Local Recurrence Significantly Increases the Risk of Metastatic Uveal Melanoma

Local Recurrence Significantly Increases the Risk of Metastatic Uveal Melanoma The Ophthalmic Oncology Task Force Purpose: To assess of the effect of local recurrence of uveal melanoma on metastasis using a multicenter international tumor registry. Design: Retrospective study using an online tumor registry. Participants: Patients with uveal melanoma diagnosed between 2001 and 2011. Methods: A committee was formed to create uveal melanoma patient-specific data fields. Ten subspecialty ophthalmic oncology centers from 4 continents shared data. Patient selection criteria included diagnosis of uveal melanoma and adequate records to allow tumor staging by American Joint Committee on Cancer (AJCC) criteria and follow-up for metastatic melanoma. Main Outcome Measures: Local tumor recurrence and metastatic uveal melanoma. Results: Of 3809 total entries, 3217 patients with ciliary body and choroidal (CBC) melanoma and 160 with iris melanoma were evaluated. There was a median follow-up of 3.7 years (95% confidence interval [CI], 3.5e3.8). One hundred fifty-two patients (4.7%) with CBC melanoma experienced local recurrence, with a cumulative incidence of 11%. Kaplan-Meier point estimates for remaining free of local recurrence were 99% (95% CI, 99e99) at 1 year, 93% (95% CI, 92e94) at 5 years, and 89% (95% CI, 86e91) at 10 years. Five- and 10-year metastasisfree Kaplan-Meier estimates for the recurrence-free group were 87% (95% CI, 86e89) and 82% (95% CI, 79e84), and those for the local recurrence group were 71% (95% CI, 62e78) and 62% (95% CI, 49e72). The difference between these 2 groups was statistically significant (P < 0.001). Furthermore, local tumor recurrence increased the risk of metastasis by a hazard ratio (HR) of 6.28 (95% CI, 4.4e8.9; P < 0.001). Local recurrence was detected up to 9.8 years after treatment. Extrascleral extension also was associated with local recurrence (HR, 3.2; 95% CI, 1.5e6.7; P ¼ 0.003), but higher AJCC T-size category was not (P ¼ 0.63). Five patients (n ¼ 5/161 [3.1%]) with iris melanoma demonstrated local recurrence and 1 metastasized. Conclusions: International multicenter data sharing was used to evaluate the effect of local tumor recurrence on metastatic rate. In that local tumor recurrence was associated with a significantly higher risk of systemic metastasis, effective initial treatment and long-term surveillance of treated uveal melanoma patients is necessary. Ophthalmology 2016;123:86-91 ª 2016 by the American Academy of Ophthalmology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Supplemental material is available at www.aaojournal.org. Local tumor control is a primary goal in the treatment of cancer. 1 In 2002, the Collaborative Ocular Melanoma Study reported that local treatment failure occurred in 10.3% of Collaborative Ocular Melanoma Study patients who underwent iodine 125 plaque brachytherapy. 2 More recent studies, including a literature review of 49 articles, revealed local recurrence rates ranging from 0% to 55.6%. 3 Local orbital recurrence rates after enucleation are poorly documented, but are believed to be less than 1%. 4 These data are important, in that local tumor recurrence is a complication thought to be associated with an increased risk of uveal metastasis and thus decreased survival. 5e8 However, this current assumption that recurrence equals increased risk is supported by a relatively few, small singlecenter studies and 1 multicenter study; all are limited to 70 or fewer patients with local recurrence. 6e8 The Ophthalmic Oncology Task Force comprises members of the American Joint Committee on Cancer (AJCC) and the Union for International Cancer Control and includes eye cancer specialists from around the world. 9 For this study, 10 centers pooled their uveal melanoma patient data in an effort to study this disease. 10 We first used this database successfully to test the seventh edition AJCC uveal melanoma staging system. 11 In this study, we analyzed the data to define the relationships among local recurrence, extrascleral extension (EXE), and AJCC tumor size (T size). However, our primary goal was to investigate the largest collection of uveal melanomas with local recurrence and the impact of local recurrence on metastasis-free survival. Methods The Ophthalmic Oncology Task Force collaborated to develop an Internet-based retrospective registry to evaluate uveal melanoma outcomes. The methods of data collection and security have been 86 Ó 2016 by the American Academy of Ophthalmology This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Published by Elsevier Inc. http://dx.doi.org/10.1016/j.ophtha.2015.09.014 ISSN 0161-6420/15

