- Email: [email protected]
Visual Outcomes after Proton Beam Irradiation for Choroidal Melanomas Involving the Fovea
Visual Outcomes after Proton Beam Irradiation for Choroidal Melanomas Involving the Fovea Avni V. Patel, MD, MBA,1 Anne Marie Lane, MPH,1 Margaux A. Morrison, MPH,2 Alexei V. Trofimov, PhD,3 Helen A. Shih, MD,3 Evangelos S. Gragoudas, MD,1 Ivana K. Kim, MD1 Purpose: To report visual outcomes in patients undergoing proton beam irradiation of tumors located within 1 disc diameter of the fovea. Design: Retrospective review. Participants: Patients with choroidal melanoma involving the fovea treated with proton beam therapy between 1975 and 2009. Methods: Three hundred fifty-one patients with choroidal melanomas located 1 disc diameter (DD) or less from the fovea and more than 1 DD away from the optic nerve were included in this study. In a subgroup of 203 of the patients with small and medium choroidal melanomas, the effect of a reduced dose of radiation, 50 Gy (relative biological effectiveness [RBE]) versus 70 Gy (RBE), on visual outcomes was analyzed. The Kaplan-Meier method and Cox regression analysis were performed to calculate cumulative rates of vision loss and to assess risk factors for vision loss, respectively. Main Outcome Measures: Visual acuity and radiation complications, which included radiation maculopathy, papillopathy, retinal detachment, and rubeosis, were assessed. Results: Three hundred fifty-one patients were included in this study with a mean follow-up time of 68.7 months. More than one-third of patients (35.5%) retained 20/200 or better vision 5 years after proton beam irradiation. For those patients with a baseline visual acuity of 20/40 or better, 16.2% of patients retained this level of vision 5 years after proton beam irradiation. Tumor height less than 5 mm and baseline visual acuity 20/40 or better were associated significantly with a better visual outcome (P < 0.001). More than two-thirds (70.4%) of patients receiving 50 Gy (RBE) and nearly half (45.1%) of patients receiving 70 Gy (RBE) retained 20/200 or better vision 5 years after treatment, but this difference was not significant. Approximately 20% of patients with these smaller macular tumors retained 20/40 vision or better 5 years after irradiation. Conclusions: The results of this retrospective analysis demonstrate that despite receiving a full dose of radiation to the fovea, many patients with choroidal melanoma with foveal involvement maintain useful vision. A radiation dose reduction from 70 to 50 Gy (RBE) did not seem to increase the proportion of patients who retain usable vision. Ophthalmology 2015;-:1e9 ª 2015 by the American Academy of Ophthalmology. Supplemental material is available at www.aaojournal.org.
Choroidal melanoma is the most common primary intraocular malignancy among adult patients and is one of the few potentially fatal ocular diagnoses. Treatment options for choroidal melanomas include enucleation, laser photocoagulation, transpupillary thermotherapy, and irradiation. Over the past few decades, radiotherapy has become used more widely in the care of these tumors.1e4 Achieving local tumor control, while preserving the eye and maintaining some useful vision, is the primary goal of radiotherapy in the management of uveal melanoma. Two major radiotherapy methods currently are used: external beam radiation therapy, most commonly with protons, and radioactive plaques, which are placed over the sclera in the area of the tumor.2,4e6 As charged particles, protons can provide highly 2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
localized radiation dose distributions, and by doing so may reduce ocular morbidity.3 Radiation maculopathy and papillopathy are common sequelae of radiation treatment and are the predominant causes of visual loss in patients after radiotherapy. Radiation maculopathy is the main contributor to vision loss in patients who have tumors with foveal involvement.7 The clinical manifestations of radiation maculopathy include capillary closure, telangiectasia, microaneurysm formation, hemorrhages, exudates, macular edema, and nerve fiber layer infarctions.8 Previous studies have shown that the risk of radiation maculopathy and papillopathy are determined primarily by the radiation dose to the macula and optic nerve, respectively.9,10 Additionally, underlying http://dx.doi.org/10.1016/j.ophtha.2015.09.031 ISSN 0161-6420/15
1
Ophthalmology Volume -, Number -, Month 2015 vascular disorders may enhance the risk of radiation maculopathy.10 In patients who received plaque radiation, a significant dose-response relationship was shown between the dose of radiation to the fovea and the development of radiation maculopathy.11 It is commonly assumed that patients with choroidal melanoma involving the fovea are destined to sustain severe vision loss after irradiation, but actual visual outcomes in patients with choroidal melanoma involving the fovea are not well described. The primary goal of this study was to review visual outcomes in patients undergoing irradiation with proton beam therapy for tumors located within 1 disc diameter (DD) of the fovea. In addition, we examined whether a dose reduction from 70 to 50 Gy (relative biological effectiveness [RBE]) had any effect on visual outcomes in patients with this tumor location. In a previous randomized study, we demonstrated that although a dose reduction from 70 to 50 Gy (RBE) did not provide a significant visual benefit, there was a trend toward a decreased incidence of radiation papillopathy.12 However, in the prior study, visual outcomes were evaluated in a subset of patients who had tumors located within 4 DD of the fovea or optic nerve. In the present study, to determine better the effects of radiation to the fovea on visual function, we restricted our analyses to patients with tumors in very close proximity to the fovea only. We believe these data are essential to guiding treatment decisions in these patients.
Methods A waiver of informed consent and Health Insurance Portability and Accountability Act authorization was granted by the institutional review board of the Massachusetts Eye and Ear Infirmary for this medical records review. Patients with choroidal melanoma involving the fovea (n ¼ 351) treated with proton beam therapy at the Harvard Cyclotron Laboratory, Cambridge, Massachusetts, or the Francis H. Burr Proton Therapy Center at Massachusetts General Hospital, Boston, Massachusetts, between 1975 and 2009 were included in this study. Most patients (n ¼ 326) were treated between 1975 and 1996. These patients were included in a previous analysis evaluating outcomes in more than 2000 patients treated with proton therapy for unilateral choroidal or ciliary body melanoma, or both.13 Additionally, a small number of patients (n ¼ 34) participated in a randomized clinical trial evaluating dose reduction.12 Twenty-five patients included in the analysis were treated between 2005 and 2009, when a change in the dosing regimen for small to medium tumors near the fovea was implemented. The patients were identified through the Melanoma Registry of the Ocular Oncology Service at the Massachusetts Eye and Ear Infirmary, Boston, Massachusetts. Patients with choroidal melanomas located 1 DD or less from the fovea and more than 1 DD away from the optic nerve were included in this study. Patients with these tumor characteristics were selected for evaluation because they were more likely to have received at least 90% of the prescribed dose of radiation to the fovea and significantly less of the dose to the optic nerve. These criteria identified tumors at highest risk for radiation maculopathy while minimizing the risk of papillopathy. Thus, they were more likely to be at risk of vision loss resulting from radiation to the fovea than other vision-compromising conditions. Patients were evaluated annually at a minimum at the Massachusetts Eye and Ear Infirmary or by their local ophthalmologist. The ophthalmologic examination at each visit included Snellen
2
Table 1. Patient Characteristics Mean age at treatment (range), yrs Female gender, no. (%) Male gender, no. (%) Diabetes, no. (%) Hypertension, no. (%) Mean largest tumor diameter (range), mm Mean largest tumor height (range), mm Tumors within 1 DD of the fovea, no. (%) Tumors involving the fovea, no. (%) Tumors >2 DD away from the nerve, no. (%) Mean follow-up (range), mos
58.09 157 194 29 72 12.8 4.9 343 90 174 68.7
(14.62e91.47) (44.7) (55.3) (8.3) (29.4) (5e23) (0.6e14.3) (97.4) (25.6) (49.6) (0.8e264.7)
DD ¼ disc diameter.
