Bispecific T-Cell Engagers, Cell Therapies, and Other Non-Checkpoint Immunotherapies for Metastatic Uveal Melanoma: A Narrative Review

Metastatic uveal melanoma (MUM) remains largely refractory to immune-checkpoint inhibition, so recent research has turned to bispecific T-cell engagers (BTCEs), adoptive-cell therapies (ACTs), and oncolytic viruses (OVs). To summarize the available clinical evidence, we performed a structured literature search across PubMed, Scopus, and Europe PMC for primary studies published between 1 January 2010 and 31 May 2025 that enrolled at least three adults with MUM, treated with one of these modalities, and that reported efficacy or grade-3+ safety outcomes; two reviewers independently performed screening, data extraction, and risk-of-bias assessment, and because of notable heterogeneity, we synthesized the findings narratively. Twenty-two studies met the criteria-thirteen phase I-III trials, eight observational cohorts, and one case series-covering fifteen BTCE cohorts, four ACT cohorts, and three OV cohorts. Tebentafusp, the dominant BTCE evaluated in roughly 1150 HLA-A*02:01-positive patients, extended median overall survival from 16.0 to 21.7 months (hazard ratio 0.51, with three-year follow-up HR 0.68) in its pivotal phase-III trial despite objective response rates of only 5-12%, with early skin rash and week-12 circulating-tumor-DNA clearance emerging as consistent markers of benefit. Tumor-infiltrating lymphocyte therapy, administered to about thirty patients, produced objective responses in 11-35% and occasional durable complete remissions, although median progression-free survival remained 2-6 months and severe cytopenias were universal. Three early-phase OV studies, totaling twenty-nine patients, yielded no radiographic responses but showed tumor-specific T-cell expansion and transient disease stabilization. Safety profiles reflected the mechanism of action: tebentafusp most often caused rash, pyrexia, and usually manageable cytokine-release syndrome with grade-3+ events in 40-70% yet discontinuation in roughly 2%; TIL therapy toxicity was driven by lymphodepleting chemotherapy and high-dose interleukin-2 with one treatment-related death; and OVs were generally well tolerated with no more than 20% grade-3 events.