Abstract
Purpose:
It has been previously demonstrated that light iris color is a risk factor for the development of uveal melanoma. This study is estimates the correlation of iris color to gene expression profile (GEP), which is an independent measure of likelihood of metastasis.
Methods:
A retropective chart review of a consecutive series of patients treated for uveal melanoma was performed. Patients were selected for study if they had been treated for uveal melanoma with either brachytherapy or enucleation, and if they had GEP classification (Castle BioSciences) into Class 1A, Class1B, or Class 2. Iris color classification was derived from drivers license information as administered by the Department of Motor Vehicles. Iris color was classified into one of four groups: Brown, Blue, Hazel, or Green. Complete data was available for 146 patients. Differences in GEP distributions in the separate classes of iris color was analyzed with students t-test and chi-squared analysis.
Results:
The population of patients was predominantly Caucasian (98%). The incidence of iris colors in the 146 patients with uveal melanoma was: Blue 66/146 (45%), Hazel 26/146 (18%), Green 14/146 (10%), Brown 40/146 (27%). Blue iris color had the following percentage distribution of GEP results: Class 1A 58%, Class 1B 14%, Class 2 29%. Brown iris color had the following percentage distribution of GEP results: Class 1A 35%, Class 1B 33%, Class 2 33%. Differences in the distributions of GEP class in different iris colors was found to be not statistically significant.
Conclusions:
This study is consistent with previous studies which show light iris color is a risk factor for uveal melanoma in Caucasians in that light iris color is predominant in this patient group, with brown iris color represented in a minority of 25%. The separate classes of iris color, however, do not appear to be associated with significant differences in GEP result. This study supports the hypothesis that iris color does not confer clinically significant information regarding risk of developing metastatic disease in patients with uveal melanoma. The limitations of this study are it's relatively small size, and the dependence on the unknown relaibility of iris color determination as present in the database of the department of motor vehicles.