visual acuity testing, slit-lamp biomicroscopy, indirect ophthalmoscopy, and fundus photography. All patients were followed up for ocular outcomes, including visual acuity and radiation complications, which included radiation maculopathy, papillopathy, retinal detachment, and rubeosis. The treatment planning protocol for proton therapy of uveal melanomas has been described previously.3 A 3-dimensional treatment planning computer program is used with selection of visual fixation direction to minimize irradiation to the lens, fovea, and optic disc. Patient immobilization is achieved via a bite block and individually contoured plastic mask mounted into a frame on the head holder. Patients are set up radiographically and monitored during treatment with a video camera to ensure that they are fixating on a predetermined point. Determination of the radiation doses to the fovea for each patient is derived from the 3dimensional treatment plan. In addition to the analysis performed in this series of 351 patients with choroidal melanomas of any size located in close proximity to the fovea, a subgroup analysis of 203 patients with small and medium choroidal melanomas also was performed. In this subgroup analysis, 156 patients received a total dose to the tumor of 70 Gy (RBE) and 47 patients received a total dose of 50 Gy (RBE) delivered in 5 equal fractions over 5 to 10 days.
Statistical Analysis Cumulative rates of vision loss were calculated according to the Kaplan-Meier method. These rates of vision loss also were assessed by tumor height and by baseline visual acuity. Cox regression analysis was performed to assess risk factors for vision loss among these patients. The risk factors analyzed included tumor height, tumor diameter, distance of tumor from the optic nerve, age at treatment, history of diabetes, and baseline visual acuity. We also evaluated the effect of ocular complications from radiation Table 2. Visual Acuity at Final Follow-up Examination Visual Acuity 20/40 or better 20/50e20/100 20/125e20/800 CFeNLP
Baseline 178 122 51 0
(50.7) (34.8) (14.5) (0.0)
CF ¼ counting fingers; NLP ¼ no light perception. Data are no. of patients (%).
Last Follow-up 66 57 90 138
(18.8) (16.2) (25.6) (39.3)
Patel et al
Visual Outcomes of Macular Melanomas
Figure 1. Fundus photographs from 2 patients with choroidal melanoma involving the fovea after proton beam irradiation. Both patients received the prescribed dose to the fovea. A, Fundus photograph of a patient 6 years after treatment. This patient maintained visual acuity of 20/150. This fundus photograph shows evidence of radiation maculopathy with an area of exudates. B, Fundus photograph from a second patient 5 years after proton beam irradiation. This patient maintained excellent visual acuity of 20/25.
including maculopathy, papillopathy, retinal detachment, and rubeosis on risk of vision loss.
Results Patient Characteristics The mean age of patients in this study was 58 years. One hundred ninety-four patients (55.3%) were men and 157 (44.7%) were women. Of the 351 patients, 8.3% had diabetes and 29.4% had hypertension. The median largest tumor diameter was 12 mm and the median largest tumor height was 4 mm. By ophthalmic examination, all patients included in this study had tumors located within 1 DD of the fovea; however, 9 patients included in our final analysis did not receive the full prescribed dose to the fovea during planned treatment. This occurred because during the treatment planning, the tumor measured slightly more than 1 DD from the fovea. As a result, 342 patients (97.7%) in the final
analysis received the full prescribed dose to the fovea and 90 patients (25.6%) had tumors with direct foveal involvement. All of the tumors were more than 1 DD from the optic nerve, and 174 tumors (50%) were more than 2 DD away from the optic nerve. The mean follow-up time was 68.7 months and the median was 53.7 months (Table 1). Of the 351 patients in this study, 178 patients (50.7%) had visual acuity of 20/40 or better at baseline, 122 patients (34.8%) had visual acuity of 20/50 to 20/ 100 at baseline, 51 patients (14.5%) had visual acuity of 20/ 125 to 20/800 at baseline, and no patients had visual acuity of counting fingers or worse.
Tumor Control At the time of final follow-up, metastasis developed in 70 patients (19.9%) who died of the disease. Approximately 6% of the cohort (n ¼ 22) underwent enucleation. Of these, the tumor had recurred (suspected or documented growth) in 6 cases.
Figure 2. Cumulative rates of visual acuity retention of 20/200 or better (solid line) and 20/40 or better (dashed line) after proton beam irradiation.
3
Ophthalmology Volume -, Number -, Month 2015
Figure 3. Cumulative rates of visual acuity retention of 20/200 or better after proton beam irradiation by tumor height: 5 mm or less (solid line) and more than 5 mm (dashed line).
Visual Outcomes At last follow-up, 66 patients (18.8%) had visual acuity of 20/40 or better, 57 patients (16.2%) had visual acuity of 20/50 to 20/100, 90 patients (25.6%) had visual acuity of 20/125 to 20/800, and 138 patients (39.3%) had visual acuity of counting fingers or worse (Table 2; Fig 1).
Cumulative Rates of Vision Retention after Irradiation Analysis of cumulative rates of vision retention in patients with baseline vision of 20/200 or better revealed that 1 year after proton beam irradiation, 77.7% of patients retained 20/200 or better vision. Three years after proton beam irradiation, 53.5% of the patients retained 20/200 or better vision. More than one third of patients (35.5%) retained 20/200 or better vision 5 years after
proton beam irradiation. For those patients with a baseline visual acuity of 20/40 or better, 61.7% and 16.2% of patients retained this level of vision 1 year and 5 years after proton beam irradiation, respectively (Fig 2).
Risk Factors Associated with Vision Loss after Irradiation Tumor height was associated with vision outcome. The proportion of patients retaining vision of 20/200 or better 5 years after irradiation was 47.5% among patients with tumor height of 5 mm or less, compared with a proportion of 13.8% among patients with tumor height of more than 5 mm (P < 0.001; Fig 3). The proportion of patients retaining 20/40 or better vision 5 years after proton beam irradiation was 19.7% among those with tumor height of 5 mm or less compared with 7.0% of patients with tumor height of more than 5 mm (P < 0.001; Fig 4).
Figure 4. Cumulative rates of visual acuity retention of 20/40 or better after proton beam irradiation by tumor height: 5 mm or less (solid line) and more than 5 mm (dashed line).
4
Patel et al
Visual Outcomes of Macular Melanomas
Figure 5. Cumulative rates of visual acuity retention of 20/200 or better after proton beam irradiation by baseline visual acuity: 20/40 or better (solid line) and worse than 20/40 (dashed line).
Baseline visual acuity also influenced visual outcome. At 5 years after proton beam irradiation, 55.0% of those patients with baseline visual acuity of 20/40 or better retained 20/200 or better vision, whereas 18.0% of those with worse than 20/40 vision at baseline retained 20/200 or better vision (P < 0.001; Fig 5). Fewer patients with tumors 1 to 2 DD from the nerve had a final visual acuity of 20/40 or better than patients with tumors located more than 2 DD from the nerve (16.4% vs. 21.3%, respectively; P ¼ 0.15). Poor visual acuity (worse than 20/200) at final followup examination was observed in 58.2% of patients with tumors 1 to 2 DD away from the disc compared with 51.7% of patients with tumors more than 2 DD away from the disc (P ¼ 0.13). These differences were not statistically significant. There was a significant difference in the occurrence of papillopathy and rubeosis between patients who retained vision of 20/ 200 or better and those who did not. Papillopathy was present in 3.2% of patients with 20/200 vision or better at final follow-up and 10.4% of patients with vision worse than 20/200 (P ¼ 0.01). Rubeosis was present in 0.6% of patients with 20/200 vision or better at final follow-up and 20.7% of patients with vision worse than 20/200 (P < 0.0001). There was no significant difference in the occurrence of maculopathy or retinal detachment between the 2 groups (Table 3). Univariate Cox regression analysis showed that tumor height more than 5 mm, largest tumor diameter more than 15 mm, and baseline visual acuity worse than 20/40 were risk factors for vision
Table 3. Incidence of Ocular Complications by Visual Acuity at Final Follow-up Complication Maculopathy Papillopathy Retinal detachment Rubeosis
Visual Acuity ‡20/200 74 5 3 1
(46.8) (3.2) (1.6) (0.6)
Visual Acuity <20/200 98 20 8 40
(50.8) (10.4) (4.1) (20.7)
Data are no. of patients (%) unless otherwise indicated. *Chi-square test.
