Purchase this article with an account.
Nathalie Cassoux, Laurence Desjardins, Corine Plancher, Bernard Asselain, Christine Levy-Gabrielle, Livia Lumbroso-Le Rouic, Manuel Rodrigues, Xavier Sastre, Sophie Piperno-Neumann, Jerome Couturier; Choroidal Melanoma pronostication: a robust and costless classifying system. Invest. Ophthalmol. Vis. Sci. 2013;54(15):4220. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
This paper investigate the capability to predict high risk patients using genetic analysis of choroidal melanoma on enucleated eyes, local resection or FNAB material.
Patients with choroidal melanoma were prospectively enrolled. Demographic data were collected including patient age, gender, tumor diameter, and tumor thickness, ciliary body invasion or extrascleral extension. Tumor samples were taken on enucleated globe, at the time of clips or plaque positioning with FNAB, or after endoresection. Array-CGH techniques were performed on two different platforms. Home-made BAC-arrays (CIT / INSERM U830, Institut Curie, Paris) and NimbleGen 4x72K arrays (Roche NimbleGen, Madison, WI). Genomic DNA was extracted with Phase Lock Gel Light and analyzed versus normal reference DNA. Gains are defined as a log2ratio ≥0.58 and deletions ≤-1. Patients were managed for their primary tumor then monitored for metastasis.
3 groups were classified by Array-CGH. “Low risk” patients were defined by disomy3, “intermediate risk” was defined by monosomy 3 alone, and “high risk” patients were defined by monosomy 3 and 8q gain. The media follow-up was 25 months [range; 22-29 months]. At the end of the study, 128 patients developed metastasis (33.7%) and 96 patients died. Univariate cox analysis showed that metastatic free interval at 2 years was 92% for low risk patients, 67% for intermediate risk and 33% for high risk patients. There is a statistical association between genomic high risk results and other known prognostic factors especially for mixt/epitheliod cell type. Univariate Cox proportional hazard analysis factors associated in our study with metastasis included basal tumor diameter p=0.001, tumor thickness p=0.03, mixt/epithelioid cell type p=0.006, genetic data p<0.0001. High risk profile was more strongly associated with metastasis than the other prognostic factor p<0.001. Multivariate cox modeling analysis showed that genetic result was the only variable that contributed independently to prognosis with p=0.001 IC 95%[1.6-7.8] for chromosome 3 monosomy alone and p<0.0001 IC 95% [5.2-20.4] for chromosome 3 monosomy and 8q gain.
Array-CGH is a robust and not too expensive method for choroidal melanoma prognostication. This method is to date currently used in France. High risk patients are included in an adjuvant therapy trial.
This PDF is available to Subscribers Only