Real world experience of molecular landscape of uveal melanomas: Implications for personalized medicine strategies.

e21572 Background: In the realm of ocular oncology, Uveal Melanoma (UM) stands out as a rare and formidable malignancy, necessitating a deeper understanding of its molecular landscape to enhance treatment strategies. This abstract delves into the real-world experiences of unraveling the molecular intricacies of UM and explores the implications for personalized medicine strategies. Methods: In this study, we conducted a comprehensive genomic analysis of tissue samples from 7 UM cases to identify key genetic alterations and molecular features, elucidating the heterogeneity of this ocular malignancy. Results: Consistent with other studies, 57% (4/7) of our cohort harbored a driver mutation in either the GNAQ or GNA11 genes. GNAQ Q209 codon mutations (Q209L or Q209P), were the most frequent variants identified. GNAQ/GNA11 mutations are known to activate the MAPK/ERK signaling pathway, contributing to uncontrolled cell proliferation; its presence qualifies for entry into clinical trial testing DYP688 in metastatic UM. Concurrent mutations were identified in EIF1AX (n = 1), SF3B1 genes (n = 1) and BAP1 (n = 1), highlighting the molecular diversity. EIF1AX, SF3B1, and BAP1 inactivating mutations indicate varying risks for distant metastasis, categorized as low, medium, and high risk, respectively. Notably, homologous recombination repair (HRR) pathway gene mutations were identified in 28% (2/7) of cases. Additionally, SF3B1 mutations have been proposed to induce a BRCA-like phenotype, resulting in synthetic lethality to PARP inhibitors (PARPi). These findings provide opportunities to explore additional targeted therapy (off label) options with PARPi. In a case where gene expression profiling (GEP) was performed, an upregulation in PRAME gene expression (PRAME+) was observed. This finding potentially indicates an increased metastatic risk and a shorter time to metastasis, given that PRAME+ status has been established as an independent risk modifier for metastasis, regardless of the GEP-derived UM class (Class 1 or 2). TMB ranged from 2 to 21 muts/mb, with 5 out of 7 specimens showing low TMB ( < 10 mutations/Mb) and two sample with high TMB (15 and 21muts/mb). This case with TMB of 21 muts/mb showed Deficient Mismatch Repair (dMMR) status. All samples evaluated for PD-L1 expression (n = 4) in our cohort showed < 1% PD-L1 expression in tumor cells, including the dMMR tumor. Conclusions: These findings underscore the need for a comprehensive genomic approach in guiding therapeutic decisions, understanding prognostic implications and advancing personalized medicine strategies for UM patients. Our study provides an overview of the molecular landscape of UMs with delineation of key molecular features, thereby enhancing our understanding of UM biology and setting the stage for the development of targeted therapies tailored to the unique genomic profile of each patient.