Skip to content
Bending the Curve Trial by Trial
?

Skin

Melanoma

How melanoma care moved from surgery and dacarbazine to checkpoint inhibitors, BRAF/MEK therapy and neoadjuvant immunotherapy.

5-year relative survival · United States, SEER 8 registries

82.34%→97.52%

1975 (observed value; first year in the table) to 2018 (observed value; last observed year in the table). Read these with the caveats under “Then and now”.Source

35 milestones, by decade

  • In US registry data on the SEER Cancer Stat Facts page, the age-adjusted melanoma death rate was 2.68 per 100,000 in 2013 and 2.00 in 2024, while the age-adjusted rate of new cases in the SEER 8 registries was 25.53 per 100,000 in 2013 and 28.26 in 2023. Observed five-year relative survival in the SEER 8 series was 82.34% for 1975 and 97.52% for 2018 (the page does not say whether the year is the year of diagnosis). The page does not say how much of the change reflects treatment, detection or case mix.
  • A review of melanoma history credits the surgeon John Hunter with the first surgical removal of a melanoma in the Western medical literature: in 1787 he removed a recurrent tumour from the jaw of a 35-year-old man, a case the review says was retold by Everard Home in 1805. The review notes that it was not clear Hunter knew what he was dealing with; the preserved tumour was later diagnosed as melanoma, in 1968, and is still held at the Hunterian Museum in London.
  • The US FDA granted accelerated approval to lifileucel (Amtagvi), a tumour-derived T-cell therapy, on 16 February 2024, on the basis of an objective response rate of 31.5% among 73 patients treated at the recommended dose in a single multicohort study. The company withdrew its application to the European Medicines Agency on 22 July 2025; the EMA page states that the Agency's provisional view was that the benefits did not outweigh the risks, citing a response rate in the main study too low to determine meaningful benefit, safety concerns, insufficient data to support the proposed dose, and the applicant not having provided all the necessary good manufacturing practice (GMP) documentation.

Evidence current to October 2026 · Checked against primary sources to the extent possible

A guided tour of 12 milestones. You can pause or step through it yourself.

Filter milestonesshowing all 35
Treatment type
Setting

Chemotherapy

  1. 1969 · Understanding · Understanding⚑ needs reviewClark levels and Breslow thickness: microstaging of melanomaTwo systems linked how deeply a melanoma had grown to a person's outlook. Clark graded invasion by the layers of the skin; Breslow related outcome to the tumour's measured thickness. Thickness remains the central measurement in modern melanoma staging, while the level of invasion has been dropped from the thin-melanoma criteria.Clark (Cancer Res, indexed March 1969) is a title-only record; a 2014 review dates Clark's level-of-invasion system (levels I-V) to 1966 and Breslow's thickness observation (Ann Surg, November 1970) to 1970. Neither paper has a PubMed abstract; no case numbers or outcomes are stated here. AJCC 2009: mitotic rate replaced level of invasion for T1b. AJCC eighth edition: thickness to the nearest 0.1 mm; T1a under 0.8 mm without ulceration; T1b 0.8-1.0 mm.
  2. c. 1975 · Treatment · Chemotherapy⚑ needs reviewDacarbazine: the reference single agent in metastatic melanomaDacarbazine, a chemotherapy drug, had its US application approved in 1975, and a later review reports it was approved for metastatic melanoma in the 1970s. For decades it was the reference treatment. Randomised trials of a four-drug regimen and of temozolomide did not show a survival advantage over it. It later served as the control arm of several trials.Dacarbazine: FDA NDA 017575 original approval 27 May 1975 (Drugs@FDA; indication not listed there). Dartmouth regimen vs dacarbazine (n=240): no survival difference; response 18.5% vs 10.2% (P=0.09). Temozolomide vs dacarbazine (n=305): median OS 7.7 vs 6.4 months; median PFS 1.9 vs 1.5 months (P=0.012).

Molecular origins

  1. 1988 · De-escalation · Surgery⚑ needs reviewNarrower excision margins: randomised trials from 1988 to 2016Surgeons once took wide margins of skin around a melanoma. Randomised trials in thinner melanomas found similar outcomes with narrower margins, and in one trial fewer skin grafts. A largely UK trial in thicker melanomas found more local and nearby-node recurrence with a 1 cm margin than with 3 cm, and a later analysis reported more melanoma deaths.Melanoma up to 2 mm: 1 cm vs 3 cm margins (305 vs 307 patients) gave similar disease-free and overall survival; 2 cm vs 5 cm (989 patients, 0.8-2.0 mm) overall survival HR 0.96 (95% CI 0.75-1.24). 1-4 mm: 2 cm vs 4 cm, 10-year survival 70% vs 77% (not significant). Melanoma of 2 mm or more, 1 cm vs 3 cm (900 patients): locoregional recurrence HR 1.26 (1.00-1.59); melanoma deaths in the long-term analysis HR 1.24 (1.01-1.53).
  2. 1988 · Treatment · Cell therapy & transplantTIL plus IL-2 in metastatic melanoma: NCI preliminary report (1988)In 1988 a US team reported treating 20 people with metastatic melanoma with immune cells taken from their own tumours, grown in the laboratory and returned with the signalling protein interleukin-2. Some tumours shrank in about half of the people, but the authors said it was too early to know whether survival improved.Uncontrolled series (NEJM 1988): 20 patients with metastatic melanoma given autologous TIL and IL-2 after cyclophosphamide. Objective regression in 9 of 15 patients (60%) not previously treated with IL-2 and 2 of 5 (40%) after IL-2 failure, lasting 2 to more than 13 months. 2011 report of 93 patients: ORR 49%, 52%, 72% across three lymphodepletion regimens; actuarial OS 36% at 3 and 29% at 5 years.
  3. 1992 · Treatment · SurgerySentinel-node biopsy: Morton 1992 to MSLT-I (2006, 2014)In 1992 surgeons described a dye technique to find the first lymph node that drains a melanoma and test it for spread. A large randomised trial later compared this approach with watching the nodes. Disease-free survival was higher with biopsy, but survival from melanoma overall was not significantly different.Morton 1992: sentinel node identified in 194 of 237 lymphatic basins; false-negative rate under 1%. MSLT-I (first report, 1269 intermediate-thickness patients): 5-year disease-free survival 78.3% with sentinel-node biopsy vs 73.1% with observation (HR for recurrence 0.74, 95% CI 0.59-0.93, P=0.009); 5-year melanoma-specific survival 87.1% vs 86.6%. Final report (2001 patients): 10-year disease-free survival 71.3% vs 64.7% (intermediate thickness); no significant overall melanoma-specific survival difference.
  4. 1994 · Treatment · ImmunotherapyHigh-dose IL-2: durable responses in a minority; FDA indication 1998High-dose interleukin 2, an immune-stimulating protein, was given to people with metastatic melanoma in single-institution and multi-trial series. A small share had their disease disappear, and most of those who did stayed in remission. Treatment was demanding and caused deaths early on. The FDA added a melanoma indication in January 1998.NCI Surgery Branch consecutive series (JAMA 1994): 283 patients with metastatic melanoma or renal cell cancer, IL-2 720,000 IU/kg every 8 hours; melanoma: complete regression in 9 patients (7%), partial in 14 (10%); 15 of 19 complete regressions (both cancers) remained in remission 7 to 91 months. Analysis of 270 patients (JCO 1999): response rate 16% (95% CI 12% to 21%), 17 complete responses (6%); 6 deaths (2%).
  5. 1996 · Treatment · Immunotherapy⚑ needs reviewAdjuvant interferon: E1684 (1996) to pegylated interferon (2011)In 1996 the ECOG E1684 trial reported that interferon alfa-2b at the maximum tolerated dose after surgery prolonged relapse-free and overall survival compared with observation. A later trial of pegylated interferon in stage III melanoma reported longer relapse-free survival but no overall survival difference, and about a third of patients stopped because of side effects.E1684 (N=287): interferon alfa-2b vs observation after resection of T4 or N1 melanoma; median follow-up 6.9 years; RFS P=0.0023 and OS P=0.0237 (one-sided); median OS 3.8 vs 2.8 years. EORTC 18991 (phase 3, N=1,256): pegylated interferon alfa-2b vs observation, 4-year RFS 45.6% vs 38.9% (HR 0.82; 95% CI 0.71-0.96); at 7.6 years HR 0.87 (95% CI 0.76-1.00; P=0.055) and no OS difference (P=0.57).

