Relationship between clinical features, GEP class, and PRAME expression in uveal melanoma

• Published in: Graefe's archive for clinical and experimental ophthalmology 2019-07, Vol.257 (7), p.1541-1545
• Main Authors: Schefler, Amy C., Koca, Emre, Bernicker, Eric H., Correa, Zelia M.
• Format: Article
• Language: English
• Subjects: Adult; Aged; Antigen (tumor-associated); Antigens; Antigens, Neoplasm - biosynthesis; Antigens, Neoplasm - genetics; Biomarkers, Tumor - biosynthesis; Biomarkers, Tumor - genetics; Biopsy, Fine-Needle; Female; Follow-Up Studies; Gene expression; Gene Expression Profiling - methods; Gene Expression Regulation, Neoplastic; Humans; Male; Medical records; Medicine; Medicine & Public Health; Melanoma; Melanoma - diagnosis; Melanoma - genetics; Melanoma - metabolism; Metastases; Metastasis; Middle Aged; Neoplasm Staging; Oncology; Ophthalmology; Patients; Retrospective Studies; RNA, Neoplasm - genetics; Statistical analysis; Tumors; Ultrasonography; Uveal Neoplasms - diagnosis; Uveal Neoplasms - genetics; Uveal Neoplasms - metabolism
• ISSN: 0721-832X; 1435-702X
• DOI: 10.1007/s00417-019-04335-w

Background Metastatic risk for uveal melanoma (UM) patients can be characterized by gene expression profiling (GEP) (Castle Biosciences, Friendswood, TX). Class 1A tumors carry low metastatic risk; class 1B tumors have intermediate risk; and class 2 tumors have high risk. Preferentially expressed antigen in melanoma (PRAME) is a tumor-associated antigen which is expressed in various neoplasms including UM. Recently, PRAME expression in uveal melanoma was first recognized to confer an additional metastatic risk beyond GEP status. Methods This was a retrospective, consecutive, multicenter chart review study. All patients diagnosed with UM at two major ocular oncology centers from August 2016 to February 2018 who underwent both GEP and PRAME mRNA expression testing were included. Patient age at diagnosis, gender, and tumor variables such as thickness, largest basal diameter (LBD), tumor volume, TNM stage, and GEP class and PRAME status were extracted from the medical records. Statistical analysis was performed to analyze the association of PRAME +/− status with all clinical and molecular variables. Results One hundred forty-eight UM patients were identified. TNM was stage I in 51 (34.5%), stage IIA in 33 (22.3%), stage IIB in 34 (23%), stage IIIA in 20 (13.5%), and stage IIIB in 10 (6.8%) patients. Fifty-five patients (37%) were PRAME-positive, a significant fraction. There was no association between higher TNM stage and positive PRAME status ( p  = 0.129). PRAME expression was found to be independent of gender, patient age, and tumor thickness. PRAME expression was statistically associated with LBD and tumor volume. Higher GEP class was associated with higher TNM staging ( p