P Value* 0.52 0.01 0.036 <0.0001
loss, whereas tumor location more than 2 DD from the optic nerve, age at treatment, and history of diabetes were not significant. Ocular complications were associated significantly with increased risk of vision loss with the exception of maculopathy. In a multivariate model, tumor height more than 5 mm, largest tumor diameter more than 15 mm, baseline visual acuity worse than 20/ 40, presence of papillopathy, and presence of rubeosis all were associated significantly with risk of vision loss to worse than 20/ 200 (Table 4). Risk factors associated significantly with the development of papillopathy and rubeosis in this cohort were tumor diameter more than 15 mm (papillopathy and rubeosis) and tumor height more than 5 mm (rubeosis; Tables 5 and 6, available at www.aaojournal.org).
Subgroup Analysis: Dose Reduction In a subgroup of 203 patients with small and medium choroidal melanomas 5 mm or less in height and 15 mm or less in diameter, the effect of a reduced dose of radiation on visual outcomes was analyzed. The tumor location for this subgroup was the same as that of the overall group, with all tumors being 1 DD or less from the fovea and more than 1 DD from the optic nerve. Forty-seven patients received a dose of 50 Gy (RBE) and 156 patients received 70 Gy (RBE) in this subgroup. There was no statistically significant difference in the age, gender, largest tumor diameter, tumor height, distance from the optic nerve, or distance from the fovea between the 2 groups. All patients had a minimum of 3 years of follow-up, with a mean follow-up time of 59.5 months for the 50 Gy (RBE) group and 79.6 months for the 70 Gy (RBE) group (P ¼ 0.12). In the 50 Gy (RBE) group, 72.3% of patients had baseline visual acuity of 20/40 or better, and 55.1% of patients in the 70 Gy (RBE) group had baseline visual acuity of 20/40 or better (P ¼ 0.03). The cumulative rates of vision retention of 20/200 or better 1 year after proton beam irradiation were 95.7% for the 50 Gy (RBE) group and 89.0% for 70 Gy (RBE) group. At 5 years after proton beam irradiation, the rates of vision retention of 20/200 or better were 70.5% and 44.1% for the 50 and 70 Gy (RBE) groups, respectively, which was not significantly different (P ¼ 0.06; Fig 6). The cumulative rates of vision retention of 20/40 or better for the 2 groups 5 years after proton beam irradiation were 16.8%
5
Ophthalmology Volume -, Number -, Month 2015 Table 4. Cox Regression Analysis of Risk Factors for Vision Loss to Worse than 20/200* Univariate Models
Full Model
Risk Factor
Referent
Tumor height >5 mm BLVA worse than 20/40 Largest tumor diameter >15 mm Papillopathy present Rubeosis present Maculopathy present Age at treatment History of diabetes Location >2 DD from nerve
5 mm 20/40 or better 15 mm
2.861 (2.171e3.769) 2.964 (2.225e3.949) 3.063 (2.268e4.1354)
8.31Ee14 1.16Ee13 <0.001
1.667 (1.173e2.370) 2.168(1.592e2.953) 1.689 (1.176e2.427)
0.004 <0.001 0.005
1.613 (1.157e2.249) 2.295 (1.699e3.100) 1.691 (1.178e2.426)
0.005 <0.001 0.004
Absent Absent Absent Continuous No history 1e2 DD from nerve
1.869 4.724 0.666 1.002 1.031 0.836
0.005 <0.001 0.004 0.6174 0.7733 0.1974
1.663 2.552 0.873 1.002 1.043 0.860
0.034 <0.001 0.370 0.569 0.693 0.301
1.636 (1.040e2.571) 2.563 (1.715e3.829)
0.033 <0.001
(1.211e2.885) (3.294e6.774) (0.506e876) (0.993e1.012) (0.836e1.273) (0.637e1.098)
P Value
Hazard Ratio (95% Confidence Interval)
Best Fit Model
Hazard Ratio (95% Confidence Interval)
(1.039e2.662) (1.707e3.816) (0.650e1.173) (0.993e1.012) (0.845e1.288) (0.646e1.144)
P Value
Hazard Ratio (95% Confidence Interval)
P Value
BLVA ¼ baseline visual acuity; DD ¼ disc diameter. *Includes only patients with baseline vision better than 20/200.
and 23.9% for 50 and 70 Gy (RBE), respectively, but this difference was not statistically significant (P ¼ 0.43; Fig 7). There was no statistically significant difference in visual acuity at last follow-up between the 2 groups. At the last follow-up, nearly one-third (31.9%) of the 50 Gy (RBE) group, and nearly one-quarter (24.3%) of the 70 Gy (RBE) group retained 20/40 or better vision. The proportion of patients with extremely poor vision (counting fingers to no light perception) was 12.8% of the 50 Gy (RBE) group and 27.6% of the 70 Gy (RBE) group (Table 7). Thirty of 47 patients (63.8%) who received 50 Gy (RBE) and 90 of 156 patients (57.6%) who received 70 Gy (RBE) demonstrated maculopathy by the last examination. Of those patients with visual acuity worse than 20/200 at final follow-up, there was a significant difference in the proportion of patients with evidence of radiation maculopathy: 26.7% of the 50 Gy (RBE) group and 55.6% of the 70 Gy (RBE) group (P ¼ 0.007).
Discussion The management of choroidal melanoma involving the fovea poses one of the most challenging scenarios for practitioners. This study presents a large series of patients who received at least 90% of the prescribed radiation dose to the fovea during the treatment of their choroidal melanoma with proton beam radiotherapy. Despite the high dose of radiation to the fovea, well above the previously observed thresholds for producing radiation maculopathy,10 the visual outcomes in this retrospective analysis suggest that a significant proportion of these patients maintain some degree of useful vision. For patients with tumors of any size within 1 DD of the fovea, more than one-third of patients (35.5%) retained
Figure 6. Cumulative rates of vision retention of 20/200 or better after proton beam irradiation by dose: 50 Gy (relative biological effectiveness [RBE]) (solid line) or 70 Gy (RBE) (dashed line).
