Choroidal melanoma (CM) is both the most common form of uveal melanoma and the most frequent primary intraocular malignancy. There is a reported yearly incidence of 4–6 per million per year in the US and Europe. Of interest, Queensland, Australia (beneath the ozone hole) has the highest reported incidence (10 per million per year) worldwide.1 While ultraviolet (UV) radiation has been proved to play a role in the pathogenesis of cutaneous melanoma, the statistical link for CM has been less clear. Factors implicating UV radiation in the incidence of CM include an increased incidence in patients with blue irises, outdoor occupations, and that these tumors are most commonly found in the posterior uvea, the choroid.1
CMs can present with or without symptoms.2 Size, location, and tumor-related secondary complications (e.g. detachment, hemorrhage, cataract, and glaucoma) are associated with a variety of clinical symptoms. Symptomless patients are typically discovered during the course of periodic ophthalmic examinations or incidentally when patients seek refraction. Once found, CMs are typically sent to regional referral centers where further evaluations include extended ophthalmoscopy and specialized photography (e.g. color imaging, fundus autofluorescent imaging [FAF], fluorescein angiography [FA], and indocyanine green angiography [ICG]), as well as optical coherence tomography (OCT).3–6 Each diagnostic method has been shown to offer unique capabilities to reveal the diagnostic characteristics of CM.
Radiographic studies, such as magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography–computed tomography (PET/CT) are also selectively employed to evaluate CM.7–11 However, these techniques are more commonly used for systemic screening for metastatic disease.12,13
CM treatment has evolved from primary enucleation and local resection to eye-, sight-, and vision-sparing radiotherapy techniques. Stallard introduced the first eye plaques.14 These first disk-shaped radiation therapy devices contained cobalt-60 and were later replaced by ruthenium-106, iodine-125, and palladium-103 sources.15–17 Each radionuclide offers a unique and different intraocular dose distribution. Select centers have also used external beam radiation therapy (EBRT). The most common eye- and vision-sparing EBRT is proton beam.18 More exotic forms of EBRT include the gamma-knife and stereotactic radiosurgery.19
The Collaborative ocular melanoma study (COMS) was a three-arm trial that included two multicenter prospective randomized clinical studies. A 16-year effort, the COMS stands as the largest evidence-based study of CM. Of the three clinical arms, the small tumor study found that 31 % of small CMs were documented to grow over five years.20 The medium-sized tumor trial found no survival benefit from enucleation (removal of the eye) versus episcleral iodine-125 plaque irradiation.16 The large CM study (an indirect test of the Zimmerman Hypothesis: that enucleation surgery disseminated metastases) found that pre-enucleation EBRT (2 Gy x five daily fractions) did not improve post-enucleation survival.21 Though the COMS had a significant impact on practice patterns, much has changed over subsequent years. Current practice patterns are based on the characteristics of the CM, visual acuity, treatment modalities available and the patients’ motivation to keep their eye, as well as regional, political, and socioeconomic factors.
The Diagnosis of Choroidal Melanoma
In the UK, 45 % of CM patients were reported to be asymptomatic at diagnosis.22 In these cases, CM is an accidental finding during a routine periodic ophthalmic examination. In all other cases, patients present with symptoms of visual loss, metamorphopsia, flashing lights, or floaters. Less common findings have been related to secondary effects, such as cataract, narrow-angle glaucoma, or vitreous hemorrhage. It is not uncommon for a CM to be missed at initial examination, particularly with small ciliary body tumors. The rates of failure to diagnose have been reported within a range of 25–37 %.2,22 Eye cancer specialists should inquire about a past medical history of primary lung, breast, or other cancers, systemic diseases, or anticoagulant therapy. These answers may help differentiate CM from simulating lesions.
Anterior Segment Examination
CMs can affect the anterior segment. Common findings include ‘sentinel’ episcleral blood vessels. Less commonly, the tumor can push on the natural lens, inducing a sector cataract (with irregular astigmatism) or angle narrowing. Secondary glaucoma can be seen as irisneovascularization, outflow invasion, or angle closure.23 CMs with massive orbital extension can be associated with proptosis, papilledema, and exposure keratopathy. These cases may also exhibit an afferent pupillary and/or color vision deficit, as well as ocular motor disturbances.