Targeted therapy

  1. 2002 · Understanding · UnderstandingBRAF mutations in melanoma (2002) and the TCGA melanoma atlas (2015)In 2002 a genome-wide search found changes in the BRAF gene in about two thirds of the melanomas it examined, most of them one specific substitution that made the protein more active. In 2015 a large tumour-sequencing project sorted 333 melanomas into four genomic groups, one of them defined by mutant BRAF.Davies 2002 (Nature): somatic BRAF missense mutations in 66% of malignant melanomas and at lower frequency in other cancers; all in the kinase domain, with V599E (V600E in later literature) accounting for 80%; elevated kinase activity, transforming in NIH3T3 cells. TCGA 2015 (Cell): 333 primary and/or metastatic melanomas from 331 patients; four subtypes (mutant BRAF, mutant RAS, mutant NF1, Triple-WT); no significant outcome correlation with genomic class.
  2. 2009 · Understanding · UnderstandingAJCC melanoma staging: 2009 version and 2017 eighth editionMelanoma staging groups people by tumour thickness, ulceration, spread to nodes and spread to other organs, to estimate outlook and guide care. The 2009 revision used a database of tens of thousands of patients; the eighth edition added a larger database and changed the thin-melanoma categories, the stage III subgroups and the brain-metastasis category.2009 version (Balch, JCO): multivariate analysis of 30,946 patients with stages I-III and 7,972 with stage IV; thickness, mitotic rate and ulceration dominant in localized disease; mitotic rate replaced level of invasion for T1b; any microscopic nodal metastasis is stage III. Eighth edition (Gershenwald 2017): over 46,000 patients with stages I-III; T1a under 0.8 mm without ulceration, T1b 0.8-1.0 mm; stage III has four subgroups; new M1d for CNS metastases.
  3. 2010 · Treatment · ImmunotherapyIpilimumab: anti-CTLA-4 survival trials and FDA approval (2010-2011)Ipilimumab, an antibody that releases a brake on T cells, was tested in two large trials. In previously treated metastatic melanoma it was linked to longer survival than a vaccine control, and in untreated disease it was linked to longer survival when added to dacarbazine. The FDA approved it in March 2011. Serious immune-related side effects occurred.MDX010-20 (n=676, previously treated, HLA-A*0201-positive): median OS 10.0 vs 6.4 months for ipilimumab 3 mg/kg plus gp100 vs gp100 alone (HR 0.68; P<0.001); ipilimumab alone 10.1 months (HR 0.66; P=0.003). CA184-024 (n=502, untreated): ipilimumab 10 mg/kg plus dacarbazine vs dacarbazine plus placebo, median OS 11.2 vs 9.1 months (HR 0.72; P<0.001); grade 3 or 4 adverse events 56.3% vs 27.5%. Pooled analysis of 1,861 patients: 3-year survival 22%.

