Five-Year Survival with Tebentafusp in Metastatic Uveal Melanoma
Ann Oncol. 2026 May 19:S0923-7534(26)00879-3. doi: 10.1016/j.annonc.2026.05.695. Online ahead of print.
ABSTRACT
BACKGROUND: Tebentafusp demonstrated an overall survival benefit compared to investigator’s choice in a phase 3 trial in HLA-A*02:01-positive adult patients with unresectable or metastatic uveal melanoma (mUM) and is now the first-line standard of care for this population. Here we report the final 5-year analysis of overall survival.
PATIENTS AND METHODS: In this international, open-label, phase 3 trial, previously untreated HLA-A*02:01-positive patients with mUM were randomized 2:1 to receive tebentafusp or investigator’s choice of pembrolizumab, ipilimumab or dacarbazine (control group), stratified by lactate dehydrogenase. The primary endpoint was overall survival; ctDNA reduction was an exploratory endpoint.
CLINICALTRIALS: gov identifier: NCT03070392.
RESULTS: After a minimum of 5 years of follow-up, median overall survival was 21.6 months in the tebentafusp group and 16.9 months in the control group (stratified hazard ratio 0.67; 95% CI 0.54-0.85). Overall survival at 5 years was 16% versus 8%, respectively. Tebentafusp improved survival even in poor-prognosis groups, such as patients with baseline tumors ≥10 cm or those whose best RECISTv1.1 response was progressive disease, including cases where target tumor growth exceeded 20%. After adjusting for covariates, a post hoc analysis showed that patients who were treated with tebentafusp beyond radiographic progression had longer overall survival than those who discontinued treatment. In the tebentafusp group, longer overall survival was associated with undetectable ctDNA at baseline or ctDNA reductions ≥50% by week 9. Deep ctDNA reductions occurred regardless of baseline tumor burden or radiographic response.
CONCLUSIONS: In the longest survival follow-up of a randomized trial in metastatic uveal melanoma, tebentafusp continues to provide long-term survival benefit in previously untreated HLA-A*02:01-positive patients.
PMID:42162665 | DOI:10.1016/j.annonc.2026.05.695