6
Patel et al
Visual Outcomes of Macular Melanomas
Figure 7. Cumulative rates of vision retention of 20/40 or better after proton beam irradiation by dose: 50 Gy (relative biological effectiveness [RBE]) (solid line) or 70 Gy (RBE) (dashed line).
vision of better than 20/200 at 5 years after proton beam irradiation. Additionally, more than half (55.0%) of patients with a baseline visual acuity of 20/40 vision or better retained vision better than 20/200 at 5 years after proton beam irradiation. At 5 years after irradiation, 16.2% of these patients retained visual acuity of 20/40 or better. Visual acuity after proton beam irradiation in this cohort was influenced most significantly by baseline visual acuity worse than 20/40 and the development of rubeosis. Each of these factors more than doubled the relative risk of vision loss to worse than 20/200. Additionally, rubeosis was related significantly to tumor height and diameter. For those patients with small and medium melanomas (tumor height 5 mm and LTD 15mm) within 1 DD of the fovea treated with proton beam irradiation, a significant proportion retained good visual acuity 5 years after treatment. More than two-thirds (70.5%) of patients receiving 50 Gy (RBE) and nearly half (44.1%) of patients receiving 70 Gy (RBE) retained 20/200 or better vision 5 years after treatment. Approximately 20% of patients with these smaller macular tumors retained 20/40 vision or better 5 years after irradiation. A decrease in prescribed radiation dose from 70 to 50 Gy (RBE) did not seem to increase the Table 7. Visual Acuity at Final Follow-up by Radiation Dose Visual Acuity 20/40 or better 20/50e20/100 20/125e20/800 CFeNLP
50 Gy (RBE) 15 11 15 6
70 Gy (RBE)
(31.9) (23.4) (31.9) (12.8)
38 26 49 43
(24.3) (16.7) (31.4) (27.6)
P ¼ 0.15 CF ¼ counting fingers; NLP ¼ no light perception; RBE ¼ relative biological effectiveness. Data are no. (%) unless otherwise indicated.
proportion of patients who retained usable vision. However, this degree of prescribed dose reduction may decrease the incidence of more severe vision loss (worse than 20/200). Few prior studies discuss visual outcomes in patients with choroidal tumors involving the fovea. Of the studies that reported such visual outcomes, all used plaque radiation as the method of radiotherapy. Gündüz et al14 reported the outcomes of plaque radiotherapy in the treatment of choroidal melanomas involving the macula. In their study of 630 patients with macular melanoma with median follow-up of 63 months, they found that visual acuity was more than 20/200 in 44% of patients and better than 20/40 in 9% of patients. The median dose to the fovea was 12 207 cGy, with a range of 1263 to 31 900 cGy. That series included tumors slightly farther away from the fovea (up to 3 mm) compared with our analysis, and only 2% of the patients had a portion of the tumor located beneath the foveola, versus 26% in our cohort. Median largest tumor diameter also was smaller at 10 mm. Although a larger range of doses delivered to the fovea and differences in inclusion criteria make comparisons difficult, the overall visual outcomes seem similar. Finally, the Collaborative Ocular Melanoma Study report15 described visual acuity outcomes over 3 years for patients treated with iodine-125 brachytherapy. Of the 623 patients included in the study, 13.2% had tumors within the foveal avascular zone (FAZ) and 24.9% had tumors within 0.1 to 2.0 mm within the FAZ. Of these, 73.9% of tumors involving the FAZ and 54.7% of patients with tumors within 2.0 mm of the FAZ had worse than 20/200 vision at 3 years. In comparison, 46.5% of patients in our series had vision worse than 20/200 at 3 years. Newman et al16 similarly reported 50 patients with subfoveal melanoma who were treated with plaque brachytherapy with palladium-103. Subfoveal melanoma was defined more strictly in this study as the posterior margin or tumor apex located beneath the fovea, and patients in this study received a
7
Ophthalmology Volume -, Number -, Month 2015 mean apical radiation dose of 82.8 Gy with a mean dose of 157 Gy to the fovea. The mean follow-up time after radiation was 62.2 months, which is comparable with that of our study. They reported that 50% (25/50) of patients had vision better than 20/ 200 at final follow-up and 36% (18/50) of patients had 20/40 visual acuity or better. The limitations of that study include the adjuvant treatments with antievascular endothelial growth factor, pars plana vitrectomy, and laser therapy, which could play a role in achieving the favorable visual outcomes in that study, as suggested by the authors. In contrast, no patients in our study received any adjuvant treatments. One significant limitation of our retrospective analysis is that despite decreasing the risk of coincident maculopathy and papillopathy by choosing tumors more than 1 DD away from the optic nerve, there is still a definite overlap in the incidence of maculopathy and papillopathy occurring in these patients. We were unable to distinguish whether the cause of vision loss in individual patients was the result of radiation effects on the macula or the optic nerve. However, our selection criteria helped to ensure that radiation dose to the fovea was larger than that to the optic nerve. There is a widely held impression that eyes with choroidal melanoma with foveal involvement are unlikely to retain any useful vision after radiotherapy. However, our series demonstrated that severe vision loss was not inevitable for these patients. We showed that patients receiving proton beam irradiation for choroidal melanoma within 1 DD of the fovea have a reasonable chance of maintaining useful vision and a small chance of retaining good vision. These visual outcomes suggest that proton irradiation at a dose of up to 70 Gy (RBE) in 5 fractions to the macula is permissive for survival of neuronal and vascular elements of the retina and choroid. Further elucidation of factors that distinguish the small minority of patients able to retain excellent vision is necessary for future improvement of visual outcomes.
References 1. Shields JA, Shields CL. Intraocular Tumors: An Atlas and Textbook. Lippincott Williams & Wilkins; 2008. 2. Shields JA, Augsburger JJ, Brady LW, Day JL. Cobalt plaque therapy of posterior uveal melanomas. Ophthalmology 1982;89:1201–7.
3. Gragoudas ES, Seddon JM, Egan K, et al. Long-term results of proton beam irradiated uveal melanomas. Ophthalmology 1987;94:349–53. 4. Packer S, Rotman M. Radiotherapy of choroidal melanoma with iodine-125. Ophthalmology 1980;87:582–90. 5. Gragoudas ES, Goitein M, Verhey L, et al. Proton beam irradiation of uveal melanomas. Results of 5 1/2-year study. Arch Ophthalmol 1982;100:928–34. 6. Lommatzsch PK. beta-Irradiation of choroidal melanoma with 106Ru/106Rh applicators. 16 Years’ experience. Arch Ophthalmol 1983;101:713–7. 7. Guyer DR, Mukai S, Egan KM, et al. Radiation maculopathy after proton beam irradiation for choroidal melanoma. Ophthalmology 1992;99:1278–85. 8. Seddon JM, Gragoudas ES, Egan KM, et al. Uveal melanomas near the optic disc or fovea. Visual results after proton beam irradiation. Ophthalmology 1987;94:354–61. 9. Kim IK, Lane AM, Egan KM, et al. Natural history of radiation papillopathy after proton beam irradiation of parapapillary melanoma. Ophthalmology 2010;117:1617–22. 10. Gragoudas ES, Li W, Lane AM, et al. Risk factors for radiation maculopathy and papillopathy after intraocular irradiation. Ophthalmology 1999;106:1571–7. discussion 1577e1578. 11. Finger PT, Chin KJ, Yu G-P. Risk factors for radiation maculopathy after ophthalmic plaque radiation for choroidal melanoma. Am J Ophthalmol 2010;149:608–15. 12. Gragoudas ES, Lane AM, Regan S, et al. A randomized controlled trial of varying radiation doses in the treatment of choroidal melanoma. Arch Ophthalmol 2000;118: 773–8. 13. Gragoudas E, Li W, Goitein M, et al. Evidence-based estimates of outcome in patients irradiated for intraocular melanoma. Arch Ophthalmol 2002;120:1665–71. 14. Gündüz K, Shields CL, Shields JA, et al. Radiation complications and tumor control after plaque radiotherapy of choroidal melanoma with macular involvement. Am J Ophthalmol 1999;127:579–89. 15. Melia BM, Abramson DH, Albert DM, et al. Collaborative Ocular Melanoma Study (COMS) randomized trial of I-125 brachytherapy for medium choroidal melanoma. I. Visual acuity after 3 years. COMS report no. 16. Ophthalmology 2001;108:348–66. 16. Newman H, Chin KJ, Finger PT. Subfoveal choroidal melanoma: pretreatment characteristics and response to plaque radiation therapy. Arch Ophthalmol 2011;129: 892–8.