The Ophthalmic Oncology Task Force Uveal Melanoma Recurrence and Metastasis described previously. 11 Each participating center obtained local ethics and institutional review board approval for participation and entry of patient data into the online database. This study conforms to the Declaration of Helsinki and the Health Insurance Portability and Accountability Act of 1996. Patient Eligibility Patient eligibility criteria included patients with primary melanoma of the iris, ciliary body, and choroid consecutively diagnosed and treated in a 10-year span from April 1, 2001 through April 30, 2011. Primary tumor treatment methods varied by center, with each type chosen at the discretion of the eye cancer specialists at each center. These were reported to include, but were not limited to, brachytherapy, enucleation or local resection (endoresection and lamellar eye wall resection), transpupillary thermotherapy with and without adjuvant radiation therapy, and stereotactic radiotherapy. However, no numerical treatment-related data were required for this database. Detection of local tumor recurrence was determined by ultrasonography, comparative fundus photography, or inspection and palpation of the anophthalmic orbit (after enucleation). Local recurrence was determined by clinical and imaging studies used at each individual center. There were no exclusions to type of recurrence. Therefore, this study represents site-determined global clinical practice at 10 centers during this decade. Patients were excluded from analysis if they had metastasis (stage 4 disease) at initial staging. Recorded outcome measures included local recurrence and metastasis. There were no limitations on the methods of screening for metastasis. However, these included abdominal ultrasound imaging, computed tomography, magnetic resonance imaging, or combined whole body positron emission tomography and computed tomography, as was the custom and practice at each center. Based on the current assumption that there is no current curative treatment for almost all patients with metastatic uveal melanoma and the recognized difficulties confirming the actual cause of death, this study defined metastasis to be synonymous with metastatic death. Statistical Analysis Categorical variables were described using frequencies and proportions. Time to local recurrence was measured from the date of initial staging. Patients who were alive and free of local recurrence at last follow-up were censored. One hundred thirty patients (4%) died without local recurrence. Because the number of competing risk events (death without local recurrence) was small relative to the size of the data set, these patients were censored at their date of death. Standard methods were used for analysis, rather than implementing competing risks methodology. Kaplan-Meier plots, a standard Cox proportional hazards model, and the log-rank test for trend were implemented to test for a trend relationship between time to local recurrence and T size (based on the AJCC 7th edition criteria). 12 The standard log-rank test was implemented to test for a relationship between time to local recurrence and EXE. Time to metastasis was measured from the date of staging until the date of metastasis; patients who were alive and free of metastasis at last follow-up were censored. Twenty-five patients died without metastasis. Because the number of patients in this latter category was small relative to the size of the data set, these patients were censored at their date of death rather than treating their death as a competing risk event. A Kaplan-Meier plot and the log-rank test were implemented to test for a relationship between time to metastasis and local recurrence at last follow-up. The effect of a local recurrence on the risk of metastasis after staging also was investigated using Cox models, with local recurrence treated as a time-varying covariate. In the univariate model, other factors were not taken into account. An unadjusted (univariate) model and a model adjusted for