Author: Claude (AI), from PubMed abstracts · Reviewed by: Dr Mina Abdelmseih · Reviewed: 2026-11-15 · Review due: 2027-11-15

Educational content only. Not personalised medical advice. Clinical decisions remain the responsibility of the treating clinician.

Metastatic uveal melanoma and tebentafusp

Metastatic uveal melanoma develops in approximately 20-30% of patients despite effective control of the primary tumour, mainly in the liver, and carries a poor prognosis [S8]. Tebentafusp is the first systemic drug shown to improve overall survival, but it can be used only in HLA-A*02:01-positive patients [S8] [S11].

Key points - In a phase 3 RCT, tebentafusp gave median overall survival of 21.6 versus 16.9 months with investigator's choice, and 5-year survival of 16% versus 8% [S11]. - HLA-A*02:01 typing is needed before tebentafusp; options are limited for HLA-negative patients [S8]. - Liver-directed therapies (isolated or percutaneous hepatic perfusion with melphalan) improve response and progression-free survival over best alternative care; a clear overall survival benefit has not been shown [S3] [S7]. - Immune checkpoint inhibitors have low response rates in uveal melanoma (pooled 5.6%), and dual therapy does better than single agents [S6]. - At current US pricing, tebentafusp was not cost-effective in one model [S4].

Evidence last checked: 2026-10-17 (PubMed abstracts).

Red flags

• Liver involvement. About half of patients with metastatic disease present with isolated liver metastases, with median survival of 6-12 months [S3].
• Short survival without effective treatment. Across 2494 patients in a pooled analysis, median overall survival was 1.07 years, and most apparent differences between treatments were attributed to bias [S1].
• Tebentafusp toxicity. Cytokine-mediated and skin events are common; in the phase 3 trial rash occurred in 83%, pyrexia in 76% and pruritus in 69%, mostly in the first three or four doses [S2].

Recognise it

• Metastases occur mainly in the liver [S8] [S9].
• A significant proportion of patients treated definitively for the primary tumour eventually develop metastatic disease, which is why systemic surveillance is recommended [S9].

Confirm it

• French consensus guidelines recommend liver biopsy and imaging, circulating tumour DNA (ctDNA) analysis and high-definition HLA typing for HLA-A*02:01 [S8].
• Surveillance studies mostly recommend imaging at least every 6 months and MRI over ultrasound for higher-risk patients, but the evidence is of variable quality [S9].

Manage it

Before acting on this section, check the national and local guidelines, product information and regulatory approvals that apply where you practise. See the medical disclaimer.

Resource tiers: metastatic uveal melanoma is managed by medical oncology and specialist liver services, so Tier A / Tier B / Tier C advice is limited to recognising the risk and referring. Eye clinics at every tier should ensure patients treated for uveal melanoma are linked to systemic surveillance [S9].

Specialist oncology management

1. Tebentafusp (HLA-A*02:01-positive). In the phase 3 trial (378 previously untreated patients), 1-year overall survival was 73% versus 59% with investigator's choice of pembrolizumab, ipilimumab or dacarbazine (HR 0.51) [S2]. At 5 years median overall survival was 21.6 versus 16.9 months (HR 0.67) [S11]. Benefit was seen even in patients whose best response was progressive disease, and undetectable or falling ctDNA was associated with longer survival [S11]. A meta-analysis of 997 patients in 12 cohorts gave median overall survival of 21.2 months and median progression-free survival of 3.9 months; the discontinuation rate was 2% [S10]. A propensity-weighted comparison also favoured tebentafusp over nivolumab plus ipilimumab (HR 0.52) [S5]. Dosing is not covered here; follow the product information and local protocols.
2. Immune checkpoint inhibitors. Pooled data from 1414 patients: objective response 5.6%, median overall survival 11.2 months; dual therapy achieved better response (12.4% vs 3.4%) and overall survival (16.3 vs 9.8 months) than single agents [S6]. For HLA-A*02:01-negative patients with extensive disease, clinical trials, percutaneous hepatic perfusion or checkpoint inhibitors alone or combined can be proposed [S8].
3. Isolated hepatic perfusion (IHP). In the SCANDIUM phase 3 RCT (87 patients with isolated liver metastases), IHP gave overall response of 40% versus 4.5% and median progression-free survival of 7.4 versus 3.3 months; there was one treatment-related death [S3].
4. Percutaneous melphalan/hepatic delivery system. In a randomised study converted to single-arm because of slow enrolment, exploratory analyses favoured melphalan/HDS over best alternative care (median overall survival 18.5 vs 14.5 months), with serious adverse events in 51.2% versus 21.9%, mostly cytopenias [S7]. Dose as reported in [S7]: 3 mg/kg ideal body weight every 6 to 8 weeks for up to 6 cycles.
5. Combined approaches. A network meta-analysis of 16 RCTs ranked liver-directed therapy combined with checkpoint inhibitors highest for overall and progression-free survival, with tebentafusp second for overall survival but worst for progression-free survival [S12]. This conflicts with the trial-based view of tebentafusp as standard first-line therapy [S8] [S11] and rests on indirect comparisons.
6. Cost. A US model estimated an incremental cost-effectiveness ratio of $953 230 per QALY for tebentafusp, far above the $200 000 threshold [S4].

Refer when

Pitfalls

• Starting tebentafusp without HLA-A*02:01 typing [S8].
• Stopping tebentafusp only because of radiological progression: patients treated beyond progression lived longer in a post hoc analysis [S11].
• Reading higher response rates after liver-directed therapy as proof of longer survival [S7] [S12].
• Using cutaneous melanoma results for uveal melanoma: response to checkpoint inhibitors is much lower [S6].

Evidence notes

• The tebentafusp phase 3 trial was open-label and funded by Immunocore [S2].
• The control arm in the tebentafusp trial was mostly single-agent therapy; comparison with combined checkpoint inhibitors is indirect [S5].
• The melphalan/HDS efficacy results are exploratory only [S7].
• Not covered by these sources: tebentafusp dosing and monitoring schedule, MEK inhibitors and other targeted agents beyond the network meta-analysis, adjuvant therapy, access to these drugs in low- and middle-income countries, and evidence-based referral timings.

Related articles

• Uveal melanoma: diagnosis and primary treatment
• Ocular adverse events of immune checkpoint inhibitors and targeted cancer drugs

References

See the numbered source list below. Each marker [S#] links to its PubMed and DOI record.