Footnotes and Financial Disclosures Originally received: June 26, 2015. Final revision: September 14, 2015. Accepted: September 24, 2015. Available online: ---.
Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2015-1079.
1
Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts. 2
Department of Ophthalmology and Visual Science, University of Utah, Salt Lake City, Utah.
3
Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts.
8
Supported by the Grimshaw-Gudewicz Charitable Foundation (Fall River, MA). Author Contributions: Conception and design: Lane, Gragoudas, Kim Analysis and interpretation: Patel, Lane, Morrison, Gragoudas, Kim Data collection: Lane, Trofimov, Shih, Gragoudas, Kim Obtained funding: Gragoudas
Patel et al
Visual Outcomes of Macular Melanomas
Overall responsibility: Patel, Lane, Morrison, Trofimov, Shih, Gragoudas, Kim Abbreviations and Acronyms: BLVA ¼ baseline visual acuity; CF ¼ counting fingers; DD ¼ disc diameter; FAZ ¼ foveal avascular zone; Gy ¼ gray; NLP ¼ no light perception; RBE ¼ relative biological effectiveness.
Correspondence: Ivana K. Kim, MD, Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA 02114. E-mail: ivana_kim@ meei.harvard.edu.
9
Choroidal melanoma is the most common primary intraocular malignancy among adult patients and is one of the few potentially fatal ocular diagnoses. Treatment options for choroidal melanomas include enucleation, laser photocoagulation, transpupillary thermotherapy, and irradiation. Over the past few decades, radiotherapy has become used more widely in the care of these tumors.1e4 Achieving local tumor control, while preserving the eye and maintaining some useful vision, is the primary goal of radiotherapy in the management of uveal melanoma. Two major radiotherapy methods currently are used: external beam radiation therapy, most commonly with protons, and radioactive plaques, which are placed over the sclera in the area of the tumor.2,4e6 As charged particles, protons can provide highly 2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
localized radiation dose distributions, and by doing so may reduce ocular morbidity.3 Radiation maculopathy and papillopathy are common sequelae of radiation treatment and are the predominant causes of visual loss in patients after radiotherapy. Radiation maculopathy is the main contributor to vision loss in patients who have tumors with foveal involvement.7 The clinical manifestations of radiation maculopathy include capillary closure, telangiectasia, microaneurysm formation, hemorrhages, exudates, macular edema, and nerve fiber layer infarctions.8 Previous studies have shown that the risk of radiation maculopathy and papillopathy are determined primarily by the radiation dose to the macula and optic nerve, respectively.9,10 Additionally, underlying http://dx.doi.org/10.1016/j.ophtha.2015.09.031 ISSN 0161-6420/15
1
Ophthalmology Volume -, Number -, Month 2015 vascular disorders may enhance the risk of radiation maculopathy.10 In patients who received plaque radiation, a significant dose-response relationship was shown between the dose of radiation to the fovea and the development of radiation maculopathy.11 It is commonly assumed that patients with choroidal melanoma involving the fovea are destined to sustain severe vision loss after irradiation, but actual visual outcomes in patients with choroidal melanoma involving the fovea are not well described. The primary goal of this study was to review visual outcomes in patients undergoing irradiation with proton beam therapy for tumors located within 1 disc diameter (DD) of the fovea. In addition, we examined whether a dose reduction from 70 to 50 Gy (relative biological effectiveness [RBE]) had any effect on visual outcomes in patients with this tumor location. In a previous randomized study, we demonstrated that although a dose reduction from 70 to 50 Gy (RBE) did not provide a significant visual benefit, there was a trend toward a decreased incidence of radiation papillopathy.12 However, in the prior study, visual outcomes were evaluated in a subset of patients who had tumors located within 4 DD of the fovea or optic nerve. In the present study, to determine better the effects of radiation to the fovea on visual function, we restricted our analyses to patients with tumors in very close proximity to the fovea only. We believe these data are essential to guiding treatment decisions in these patients.
Methods A waiver of informed consent and Health Insurance Portability and Accountability Act authorization was granted by the institutional review board of the Massachusetts Eye and Ear Infirmary for this medical records review. Patients with choroidal melanoma involving the fovea (n ¼ 351) treated with proton beam therapy at the Harvard Cyclotron Laboratory, Cambridge, Massachusetts, or the Francis H. Burr Proton Therapy Center at Massachusetts General Hospital, Boston, Massachusetts, between 1975 and 2009 were included in this study. Most patients (n ¼ 326) were treated between 1975 and 1996. These patients were included in a previous analysis evaluating outcomes in more than 2000 patients treated with proton therapy for unilateral choroidal or ciliary body melanoma, or both.13 Additionally, a small number of patients (n ¼ 34) participated in a randomized clinical trial evaluating dose reduction.12 Twenty-five patients included in the analysis were treated between 2005 and 2009, when a change in the dosing regimen for small to medium tumors near the fovea was implemented. The patients were identified through the Melanoma Registry of the Ocular Oncology Service at the Massachusetts Eye and Ear Infirmary, Boston, Massachusetts. Patients with choroidal melanomas located 1 DD or less from the fovea and more than 1 DD away from the optic nerve were included in this study. Patients with these tumor characteristics were selected for evaluation because they were more likely to have received at least 90% of the prescribed dose of radiation to the fovea and significantly less of the dose to the optic nerve. These criteria identified tumors at highest risk for radiation maculopathy while minimizing the risk of papillopathy. Thus, they were more likely to be at risk of vision loss resulting from radiation to the fovea than other vision-compromising conditions. Patients were evaluated annually at a minimum at the Massachusetts Eye and Ear Infirmary or by their local ophthalmologist. The ophthalmologic examination at each visit included Snellen
2
Table 1. Patient Characteristics Mean age at treatment (range), yrs Female gender, no. (%) Male gender, no. (%) Diabetes, no. (%) Hypertension, no. (%) Mean largest tumor diameter (range), mm Mean largest tumor height (range), mm Tumors within 1 DD of the fovea, no. (%) Tumors involving the fovea, no. (%) Tumors >2 DD away from the nerve, no. (%) Mean follow-up (range), mos
58.09 157 194 29 72 12.8 4.9 343 90 174 68.7
(14.62e91.47) (44.7) (55.3) (8.3) (29.4) (5e23) (0.6e14.3) (97.4) (25.6) (49.6) (0.8e264.7)
DD ¼ disc diameter.
visual acuity testing, slit-lamp biomicroscopy, indirect ophthalmoscopy, and fundus photography. All patients were followed up for ocular outcomes, including visual acuity and radiation complications, which included radiation maculopathy, papillopathy, retinal detachment, and rubeosis. The treatment planning protocol for proton therapy of uveal melanomas has been described previously.3 A 3-dimensional treatment planning computer program is used with selection of visual fixation direction to minimize irradiation to the lens, fovea, and optic disc. Patient immobilization is achieved via a bite block and individually contoured plastic mask mounted into a frame on the head holder. Patients are set up radiographically and monitored during treatment with a video camera to ensure that they are fixating on a predetermined point. Determination of the radiation doses to the fovea for each patient is derived from the 3dimensional treatment plan. In addition to the analysis performed in this series of 351 patients with choroidal melanomas of any size located in close proximity to the fovea, a subgroup analysis of 203 patients with small and medium choroidal melanomas also was performed. In this subgroup analysis, 156 patients received a total dose to the tumor of 70 Gy (RBE) and 47 patients received a total dose of 50 Gy (RBE) delivered in 5 equal fractions over 5 to 10 days.
