Abstract Background: Metastatic uveal melanoma (mUM) has a very poor prognosis with a historical survival rate of 10% at 5 yrs. Tebentafusp, a first in class ImmTAC bispecific (gp100 x CD3), is approved for adult HLA-A*02: 01+ patients with unresectable or mUM based on a phase 3 study demonstrating improved overall survival (OS) compared to investigator’s choice (IC) (HR 0. 51; IMCgp100-202; NCT03070392). This benefit was maintained at the 3-yr follow-up (HR 0. 68), with a 3-yr OS rate of 27% for tebentafusp and 18% for IC. Molecular response assessed by ctDNA reduction was a better indicator of OS benefit than traditional RECISTv1. 1 measurements. Here we report updated analyses of OS after a minimum follow-up of 5 yrs. Methods: In this randomized, open-label, Phase 3 trial, first line HLA-A*02: 01+ mUM patients were randomized 2: 1 to receive tebentafusp or IC of single-agent pembrolizumab, ipilimumab or dacarbazine, stratified by lactate dehydrogenase. Primary endpoint was OS. ctDNA reduction was an exploratory endpoint. OS was estimated using Kaplan-Meier methods and treatment effects compared using Cox proportional hazards model. A Cox model, adjusted for baseline and time-varying covariates at progression, compared post-progression survival in tebentafusp patients with versus without treatment beyond radiographic progression (TBP). Results: 378 patients were randomized to tebentafusp (252) or IC (126; 82% pembrolizumab). With extended follow-up, the OS continued to favor tebentafusp, with a stratified hazard ratio of 0. 67 (95% CI, 0. 54-0. 85). The 5-year OS rate for tebentafusp was 16% (95% CI, 11-21) vs 8% (95% CI, 4-14) for IC. The OS benefit was evident even in patients with known poor prognostic factors, including those with large tumors ≥ 10 cm. OS benefit was also seen in patients who did not have radiographic response including those with best response of progressive disease (PD) or those with a best change of tumor growth (20%). Notably, in the tebentafusp arm, TBP was associated with better OS compared to no TBP, even after adjusting for covariates (HR 0. 61; 95% CI, 0. 44-0. 83). In tebentafusp-treated patients, longer OS was associated with undetectable ctDNA at baseline or ctDNA reductions ≥50% by week 9. Among 21 ctDNA-evaluable patients who survived ≥ 5 years, 15 had undetectable baseline ctDNA and the remaining 6 had ctDNA clearance. Deep reductions in ctDNA were seen in patients regardless of baseline tumor burden and across all RECIST categories. Conclusions: Tebentafusp demonstrates durable, long-term OS benefit in first line HLA-A*02: 01+ patients, which at 5 years is the longest OS follow-up in a randomized trial in mUM. OS benefit is evident in those with poor prognostic factors and remains independent of radiographic response, with ctDNA levels proving to be a better indicator of activity. This is the first report of long-term OS benefit in a solid tumor treated with an ImmTAC therapy. Citation Format: Paul Nathan, Sophie Piperno-Neumann, Jessica C. Hassel, Marcus O. Butler, Max Schlaak, Ryan J. Sullivan, Reinhard Dummer, John M. Kirkwood, Joseph J. Sacco, Alexander N. Shoushtari, Josep M. Piulats, April KS Salama, Marlana Orloff, Anthony M. Joshua, Sebastian Ochsenreither, Lauris Gastaud, Brendan Curti, Lev Demidov, Mohammed Milhem, Bartosz Chmielowski, Kari Kendra, Paolo Antonio Ascierto, Eric H. Bernicker, Richard D. Carvajal, Omid Hamid, Laura Collins, Sarah Lockwood, Jaymin M. Patel, Jean-Francois Baurain, Piotr Rutkowski. Five-year survival with tebentafusp in previously untreated metastatic uveal melanoma in a phase 3 trial abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts) ; 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86 (8Suppl): Abstract nr CT029.
Nathan et al. (Fri,) studied this question.