Purpose
To assess the concordance between results of DecisionDx-UM uveal melanoma specific Gene Expression Profiling (GEP) (Castle Biosciences, Phoenix, AZ) and Fluorescent in Situ Hybridization (FISH) for chromosome 3 analysis in uveal melanomas undergoing intraoperative fine needle aspiration biopsy (FNAB) for metastatic prognostication during brachytherapy.
Methods
Consecutive patients diagnosed with uveal melanoma who underwent intraoperative transscleral or transvitreal fine needle aspiration biopsy with 30-gauge needle prior to placement of iodine-125 radioactive plaque between November 2012 and January 2014, were retrospectively reviewed. Patient demographics and tumor characteristics were obtained. The results of biopsy specimens, analyzed by cytopathology, FISH for monosomy 3 and 6p gain, and for GEP analysis with the DecisionDx-UM assay, were reviewed.
Results
A total of 99 intraocular tumors underwent brachytherapy with intraoperative biopsy, including 90 choroidal melanomas, 6 iris melanomas, and 3 uveal metastatic lesions. FISH and GEP results were both available in 44 (44%) patients all with iris or choroidal melanoma. Of these 44, FISH and GEP results were discordant in 7 tumors (15.9%). Six tumors were classified as “Class 1” (four 1A, two 1B), or low risk-for metastasis designation, by GEP but monosomy 3 by FISH; and one tumor was found to be “Class 2” by GEP and disomy 3 by FISH analysis. There was no significant difference with regard to tumor height (p=0.94), patient age (p=0.95), or ciliary body involvement (p=0.97) between discordant and concordant cases. No patients with discordant tumors had confirmed metastatic disease with limited follow up (mean 8.4 months, range 1- 15.5 months).
Conclusions
Discordance between GEP and Chromosome 3 status by FISH occurred in our series at a rate of 15.9%, which is similar to previously published reports. No patient or tumor characteristics were identified as risk factors for discordance. The metastatic prognosis of uveal melanoma patients with discordant results is unclear from our study due to limited follow up.