Statistical Analysis Cumulative rates of vision loss were calculated according to the Kaplan-Meier method. These rates of vision loss also were assessed by tumor height and by baseline visual acuity. Cox regression analysis was performed to assess risk factors for vision loss among these patients. The risk factors analyzed included tumor height, tumor diameter, distance of tumor from the optic nerve, age at treatment, history of diabetes, and baseline visual acuity. We also evaluated the effect of ocular complications from radiation Table 2. Visual Acuity at Final Follow-up Examination Visual Acuity 20/40 or better 20/50e20/100 20/125e20/800 CFeNLP
Baseline 178 122 51 0
(50.7) (34.8) (14.5) (0.0)
CF ¼ counting fingers; NLP ¼ no light perception. Data are no. of patients (%).
Last Follow-up 66 57 90 138
(18.8) (16.2) (25.6) (39.3)
Patel et al
Visual Outcomes of Macular Melanomas
Figure 1. Fundus photographs from 2 patients with choroidal melanoma involving the fovea after proton beam irradiation. Both patients received the prescribed dose to the fovea. A, Fundus photograph of a patient 6 years after treatment. This patient maintained visual acuity of 20/150. This fundus photograph shows evidence of radiation maculopathy with an area of exudates. B, Fundus photograph from a second patient 5 years after proton beam irradiation. This patient maintained excellent visual acuity of 20/25.
including maculopathy, papillopathy, retinal detachment, and rubeosis on risk of vision loss.
Results Patient Characteristics The mean age of patients in this study was 58 years. One hundred ninety-four patients (55.3%) were men and 157 (44.7%) were women. Of the 351 patients, 8.3% had diabetes and 29.4% had hypertension. The median largest tumor diameter was 12 mm and the median largest tumor height was 4 mm. By ophthalmic examination, all patients included in this study had tumors located within 1 DD of the fovea; however, 9 patients included in our final analysis did not receive the full prescribed dose to the fovea during planned treatment. This occurred because during the treatment planning, the tumor measured slightly more than 1 DD from the fovea. As a result, 342 patients (97.7%) in the final
analysis received the full prescribed dose to the fovea and 90 patients (25.6%) had tumors with direct foveal involvement. All of the tumors were more than 1 DD from the optic nerve, and 174 tumors (50%) were more than 2 DD away from the optic nerve. The mean follow-up time was 68.7 months and the median was 53.7 months (Table 1). Of the 351 patients in this study, 178 patients (50.7%) had visual acuity of 20/40 or better at baseline, 122 patients (34.8%) had visual acuity of 20/50 to 20/ 100 at baseline, 51 patients (14.5%) had visual acuity of 20/ 125 to 20/800 at baseline, and no patients had visual acuity of counting fingers or worse.
Tumor Control At the time of final follow-up, metastasis developed in 70 patients (19.9%) who died of the disease. Approximately 6% of the cohort (n ¼ 22) underwent enucleation. Of these, the tumor had recurred (suspected or documented growth) in 6 cases.
Figure 2. Cumulative rates of visual acuity retention of 20/200 or better (solid line) and 20/40 or better (dashed line) after proton beam irradiation.
3
Ophthalmology Volume -, Number -, Month 2015
Figure 3. Cumulative rates of visual acuity retention of 20/200 or better after proton beam irradiation by tumor height: 5 mm or less (solid line) and more than 5 mm (dashed line).
Visual Outcomes At last follow-up, 66 patients (18.8%) had visual acuity of 20/40 or better, 57 patients (16.2%) had visual acuity of 20/50 to 20/100, 90 patients (25.6%) had visual acuity of 20/125 to 20/800, and 138 patients (39.3%) had visual acuity of counting fingers or worse (Table 2; Fig 1).
Cumulative Rates of Vision Retention after Irradiation Analysis of cumulative rates of vision retention in patients with baseline vision of 20/200 or better revealed that 1 year after proton beam irradiation, 77.7% of patients retained 20/200 or better vision. Three years after proton beam irradiation, 53.5% of the patients retained 20/200 or better vision. More than one third of patients (35.5%) retained 20/200 or better vision 5 years after
proton beam irradiation. For those patients with a baseline visual acuity of 20/40 or better, 61.7% and 16.2% of patients retained this level of vision 1 year and 5 years after proton beam irradiation, respectively (Fig 2).
Risk Factors Associated with Vision Loss after Irradiation Tumor height was associated with vision outcome. The proportion of patients retaining vision of 20/200 or better 5 years after irradiation was 47.5% among patients with tumor height of 5 mm or less, compared with a proportion of 13.8% among patients with tumor height of more than 5 mm (P < 0.001; Fig 3). The proportion of patients retaining 20/40 or better vision 5 years after proton beam irradiation was 19.7% among those with tumor height of 5 mm or less compared with 7.0% of patients with tumor height of more than 5 mm (P < 0.001; Fig 4).
Figure 4. Cumulative rates of visual acuity retention of 20/40 or better after proton beam irradiation by tumor height: 5 mm or less (solid line) and more than 5 mm (dashed line).
4
Patel et al
Visual Outcomes of Macular Melanomas
Figure 5. Cumulative rates of visual acuity retention of 20/200 or better after proton beam irradiation by baseline visual acuity: 20/40 or better (solid line) and worse than 20/40 (dashed line).
Baseline visual acuity also influenced visual outcome. At 5 years after proton beam irradiation, 55.0% of those patients with baseline visual acuity of 20/40 or better retained 20/200 or better vision, whereas 18.0% of those with worse than 20/40 vision at baseline retained 20/200 or better vision (P < 0.001; Fig 5). Fewer patients with tumors 1 to 2 DD from the nerve had a final visual acuity of 20/40 or better than patients with tumors located more than 2 DD from the nerve (16.4% vs. 21.3%, respectively; P ¼ 0.15). Poor visual acuity (worse than 20/200) at final followup examination was observed in 58.2% of patients with tumors 1 to 2 DD away from the disc compared with 51.7% of patients with tumors more than 2 DD away from the disc (P ¼ 0.13). These differences were not statistically significant. There was a significant difference in the occurrence of papillopathy and rubeosis between patients who retained vision of 20/ 200 or better and those who did not. Papillopathy was present in 3.2% of patients with 20/200 vision or better at final follow-up and 10.4% of patients with vision worse than 20/200 (P ¼ 0.01). Rubeosis was present in 0.6% of patients with 20/200 vision or better at final follow-up and 20.7% of patients with vision worse than 20/200 (P < 0.0001). There was no significant difference in the occurrence of maculopathy or retinal detachment between the 2 groups (Table 3). Univariate Cox regression analysis showed that tumor height more than 5 mm, largest tumor diameter more than 15 mm, and baseline visual acuity worse than 20/40 were risk factors for vision
Table 3. Incidence of Ocular Complications by Visual Acuity at Final Follow-up Complication Maculopathy Papillopathy Retinal detachment Rubeosis
Visual Acuity ‡20/200 74 5 3 1
(46.8) (3.2) (1.6) (0.6)
Visual Acuity <20/200 98 20 8 40
(50.8) (10.4) (4.1) (20.7)
Data are no. of patients (%) unless otherwise indicated. *Chi-square test.
P Value* 0.52 0.01 0.036 <0.0001
loss, whereas tumor location more than 2 DD from the optic nerve, age at treatment, and history of diabetes were not significant. Ocular complications were associated significantly with increased risk of vision loss with the exception of maculopathy. In a multivariate model, tumor height more than 5 mm, largest tumor diameter more than 15 mm, baseline visual acuity worse than 20/ 40, presence of papillopathy, and presence of rubeosis all were associated significantly with risk of vision loss to worse than 20/ 200 (Table 4). Risk factors associated significantly with the development of papillopathy and rubeosis in this cohort were tumor diameter more than 15 mm (papillopathy and rubeosis) and tumor height more than 5 mm (rubeosis; Tables 5 and 6, available at www.aaojournal.org).