Most CMs can be diagnosed by ophthalmoscopy (see Figure 1). Direct, indirect, and contact lens techniques are employed. Visualization of a pigmented choroidal tumor with the orange pigment lipofuscin (OPL) on the tumor’s surface, significant thickness, and exudative retinal detachment are diagnostic of CM.24 For smaller tumors, photography and ultrasound-assisted documentation of growth may be required.
Over the last 30 years, the misdiagnosis rate for CM has been greatly reduced. For example, the use of experienced examiners, indirect ophthalmoscopy, FA, and ultrasound imaging reduced the COMS misdiagnosis rate to less than 1 %.25,26
Ophthalmic Imaging Techniques
Fundus photography, FAF, and OCT have significantly contributed to the diagnosis and understanding of CM and its associated comorbidities.
Fundus photography plays a unique, central, and indispensable role in the diagnosis and follow-up of CM. Photography should be used to document the initial tumor size, to measure its basal dimensions, and to document its surface characteristics. Furthermore, the distances from the posterior tumor edge to the fovea and optic disk should be recorded. These measurements can be used to help the clinician determine radiation dose to critical structures and risks for vision loss. Photographic evidence of tumor-associated exudative subretinal fluid, choroidal neovascularization, OPL deposition, and retinal pigment epithelial metaplasia at first offers diagnostic significance and post-treatment measures of tumor response. For example, after treatment exudative fluid and OPL typically regress and the tumor darkens. Photographic documentation of these changes allows the eye cancer specialists to establish the pattern of CM regression. Late radiation changes (including radiation retinopathy and optic neuropathy) can be initially documented and comparatively evaluated for progression or regression by fundus photography.27
FAF has emerged as a powerful tool to document and discover occult OPL. Exudative retinal detachments can also be seen with FAF, particularly when there is an associated acute or chronic retinal pigment epithelial degeneration.3,28
OCT is the most sensitive method of detecting subretinal fluid and intraretinal edema.29 OPL can also be seen as deposits on the retinal pigment epithelium. Some studies suggest that choroidal nevi and their overlying intact retinal pigment epithelium are less likely to allow transmission of light into the tumor’s stroma and subjacent choroid.
FA and ICG are used to evaluate CM. Angiographic imaging can reveal unique patterns of tumor circulation. For example, a ‘double circulation’ with well-defined blood vessels is rarely seen in metastatic tumors and a ‘coarse vascular pattern’ is characteristic of choroidal hemangioma. Metastatic choroidal tumors typically have poor circulation, characterized by slow uptake, leakage, and stromal dot-like microaneurysms. In select cases, FA or ICG may be the best methods to determine largest basal tumor dimensions prior to the choice of plaque size.
Beyond diagnosis, post-treatment surveillance of tumor circulation can help determine response to radiation, adjuvant laser therapies, and the onset of secondary vasculopathy (e.g. radiation maculopathy and optic neuropathy).
Ultrasound imaging has come to play an essential role in the diagnosis, treatment, and follow-up of patients with CM. Pioneered by Karl Ossoinig and Sandra Frazier Byrne, standardized ophthalmic ultrasound techniques were employed by the COMS and thus established low internal reflectivity and dome shape as the most common presenting characteristics of CM (see Figure 2a).30 Mushroom-shaped CM were noted in only 25 % of cases and found to be more common among larger tumors (see Figure 2b). The COMS report number 29, describing baseline echographic characteristics of 2,320 patients with medium and large tumors, found that 1,268 patients (54 %) had a tumor apex located posterior to the equator (27 % equator-posterior, 22 % posterior-equator, and only 5 % posterior).30
A-scan imaging is the most sensitive method to assess a tumor’s internal acoustic reflectivity. This characteristic has been exploited to differentiate between types of tumors and to monitor response to therapy. For example, internal reflectivity is typically high with choroidal hemangioma, moderate with metastases, and low with CM.31 Non-hemorrhagic retinal pigment epithelial detachments and choroidal effusions are typically echolucent. During the years after radiation therapy for choroidal melanoma, A-scan typically reveals increasing internal reflectivity.
B-scan imaging has become an essential tool for measuring tumor dimensions. Commonly, CMs are measured to be larger by ultrasound compared with fundus photography. This information is vital for treatment planning (accurate plaque size selection) and for assessing response to radiotherapy. Furthermore, both dynamic A- and B-scan imaging can be used to assess a tumor’s intrinsic vascularity at diagnosis and document post-treatment reductions over time.