stage, T-size category, and ciliary body involvement or EXE grouping were implemented. Kaplan-Meier plots, a standard Cox proportional hazards model, and the log-rank test were implemented to test for a relationship between time to local recurrence and EXE. In the analysis of time to metastasis after a local recurrence, only patients who experienced a local recurrence were included. Patients who experienced metastasis before a local recurrence or on the same date as local recurrence were included with their time to metastasis set to 0.00001 years. Those who had no metastasis and no follow-up after their local recurrence were excluded. The related estimates are generalizable to all patients who had a local recurrence. The Kaplan-Meier curves estimate the probability of being free of metastasis after a local recurrence. The 95% confidence limits for this sub-analysis were calculated using the method of Greenwood 13 and the complementary log-log transformation. The survival package in R software (R Core Team, Vienna, Austria) version 2.15.1 was used to generate Kaplan-Meier plots; SAS TS level 1M1 version 9.3 software (SAS Inc, Cary, NC) was used to perform all other statistical analyses. Statistical significance was set to 0.05. Table 1. Summary of Studies Focused on Local Treatment Failure Rates of Choroidal Melanoma and Effect on Metastasis or Death Treatment Authors Method Jampol et al Iodine 125 (COMS) 2 plaque Vrabec et al 8 Cobalt 60 plaque Single or Multicenter Median Follow-up (Range), Years Cajoulle et al 6 Proton beam Single No LR: 3.93 (0e17.84) LR: 5.11 (0.22e17.86) No. of Patients No. of Patients with Local Recurrence Local Failure Incidence (%) Risk of Metastasis or Death Multicenter 5.6 (0e12.8) 650 57 10.3* Adjusted risk ratio, 1.5 for LR (P ¼ 0.08) Single 4.9 (1.8e10.7; LR group) 445 70 15.7 Estimated 5-yr survival: 87% for LR-free, 58% for LR (P < 0.0001) 1102 61 6.1 Overall survival at 10 years: 83.6% for LR-free, 43.1% for LR Gragoudas et al 7 Proton beam Single 5.2 (n/a) 1922 62 3.2 Relative risk, 4.1 for LR Present study Varied Multicenter 3.7 (0.1e12.6) 3217 152 4.7, 11* Hazard ratio, 6.3 for LR (P < 0.001) COMS ¼ Collaborative Ocular Melanoma Study; LR ¼ local recurrence; n/a ¼ not available. *cumulative estimate. 87

Ophthalmology Volume 123, Number 1, January 2016 Results Participating Centers Ten ophthalmic oncology centers from 8 countries (Argentina, Canada, Japan, Russia, Spain, Sweden, The Netherlands, and the United States) in 4 continents (North and South America, Europe, and Asia) participated in this study. Patients Three thousand eight hundred nine patients with uveal melanoma were entered into the online database and staged according to the 7th edition of the AJCC system. Of these, 3377 (89%) records were complete. One hundred eighty-one patient entries were eliminated because of lack of follow-up and 67 were eliminated because of initial presentation with stage 4 disease (metastasis). One hundred eighty-four were removed because of missing or inconsistent key variables. This included 118 with unknown metastasis status, 35 with unknown EXE status, 11 with unknown ciliary body involvement status, and 128 with incomplete or inconsistent data that prevented being assigned a T-size category or prognostic stage. Some patient entries were eliminated for more than 1 of the above reasons. Primary ciliary body and choroidal melanoma was the diagnosis for 3217 patients, and 160 patients had primary iris melanoma. The median follow-up from date of initial diagnosis and staging for the 3217 patients was 3.7 years (95% confidence interval [CI], 3.5e3.8; range, 0.1e12.6 years; Table 1). Ciliary Body and Choroidal Melanoma Recurrence One hundred fifty-two patients (4.7%; n ¼ 152/3217) with ciliary body and choroidal melanoma experienced local recurrence by their last follow-up. Kaplan-Meier point estimates for remaining free of local recurrence were 99% (95% CI, 99%e99%) at 1 year, 93% (95% CI, 92%e94%) at 5 years, and 89% (95% CI, 86%e91%) at 10 years after treatment. Kaplan-Meier estimates of time to local treatment failure are shown graphically in Figure 1A and numerically in Figure 1B. Five years after initial tumor staging, an estimated 7.3% of eyes (95% CI, 6.1%e8.7%) exhibited treatment failure, and an estimated 11% of eyes exhibited treatment failure at 10 years (95% CI, 8.9%e 13.7%). Local recurrence