Subgroup Analysis: Dose Reduction In a subgroup of 203 patients with small and medium choroidal melanomas 5 mm or less in height and 15 mm or less in diameter, the effect of a reduced dose of radiation on visual outcomes was analyzed. The tumor location for this subgroup was the same as that of the overall group, with all tumors being 1 DD or less from the fovea and more than 1 DD from the optic nerve. Forty-seven patients received a dose of 50 Gy (RBE) and 156 patients received 70 Gy (RBE) in this subgroup. There was no statistically significant difference in the age, gender, largest tumor diameter, tumor height, distance from the optic nerve, or distance from the fovea between the 2 groups. All patients had a minimum of 3 years of follow-up, with a mean follow-up time of 59.5 months for the 50 Gy (RBE) group and 79.6 months for the 70 Gy (RBE) group (P ¼ 0.12). In the 50 Gy (RBE) group, 72.3% of patients had baseline visual acuity of 20/40 or better, and 55.1% of patients in the 70 Gy (RBE) group had baseline visual acuity of 20/40 or better (P ¼ 0.03). The cumulative rates of vision retention of 20/200 or better 1 year after proton beam irradiation were 95.7% for the 50 Gy (RBE) group and 89.0% for 70 Gy (RBE) group. At 5 years after proton beam irradiation, the rates of vision retention of 20/200 or better were 70.5% and 44.1% for the 50 and 70 Gy (RBE) groups, respectively, which was not significantly different (P ¼ 0.06; Fig 6). The cumulative rates of vision retention of 20/40 or better for the 2 groups 5 years after proton beam irradiation were 16.8%
5
Ophthalmology Volume -, Number -, Month 2015 Table 4. Cox Regression Analysis of Risk Factors for Vision Loss to Worse than 20/200* Univariate Models
Full Model
Risk Factor
Referent
Tumor height >5 mm BLVA worse than 20/40 Largest tumor diameter >15 mm Papillopathy present Rubeosis present Maculopathy present Age at treatment History of diabetes Location >2 DD from nerve
5 mm 20/40 or better 15 mm
2.861 (2.171e3.769) 2.964 (2.225e3.949) 3.063 (2.268e4.1354)
8.31Ee14 1.16Ee13 <0.001
1.667 (1.173e2.370) 2.168(1.592e2.953) 1.689 (1.176e2.427)
0.004 <0.001 0.005
1.613 (1.157e2.249) 2.295 (1.699e3.100) 1.691 (1.178e2.426)
0.005 <0.001 0.004
Absent Absent Absent Continuous No history 1e2 DD from nerve
1.869 4.724 0.666 1.002 1.031 0.836
0.005 <0.001 0.004 0.6174 0.7733 0.1974
1.663 2.552 0.873 1.002 1.043 0.860
0.034 <0.001 0.370 0.569 0.693 0.301
1.636 (1.040e2.571) 2.563 (1.715e3.829)
0.033 <0.001
(1.211e2.885) (3.294e6.774) (0.506e876) (0.993e1.012) (0.836e1.273) (0.637e1.098)
P Value
Hazard Ratio (95% Confidence Interval)
Best Fit Model
Hazard Ratio (95% Confidence Interval)
(1.039e2.662) (1.707e3.816) (0.650e1.173) (0.993e1.012) (0.845e1.288) (0.646e1.144)
P Value
Hazard Ratio (95% Confidence Interval)
P Value
BLVA ¼ baseline visual acuity; DD ¼ disc diameter. *Includes only patients with baseline vision better than 20/200.
and 23.9% for 50 and 70 Gy (RBE), respectively, but this difference was not statistically significant (P ¼ 0.43; Fig 7). There was no statistically significant difference in visual acuity at last follow-up between the 2 groups. At the last follow-up, nearly one-third (31.9%) of the 50 Gy (RBE) group, and nearly one-quarter (24.3%) of the 70 Gy (RBE) group retained 20/40 or better vision. The proportion of patients with extremely poor vision (counting fingers to no light perception) was 12.8% of the 50 Gy (RBE) group and 27.6% of the 70 Gy (RBE) group (Table 7). Thirty of 47 patients (63.8%) who received 50 Gy (RBE) and 90 of 156 patients (57.6%) who received 70 Gy (RBE) demonstrated maculopathy by the last examination. Of those patients with visual acuity worse than 20/200 at final follow-up, there was a significant difference in the proportion of patients with evidence of radiation maculopathy: 26.7% of the 50 Gy (RBE) group and 55.6% of the 70 Gy (RBE) group (P ¼ 0.007).
Discussion The management of choroidal melanoma involving the fovea poses one of the most challenging scenarios for practitioners. This study presents a large series of patients who received at least 90% of the prescribed radiation dose to the fovea during the treatment of their choroidal melanoma with proton beam radiotherapy. Despite the high dose of radiation to the fovea, well above the previously observed thresholds for producing radiation maculopathy,10 the visual outcomes in this retrospective analysis suggest that a significant proportion of these patients maintain some degree of useful vision. For patients with tumors of any size within 1 DD of the fovea, more than one-third of patients (35.5%) retained
Figure 6. Cumulative rates of vision retention of 20/200 or better after proton beam irradiation by dose: 50 Gy (relative biological effectiveness [RBE]) (solid line) or 70 Gy (RBE) (dashed line).
6
Patel et al
Visual Outcomes of Macular Melanomas
Figure 7. Cumulative rates of vision retention of 20/40 or better after proton beam irradiation by dose: 50 Gy (relative biological effectiveness [RBE]) (solid line) or 70 Gy (RBE) (dashed line).
vision of better than 20/200 at 5 years after proton beam irradiation. Additionally, more than half (55.0%) of patients with a baseline visual acuity of 20/40 vision or better retained vision better than 20/200 at 5 years after proton beam irradiation. At 5 years after irradiation, 16.2% of these patients retained visual acuity of 20/40 or better. Visual acuity after proton beam irradiation in this cohort was influenced most significantly by baseline visual acuity worse than 20/40 and the development of rubeosis. Each of these factors more than doubled the relative risk of vision loss to worse than 20/200. Additionally, rubeosis was related significantly to tumor height and diameter. For those patients with small and medium melanomas (tumor height 5 mm and LTD 15mm) within 1 DD of the fovea treated with proton beam irradiation, a significant proportion retained good visual acuity 5 years after treatment. More than two-thirds (70.5%) of patients receiving 50 Gy (RBE) and nearly half (44.1%) of patients receiving 70 Gy (RBE) retained 20/200 or better vision 5 years after treatment. Approximately 20% of patients with these smaller macular tumors retained 20/40 vision or better 5 years after irradiation. A decrease in prescribed radiation dose from 70 to 50 Gy (RBE) did not seem to increase the Table 7. Visual Acuity at Final Follow-up by Radiation Dose Visual Acuity 20/40 or better 20/50e20/100 20/125e20/800 CFeNLP
50 Gy (RBE) 15 11 15 6
70 Gy (RBE)
(31.9) (23.4) (31.9) (12.8)
38 26 49 43
(24.3) (16.7) (31.4) (27.6)
P ¼ 0.15 CF ¼ counting fingers; NLP ¼ no light perception; RBE ¼ relative biological effectiveness. Data are no. (%) unless otherwise indicated.