Immunotherapy

  1. 2011 · Treatment · TargetedBRAF inhibition: vemurafenib (BRIM-3) and companions, 2011-2012In melanoma carrying a BRAF V600E mutation, vemurafenib was compared with dacarbazine chemotherapy. Survival at six months and the time before progression were better with vemurafenib, and the FDA approved it in August 2011. Dabrafenib and trametinib each improved progression-free survival over chemotherapy in 2012 reports and were approved as single agents in 2013.BRIM-3 (n=675, untreated BRAF V600E): vemurafenib 960 mg twice daily vs dacarbazine; 6-month OS 84% vs 64% (HR for death 0.37; 95% CI 0.26 to 0.55); median PFS 5.3 vs 1.6 months (HR 0.26); response 48% vs 5%. BREAK-3 (n=250): dabrafenib vs dacarbazine, median PFS 5.1 vs 2.7 months (HR 0.30). METRIC (n=322): trametinib vs chemotherapy, median PFS 4.8 vs 1.5 months (HR 0.45).
  2. 2012 · Treatment · ImmunotherapyAnti-PD-1 in melanoma: phase 1 results to FDA approvals (2012-2014)Early studies of antibodies against PD-1 reported responses in some people with advanced melanoma, with some lasting a year or more. Randomised trials in melanoma that had progressed after ipilimumab followed. The FDA granted accelerated approval to pembrolizumab in September 2014 and nivolumab in December 2014.Phase 1 nivolumab (n=296 treated, several tumour types): melanoma objective response 28% (26 of 94). Phase 1 lambrolizumab (135 melanoma patients treated): confirmed response 38% (44 of 117 by central review), 52% at 10 mg/kg every 2 weeks. KEYNOTE-002 (n=540, post-ipilimumab, phase 2): PFS HR 0.57 (2 mg/kg) and 0.50 (10 mg/kg) vs chemotherapy. CheckMate 037 (n=405): response 31.7% vs 10.6%, first interim analysis. FDA accelerated approvals 4 September and 22 December 2014.
  3. 2014 · Treatment · ImmunotherapyNivolumab monotherapy: CheckMate 066 and CheckMate 037 (2014-2015)Two trials tested nivolumab against chemotherapy. In untreated melanoma without a BRAF mutation, people on nivolumab were more likely to be alive at one year and had more responses than those on dacarbazine. After ipilimumab, nivolumab produced more responses than chemotherapy at an interim analysis. The two groups are different populations.CheckMate 066 (n=418, untreated, BRAF wild-type): nivolumab vs dacarbazine, 1-year OS 72.9% vs 42.1% (HR 0.42; 99.79% CI 0.25 to 0.73); median PFS 5.1 vs 2.2 months (HR 0.43); response 40.0% vs 13.9%. CheckMate 037 (n=405, after ipilimumab): response 31.7% vs 10.6% in the first interim analysis.
  4. 2014 · Treatment · TargetedDabrafenib plus trametinib: COMBI-d and COMBI-v (2014)Two large trials added a MEK inhibitor, trametinib, to a BRAF inhibitor in melanoma with a BRAF mutation. The combination delayed progression compared with dabrafenib alone and was linked to longer survival than dabrafenib alone or vemurafenib. The FDA had approved the combination in January 2014, before either trial appeared in a journal.COMBI-d (n=423, untreated BRAF V600E/K): dabrafenib plus trametinib vs dabrafenib plus placebo, median PFS 9.3 vs 8.8 months (HR 0.75; 95% CI 0.57 to 0.99); OS in the Lancet report 25.1 vs 18.7 months (HR 0.71; 95% CI 0.55 to 0.92). COMBI-v (n=704): vs vemurafenib, 12-month OS 72% vs 65% (HR 0.69; 95% CI 0.53 to 0.89; P=0.005). Pooled five-year OS 34%.
  5. 2015 · Treatment · ImmunotherapyAdjuvant ipilimumab 10 mg/kg: EORTC 18071 and FDA approval (2015)In a placebo-controlled phase 3 trial after surgery for stage III melanoma, ipilimumab at a high dose (10 mg/kg) delayed recurrence, and a later report found longer overall survival too. The cost was many more serious side effects, including immune-related deaths in about one in a hundred patients.Phase 3, double-blind; completely resected stage III melanoma (N=951): ipilimumab 10 mg/kg (n=475) vs placebo (n=476). Primary report: median RFS 26.1 vs 17.1 months (HR 0.75; 95% CI 0.64-0.90; P=0.0013). At median 5.3 years: 5-year OS 65.4% vs 54.4% (HR 0.72; 95.1% CI 0.58-0.88); grade 3-4 adverse events 54.1% vs 26.2%; 5 patients (1.1%) died of immune-related adverse events.
  6. 2015 · Treatment · ImmunotherapyNivolumab plus ipilimumab: CheckMate 069/067 and the 10-year follow-upIn untreated advanced melanoma, nivolumab combined with ipilimumab, and nivolumab alone, each gave better results than ipilimumab alone. After ten years of follow-up, survival remained longer with the combination and with nivolumab. Serious side effects were more frequent with the combination. The trial was not designed to compare the combination with nivolumab alone.CheckMate 067 (n=945, untreated, phase 3): median PFS 11.5 months nivolumab plus ipilimumab, 6.9 nivolumab, 2.9 ipilimumab (HR 0.42 and 0.57 vs ipilimumab; 99.5% CIs). Median OS 71.9, 36.9 and 19.9 months at 10 years (HR 0.53 and 0.63); 10-year OS 43%, 37% and 19%. Not designed to compare the combination with nivolumab alone. CheckMate 069 (n=142, phase 2): response 61% vs 11% in BRAF wild-type.
  7. 2015 · Treatment · ImmunotherapyKEYNOTE-006: pembrolizumab versus ipilimumab (2015, 10-year follow-up)In advanced melanoma, pembrolizumab given every two or three weeks was compared with ipilimumab. People on pembrolizumab went longer without progression and lived longer, with fewer severe side effects. Follow-up reports at five and ten years kept showing a survival difference, although the later reports combined the two pembrolizumab schedules.KEYNOTE-006 (n=834, phase 3, ipilimumab-naive): pembrolizumab 10 mg/kg every 2 or 3 weeks vs ipilimumab. 2015: 6-month PFS 47.3% and 46.4% vs 26.5% (HR 0.58); 12-month OS 74.1% and 68.4% vs 58.2%. Final OS: median 16.0 months with ipilimumab, not reached with pembrolizumab (HR 0.68 for each schedule). Pooled pembrolizumab at 10 years: median OS 32.7 vs 15.9 months (HR 0.71); 10-year OS 34.0% vs 23.6%.