was detected as early as 1 month and as late as 9.8 years after initial tumor staging (Fig 1B). Figure 1. A, The cumulative proportion of patients and (B) a table showing Kaplan-Meier estimates of local recurrence of primary ciliary body and choroidal melanoma after initial tumor staging and treatment. Local Recurrence and Metastasis Kaplan-Meier point estimates at 5 and 10 years after initial tumor staging show that there is a statistically significant difference in the metastasis-free survival for the recurrence and recurrence-free groups (P < 0.001). Forty-three of the 152 patients (28.3%) with local recurrence demonstrated metastasis, compared with 287 (9.4%) of those who were free of recurrence. For the recurrencefree group, the 5- and 10-year metastasis-free estimates were 87% (95% CI, 86%e89%) and 82% (95% CI, 79%e84%), whereas the recurrence group estimates were 71% (95% CI, 62%e78%) and 62% (95% CI, 49%e72%), respectively (Fig 2). In a univariate Cox model, local recurrence increased the risk of metastasis by a hazard ratio (HR) of 6.28 (95% CI, 4.42e8.92). Thus, failure of local control was shown to be associated with a statistically significant (P < 0.001) increase in metastatic disease. Multivariate analysis adjusting for AJCC stage, T-size category, and presence of ciliary body involvement or EXE 88

The Ophthalmic Oncology Task Force Uveal Melanoma Recurrence and Metastasis demonstrated that local recurrence was still associated with an increased risk of metastasis (HR, 6.46; 95% CI, 4.52e9.24; P < 0.001). Local Recurrence According to American Joint Committee on Cancer T-Size Category According to T-size category and using the entire 3217-patient cohort, 4.6% of patients in size category T1 (51/1115), 5.7% of patients in size category T2 (64/1128), 4.3% of patients in size category T3 (34/789), and 1.6% of patients in size category T4 (3/185) had local recurrence. The evidence failed to show a trend relationship between T-size category and local recurrence (P ¼ 0.63, log-rank test for trend; Kaplan-Meier estimates did not show a trend). No HRs from a Cox proportional hazards model were statistically significant (all P ¼ 0.12 and higher). Extrascleral Extension and Local Recurrence Of the 152 patients with local recurrence, 7 patients had EXE. Using the entire cohort of 3217, 12.5% (7/56) with EXE demonstrated local recurrence as compared with 4.6% (145/3161) without EXE who demonstrated local recurrence. The HR from a Cox proportional hazards model indicated that the presence of EXE was associated with a higher risk of local recurrence (HR, 3.2; 95% CI, 1.5e6.7; P ¼ 0.003). Time from Local Recurrence to Metastasis Only patients who experienced a local recurrence (n ¼ 152) were included in the analysis of time from local recurrence to metastasis. Of these, 43 patients (28% [43/152]) with local recurrence demonstrated metastasis. Six (3.9%) had already experienced metastasis before local recurrence, 9 (5.9%) had metastasis and local recurrence detected on the same date, 8 (5.3%) had no followup after their local recurrence, and 28 (18.4%) experienced metastasis after local recurrence. One hundred one patients (66.4%) had additional follow-up after their local recurrence date and did not demonstrate metastasis (during that time). After excluding 8 patients because of a lack of follow-up after local recurrence, a cohort of 144 was used to evaluate time to metastasis, with a median follow-up after local recurrence of 2.3 years (95% CI, 1.8e3.0 years). Forty-three of the 144 patients (29.9%) who had a local recurrence also had a metastasis either before, at the same time as, or after local recurrence; the remaining 70.1% of the sample remained free of metastasis by their last follow-up date. The median time to metastasis was 6.8 years; however, in Figure 3, note this estimate was based on a small number remaining at risk for local recurrence with a large 95% CI (95% CI, 6.2eupper limit cannot be calculated). The longest interval from local recurrence to detection of metastasis was 9.8 years. Two (1.4%) patients died without metastasis. Local Recurrence of Iris Melanoma Of the 160 patients with iris melanoma, 5 (3.1%) demonstrated local recurrence by their last follow-up. Of these 5 patients, 2 were in T-size category T1, and 3 were T2. One experienced metastasis 2 years after local recurrence, 3 had follow-up