathy). (d) Another juxtapapillary choroidal melanoma after stereotactic radiotherapy with swelling of optic disc and overlying cotton wool spots and intraretinal hemorrhage (radiation optic neuropathy).

globe, lens, retina surrounding the tumor volume within a distance of 3 mm, and anterior chamber. The Student t test and the Wilcoxon rank-sum test were used to compare dosimetric indices between the NVG and tumor progression groups, and p < 0.05 was considered statistically significant.

patient had died of metastatic disease, 43 months after diagnosis of the primary tumor. The clinical complications are listed in Table 1, along with their respective time of onset. Radiation-related complications including radiation retinopathy, tumor vasculopathy, optic neuropathy, maculopathy and cataract were seen more commonly in patients with NVG.

RESULTS Clinical data Six patients underwent enucleation because of NVG and four patients because of tumor progression. The median age, median tumor height, median tumor diameter, and median follow-up for the tumor progression group as opposed to the NVG group were 71.5 years (range, 56–83 years) and 57.3 years (range, 40–79 years), 6.2 mm (range, 3.9– 8.1 mm) and 5.4 mm (range, 2.7–7.6 mm), 12.3 mm (range, 2.7–7.6 mm) and 11.1 mm (range, 7.0–17.0 mm), and 11.3 months (range, 6–25 months) and 31.7 months (range, 12– 48 months), respectively. In the tumor progression group tumors were classified as medium (2 of 4) or large (2 of 4), and in the NVG group medium (5 of 6) or large (1 of 6). No patient had evidence of elevated intra-ocular pressure (>21 mm Hg) at baseline. At the time of our most recent follow-up, one

Histopathologic changes A comprehensive histopathologic evaluation of the tumor, retina, and anterior segment for all enucleated eyes is presented in Tables 2, 3, and 4, respectively. Figure 3 illustrates examples of histopathologic features of eyes with and without NVG. Compared with the NVG group, tumors from the progression group showed more aggressive features including invasion of the sclera and epithelioid cell type (Table 2). In contrast, there was more retinal pathology in eyes enucleated because of NVG (Table 3). In particular, fibrinoid necrosis and hyalinization of blood vessels were observed in the retina only in patients in the NVG group (Table 3). In the anterior segment, angle closure with neovascularization of the iris or angle was observed in all patients with NVG. However, hyalinization of blood vessels within the

380

I. J. Radiation Oncology d Biology d Physics

Table 1. Complications after stereotactic radiotherapy for juxtapapillary choroidal melanoma

Time to enucleation (mo) Median (range) Complications NVG Median time to onset (mo) Cataract Median onset (range) (mo) Tumor vasculopathy Median onset (range) (mo) Radiation retinopathy Median onset (range) (mo) Maculopathy Median onset (range) (mo) Optic neuropathy Median onset (range) (mo) Tumor progression Median onset (range) (mo)

Recurrence (n = 4)

NVG (n = 6)

15 (7–24)

22.8 (15–30)

0 N/A 0 N/A 2 7 (6–8) 1 8 (N/A) 1 8 (N/A) 1 8 (N/A) 4 15 (6–24)

6 19.7 (13–26) 3 16.7 (12–24) 6 12.8 (5–18) 5 14.2 (8–18) 3 12 (9–15) 2 20.5 (19–22) 0 N/A

Abbreviations: NVG = neovascular glaucoma; N/A = not applicable.