  8. 2015 · Treatment · ImmunotherapyTalimogene laherparepvec (T-VEC): OPTiM trial and approval (2015)Talimogene laherparepvec is a modified herpes virus injected into melanoma lesions. In a randomised trial it produced lasting responses in about 16 in 100 people, against about 2 in 100 with the comparison treatment. Survival was not clearly longer, and regulators approved it for local treatment of lesions in the skin and lymph nodes.OPTiM, randomised open-label phase 3 (436 patients, stage IIIB to IV, unresected), intralesional T-VEC vs subcutaneous GM-CSF 2:1. Durable response rate 16.3% vs 2.1% (OR 8.9; P<.001). ORR 26.4% vs 5.7%. Median OS 23.3 vs 18.9 months (HR 0.79, 95% CI 0.62 to 1.00; P=.051). FDA and EMA approvals 2015 for local or regional lesions.
  9. 2016 · De-escalation · SurgeryCompletion node dissection vs observation: DeCOG-SLT and MSLT-IIWhen a sentinel-node biopsy finds melanoma in a lymph node, surgeons had routinely removed the remaining nodes in that area. Two randomised trials compared that operation with close observation, by ultrasound in the larger trial. Survival did not differ, and in the larger trial lymphedema (swelling) was much more common after surgery.MSLT-II (1939 randomised; 1934 evaluable ITT; 1755 per-protocol): completion dissection vs ultrasound observation after sentinel-node metastasis; per-protocol 3-year melanoma-specific survival 86% vs 86% (P=0.42, median follow-up 43 months); disease-free survival 68% vs 63%; regional nodal control 92% vs 77%; lymphedema 24.1% vs 6.3%. DeCOG-SLT (483 randomised, closed early, underpowered): 3-year distant metastasis-free survival 74.9% (dissection) vs 77.0% (observation).
  10. 2017 · Treatment · TargetedAdjuvant dabrafenib plus trametinib: COMBI-AD (2017)For people with stage III melanoma carrying a BRAF V600E or V600K mutation, a year of dabrafenib plus trametinib after surgery kept more people free of relapse than placebo. At five years about half were alive without relapse, against about a third on placebo. Overall survival was not analysed at five years.Phase 3, double-blind, placebo-controlled; resected stage III BRAF V600E/K melanoma (N=870): dabrafenib 150 mg twice daily plus trametinib 2 mg daily for 12 months. 3-year RFS 58% vs 39% (HR 0.47; 95% CI 0.39-0.58; P<0.001). At 5 years RFS 52% vs 36% (HR 0.51; 95% CI 0.42-0.61) and DMFS 65% vs 54% (HR 0.55; 95% CI 0.44-0.70). 3-year OS 86% vs 77% did not cross the interim boundary.
  11. 2017 · Treatment · Immunotherapy⚑ needs reviewAdjuvant nivolumab versus ipilimumab: CheckMate 238 (2017)After surgery for advanced melanoma, people given the immunotherapy nivolumab stayed free of recurrence more often than those given ipilimumab at a high dose, and had far fewer serious side effects. A nine-year report did not show a clear difference in overall survival between the two drugs.Phase 3, double-blind; resected stage IIIB-C or IV melanoma (N=906): nivolumab 3 mg/kg every 2 weeks vs ipilimumab 10 mg/kg, up to 1 year. 12-month RFS 70.5% vs 60.8% (HR 0.65; 97.56% CI 0.51-0.83; P<0.001); grade 3-4 treatment-related adverse events 14.4% vs 45.9%. At minimum 107 months: median RFS 61.1 vs 24.2 months (HR 0.76; 95% CI 0.63-0.90); 9-year OS 69% vs 65% (HR 0.88; 95.03% CI 0.69-1.11).
  12. 2018 · Treatment · ImmunotherapyAdjuvant pembrolizumab: EORTC 1325/KEYNOTE-054 (2018)After surgery for high-risk stage III melanoma, a year of pembrolizumab kept more people free of recurrence than placebo, and the benefit was still seen after a median follow-up of almost seven years. Serious side effects were more common with pembrolizumab. The reports summarised here give no overall survival result.Phase 3, double-blind; resected stage III melanoma (N=1,019): pembrolizumab 200 mg every 3 weeks (n=514) vs placebo (n=505), up to 18 doses. 12-month RFS 75.4% vs 61.0% (HR 0.57; 98.4% CI 0.43-0.74; P<0.001). Distant metastasis-free survival at 3.5 years 65.3% vs 49.4% (HR 0.60; 95% CI 0.49-0.73). At median 6.9 years, 7-year RFS 50% vs 36% (HR 0.63; 95% CI 0.53-0.74).
  13. 2018 · Treatment · ImmunotherapyBrain metastases: ipilimumab-nivolumab trials (ABC, CheckMate 204)Brain metastases had been excluded from many melanoma trials. Two phase 2 studies gave nivolumab with ipilimumab to people with untreated brain metastases and saw shrinkage or disease control in about half. In a small randomised comparison, the combination had a higher intracranial response rate than nivolumab alone, without a formal test. Side effects were frequent.ABC (randomised phase 2, n=79 enrolled): intracranial response 46% (nivolumab plus ipilimumab), 20% (nivolumab), 6% (non-randomised cohort C); at 7 years OS 48%, 26% and 13%. CheckMate 204 (single-arm, n=94): intracranial clinical benefit 57% (95% CI 47 to 68); grade 3 or 4 treatment-related adverse events 55%.
  14. 2018 · Treatment · Immunotherapy⚑ needs reviewNeoadjuvant checkpoint blockade: OpACIN, OpACIN-neo and PRADOSmall early studies gave ipilimumab plus nivolumab before surgery to people with stage III melanoma. Most had a pathological response, but the first schedule caused frequent serious side effects. A lower-ipilimumab schedule had fewer, though the small comparison could not confirm this, and a later cohort used response to guide treatment.OpACIN (n=20, 1:1): neoadjuvant vs adjuvant ipilimumab 3 mg/kg plus nivolumab 1 mg/kg; pathological response 7/9 (78%) in the neoadjuvant arm. OpACIN-neo (phase 2, 89 randomised): grade 3-4 immune-related adverse events 40%, 20%, 50% and pathological response 80%, 77%, 65% in groups A, B (ipilimumab 1 mg/kg plus nivolumab 3 mg/kg) and C. PRADO (n=99): pathological response 72%; 24-month RFS 93% (major response).
  15. 2019 · Treatment · ImmunotherapyE1609: adjuvant ipilimumab versus high-dose interferon (2019)In a large trial after surgery for high-risk melanoma, ipilimumab at a lower dose (3 mg/kg) was linked to longer overall survival than high-dose interferon, with fewer serious side effects. The approved higher dose (10 mg/kg) was not shown to be better than interferon, and more patients stopped it because of side effects.Phase 3, three arms; resected AJCC-7 stage IIIB, IIIC, M1a or M1b melanoma (N=1,670): ipilimumab 3 mg/kg (n=523), high-dose interferon alfa-2b (n=636), ipilimumab 10 mg/kg (n=511). Ipilimumab 3 mg/kg vs interferon (n=1,051, concurrent randomisation): OS HR 0.78 (95.6% repeated CI 0.61-0.99; P=0.044); RFS HR 0.85 (99.4% CI 0.66-1.09; P=0.065). Ipilimumab 10 mg/kg vs interferon did not reach significance.