after their local recurrence date (6 months, 6 months, and 3 years) and did not experience metastasis during that time, and 1 had no follow-up after local recurrence (and no metastasis). Kaplan-Meier point estimates (n ¼ 160) for remaining free of local recurrence were 99% (95% CI, 95%e100%) at 1 year, 95% (95% CI, 88%e98%) at 5 years, and 93% (95% CI, 82%e97%) at 10 years. This subset was too small for further statistical analysis. Figure 2. Kaplan-Meier metastasis-free probability for 3217 ciliary body and choroidal melanoma patients with versus without local recurrence (LR) (P < 0.001). Discussion Our study revealed that local tumor recurrence significantly increases the risk of metastatic melanoma (P < 0.001; HR, 6.28). This is the highest HR reported to date. Previous single-center studies evaluating posttreatment local recurrence also have found an increased risk of metastasis (Table 1). For example, Gragoudas et al 7 found a relative risk of 4.1 for those with local recurrence (n ¼ 62) after proton beam irradiation. Vrabec et al 8 reported on cobalt 60 plaque brachytherapy (n ¼ 70) and found a statistically significant decrease in the estimated 5-year survival between their local recurrence-free (87%) and recurrence (58%) groups. In another single-center study, Caujolle et al 6 Figure 3. Kaplan-Meier metastasis-free probability (solid line) and 95% confidence limits (dashed lines) for 144 ciliary body and choroidal melanoma patients with local recurrence. Note the large confidence interval after year 5 resulting from the small number of patients. 89

Ophthalmology Volume 123, Number 1, January 2016 (n ¼ 61) noted that survival rates in uveal melanoma patients after proton beam therapy decreased from 83.6% (in those without recurrence) to 55% in those with local recurrence at 10 years. When compared with our relatively large (n ¼ 152) multicenter study, we found that patients with local recurrence fared slightly better, with 71% (5 years) and 62% (10 years) demonstrating metastasis-free survival. Unlike the aforementioned studies, we analyzed local tumor recurrence following the standard of care in multiple international eye cancer centers using their bestrecommended treatment strategies. That is, the study cohort included patients treated with varying types of primary tumor treatment and was not limited to a single treatment method. Thus, local recurrence rates best reflect the standard of care during that decade. This study provides evidence that local recurrence significantly increases the risk of metastasis despite the type of primary tumor treatment. Local radiation failure has been attributed to radiation resistance, misplacement of radioactive plaque, insufficient radiotherapeutic margins, and eye movement (primarily during charged-particle beam therapy). 5,14 Therefore, our findings support the use of methods to improve local control (e.g., intraoperative ultrasound to evaluate plaque placement, slotted plaque radiotherapy, 2e3-mm radiotherapeutic margins, adequate tumor dose and eye fixation during external beam techniques). 15e20 Transpupillary thermotherapy and resection can be augmented by adjuvant radiation treatment. 21e23 Clearly, our findings suggest that all uveal melanoma patients be counseled of greater risks for metastatic disease should their primary tumor treatment fail. This study does not examine the exact mechanism of how local recurrence increased the risk of metastasis. In addition to therapeutic failure, theories include that recurrent tumors may have physiologic, genetic, or anatomic characteristics that made them more malignant or early to metastasize. 11,24e28 It also has been proposed that these biomarkers may contribute both to outcomes of local recurrence and to metastasis. However, this study was not designed to analyze these complex genetic or biological predisposing factors. Rather, it was designed to evaluate the relationship between local recurrence and the incidence of metastatic disease. Certainly, those determining the prognostic efficacy of such tumor-specific biomarkers should consider failure of local control in their calculations. In conclusion, the Ophthalmic Oncology Task Force s uveal melanoma registry clearly demonstrates that multicenter, international data sharing can be used to examine statistically important clinical questions such as the impact of failure of local