Use of 3D B-scan ultrasound imaging has proved invaluable for evaluating disorganized eyes (e.g. those containing occult CM). Interactive sectioning of the 3D block has allowed for post-enucleation ultrasonographic-histopathologic reconstruction. Lastly, such sectioning also allows for simultaneous longitudinal and transverse views of ophthalmic plaques in vivo, aiding intra-operative evaluations of plaque placement.25 Ophthalmic ultrasound imaging has been recognized as an integral method for evaluating and T-staging uveal melanomas by both the American Joint Committee on Cancer and the International Union Against Cancer.20,32–34
Intraocular Radiographic Imaging
MRI and radiographic CT have limited diagnostic value for the diagnosis of CM. In special circumstances, MRI can be used to help differentiate between choroidal hemorrhage and CM. It is also used to localize CM prior to endoresection when there are opaque media. Both MRI and CT are essential in evaluating extrascleral, intraneural, or intracranial CM extension.
PET/CT has recently been used to characterize CM. Using 18-fluorodeoxyglucose uptake, PET/CT allows in vivo measurements of metabolic activity within CM, though only relatively large CMs were found to be specific uptake value (SUV)-positive. However, this characteristic may be a biomarker for metastasis. In one study, an SUV greater than 4.0 was correlated to such currently accepted tumor high-risk factors as epithelioid cell type, high-risk vascular loops, extrascleral tumor extension and metastases.10 A recent report found that plaque irradiation extinguished SUV positivity and therefore metabolic activity of CM.35
Intraocular Tumor Biopsy
Since CM can be diagnosed by clinical evaluation and testing in over 99 % of cases, one must question the need for biopsy. On the other hand, if there existed a risk-free method to retrieve a piece of tumor, such tissue would serve to confirm the clinical diagnosis as well as offering helpful histopathologic, genetic, and molecular information. Unfortunately, choroidal biopsy carries the risks (albeit small) of endophthalmitis, vitreous hemorrhage, retinal detachment, and CM dissemination. Therefore, we must weigh the risks versus such potential benefits. Most ophthalmic oncologists agree that biopsy is indicated for atypical tumors where the diagnosis is unclear, when the tumor is probably metastatic with no known primary, and when the patient requires a biopsy to proceed with treatment.36
With advances in cytogenetics, in recent years it appeared likely that biopsy would also be used to grade uveal melanomas. In the 1990s, it was discovered that uveal melanomas tend to develop several chromosomal abnormalities, the most important of which re monosomy 3, isochromosome 6p, trisomy 8, and isochromosome 8q.37 A landmark article by Prescher et al. showed monosomy 3 to correlate strongly with metastatic death, data indicating a reduction in the five-year survival from 100 to less than 50 %.38 However, these initial expectations regarding the sensitivity and specificity of monosomy 3 tests in predicting metastatic death were found to be overly optimistic. Perhaps cytogenetic prognostication may require taking into account the complex pathway we call metastasis and that immune status, tumor size, and access to routes of egress may also play a role.
More recently, Harbour and Onken have introduced the concept of RNA-based analysis, dividing CM into ‘type 1’ and ‘type 2’ tumors. According to their study, type 1 tumors carry a low risk of metastasis, while type 2 are mostly lethal.39 These data are admittedly preliminary and require validation by a large clinical study. In the interim, we know of no ophthalmic oncologist who would deny treatment to a type 1 CM based on genetic testing. In summary, cytogenetic classification (both monosomy 3 and RNA analysis) should be considered investigational.
Biopsy methods have been reported to employ a fine needle or vitreous cutter. Approaches include the pars plana, cornea (for anterior segment tumors), and trans-scleral aspiration into the tumor’s base. Ocular oncologists are not unanimous about which option is better, or whether a stepwise approach starting with a fine-needle aspiration biopsy (FNAB) is possible. No matter which approach is taken, potential dissemination of tumor cells remains a concern.40
Biomarkers for Metastases
The first biomarkers for CM were histopathologic as offered by Callender’s classification, in which predominantly spindle cell melanomas carried the best prognosis, mixed-cell tumors were worse, and epithelioid cell tumors the most likely to metastasize.41 Other features associated with systemic prognosis include intrinsic tumor vascular patterns, tumor vascular density, tumor size, location of the anterior margin of the tumor, and degree of ciliary body involvement.42,43 It is important to note that in consideration of most of the above, as well as epidemiologic and other clinical factors, the COMS offered the first evidence-based results that patient age and largest tumor dimension were the most significant biomarkers for metastasis by multivariate analysis.16 In 2006, Finger and colleages suggested that PET/CT imaging specfic uptake value (SUV) could be used as a biomarker for metastatic risk.10 This observation was recently confirmed by Lee et al., 2011.44
Metastatic Disease Detection
Although CM is not as aggressive as skin melanoma, Kujala et al. reported mortality rates of 31 % by five years, 45 % by 10 years, 49 % by 25 years, and 52 % by 35 years.45 In this article, we found that the diagnosis and treatment of metastatic CM varies by country, region, and clinical centers. It is also driven by patient and social and economic factors.