Precision & engineered

  1. 2021 · Treatment · ImmunotherapyAdjuvant therapy in stage IIB/IIC: KEYNOTE-716 and CheckMate 76KIn two separate placebo-controlled phase 3 trials of people with resected stage IIB or IIC melanoma, a year of pembrolizumab (KEYNOTE-716) or nivolumab (CheckMate 76K) was followed by fewer recurrences than placebo at early follow-up. Each trial stands alone, and neither report gives an overall survival result yet.Two phase 3 placebo-controlled trials in resected stage IIB/IIC melanoma. KEYNOTE-716 (N=976, 1:1): pembrolizumab vs placebo, RFS HR 0.65 (95% CI 0.46-0.92) at the first interim analysis and 0.61 (95% CI 0.45-0.82) at the second. CheckMate 76K (N=790, 2:1): nivolumab vs placebo, 12-month RFS 89.0% vs 79.4% (HR 0.42; 95% CI 0.30-0.59; P<0.0001). The trials are not directly comparable.
  2. 2021 · Treatment · Cell therapy & transplantLifileucel: from phase 2 to FDA accelerated approval (2024)Lifileucel is made from a person's own immune cells taken from their tumour, grown in a laboratory and given back by infusion. In a single-arm study of people whose melanoma had progressed after earlier treatment, just over a third responded. The US regulator granted accelerated approval in 2024; the European application was withdrawn in 2025.Single-arm phase 2 (n=66, all with prior anti-PD-1/PD-L1): investigator-assessed ORR 36% (95% CI 25 to 49), 2 complete and 22 partial responses; median DOR not reached at 18.7 months' follow-up. FDA accelerated approval 16 Feb 2024 on ORR and DOR (ORR 31.5% in 73 patients at the recommended dose). EMA application withdrawn 22 Jul 2025.
  3. 2021 · Treatment · ImmunotherapyTebentafusp: survival benefit in metastatic uveal melanoma (2021)Uveal melanoma starts in the eye and behaves differently from skin melanoma. Tebentafusp is an engineered protein that directs the body's T cells to melanoma cells. In a randomised trial of people with a particular tissue type, 73 in 100 were alive at one year, against 59 in 100 given other drugs.Open-label phase 3: 378 previously untreated HLA-A*02:01-positive patients randomised 2:1 to tebentafusp (252) or investigator's choice (126; pembrolizumab, ipilimumab or dacarbazine). OS at 1 year 73% vs 59% (HR 0.51, 95% CI 0.37 to 0.71; P<0.001, ITT). 3-year update: median OS 21.6 vs 16.9 months (HR 0.68, 0.54 to 0.87); 27% vs 18% alive at 3 years.
  4. 2022 · Treatment · ImmunotherapySequencing in BRAF-mutant disease: DREAMseq, SECOMBIT and EBINFor melanoma with a BRAF mutation, two sets of drugs are available, and trials asked which to give first. In DREAMseq, starting with nivolumab plus ipilimumab was associated with better two-year survival than starting with dabrafenib plus trametinib. SECOMBIT did not compare arms. In EBIN, a targeted lead-in before immunotherapy did not improve progression-free survival.DREAMseq (phase 3, n=265, stopped early): 2-year OS 71.8% (nivolumab plus ipilimumab first) vs 51.5% (dabrafenib plus trametinib first), P=0.010. SECOMBIT (noncomparative phase 2, n=209): 3-year OS 54%, 62% and 60% across the three sequences. EBIN (phase 2, n=271): encorafenib plus binimetinib induction did not improve PFS (median 9 vs 9 months; P=0.36).
  5. 2022 · Treatment · ImmunotherapyRelatlimab plus nivolumab: RELATIVITY-047 and LAG-3 blockade (2022)In untreated advanced melanoma, adding relatlimab, an antibody against the immune checkpoint LAG-3, to nivolumab delayed progression compared with nivolumab alone. Survival differences at three and four years were reported as descriptive. The FDA approved the combination in March 2022; in Europe it is limited to tumours with low PD-L1.RELATIVITY-047 (phase 2/3, n=714, untreated): relatlimab plus nivolumab vs nivolumab, median PFS 10.1 vs 4.6 months (HR 0.75; 95% CI 0.62 to 0.92; P=0.006); 12-month PFS 47.7% vs 36.0%. Grade 3 or 4 treatment-related adverse events 18.9% vs 9.7%. At 3 years median OS 51.0 vs 34.1 months (HR 0.80; 95% CI 0.66 to 0.99), descriptive. EMA indication restricted to tumour PD-L1 below 1%.
  6. 2022 · Treatment · Cell therapy & transplantTIL therapy versus ipilimumab: randomised phase 3 trial (2022)A randomised trial led by the Netherlands Cancer Institute compared infusion of a patient's own tumour-infiltrating immune cells with the antibody ipilimumab in advanced melanoma, mostly refractory to earlier PD-1 treatment. Time without progression was longer with the cell therapy, but everyone who received it had a severe treatment-related side effect, mainly from the preparatory chemotherapy.Phase 3, open-label, 168 patients with unresectable stage IIIC or IV melanoma (86% anti-PD-1 refractory), TIL vs ipilimumab 3 mg/kg, 1:1. Median PFS 7.2 vs 3.1 months (HR 0.50, 95% CI 0.35 to 0.72; P<0.001, ITT); ORR 49% vs 21%. Median OS 25.8 vs 18.9 months. Grade 3 or higher treatment-related events: all TIL recipients, 57% with ipilimumab.
  7. 2023 · Treatment · ImmunotherapySWOG S1801: neoadjuvant-adjuvant versus adjuvant-only pembrolizumabThis randomised phase 2 trial compared pembrolizumab given before and after surgery with pembrolizumab given only after surgery in people with resectable stage III or IV melanoma. Event-free survival was longer with the before-and-after schedule: about seven in ten were event-free at two years against about half. Follow-up was short.Randomised, open-label phase 2 (N=313); clinically detectable, measurable stage IIIB-IVC melanoma amenable to resection. Pembrolizumab 3 doses before surgery plus 15 after (n=154) vs surgery then 18 doses (n=159). Median follow-up 14.7 months: EFS longer with neoadjuvant-adjuvant (P=0.004, log-rank); landmark 2-year EFS 72% (95% CI 64-80) vs 49% (95% CI 41-59). Grade 3 or higher treatment-related adverse events 12% vs 14%.
  8. 2024 · Treatment · ImmunotherapyIndividualised mRNA neoantigen therapy plus pembrolizumab: KEYNOTE-942In KEYNOTE-942, a phase 2 trial of 157 people with resected high-risk melanoma, adding an individualised mRNA neoantigen therapy (mRNA-4157, also called V940 or intismeran) to pembrolizumab was followed by fewer recurrences than pembrolizumab alone. The hazard ratio interval included 1 (two-sided P=0.053). A phase 3 trial has been announced; results are unpublished.Open-label randomised phase 2b (N=157, 2:1): mRNA-4157 plus pembrolizumab (n=107) vs pembrolizumab (n=50), resected stage IIIB-IV cutaneous melanoma. RFS HR 0.561 (95% CI 0.309-1.017; two-sided P=0.053); events 22% vs 40%; 18-month RFS 79% vs 62%. Descriptive 5-year update: RFS HR 0.510 (95% CI 0.294-0.887), DMFS HR 0.411 (95% CI 0.200-0.843), OS HR 0.471 (95% CI 0.165-1.345).
  9. 2024 · Treatment · ImmunotherapyNADINA: neoadjuvant ipilimumab plus nivolumab in stage III (2024)In NADINA, a phase 3 trial of people with stage III melanoma that could be removed by surgery, two doses of ipilimumab plus nivolumab before surgery (with more treatment afterwards only for those without a major response) were followed by longer event-free survival than surgery followed by 12 cycles of nivolumab. Follow-up was short, a median of about ten months.Phase 3, randomised; resectable macroscopic stage III melanoma (N=423): two cycles of neoadjuvant ipilimumab plus nivolumab, surgery, and adjuvant therapy only for partial response or nonresponse, vs surgery then 12 cycles of adjuvant nivolumab. Median follow-up 9.9 months: 12-month EFS 83.7% vs 57.2% (HR 0.32; 99.9% CI 0.15-0.66; P<0.001). Major pathological response 59.0%. Grade 3 or higher treatment-related adverse events 29.7% vs 14.7%.
  10. 2025 · Treatment · ImmunotherapyRP1 plus nivolumab after anti-PD-1 failure: IGNYTE and FDA approvalRP1 is a modified herpes virus injected into tumours and given with nivolumab. In a single-arm trial of people whose melanoma had progressed on anti-PD-1 treatment, about one in three responded in the journal report. The US regulator granted accelerated approval in August 2026.IGNYTE: single-arm trial of RP1 plus nivolumab after anti-PD-1 progression (140 enrolled). Journal report: confirmed ORR 32.9% (95% CI 25.2 to 41.3; RECIST 1.1, independent review), 15.0% complete response; median DOR 33.7 months. FDA accelerated approval 6 Aug 2026 on ORR 24.2% (95% CI 15.8 to 34.3) and median DOR 14.1 months in 91 evaluable patients.