control on metastatic risk. This study revealed that local recurrence of uveal melanoma is associated with a statistically significant increase in metastasis and that this subgroup of high-risk patients deserves increased surveillance for metastatic disease. Acknowledgment. Princess Margaret Cancer Centre, Toronto, Canada, served as the coordinating center and provided statistical analysis through Christine Massey; Dr. Priya Durairaj provided extensive data collection and oversaw overall data quality; Dr. Yuliya Gavrylyuk and the Health Informatics Research team developed the algorithms and database and managed the Internet website. The Institut Catala Oncologia acknowledges Dr. Juan Pera, Department of Radiotherapy, and Dr. Josep M. Piulats, Department of Medical Oncology, for their participation in uveal melanoma treatment and research. The Leiden University Medical Center acknowledges Professor Stefan de Geus for his contribution to patient care. Writing Committee: Paul T. Finger, MD and Kimberly Chin, OD; The Ophthalmic Oncology Task Force (in order of number of patients contributed): Brenda L. Gallie, MD, E. Rand Simpson, MD, Princess Margaret Cancer Research Centre, University Health Network, Toronto, Canada; Svetlana Saakyan, MD, Anush Amiryan, MD, Vladimir Valskiy, MD, PhD, Moscow Helmholtz Research Institute of Eye Diseases, Moscow, Russia; Paul T. Finger, MD, Kimberly J. Chin, OD, Ekaterina Semenova, MD, The New York Eye Cancer Center, New York, NY, USA; Stefan Seregard, MD, Maria Fili, MD, St. Eriks Eye Hospital, Karolinska Institute, Stockholm, Sweden; Matthew Wilson, MD, Barrett Haik, MD, University of Tennessee Health Science Center Hamilton Eye Institute, Memphis, TN, USA; Jose M. Caminal, MD, Jaume Català, MD, Cristina Gutierrez, MD, Bellvitge University Hospital, Barcelona, Spain; David E. Pelayes, MD, PhD, Anibal Martín Folgar, MD, Carlos G. Durand Hospital, Buenos Aires University, Buenos Aires, Argentina; Martine J. Jager, MD, Mehmet Dogrusöz, MD, Gregorius P.M. Luyten, MD, Leiden University Medical Center, Leiden, The Netherlands; Arun Singh, MD, Andrew P. Schachat, MD, Cole Eye Institute, Cleveland Clinic, Cleveland, OH, USA; Shigenobu Suzuki, MD, Yukiko Aihara, MD; National Cancer Center, Tokyo, Japan. References 1. DeVita Jr VT, DeVita J, Theodore S, et al., eds. Cancer: Principles and Practice of Oncology. 10th ed. Philadelphia: Wolters Kluwer; 2014. 2. Jampol LM, Moy CS, Murray TG, et al. The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma: IV. Local treatment failure and enucleation in the first 5 years after brachytherapy. COMS report no. 19. Ophthalmology 2002;109:2197 206. 3. Chang MY, McCannel TA. Local treatment failure after globeconserving therapy for choroidal melanoma. Br J Ophthalmol 2013;97:804 11. 4. Moshfeghi DM, Moshfeghi AA, Finger PT. Enucleation. Surv Ophthalmol 2000;44:277 301. 5. Finger PT. Intraocular melanoma. In: Vincent Jr VT, DeVita J, Theodore S, et al., eds. Cancer: Principles and Practice of Oncology. 10th ed. Philadelphia: Wolters Kluwer; 2015:1770 9. 6. Caujolle JP, Paoli V, Chamorey E, et al. Local recurrence after uveal melanoma proton beam therapy: recurrence types and prognostic consequences. Int J Radiat Oncol Biol Phys 2013;85:1218 24. 7. Gragoudas ES, Lane AM, Munzenrider J, et al. Long-term risk of local failure after proton therapy for choroidal/ciliary body melanoma. Trans Am Ophthalmol Soc 2002;100:43 8. discussion 48e49. 8. Vrabec TR, Augsburger JJ, Gamel JW, et al. Impact of local tumor relapse on patient survival after cobalt 60 plaque radiotherapy. Ophthalmology 1991;98:984 8. 9. Finger PT; 7th Edition AJCC-UICC Ophthalmic Oncology Task Force. The 7th edition AJCC staging system for eye cancer: an international language for ophthalmic oncology. Arch Pathol Lab Med 2009;133:1197 8. 10. The Ophthalmic Oncology Task Force. Ophthalmic sites. In: Edge SB, Byrd DR, Compton CC, et al., eds. AJCC Cancer 90