iris stroma was observed in only 1 patient from the NVG group (Patient 1). For this patient, the anterior chamber maximum point dose (Pmax) was well below the threshold for late radiation-related vascular effects at 2 Gy (Fig. 4, Table 5). In only 1 patient with tumor progression (Patient 3), upon pathologic examination, iris neovascularization was incidentally observed as was hyalinization of blood vessels within the iris stroma. In this patient NVG was not clinically diagnosed at the time of enucleation based on criteria. The DVH analysis of the anterior chamber of this eye showed a high Pmax of 26.3 Gy (Table 5). Histopathologically, at the time of enucleation, tumor had also extended into the anterior chamber and most of the interior surfaces of the eye with free-floating cells present in the vitreous cavity and anterior chamber. Figure 3D illustrates a layer of malignant cells extending over the iris and angle, as well as along the endothelial surface of the cornea, in this patient. DVH data Dosimetric data for the tumor, anterior chamber, retina, and optic nerve head are detailed in Table 5 and separated based on reason for enucleation. We recorded the Pmax and the dose to 0.1 cc (D0.1 cc) for each anatomic structure. In all cases tumor was abutting the optic nerve head; therefore the doses to the optic nerve head and tumor were similar for both Pmax and D0.1 cc. The Pmax for the retina surrounding the tumor was also similar to the tumor Pmax, because the Pmax represents a small volume. Even at the larger 0.1-cc volume, the retinal doses are beyond the threshold for late radiation-related retinal damage. For the anterior chamber, the Pmax in the tumor progression group was beyond the tolerance limit for late radiation toxicity for only 2 cases (Patients 3 and 10) whereas the D0.1 cc was beyond the tolerance threshold for only 1 case (Patient 10). In

Volume 80, Number 2, 2011

Table 2. Histology of eyes enucleated after stereotactic radiotherapy for juxtapapillary choroidal melanoma: Tumor features Reason for enucleation

Bruch membrane ruptured Invasion of retina Invasion of tumor vessels Optic nerve head invasion None Anterior to lamina cribrosa Beyond lamina cribrosa Scleral extension None or innermost layers Within sclera Extrascleral extension Cell type Spindle Mixed Epithelioid Total mitosis in 40 0 1 >2 Inflammation None to minimal Moderate Marked Degree of necrosis None 50% Prominent tumor vessels Extravasation of blood None >1 focal hemorrhage Diffuse

Recurrence (n = 4)

NVG (n = 6)

3 (75%) 2 (50%) 0

4 (67%) 2 (33%) 0

4 (100%) 0 0

5 (83%) 0 1 (17%)

2 (50%) 1 (25%) 1 (25%)

6 (100%) 0 0

0 2 (50%) 2 (50%)

2 (33%) 4 (67%) 0

3 (75%) 1 (25%) 0

5 (83%) 1 (17%) 0

3 (75%) 0 1 (25%)

4 (67%) 2 (33%) 0

3 (75%) 0 1 (25%) 0 3 (75%)

1 (17%) 2 (33%) 3 (50%) 0 5 (83%)

2 (50%) 1 (25%) 1 (25%)

5 (83%) 1 (17%) 0

Abbreviation: NVG = neovascular glaucoma.

the NVG group similar results were observed. The Pmax exceeded the tolerance threshold for late radiation toxicity in only 2 patients (Patients 4 and 5); however, the D0.1 cc was below the tolerance threshold for each of these 2 patients. In Patients 5 and 10 the exceptionally high dose to the anterior chamber was the result of greater tumor height. Neither of these patients were found to have radiation-induced vascular changes in the anterior segment on histopathologic examination. When comparing the mean doses for each anatomic structure in the NVG vs. tumor progression group, we did not observe any significant differences. However, one can appreciate that the mean Pmax and D0.1 cc for the anterior chamber are certainly much lower than the respective mean doses to the retina (Table 5). DISCUSSION Several centers have reported on histopathologic changes in enucleated eyes after radiation. The Collaborative Ocular Melanoma Study provided two reports. The first study

Stereotactic radiotherapy and neovascular glaucoma d B. F. FERNANDES et al.