Then and now

  • 5-year relative survival · United States, SEER 8 registries

    • 82.34%

      1975 (observed value; first year in the table)

    People with melanoma of the skin, all races, both sexes (page note); SEER 8 five-year relative survival series, 1975-2018, on the SEER Cancer Stat Facts melanoma page. Age range and stage mix are not stated for this series, and the page does not say whether in situ melanoma is included.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modelled trend value in the adjacent column (80.99% for this year). The table column is headed 'Year' and the caption beneath the chart reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'; neither labels the year as year of diagnosis. Single-year values fluctuate (for example 84.91% in 1980 and 80.90% in 1981 in the same series). The page does not say how much of the change reflects treatment, stage mix, detection or follow-up methods. Incidence in the same registries rose over the period (SEER 8 observed rate of new cases 8.80 per 100,000 in 1975 and 28.26 in 2023), so survival cannot be read without the incidence and death-rate values. Vemurafenib was approved in 2011 and anti-PD-1 antibodies in 2014 (FDA letters cited under Today), so most of this series predates modern systemic therapy and cannot show its effect. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 89.11%

      1990 (observed value)

    People with melanoma of the skin, all races, both sexes (page note); SEER 8 five-year relative survival series, 1975-2018, on the SEER Cancer Stat Facts melanoma page. Age range and stage mix are not stated for this series, and the page does not say whether in situ melanoma is included.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modelled trend value in the adjacent column (88.93% for this year). The table column is headed 'Year' and the caption beneath the chart reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'; neither labels the year as year of diagnosis. Single-year values fluctuate (for example 84.91% in 1980 and 80.90% in 1981 in the same series). The page does not say how much of the change reflects treatment, stage mix, detection or follow-up methods. Incidence in the same registries rose over the period (SEER 8 observed rate of new cases 8.80 per 100,000 in 1975 and 28.26 in 2023), so survival cannot be read without the incidence and death-rate values. Vemurafenib was approved in 2011 and anti-PD-1 antibodies in 2014 (FDA letters cited under Today), so most of this series predates modern systemic therapy and cannot show its effect. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 92.52%

      2000 (observed value)

    People with melanoma of the skin, all races, both sexes (page note); SEER 8 five-year relative survival series, 1975-2018, on the SEER Cancer Stat Facts melanoma page. Age range and stage mix are not stated for this series, and the page does not say whether in situ melanoma is included.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modelled trend value in the adjacent column (92.36% for this year). The table column is headed 'Year' and the caption beneath the chart reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'; neither labels the year as year of diagnosis. Single-year values fluctuate (for example 84.91% in 1980 and 80.90% in 1981 in the same series). The page does not say how much of the change reflects treatment, stage mix, detection or follow-up methods. Incidence in the same registries rose over the period (SEER 8 observed rate of new cases 8.80 per 100,000 in 1975 and 28.26 in 2023), so survival cannot be read without the incidence and death-rate values. Vemurafenib was approved in 2011 and anti-PD-1 antibodies in 2014 (FDA letters cited under Today), so most of this series predates modern systemic therapy and cannot show its effect. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 94.62%