In general, pre-operative metastatic surveys should include a physical examination looking for cutaneous and subcutaneous nodularity as well as hepatomegaly. Radiographic imaging can range from contrast-enhanced abdominal MRI, CT or ultrasound imaging to total body PET/CT.
In that over 90 % of patients present with hepatic involvement, abdominal imaging should be integral to pre-operative staging and subsequent surveillance. However, in a recent study using whole-body PET/CT, 75 % of patients with metastatic CM were noted to have extrahepatic – primarily bone (50 %), lung (25 %), lymph node (25 %), and skin (25 %) – metastases (see Figure 3). This distribution suggests that liver function tests alone are inadequate.24,46,47
There are centers that do not perform metastatic surveys due to a lack of definitive treatment for metastatic CM. At The New York Eye Cancer Center, we currently perform pre-operative surveys to find the <4 % that have detectable disease at presentation.12 Such discovery can eliminate the need for surgery for the primary CM, reveals synchronous non-ocular tumors, and allows for both systemic staging and early intervention. Follow-up metastatic surveys allow for both early intervention and end-of-life planning.
Clinical practice varies greatly from one center to another. Though the COMS defined CMs as small, medium, and large, the Ophthalmic Oncology Task Force of the AJCC, together with the UICC, has recently produced a universally accepted seventh edition AJCC staging system. This five-year effort involved 45 eye cancer specialists from 11 countries.48
Specifically, Chapter 51, Malignant Melanoma of the Uvea, was the result of a retrospective multinational evaluation of over 8,000 patients. This system is now considered ‘universal’ after being widely accepted as a requirement for publication by over 10 major ophthalmic journals and their associated societies. They recognized that standardized reporting would allow for more accurate comparison of diagnostic and therapeutic techniques, as well as the development of biomarkers for CM.
Diagnosis of Choroidal Melanoma—Summary
Findings of OPL, thickness greater than 2 mm and subretinal fluid remain the most important diagnostic characteristics of CM. These characteristics are evaluated by ophthalmoscopy, photography with and without angiography, and ophthalmic ultrasound imaging.
Specialized FAF and OCT imaging offer unique and often helpful information about the presence of OPL and subretinal fluid. Ultrasound is best at uncovering obscure basal extensions, vascularity, shifting exudative fluid, scleral extension, and tumor height. Despite all this technology, small CM and indeterminate lesions are typically considered too small for biopsy and are followed for evidence of growth prior to intervention. In summary, clinical, photographic, and ultrasound imaging techniques are currently employed for initial assessment, evaluation of post-treatment regression, and side effects.
The Treatment of Choroidal Melanoma
CM treatment is performed to prevent metastatic disease, to maximize visual acuity, and to preserve the eye. With these goals in mind, there has been an evolution of treatment from primary enucleation and local resection to laser photocoagulation and radiation therapy. At this time, radiation therapy is the most widely accepted alternative to enucleation.49
In most developed countries, enucleation surgery is performed when the CM is considered to be too large for vision-sparing techniques, for extensive extrascleral tumor extension, and for neovascular glaucoma.50 Alternatively, in undeveloped countries eye- and vision-sparing radiation techniques may not be available. The COMS large tumor trial found that 20 Gy (4 Gy per day x five 5 days) did not improve survival after enucleation surgery.51 This finding suggests that the Zimmerman Hypothesis (that we disseminated metastasis by manipulation during enucleation surgery) was incorrect.52 It is important to note that orbital tumor recurrence was reported in less than 1 % of cases. If this occurs, it can be treated by local resection of all visible tumor, followed by 50 Gy external beam radiotherapy.53
At The New York Eye Cancer Center, most patients can be fitted with an ocular prosthesis that requires minimal maintenance and has excellent cosmesis.54 Prosthetic ocular motility is typically less than the natural eye and mucus discharge is common. Patients should be counseled to wear unbreakable polycarbonate glasses to protect the remaining eye.
