Comprehensive clinical imaging, histopathological analysis and liquid biopsy-based surveillance of human uveal melanoma in a prolonged rabbit xenograft model.

Uveal melanoma is the most common intraocular tumor in adults. Our group has previously developed a human uveal melanoma animal model; however, adverse effects caused by the immunosuppressive agent, cyclosporine A, prevented animals from surviving more than 12 weeks. In this study, we tested multiple cyclosporine A doses over an extended disease course up to 20 weeks, providing complete clinical imaging of intraocular tumors, histopathological analysis and liquid biopsy biomarker analysis. Twenty albino rabbits were divided into four groups with different daily cyclosporine A schedules (0-10 mg/kg) and inoculated with human uveal melanoma cell lines, 92.1 or MP41, into the suprachoroidal space. Rabbits were monitored with fundoscopy, ultrasound and optical coherence tomography. Intraocular tumors (macroscopic or microscopic) were detected in all study animals. Tumor size and growth were correlated to cyclosporine A dose, with tumors regressing when cyclosporine A was arrested. All tumors expressed HMB-45 and MelanA; however, tumor size, pigmentation and cell morphology differed in 92.1 vs. MP41 tumors. Finally, across all groups, circulating tumor DNA from plasma and aqueous humor was detected earlier than tumor detection by imaging and correlated to tumor growth. In conclusion, using three clinically relevant imaging modalities (fundoscopy, ultrasonography and optical coherence tomography) and liquid biopsy, we were successfully able to monitor tumor progression in our rabbit xenograft model of human uveal melanoma.