Purchase this article with an account.
Ryan Kim, Brandon Nguyen, Maria Eugene Bretana, Eric Kegley, Amy C Schefler; Relationship between traditional clinical risk factors and GEP classification for uveal melanoma. Invest. Ophthalmol. Vis. Sci. 2017;58(8):4407.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To examine the association between traditional clinical high-risk features of choroidal nevi/small uveal melanoma and genomic analysis with the gene expression profile (GEP) assay.
This was a retrospective, single-center, observational case-series that included patients with small to medium-sized uveal melanomas treated from 2012-2016 with adequate data for analysis. A total of 83 patients were identified to have met the criteria for the study. Patients were clinically examined for the following clinical risk factors: drusen/retinal pigment epithelium (RPE) changes, vascularity on B-scan ultrasound, internal reflectivity on A-scan ultrasound, subretinal fluid (SRF), orange pigment, and apical tumor height. A point system was created to assess the degree of high-risk features each tumor had: one point was given each for a tumor height greater than or equal to 2 mm, lack of drusen/RPE changes, presence of vascularity, a low to medium internal reflectivity, presence of SRF, and orange pigment for a total of six possible points. The association of the clinical risk score with the GEP classification (1A, 1B, 2) was then assessed statistically using the Chi-Square Test and the Fisher Exact Test.
Sixty-nine patients had complete data on clinical risk factors and were included in the final analysis. Thirty-two patients were classified as GEP Class 1A; 15 as Class 1B; and 22 as Class 2. Tumors with a Class 2 GEP signature were more likely to lack drusen/RPE changes than Class 1A tumors (p<0.001) and 1B tumors (p=0.02). All other clinical risk factors as well as a cumulative total risk score did not have a statistically significant association with GEP signature.
In this cohort, lack of drusen/RPE changes was statistically significantly associated with GEP Class 2 lesions. Other than this feature, no other traditional clinical high-risk features were associated with GEP Class. Larger studies are needed, but clinicians should keep in mind that modern genomic data is much more predictive of metastatic death than traditional clinical funduscopic assessment in uveal melanoma patients.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.
This PDF is available to Subscribers Only