Castle Biosciences announces new data on DecisionDx-UM

Castle Biosciences (CSTL) announced new data from the first independent validation of the recently published Collaborative Ocular Oncology Group Study No. 2 by Harbour et al.1 The data, from a real-world cohort of 1,297 patients with uveal melanoma, was presented at the Association for Research in Vision and Ophthalmology, ARVO, 2025 Annual Meeting in Salt Lake City. The findings provide further support for adding Preferentially Expressed Antigen in Melanoma, PRAME, gene expression information to the DecisionDx-UM test result to further refine metastatic risk prediction for patients with UM, which is a rare but aggressive eye cancer. Summary: In a large, real-world, population-based cohort of patients with UM, this study aimed to validate the findings from the prospective COOG2.1 study, which found that adding the reported expression of the PRAME gene to the DecisionDx-UM class result can further refine risk by subdividing Class 1 and Class 2 tumors based on PRAME positive (+) versus negative (-) status. Consistent with the COOG2.1 study, the findings shared at ARVO demonstrate that co-reporting of DecisionDx-UM class and PRAME status provides additional insights into a patient’s likelihood of metastasis to better inform treatment pathway decisions.