Local Resection Techniques
Endoresection comprises the removal of a posteriorly located uveal melanoma through a pars plana vitrectomy, typically after radiation therapy.55 Surgery is performed with a vitrector through a retinotomy or beneath a retinal flap. An air–fluid exchange drains residual subretinal fluid; endolaser photocoagulation is used to destroy any residual intrascleral tumor and to achieve retinopexy. Silicone oil is typically required. Theory suggests that endoresection reduces tumor-related ocular morbidity but adds the risk of vitreoretinal complications (silicone oil removal, secondary rhegmatogenous retinal detachment, intraocular hemorrhage) and tumor dissemination.56
Trans-scleral Tumor Resection
Trans-scleral tumor resection (TSR) can be lamellar or full thickness. The more widely used partial thickness technique involves the preparation of a lamellar scleral flap, ocular decompression by limited pars plana vitrectomy, resection of the tumor together with the deep scleral lamella, suturing of the scleral flap, and intraocular injection of balanced salt solution. Due to initially high tumor regrowth rates, adjunctive radiotherapy has been added before or after TSR.57
Primary indications for TSR include large anterior melanomas, where endoresection is not possible, and the ability to tolerate hypotensive anesthesia. Poor local control rates, coupled with a high incidence of post-operative complications (secondary rhegmatogenous retinal detachment, intraocular hemorrhage, ocular hypotony, and phthisis) and concerns about CM dissemination, have limited the widespread use of TSR.58–60 TSR can be considered a relatively high-risk alternative to enucleation of eyes with very large anteriorly placed CM.
In the 1950s, Dr Meyer-Schwickerath attempted primary treatment of CM with the xenon-arc laser.61 Since that time others have used argon, krypton, dye, and, most recently, the infrared spectrum. To date, no laser method has produced acceptable local control, but has added risks including hemorrhage, retinal detachment, edema and traction, scleromalacia with orbital tumor extension, and laser-induced optic neuropathy.
Transpupillary thermotherapy (TTT), as originally described by Oosterhuis and popularized by Shields,62–65 was suggested for the treatment of small tumors, up to 4 mm in height, lying near the optic disk or the fovea (in an effort to spare these structures from damage). With a modified delivery system beam widths between 1 and 3 mm and exposure times of up to one minute are typically generated.66 This method was reported to offer a local tumor control rate of 76 % (not as successful as radiation therapy).67 However, due to failures of local control (intraocular and orbital), TTT has largely been abandoned as a primary treatment for CM.68 Current patient selection includes very thin tumors, treatment of circumpapillary tumors that cannot be reached by plaque, and the sandwich technique, where TTT is used in combination with a radiation plaque. At The New York Eye Cancer Center, we use TTT laser for marginal tumor recurrences and in patients who cannot tolerate plaque surgery.
Radiation Therapy for Choroidal Melanoma
You can destroy a choroidal melanoma with any form of radiation therapy, but dose-related differences (side effects) will be seen in the normal ocular structures. Therefore, our primary goal of destroying the tumor will be accomplished, but there will be distinct differences in resultant rates of visual acuity and eye preservation.