The Ophthalmic Oncology Task Force Uveal Melanoma Recurrence and Metastasis Staging Manual. 7th ed. New York: Springer-Verlag; 2009: 523 89. 11. The Ophthalmic Oncology Task Force. International validation of the American Joint Committee on Cancer s 7th Edition classification of uveal melanoma. JAMA Ophthalmol 2015;133:376 83. 12. The Ophthalmic Oncology Task Force. Malignant melanoma of the uvea. In: Edge SB, Byrd DR, Compton CC, et al., eds. AJCC Cancer Staging Manual. 7th ed. New York: Springer- Verlag; 2009:547 59. 13. Greenwood M. The natural duration of cancer. Reports on Public Health and Medical Subjects. London: Her Majesty s Stationery Office. 1926;33:1e26. 14. Char DH, Kroll S, Phillips TL, Quivey JM. Late radiation failures after iodine 125 brachytherapy for uveal melanoma compared with charged-particle (proton or helium ion) therapy. Ophthalmology 2002;109:1850 4. 15. Finger PT, Chin KJ, Tena LB. A five-year study of slotted eye plaque radiation therapy for choroidal melanoma: near, touching, or surrounding the optic nerve. Ophthalmology 2012;119:415 22. 16. Chiu-Tsao ST, Astrahan MA, Finger PT, et al. Dosimetry of (125)I and (103)Pd COMS eye plaques for intraocular tumors: report of Task Group 129 by the AAPM and ABS. Med Phys 2012;39:6161 84. 17. The Ophthalmic Oncology Task Force. The American Brachytherapy Society consensus guidelines for plaque brachytherapy of uveal melanoma and retinoblastoma. Brachytherapy 2014;13:1 14. 18. Harbour JW, Murray TG, Byrne SF, et al. Intraoperative echographic localization of iodine 125 episcleral radioactive plaques for posterior uveal melanoma. Retina 1996;16: 129 34. 19. Barker CA, Francis JH, Cohen GN, et al. (106)Ru plaque brachytherapy for uveal melanoma: factors associated with local tumor recurrence. Brachytherapy 2014;13: 584 90. 20. Rivard MJ, Chiu-Tsao ST, Finger PT, et al. Comparison of dose calculation methods for brachytherapy of intraocular tumors. Med Phys 2011;38:306 16. 21. Damato B. Adjunctive plaque radiotherapy after local resection of uveal melanoma. Front Radiat Ther Oncol 1997;30: 123 32. 22. Damato BE, Paul J, Foulds WS. Risk factors for residual and recurrent uveal melanoma after trans-scleral local resection. Br J Ophthalmol 1996;80:102 8. 23. Sagoo MS, Shields CL, Mashayekhi A, et al. Plaque radiotherapy for juxtapapillary choroidal melanoma: tumor control in 650 consecutive cases. Ophthalmology 2011;118: 402 7. 24. Onken MD, Worley LA, Char DH, et al. Collaborative Ocular Oncology Group report number 1: prospective validation of a multi-gene prognostic assay in uveal melanoma. Ophthalmology 2012;119:1596 603. 25. Finger PT, Chin K, Iacob CE. 18-Fluorine-labelled 2-deoxy-2- fluoro-d-glucose positron emission tomography/computed tomography standardised uptake values: a non-invasive biomarker for the risk of metastasis from choroidal melanoma. Br J Ophthalmol 2006;90:1263 6. 26. Kujala E, Damato B, Coupland SE, et al. Staging of ciliary body and choroidal melanomas based on anatomic extent. J Clin Oncol 2013;31:2825 31. 27. Kivela T, Kujala E. Prognostication in eye cancer: the latest tumor, node, metastasis classification and beyond. Eye (Lond) 2013;27:243 52. 28. Finger PT. Eye: choroidal melanoma, retinoblastoma, ocular adnexal lymphoma and eyelid cancers. In: O Sullivan B, Brierly J, D Cruz A, et al., eds. UICC Manual of Clinical Oncology. 9th ed. Chichester: John Wiley & Sons, Ltd; 2015: 726 44. Footnotes and Financial Disclosures Originally received: July 27, 2015. Final revision: September 11, 2015. Accepted: September 11, 2015. Available online: October 21, 2015. Manuscript no. 2015-1263. The New York Eye Cancer Center, New York, New York. Presented at: American Academy of Ophthalmology Annual Meeting, November 2015, Las Vegas, Nevada, Paul T. Finger, MD, on behalf of the Ophthalmic Oncology Task Force. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Supported by The Eye Cancer Foundation; the Paul T. Finger, MD, Fund at Princess Margaret Cancer Centre, Toronto, Canada; and John and Myrna Daniels (Toronto, Ontario, Canada, and New York, NY). The sponsors had no role in the design or conduct of this research. Author Contributions: Conception and design: Finger, Simpson, Ophthalmic Oncology Task Force Analysis and interpretation: Finger, Chin, Ophthalmic Oncology Task Force Data collection: Ophthalmic Oncology Task Force Obtained funding: Finger Overall responsibility: Finger, Ophthalmic Oncology Task Force Abbreviations and Acronyms: AJCC ¼ American Joint Committee on Cancer; CBC ¼ ciliary body and choroidal; CI ¼ confidence interval; EXE ¼ extrascleral extension; HR ¼ hazard ratio; T size ¼ tumor size (T-size refers to AJCC ct stage). Correspondence: Paul T. Finger, MD, New York Eye Cancer Center, 115 East 61st Street, Suite 5A/B, New York, NY 10065. E-mail: pfinger@eyecancer.com. 91