      2010 (observed value)

    People with melanoma of the skin, all races, both sexes (page note); SEER 8 five-year relative survival series, 1975-2018, on the SEER Cancer Stat Facts melanoma page. Age range and stage mix are not stated for this series, and the page does not say whether in situ melanoma is included.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modelled trend value in the adjacent column (94.77% for this year). The table column is headed 'Year' and the caption beneath the chart reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'; neither labels the year as year of diagnosis. Single-year values fluctuate (for example 84.91% in 1980 and 80.90% in 1981 in the same series). The page does not say how much of the change reflects treatment, stage mix, detection or follow-up methods. Incidence in the same registries rose over the period (SEER 8 observed rate of new cases 8.80 per 100,000 in 1975 and 28.26 in 2023), so survival cannot be read without the incidence and death-rate values. Vemurafenib was approved in 2011 and anti-PD-1 antibodies in 2014 (FDA letters cited under Today), so most of this series predates modern systemic therapy and cannot show its effect. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 97.52%

      2018 (observed value; last observed year in the table)

    People with melanoma of the skin, all races, both sexes (page note); SEER 8 five-year relative survival series, 1975-2018, on the SEER Cancer Stat Facts melanoma page. Age range and stage mix are not stated for this series, and the page does not say whether in situ melanoma is included.

    Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modelled trend value in the adjacent column (96.14% for this year). The table column is headed 'Year' and the caption beneath the chart reads 'SEER 8 5-Year Relative Survival Percent from 1975-2018, All Races, Both Sexes'; neither labels the year as year of diagnosis. Single-year values fluctuate (for example 84.91% in 1980 and 80.90% in 1981 in the same series). The page does not say how much of the change reflects treatment, stage mix, detection or follow-up methods. Incidence in the same registries rose over the period (SEER 8 observed rate of new cases 8.80 per 100,000 in 1975 and 28.26 in 2023), so survival cannot be read without the incidence and death-rate values. Vemurafenib was approved in 2011 and anti-PD-1 antibodies in 2014 (FDA letters cited under Today), so most of this series predates modern systemic therapy and cannot show its effect. After 2018 the table shows only modelled trend values (96.29% for 2019 to 96.82% for 2023); the page does not say why observed values stop at 2018. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 94.7%

      2016-2022

    People with melanoma of the skin, all stages (headline figure on the SEER Cancer Stat Facts melanoma page, labelled 'Based on data from SEER 21 (Excluding IL) 2016-2022'). The note beside the headline figure names only the registry set and period; the stage table below it is labelled 'All Races, Both Sexes'.

    A different registry set and period from the SEER 8 series, so it is not a continuation of the 1975-2018 values and must not be subtracted from them. It is lower than the SEER 8 observed value for 2018 (97.52%); the page does not explain the difference, so it must not be read as a decline in survival. The page labels the period '2016-2022' without saying whether it is the year of diagnosis. Relative survival excludes deaths from other causes. The page does not say whether in situ melanoma is included.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 100%

      2016-2022; localized stage at diagnosis (77% of cases)

    People with melanoma of the skin, all races, both sexes, by SEER Combined Summary Stage; the page's label for localized is 'Confined to Primary Site'.

    Stage-specific value from the page's 'Percent of Cases & 5-Year Relative Survival by Stage at Diagnosis' table (stage mix on the page: 77% localized, 10% regional, 5% distant, 9% unknown; the percentages sum to 101% because of rounding). The page's text gives 76.9% for the localized share. The page defines relative survival as an estimate of the percentage of patients expected to survive the effects of their cancer, excluding the risk of dying from other causes, and notes that population statistics cannot be used to predict what will happen to an individual patient; 100.0% is a rounded population estimate for this stage group. This is SEER Combined Summary Stage, not AJCC stage.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 76%

      2016-2022; regional stage at diagnosis (10% of cases)

    People with melanoma of the skin, all races, both sexes, by SEER Combined Summary Stage; the page's label for regional is 'Spread to Regional Lymph Nodes'.

    SEER Combined Summary Stage, not AJCC stage. The page labels the period '2016-2022' without saying whether it is the year of diagnosis and gives no breakdown by treatment, so the figure is not a prediction for people treated with current regimens. The 'unknown' stage group (9% of cases) had 95.4%, which should not be read as a stage.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 34%

      2016-2022; distant stage at diagnosis (5% of cases)

    People with melanoma of the skin, all races, both sexes, by SEER Combined Summary Stage; the page's label for distant is 'Cancer Has Metastasized'.

    SEER Combined Summary Stage, not AJCC stage. The page labels the period '2016-2022' without saying whether it is the year of diagnosis and gives no breakdown by treatment, so the figure is not a prediction for people treated with current regimens. Distant-stage cases are 5% of the total, so the headline figure is dominated by localized disease.

    Source

  • Age-standardised mortality · United States (U.S. Mortality, as tabulated by SEER)

    • 2.06per 100,000

      1975 (observed)

    People with melanoma of the skin, all races, both sexes, age-adjusted death rate from U.S. Mortality (NCHS) as shown in the SEER Cancer Stat Facts table; observed value. Deaths are not drawn from the SEER registries.

    The page states that the rates are age-adjusted and does not name the standard population, so they should not be compared directly with rates age-standardised to another (for example European) standard population. Observed rate of new cases in the SEER 8 column was 8.80 per 100,000 in the same year. The page does not say how much of later changes in mortality reflect detection, case mix or treatment.

    Source

  • Age-standardised mortality · United States (U.S. Mortality, as tabulated by SEER)

    • 2.75per 100,000

      1990 (observed)

    People with melanoma of the skin, all races, both sexes, age-adjusted death rate from U.S. Mortality (NCHS) as shown in the SEER Cancer Stat Facts table; observed value. Deaths are not drawn from the SEER registries.