Table 3. Histology of eyes enucleated after stereotactic radiotherapy for juxtapapillary choroidal melanoma: Retinal features Reason for enucleation

Retinal detachment Retinal atrophy Retinal gliosis Neovascularization of retina RPE metaplasia RPE hypertrophy or atrophy Fibrinoid necrosis of blood vessels Hyalinization of blood vessels Retinal exudation

Recurrence (n = 4)

NVG (n = 6)

3 (75%) 3 (75%) 3 (75%) 0 1 (25%) 4 (100%) 0

4 (67%) 6 (100%) 5 (83%) 1 (17%) 1 (17%) 6 (100%) 1 (17%)

0 2 (50%)

3 (50%) 4 (67%)

Abbreviations: NVG = neovascular glaucoma; RPE = retinal pigment epithelium.

investigated eyes treated with enucleation alone or preenucleation external beam radiation therapy (10). For large tumors, the mitotic activity was found to be significantly lower in eyes receiving pre-enucleation external beam radiation therapy compared with those that were not irradiated. The other study examined eyes that underwent plaque brachytherapy and confirmed the reduced mitotic activity after radiation (11). In addition, irradiated eyes had a smaller proportion of histologically intact tumor; more inflammation, fibrosis, and vascular damage within the tumor; a higher frequency of retinal invasion; greater damage to the retinal vasculature; and more cases with neovascularization of the iris and vitreous hemorrhage. Similar histopathologic fea-

Table 4. Histology of eyes enucleated after stereotactic radiotherapy for juxtapapillary choroidal melanoma: Anterior segment features Reason for enucleation

NVI/NVA Fibrinoid necrosis of blood vessels Hyalinization of blood vessels in iris Hyalinization of blood vessels in ciliary body PAS Anterior chamber angle closure Posterior synechiae Hemorrhage in AC Inflammatory cells in AC Tumor cells in AC

Recurrence (n = 4)

NVG (n = 6)

1 (25%) 0

6 (100%) 0

1 (25%)

1 (17%)

0

0

1 (25%) 1 (25%)

5 (83%) 6 (100%)

1 (25%) 0 0

4 (67%) 0 1 (17%)

1 (25%)

0

Abbreviations: NVG = neovascular glaucoma; NVI = neovascularization of iris; NVA = neovascularization of angle; PAS = peripheral anterior synechiae; AC = anterior chamber.

381

tures have been described with iodine 125 plaque brachytherapy (14), proton beam therapy (13), and charged-particle radiation (12). In our study the histopathology due to SRT was comparable to that described based on other methods of radiation delivery (10–15). We found a low tumor mitotic rate and a high frequency of retinal detachment, atrophy, and gliosis. However, unlike other studies, we separated eyes enucleated for NVG as opposed to tumor progression in an attempt to determine whether any histopathologic differences could be observed. Although our small sample size does not permit definitive conclusions, our histopathologic analysis showed more retinal pathology in eyes enucleated for NVG (Table 3), radiation-related retinal vascular changes exclusively in these same eyes, and a lack of conclusive anterior segment changes attributable to radiation (Table 4). Radiation to the anterior segment has been thought to be a risk factor in the development of NVG, but this has never been proven histologically (16, 17). In our series only 1 of 6 eyes enucleated for NVG (Patient 1) had evidence of iris vessel hyalinization. The dose to the anterior chamber in this patient was 2 Gy, well below the threshold for radiation-induced late vascular effects. One eye from the tumor progression group (Patient 3) had evidence of iris neovascularization histopathologically (but not clinically), and in this patient hyalinization of iris stromal vessels was also observed. Although the dose to the anterior segment was high (26.3 Gy), we cannot directly attribute the cause of neovascularization and vessel hyalinization to radiation. Extensive tumor infiltration in the anterior segment was also observed, which could have resulted in the observed neovascularization and vessel hyalinization. Considering that eyes enucleated for NVG showed greater evidence of post-SRT retinal pathology and radiation-related vascular changes seen only in the retina in these patients, we support the hypothesis that NVG results secondarily from radiationinduced effects in the posterior segment as opposed to direct radiation damage to the anterior segment. We postulate that the release of proangiogenic factors from direct radiation damage to tumor endothelial cells, as well as secondary ischemic changes from injury to the surrounding normal retinal microvasculature, is fundamental to the pathogenesis of radiation-induced NVG (18). Endothelial cells have been shown to be particularly sensitive to high dose– per–fraction radiation (>8 Gy per fraction as opposed to low conventional doses of 1.8–3.0 Gy per fraction) resulting in endothelial apoptosis and microvascular dysfunction (19, 20). Angiogenic factors like vascular endothelial growth factor and fibroblast growth factor (FGF2) have been shown to be upregulated after high-dose radiation exposure in normal brain tissue (21), and repeated administration of high-dose radiation serves to enhance vascular destruction and endothelial apoptosis and compromise blood flow (22). The hypofractionated regimen used by our center (14 Gy per fraction) poises the tissue in the high-dose volume (tumor, surrounding retina, and optic nerve) to develop tumor vasculopathy, radiation retinopathy, and optic neuropathy,