proportion of patients who retained usable vision. However, this degree of prescribed dose reduction may decrease the incidence of more severe vision loss (worse than 20/200). Few prior studies discuss visual outcomes in patients with choroidal tumors involving the fovea. Of the studies that reported such visual outcomes, all used plaque radiation as the method of radiotherapy. Gündüz et al14 reported the outcomes of plaque radiotherapy in the treatment of choroidal melanomas involving the macula. In their study of 630 patients with macular melanoma with median follow-up of 63 months, they found that visual acuity was more than 20/200 in 44% of patients and better than 20/40 in 9% of patients. The median dose to the fovea was 12 207 cGy, with a range of 1263 to 31 900 cGy. That series included tumors slightly farther away from the fovea (up to 3 mm) compared with our analysis, and only 2% of the patients had a portion of the tumor located beneath the foveola, versus 26% in our cohort. Median largest tumor diameter also was smaller at 10 mm. Although a larger range of doses delivered to the fovea and differences in inclusion criteria make comparisons difficult, the overall visual outcomes seem similar. Finally, the Collaborative Ocular Melanoma Study report15 described visual acuity outcomes over 3 years for patients treated with iodine-125 brachytherapy. Of the 623 patients included in the study, 13.2% had tumors within the foveal avascular zone (FAZ) and 24.9% had tumors within 0.1 to 2.0 mm within the FAZ. Of these, 73.9% of tumors involving the FAZ and 54.7% of patients with tumors within 2.0 mm of the FAZ had worse than 20/200 vision at 3 years. In comparison, 46.5% of patients in our series had vision worse than 20/200 at 3 years. Newman et al16 similarly reported 50 patients with subfoveal melanoma who were treated with plaque brachytherapy with palladium-103. Subfoveal melanoma was defined more strictly in this study as the posterior margin or tumor apex located beneath the fovea, and patients in this study received a
7
Ophthalmology Volume -, Number -, Month 2015 mean apical radiation dose of 82.8 Gy with a mean dose of 157 Gy to the fovea. The mean follow-up time after radiation was 62.2 months, which is comparable with that of our study. They reported that 50% (25/50) of patients had vision better than 20/ 200 at final follow-up and 36% (18/50) of patients had 20/40 visual acuity or better. The limitations of that study include the adjuvant treatments with antievascular endothelial growth factor, pars plana vitrectomy, and laser therapy, which could play a role in achieving the favorable visual outcomes in that study, as suggested by the authors. In contrast, no patients in our study received any adjuvant treatments. One significant limitation of our retrospective analysis is that despite decreasing the risk of coincident maculopathy and papillopathy by choosing tumors more than 1 DD away from the optic nerve, there is still a definite overlap in the incidence of maculopathy and papillopathy occurring in these patients. We were unable to distinguish whether the cause of vision loss in individual patients was the result of radiation effects on the macula or the optic nerve. However, our selection criteria helped to ensure that radiation dose to the fovea was larger than that to the optic nerve. There is a widely held impression that eyes with choroidal melanoma with foveal involvement are unlikely to retain any useful vision after radiotherapy. However, our series demonstrated that severe vision loss was not inevitable for these patients. We showed that patients receiving proton beam irradiation for choroidal melanoma within 1 DD of the fovea have a reasonable chance of maintaining useful vision and a small chance of retaining good vision. These visual outcomes suggest that proton irradiation at a dose of up to 70 Gy (RBE) in 5 fractions to the macula is permissive for survival of neuronal and vascular elements of the retina and choroid. Further elucidation of factors that distinguish the small minority of patients able to retain excellent vision is necessary for future improvement of visual outcomes.
References 1. Shields JA, Shields CL. Intraocular Tumors: An Atlas and Textbook. Lippincott Williams & Wilkins; 2008. 2. Shields JA, Augsburger JJ, Brady LW, Day JL. Cobalt plaque therapy of posterior uveal melanomas. Ophthalmology 1982;89:1201–7.
3. Gragoudas ES, Seddon JM, Egan K, et al. Long-term results of proton beam irradiated uveal melanomas. Ophthalmology 1987;94:349–53. 4. Packer S, Rotman M. Radiotherapy of choroidal melanoma with iodine-125. Ophthalmology 1980;87:582–90. 5. Gragoudas ES, Goitein M, Verhey L, et al. Proton beam irradiation of uveal melanomas. Results of 5 1/2-year study. Arch Ophthalmol 1982;100:928–34. 6. Lommatzsch PK. beta-Irradiation of choroidal melanoma with 106Ru/106Rh applicators. 16 Years’ experience. Arch Ophthalmol 1983;101:713–7. 7. Guyer DR, Mukai S, Egan KM, et al. Radiation maculopathy after proton beam irradiation for choroidal melanoma. Ophthalmology 1992;99:1278–85. 8. Seddon JM, Gragoudas ES, Egan KM, et al. Uveal melanomas near the optic disc or fovea. Visual results after proton beam irradiation. Ophthalmology 1987;94:354–61. 9. Kim IK, Lane AM, Egan KM, et al. Natural history of radiation papillopathy after proton beam irradiation of parapapillary melanoma. Ophthalmology 2010;117:1617–22. 10. Gragoudas ES, Li W, Lane AM, et al. Risk factors for radiation maculopathy and papillopathy after intraocular irradiation. Ophthalmology 1999;106:1571–7. discussion 1577e1578. 11. Finger PT, Chin KJ, Yu G-P. Risk factors for radiation maculopathy after ophthalmic plaque radiation for choroidal melanoma. Am J Ophthalmol 2010;149:608–15. 12. Gragoudas ES, Lane AM, Regan S, et al. A randomized controlled trial of varying radiation doses in the treatment of choroidal melanoma. Arch Ophthalmol 2000;118: 773–8. 13. Gragoudas E, Li W, Goitein M, et al. Evidence-based estimates of outcome in patients irradiated for intraocular melanoma. Arch Ophthalmol 2002;120:1665–71. 14. Gündüz K, Shields CL, Shields JA, et al. Radiation complications and tumor control after plaque radiotherapy of choroidal melanoma with macular involvement. Am J Ophthalmol 1999;127:579–89. 15. Melia BM, Abramson DH, Albert DM, et al. Collaborative Ocular Melanoma Study (COMS) randomized trial of I-125 brachytherapy for medium choroidal melanoma. I. Visual acuity after 3 years. COMS report no. 16. Ophthalmology 2001;108:348–66. 16. Newman H, Chin KJ, Finger PT. Subfoveal choroidal melanoma: pretreatment characteristics and response to plaque radiation therapy. Arch Ophthalmol 2011;129: 892–8.
Footnotes and Financial Disclosures Originally received: June 26, 2015. Final revision: September 14, 2015. Accepted: September 24, 2015. Available online: ---.
Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2015-1079.
1
Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts. 2
Department of Ophthalmology and Visual Science, University of Utah, Salt Lake City, Utah.
3
Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts.
8
Supported by the Grimshaw-Gudewicz Charitable Foundation (Fall River, MA). Author Contributions: Conception and design: Lane, Gragoudas, Kim Analysis and interpretation: Patel, Lane, Morrison, Gragoudas, Kim Data collection: Lane, Trofimov, Shih, Gragoudas, Kim Obtained funding: Gragoudas
Patel et al
Visual Outcomes of Macular Melanomas
Overall responsibility: Patel, Lane, Morrison, Trofimov, Shih, Gragoudas, Kim Abbreviations and Acronyms: BLVA ¼ baseline visual acuity; CF ¼ counting fingers; DD ¼ disc diameter; FAZ ¼ foveal avascular zone; Gy ¼ gray; NLP ¼ no light perception; RBE ¼ relative biological effectiveness.
Correspondence: Ivana K. Kim, MD, Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA 02114. E-mail: ivana_kim@ meei.harvard.edu.
9