A 43 year-old male applies for life insurance. He recently complained of seeing “floaters” when in bright sunlight and consulted an ophthalmologist. The examination revealed a choroidal nevus in the left eye. The nevus measured 6.2 mm in diameter and .25 mm in thickness. The eye examination was otherwise normal. Plans were made for follow-up in one year. Personal and family history were negative for melanoma.
What is the relationship of choroidal nevus to ocular melanoma, if any? Is there any extra mortality anticipated in this case?
The choroid is a vascular layer of the eye that is sandwiched between the retina and the sclera. It is part of the uveal tract which contains melanocytes (pigmented cells containing melanin). The uveal tract also includes the iris and the ciliary body. Any part of the uveal tract can develop melanoma, however, 86% of reported cases arise in the choroid.
Figure 1 - Schematic of the Human Eye
The uveal tract is made up of the iris, the ciliary body and the choroid. The choroid is the vascular layer between the retina and the sclera.
Choroidal nevi were found in 6.5% of Caucasian adults participating in a screening study in Australia. Estimates in the white US population range from 4.6%-7.9%. A recent review of participants (n=5575) undergoing retinal imaging as part of the 2005-2008 National Health and Nutrition Examination Survey (NHANES) provided a more detailed breakdown of the prevalence of choroidal nevi by age, sex, and race. The prevalence is higher in males, Caucasians and those over age 60 (Figure 2). Overall, the prevalence was 4.7%.
Figure 2 - Prevalence of Choroidal Nevi by Age, Gender & Race
Observed prevalence of choroidal nevi in US adults from the 2005-2008 NHANES study, stratified by age, gender and race. The prevalence is likely underestimanted because the imaged areas were focused on the macula and the optic nerve, which represent about 40%-45% of th total choroidal area.
The obvious underwriting concern about a choroidal nevus is that it may transform into an ocular melanoma. Fortunately, this is a rare occurrence. It is estimated that uveal tract melanoma occurs in six to seven per million persons in the US. Analysis by Singh, et. al. used the prevalence of choroidal nevus and the incidence of choroidal melanoma from the Surveillance, Epidemiology and End Result (SEER) database and calculated a 1 per 8845 annual conversion rate from choroidal nevus to choroidal melanoma.
The authors of the study note that the estimate may be high due to the potential for underestimation of the prevalence of choroidal nevi. Indeed, cancer of the eye is rare, since the most recent (2012) SEER data places the incidence of cancer of the eye and orbit for Caucasians age 75+ at 3.8/100,000; 65-74 at 3.26/100,000; and 50-64 at 1.7/100,000.
Much research has gone into identifying characteristics of cutaneous nevi to identify or predict current or future melanoma. This has resulted in the mnemonic ABCDE:
Borders that are irregular
Color that is variegated or uneven
Diameter over 6 mm
Evolution denoted by changes in color or size
The presence of any of these characteristics should prompt a consult with a dermatologist and consideration for possible biopsy. In another study, Shields, et al suggest another helpful mnemonic: “To Find Small Ocular Melanoma Using Helpful Hints.” The beginning letters represent:
Thickness > 2 mm
Fluid (presence of subretinal)
Symptoms (flashing lights and floaters)
Margin (within 3 mm of the optic disc)
Halo (absence of)
Hazard Ratio for Developing Ocular Melanoma
The hazard ratio for choroidal nevus transforming to ocular melanoma for each characteristic is listed in Figure 3.
Figure 3 - Factors Predictive of Growth of Choroidal Nevi into Melanoma
|Halo||Absence of surrounding halo||6.48||.009|
|Fluid||Presence of subretinal fluid||3.16||.002|
|Hollowness||Hollow by ultrasound||2.92||<.001|
|Orange Pigment||Presence of orange pigment||2.75||<.001|
|Symptoms||Flashing lights / floaters||2.34||.002|
|Margin||Within 3 mm of the optic disc||1.82||.001|
As a continuous variable, there is an increase in hazard ratio of 2.75 for each 1-mm increase in thickness
Based on the 2,514 patients with choroidal nevi followed at a tertiary care center. This may have been a high-risk group with a history of dysplastic nevus syndrome in 1%, cutaneous melanoma in the opposite eye in 4%. Arch Opthalmol. 2009; 127(8): 981-987)
When combinations of factors were studied it was determined that the hazard ratio for the development of ocular melanoma with one or two factors was ~3; for three or four factors, ~5; for five or six factors, ~9; and for all seven factors, 21.
Growth of a lesion over time is recognized as an important feature of melanoma, especially within a short period. However, it should be noted that slow growth can occur in a benign nevus, and lesions that eventually develop into melanoma can be stable for decades prior to conversion. In one study the mean growth rate in lesions that developed into melanoma were .96 mm/year in diameter and 1.12 mm/year in thickness.