382

I. J. Radiation Oncology d Biology d Physics

Volume 80, Number 2, 2011

Fig. 3. (a) Eye enucleated because of tumor progression. Normal angle architecture is seen with no evidence of neovascularization of the iris and/or angle (hematoxylin–eosin [H&E] stain, original magnification 100). (b) Eye with neovascular glaucoma showing peripheral anterior synechiae with complete angle closure (H&E stain, original magnification 100). (c) High magnification of same case detailing thin-walled newly formed vessels over surface of iris (H&E stain, original magnification 200). (d) High magnification of the only eye from the tumor progression group that showed neovascularization during histopathologic evaluation. Melanoma cells are seen covering the surface of the iris, the angle, and the endothelial side of the cornea (H&E stain, original magnification 200).

and high rates of these complications have been reported in our series of 64 patients treated with SRT (7). The proangiogenic factors then diffuse through the vitreous to reach the anterior segment, promoting the formation of neovascularization on the iris and in the angle (23). This newly formed fibrovascular tissue prevents aqueous humor outflow and leads to elevated intraocular pressure and NVG. This is the accepted

Fig. 4. Dose–volume histogram for Patient 1. The dose to the posterior segment structures including the tumor, surrounding retina, and optic disc is high compared with the low dose to the anterior segment structures including the lens and anterior (Ant) chamber.

pathophysiologic mechanism for NVG due to other retinal vascular diseases such as proliferative diabetic retinopathy and central retinal vein occlusion (24). Furthermore, treatment with intravitreal injections of anti–vascular endothelial growth factor agents, such as bevacizumab, has been used with variable success in the management of radiation retinopathy, providing added supportive evidence of the proposed mechanism (25). Our study is unique in that we have DVH data correlated to a histopathologic analysis in eyes enucleated for NVG and tumor progression. In the literature the most comprehensive study of NVG in the computed tomography planning era was reported after carbon ion therapy by Hirasawa et al. (16). They reported an analysis of NVG in their doseescalation study for unfavorable tumors (tumors with a large size or adjacent to the optic disc). Tumor doses ranged from 60 to 85 gray-equivalent units in 5 fractions. Although no histopathologic assessment was performed, two theories regarding potential mechanisms for radiation-induced NVG were generated. One mechanism postulated that direct radiation damage to the iris–ciliary body complex can lead to NVG, whereas a second mechanism proposed that a high dose to the central retinal vessels can lead to ischemic changes that ultimately result in neovascularization. Our analysis is more consistent with the latter hypothesis, that posterior segment radiation damage leads to neovascularization.

Stereotactic radiotherapy and neovascular glaucoma d B. F. FERNANDES et al.

383

Table 5. Dosimetric data for the primary tumor and specified contoured organs at risk Primary tumor

Anterior chamber

Optic nerve head

Retina

Dose to Pmax Dose to 0.1 cc Dose to Pmax Dose to 0.1 cc Dose to Pmax Dose to 0.1 cc Dose to Pmax Dose to 0.1 cc (cGy) (cGy) (cGy) (cGy) (cGy) (cGy) (cGy) (cGy) NVG Patient 1 Patient 2 Patient 4 Patient 5 Patient 6 Patient 7 Mean Tumor progression Patient 3 Patient 8 Patient 9 Patient 10 Mean Wilcoxon rank-sum test t Test