Teletherapy means treatment from a distance. In ophthalmic oncology CM has been treated with cylindrically shaped beams of charged particles – protons or helium ions. Prior to treatment, tantalum clips are sewn to the episclera around the base of the tumor to localize the targeted zone. The particle-generating cyclotron is directed toward the eye in line with the CM as defined by the clips (see Figure 4). Eye movement is monitored on a screen and the beam turned off if the eye wanders. Typical daily fractions are 10–16 Gy, for five consecutive days (50–80 Gy). Each relatively high-dose-rate fraction is given during less than five minutes, compared with continuously over 5–7 days with brachytherapy. Therefore, in part due to high-dose-rate effects, the proton dose is much higher than numerically equivalent brachytherapy apex doses. Therefore, it is understandable that proton-beam local control rates are among the highest.49,75,76
However, the cylindrically shaped beam must traverse the anterior segment to reach the CM. As Packer predicted, proton-beam-related anterior segment doses were relatively high, resulting in more reports of eyelash loss, dry eye, neovascularization of the iris, neovascular glaucoma, and cataract. These complications became even greater in treatment of large and anteriorly located tumors. Similarly, radiation retinopathy and optic neuropathy were more commonly reported in treatment of posteriorly located CM.49
Brachytherapy means treatment up close. Stallard introduced the first ophthalmic plaques for the treatment of CM.14 The technique involves sewing the disk-shaped plaque device onto the sclera so as to cover the base of the intraocular tumor and a free margin (target zone).17,71,77–79 Radiation travels from the plaque through the sclera and into the CM. Radiation deposition is primarily governed by the inverse square law, Compton scattering, and absorption as it continues through the vitreous and exits through normal ocular structures. The original cobalt-60-based plaques produced high-energy gamma rays with greater irradiation of normal structures, the patient, and the surgeon. No longer in clinical use, cobalt-60 plaques were replaced with ruthenium-106 plaques by Lommatzsch in East Germany, and by Packer, Rotman, and Sealy, who originally used iodine-125 plaques.79,81
In 1985, the COMS chose to use iodine-125 seeds in standardized gold alloy plaques as the alternative treatment versus enucleation in the medium-sized CM trial. This decision made iodine-125 a clinical standard in North America for more than 16 years. In 1990, Finger realized that the lower-energy photons emitted from palladium-103 seeds would result in less irradiation of most normal ocular structures (and a greater likelihood of vision and eye preservation).81,82 Since that time, select centers have adopted palladium-103 plaque therapy. At The New York Eye Cancer Center, we suggest pre-operative comparative dosimetry (iodine-125, palladium-103, ruthenium-106) to monitor doses to critical intraocular structures (fovea, optic nerve, lens, opposite eye wall), prior to radionuclide selection for each plaque.17
It is important to note that iodine-125 and palladium-103 plaques deliver photons that can extend deeply into the eye. In contrast, ruthenium-106 plaques emit beta-particles that can only extend far enough to treat 5 mm high tumors reliably.22,25 Furthermore, the dose to the tumor base and tumor margin can be four times higher in treatment of the same 5 mm-tall CM using iodine-125 versus ruthenium-106. The high base dose will cause earlier chorioretinal atrophy.17,77,83 Like proton therapy, low-energy plaque irradiation of larger tumors has been associated with more radiation complications. However dry eye, eyelash loss and neovascular glaucoma are relatively uncommon (see Table 1).17
The most frequent low-energy plaque-related complications are radiation retinopathy, radiation optic neuropathy, and cataract.17,68–70 At The New York Eye Cancer Center, the largest CM basal dimension treated with ophthalmic plaque is 22 mm and the tallest has been 16 mm. Dr Finger’s slotted plaques successfully control tumors that encircle the optic nerve (circumpapillary).71 Therefore, few patients cannot be treated with eye- and vision-conserving ophthalmic plaque radiation therapy.
Anti-Vascular Endothelial Growth Factor Therapy
Prior to the advent of bevacizumab and ranibizumab (Avastin® or Lucentis®, Roche-Genentech, South San Francisco, California, US), more than 50 % of patients were reported to have less than 20/200 vision five years after CM radiation therapy (see Table 1). Though this was all-cause visual morbidity, the most common causes of severe irreversible loss of vision were radiation maculopathy and radiation optic neuropathy. In 1997, Finger reported on short-term findings of reductions in retinal and optic nerve edema, as well as decreased retinal hemorrhage, associated with periodic intravitreal injections of bevacizumab. Later reports, and those of others, have supported these findings.84–86 It has been Finger’s impression that higher radiation doses to fovea and optic nerve are harder to overcome with this method.
The primary goal of CM treatment is to destroy or remove the malignancy to prevent metastasis. Though enucleation and surgical resection are available, most patients are currently treated with plaque radiation therapy. The most widely used sources are iodine-125 and ruthenium-106 plaques, with a few referral centers using proton beam and palladium-103. Cyberknife and stereotactic radiosurgery are most exotic forms of EBRT and should be considered investigational. We know that all methods of radiation therapy can destroy a CM. However, evidence now exists that radiation dose to fovea, lens, and optic nerve can be used to predict rates of complications.81,82 Therefore, it is reasonable to compare the available radiation modalities, perform computer-aided simulations, calculate doses to critical structures, and choose the ‘best’ method prior to treatment.
Ophthalmic oncology has progressed from primary enucleation to modern and exotic methods of eye- and vision-sparing radiation therapy. Thankfully, in the modern era, fewer patients have to lose their eye due to the presence of CM. ■