Choroidal Melanoma: A Dangerous Malignancy
If it should develop, choroidal melanoma is a dangerous malignancy. Disease-specific survival rates after the diagnosis of uveal melanoma have been reported as 69% at five years, 55% at 15 years and 48% at 25 years.
Tumor size, both diameter and thickness, have a relationship to mortality. A meta analysis of uveal melanoma found all-cause mortality at five years for tumors <3 mm thickness to be 16%, for tumors 3 mm-8 mm thickness, 32% and at >8 mm thickness, 53% (Figure 4).
Figure 4 - Ocular Melanoma
In this picture of a retina, arrows indicate presence of ocular melanoma involving the optic disc.
Tumor thickness is strongly positively correlated to metastatic disease, with thicker tumors being more likely to metastasize. Another study (Lane, et al) included a cohort of 1063 patients who had either a definite choroidal nevus, an indeterminate pigmented lesion, or a small melanoma. Choroidal nevi were <6 mm in diameter, <0.5 mm thick and did not have orange pigment, subretinal fluid, or symptoms at presentation (n=256). Indeterminate lesions were between 0.5 mm-2 mm thick and did not have orange pigment, subretinal fluid or symptoms at presentation (n=334). The remaining group was diagnosed with small melanomas (lesions <10 mm in diameter and <5 mm thick, n=373).
Small melanomas were treated with radiation, as were nevi (n=3) and indeterminate lesions that evolved over time (n=39). Authors of the study followed all subjects using the Social Security Death Master File and the National Death Index to determine cause of death. No patients in the choroidal nevus group died of ocular melanoma in the average 8.4 years of follow-up. In contrast the indeterminate group recorded two deaths, and 13 subjects within the small melanoma group died during the same observation period. In the indeterminate group cumulative disease-specific death rates by years of follow-up were zero at five years, 1% at 10 years and 3% at 15 years.
One interesting finding was that the combined nevus and indeterminate lesion groups had four deaths from cutaneous melanoma during the follow-up period. This was 13 times the expected death rate from cutaneous melanoma in the general population.
Diagnosis and Treatment
99% of choroidal pigmented lesions are diagnosed visually with the ophthalmoscope with occasional addition of ultrasonography. Studies are under way to determine other methods to distinguish between indeterminate lesions that have a high probability of progressing to melanoma from those that do not.
Single-photon emission computed tomography (SPECT) using a tracer that accumulates in cells producing melanin has shown promise. Analysis of high-resolution digital images and quantification of the autofluorescence of lipofuscin, the source of the orange pigment seen with ocular melanoma, has also shown promise in the detection of early ocular melanoma.
Ocular melanoma usually is treated with radiation or enucleation. Indeterminate or suspicious lesions are often followed closely, since treatment would likely result in the loss of functional vision (60%) or endanger the viability of the globe.
Returning to the Case
By the numbers one can see that the overwhelming majority of choroidal nevi do not develop into melanoma. However, the presence of symptoms, absence of a surrounding halo and the lesion diameter of 6.2 mm would make this case a slightly higher risk for future evolution to ocular melanoma.
Based on two of these factors (symptoms and diameter) the case would have been classified as an indeterminate lesion in the study by Lane et al. They observed a disease-specific death rate of 1% at 10 years and 3% at 15 years. The thinness of the lesion and the overall rarity of ocular melanoma temper the overall risk.
From the 2008 VBT, a 43 year-old male nonsmoker would expect 18.57 deaths per 1000 through duration 15 years. The additional disease specific deaths based on the study would represent an additional 162% mortality. Mild to moderate excess mortality would be the initial assessment. With compliance on future follow-up and demonstrated stability of the lesion, the risk should decrease.
http://seer.cancer.gov/faststats/ last accessed 8/14/2015.
Lane AM, et al. “Mortality After Diagnosis of Small Melanocytic Lesions of the Choroid.” Arch Ophthalmol 2010. 128(8):996-1000.
Sheilds C, et al. “Choroidal Nevus Transformation Into Melanoma: Analysis of 2514 Consecutive Cases.” Arch Ophthalmol 2009. 127(8):981-987. Grassetto G, et al, “Ocular Melanoma and Other Unusual Sites.” PET Clin. 6 (2011) 79–89.
Grassetto G, et al, “Ocular Melanoma and Other Unusual Sites.” PET Clin. 6 (2011) 79–89.
Albertus D, et al. “Autofluorescence Quantification of Benign and Malignant Choroidal Nevomelanocytic Tumors.” JAMA Ophthalmol. 2013. 131(8):1004-1008.
Qiu M, et al. “Choroidal Nevus in the United States Adult Population.” Ophthalmology 2015. Published online awaiting print: 1-13.
Singh A, et al. “Estimating the Risk of Malignant Transformation of a Choroidal Nevus.” Ophthalmology 2005. 112(10):1784-1789.
UpToDate® last accessed 8/14/2015.