7774.5 7386.7 7549.8 7466.2 7423.4 7400.0 7500.1

7579.2 7349.6 7511.8 7428.7 7348.8 7325.6 7424.0

200.2 340.7 3842.9 5848.8 331.0 228.2 1798.6

93.4* 199.6* 1807.1 552.4 168.8* 127.6 491.5

7334.8 7250.7 7019.4 7179.0 6979.1 7307.9 7178.5

6597.6 6963.3 5837.9 7106.9 6486.3 7135.5 6687.9

7684.0 7169.0 7562.4 7387.8 7396.8 7399.4 7433.2

5715.4 6916.9 7333.4 7239.3 7210.9 7250.7 6944.6

7974.2 7490.2 7441.6 7469.9 7594.0 0.29

7894.1 7377.3 7404.3 7451.0 7531.7 0.39

2626.8 1925.7 466.2 7149.7 3042.1 0.20

1211.1 285.7* 148.3* 5272.0 1729.3 0.29

6338.6 7238.3 7046.3 6576.9 6800.0 0.14

5506.8 6779.6 6528.3 6437.4 6313.0 0.20

7918.2 7410.7 7395.9 7446.3 7542.8 0.39

7639.7 7178.6 7321.5 7259.2 7349.8 0.20

0.48

0.35

0.48

0.26

0.07

0.30

0.44

0.25

Abbreviations: Pmax = maximum point dose; NVG = neovascular glaucoma. * Dose to 0.05 cc of volume reported because of limited anatomic volume.

A strength of our study is the homogeneous patient population such that all tumors were in the same location and all eyes were treated uniformly to the same dose of radiation. Furthermore, a detailed histopathologic analysis was performed in a masked fashion under the supervision of a dedicated ocular pathologist, and dosimetric data were collected from the original treatment plan with anatomy contoured by an ocular radiation oncologist. A limitation of our report is the small sample size, which is inherent to the subject being studied. Another limitation lay in the difficulty in determining whether changes in the retina adjacent to the tumor, as well as in the anterior chamber, were due to radiation alone or tumor growth as well. Retinal detachment, atrophy, gliosis, neovascularization, and retinal pigment epithelial metaplasia can all be a consequence of tumor growth and are seen in eyes with large tumors that have not received radia-

tion before enucleation (10). Furthermore, although vessel hyalinization is a late radiation vascular effect, it is a nonspecific feature that can occur as a result of chronic vascular stress from inflammation or disruption of blood flow due to tumor cell infiltration. Lastly, although angiogenic factor release is implicated in the pathogenesis of NVG induced by benign ocular conditions, further research is required to validate our hypothesis to confirm its role in radiation-induced NVG.

CONCLUSION Our findings suggest that radiation-induced damage to the posterior segment may have an important role in the development of radiation-induced NVG.

REFERENCES 1. Boldt HC, Melia BM, Liu JC, et al. I-125 brachytherapy for choroidal melanoma photographic and angiographic abnormalities: The Collaborative Ocular Melanoma Study: COMS Report No. 30. Ophthalmology 2009;116:106–115.e1. 2. Sagoo MS, Shields CL, Mashayekhi A, et al. Plaque radiotherapy for juxtapapillary choroidal melanoma overhanging the optic disc in 141 consecutive patients. Arch Ophthalmol 2008;126: 1515–1522. 3. Williams DF, Mieler WF, Jaffe GJ, et al. Magnetic resonance imaging of juxtapapillary plaques in cadaver eyes. Br J Ophthalmol 1990;74:43–46. 4. Seddon JM, Gragoudas ES, Egan KM, et al. Uveal melanomas near the optic disc or fovea. Visual results after proton beam irradiation. Ophthalmology 1987;94:354–361. 5. Schirmer CM, Chan M, Mignano J, et al. Dose de-escalation with Gamma Knife radiosurgery in the treatment of choroidal melanoma. Int J Radiat Oncol Biol Phys 2009;75:170–176.

6. Emara K, Weisbrod DJ, Sahgal A, et al. Stereotactic radiotherapy in the treatment of juxtapapillary choroidal melanoma: Preliminary results. Int J Radiat Oncol Biol Phys 2004;59:94–100. 7. Krema H, Somani S, Sahgal A, et al. Stereotactic radiotherapy for treatment of juxtapapillary choroidal melanoma: 3-year follow-up. Br J Ophthalmol 2009;93:1172–1176. 8. Somani S, Sahgal A, Krema H, et al. Stereotactic radiotherapy in the treatment of juxtapapillary choroidal melanoma: 2-year follow-up. Can J Ophthalmol 2009;44:61–65. 9. Weinhaus RS, Seddon JM, Albert DM, et al. Prognostic factor study of survival after enucleation for juxtapapillary melanomas. Arch Ophthalmol 1985;103:1673–1677. 10. Histopathologic characteristics of uveal melanomas in eyes enucleated from the Collaborative Ocular Melanoma Study. COMS report no. 6. Am J Ophthalmol 1998;125:745–766. 11. Avery RB, Diener-West M, Reynolds SM, et al. Histopathologic characteristics of choroidal melanoma in eyes enucleated

384

12. 13. 14. 15. 16.

17. 18.

I. J. Radiation Oncology d Biology d Physics

after iodine 125 brachytherapy in the collaborative ocular melanoma study. Arch Ophthalmol 2008;126:207–212. Crawford JB, Char DH. Histopathology of uveal melanomas treated with charged particle radiation. Ophthalmology 1987; 94:639–643. Gragoudas ES, Egan KM, Saornil MA, et al. The time course of irradiation changes in proton beam-treated uveal melanomas. Ophthalmology 1993;100:1555–1559. discussion 1560. Saornil MA, Fisher MR, Campbell RJ, et al. Histopathologic study of eyes after iodine I 125 episcleral plaque irradiation for uveal melanoma. Arch Ophthalmol 1997;115:1395–1400. Shields CL, Shields JA, Karlsson U, et al. Enucleation after plaque radiotherapy for posterior uveal melanoma. Histopathologic findings. Ophthalmology 1990;97:1665–1670. Hirasawa N, Tsuji H, Ishikawa H, et al. Risk factors for neovascular glaucoma after carbon ion radiotherapy of choroidal melanoma using dose-volume histogram analysis. Int J Radiat Oncol Biol Phys 2007;67:538–543. Daftari IK, Char DH, Verhey LJ, et al. Anterior segment sparing to reduce charged particle radiotherapy complications in uveal melanoma. Int J Radiat Oncol Biol Phys 1997;39:997–1010. Fajardo LF. The pathology of ionizing radiation as defined by morphologic patterns. Acta Oncol 2005;44:13–22.

Volume 80, Number 2, 2011

19. Garcia-Barros M, Paris F, Cordon-Cardo C, et al. Tumor response to radiotherapy regulated by endothelial cell apoptosis. Science 2003;300:1155–1159. 20. Moeller BJ, Cao Y, Li CY, et al. Radiation activates HIF-1 to regulate vascular radiosensitivity in tumors: Role of reoxygenation, free radicals, and stress granules. Cancer Cell 2004;5:429–441. 21. Kim JH, Chung YG, Kim CY, et al. Upregulation of VEGF and FGF2 in normal rat brain after experimental intraoperative radiation therapy. J Korean Med Sci 2004;19:879–886. 22. Kim DW, Huamani J, Niermann KJ, et al. Noninvasive assessment of tumor vasculature response to radiation-mediated, vasculature-targeted therapy using quantified power Doppler sonography: Implications for improvement of therapy schedules. J Ultrasound Med 2006;25:1507–1517. 23. Irvine AR, Wood IS. Radiation retinopathy as an experimental model for ischemic proliferative retinopathy and rubeosis iridis. Am J Ophthalmol 1987;103:790–797. 24. Aiello LP, Avery RL, Arrigg PG, et al. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med 1994;331:1480–1487. 25. Finger PT. Radiation retinopathy is treatable with anti-vascular endothelial growth factor bevacizumab (Avastin). Int J Radiat Oncol Biol Phys 2008;70:974–977.