    The page states that the rates are age-adjusted and does not name the standard population, so they should not be compared directly with rates age-standardised to another (for example European) standard population. Observed rate of new cases (SEER 8) was 14.47 per 100,000 in 1990. Observed death rates in the table ranged from 2.61 to 2.81 per 100,000 over 1990-2013, while the SEER 8 rate of new cases rose from 14.47 to 25.53.

    Source

  • Age-standardised mortality · United States (U.S. Mortality, as tabulated by SEER)

    • 2.66per 100,000

      2000 (observed)

    People with melanoma of the skin, all races, both sexes, age-adjusted death rate from U.S. Mortality (NCHS) as shown in the SEER Cancer Stat Facts table; observed value. Deaths are not drawn from the SEER registries.

    The page states that the rates are age-adjusted and does not name the standard population, so they should not be compared directly with rates age-standardised to another (for example European) standard population. Observed rate of new cases (SEER 8) was 19.82 per 100,000 in 2000.

    Source

  • Age-standardised mortality · United States (U.S. Mortality, as tabulated by SEER)

    • 2.68per 100,000

      2013 (observed)

    People with melanoma of the skin, all races, both sexes, age-adjusted death rate from U.S. Mortality (NCHS) as shown in the SEER Cancer Stat Facts table; observed value. Deaths are not drawn from the SEER registries.

    The page states that the rates are age-adjusted and does not name the standard population, so they should not be compared directly with rates age-standardised to another (for example European) standard population. Observed rate of new cases (SEER 8) was 25.53 per 100,000 in 2013. The decline in the death rate starts after 2013 (2.60 in 2014). The FDA approvals of dabrafenib with trametinib (January 2014) and of pembrolizumab and nivolumab (September and December 2014), read in the letters cited under Today, are given here as dates only, in the same period as the start of the decline; the page does not attribute the fall to treatment, and uptake, detection, case mix and non-drug factors are not excluded. This is a coincidence of dates, not evidence of an effect.

    Source

  • Age-standardised mortality · United States (U.S. Mortality, as tabulated by SEER)

    • 2per 100,000

      2024 (observed; last year in the table)

    People with melanoma of the skin, all races, both sexes, age-adjusted death rate from U.S. Mortality (NCHS) as shown in the SEER Cancer Stat Facts table; observed value. Deaths are not drawn from the SEER registries.

    The page states that the rates are age-adjusted and does not name the standard population, so they should not be compared directly with rates age-standardised to another (for example European) standard population. SEER text: death rates fell on average 2.2% per year over 2015-2024. Incidence did not fall: the SEER 8 rate of new cases was 28.26 per 100,000 in 2023 (rising on average 1.1% per year over 2014-2023 per the page; the 2020 value of 24.14 is displayed by SEER but not used in its trend fit, and the page links a note on the impact of COVID-19 on 2020 incidence data). The United Kingdom shows a different pattern (see the Cancer Research UK entries and the notes).

    Source

  • 10-year relative survival · United Kingdom (as worded on the Cancer Research UK page)

    • 46.4%

      People diagnosed in the 1970s (period wording from the page)

    People diagnosed with melanoma skin cancer in the UK who 'survived their disease beyond ten years' (page wording); Cancer Research UK health-professional survival page. Age range and exact years are not stated.

    The page does not name the measure (net, relative or observed survival), the age range, or the exact years of the 1970s figure, and the 1970s sentence itself names no geography (the page's headline sentence for the 2018 value refers to the UK). It is entered under the nearest metric (10-year relative survival) and should be treated as illustrative. Case mix changed over the period and the page says survival trends reflect a combination of treatment and stage distribution. Not comparable with the US SEER series. The page has no visible update date.

    Source

  • 10-year relative survival · United Kingdom (as worded on the Cancer Research UK page)

    • 92.7%

      People diagnosed in 2018 (predicted)

    People diagnosed with melanoma skin cancer in the UK; the page says it is 'predicted' that 92.7% survive their disease for ten years or more (2018). Cancer Research UK health-professional survival page.

    A predicted value, not an observed 10-year follow-up, and the 1970s and 2018 values may not use the same method. The measure is not named on the page. Cancer Research UK's separate mortality page reports that UK melanoma mortality rates have 'increased by around two-and-a-half times (153%)' since the early 1970s and 'remained stable in the last decade (2022-2024)', so the UK mortality pattern differs from the US death-rate series. Not comparable with the US SEER series.

    Source

  • 5-year relative survival · England (Cancer Research UK)

    • 90.5%

      2015-2019

    People diagnosed with melanoma skin cancer in England; the page words it as 'survive their disease for five years or more' (2015-2019).

    The page cites 'five-year relative survival' for the European comparison but does not label this England figure's measure beyond 'survive their disease'. For a different period (2018-2022) the page gives 89.5% in the most deprived group and 93.1% in the least deprived group in England; these are not comparable with the 2015-2019 figure. Not comparable with the US SEER series.

    Source

Survival figures are affected by earlier detection (lead time) and overdiagnosis, and by changes in who is diagnosed. For screened cancers, read them alongside mortality.

Evidence chains

A lane follows one defined population and setting, where each new trial tests against what was then standard. Compare only within a lane, and read every comparison as illustrative.

BRAF V600-mutant metastatic melanoma, targeted therapy

Metastatic or unresectable melanoma with a BRAF V600 mutation (V600E in BRIM-3 and BREAK-3; V600E or V600K in COMBI-d, COMBI-v and METRIC); BRIM-3, BREAK-3, COMBI-d and COMBI-v enrolled previously untreated patients · Metastatic · endpoint: PFS

Results are not comparable across trials.

Controls differ: dacarbazine in BRIM-3 and BREAK-3, dabrafenib plus placebo in COMBI-d, and vemurafenib in COMBI-v, so each pair is a different question. COMBI-d and COMBI-v are drawn with their first reported progression-free survival (COMBI-v from an interim analysis); COMBI-v's primary endpoint was overall survival. Eras, crossover and later immunotherapy differ. METRIC (trametinib versus chemotherapy) is listed in the milestone but not plotted because its record does not state prior-treatment status. Compare only each trial's own two arms; the chart is illustrative, not a pooled estimate.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

02468101214PFS, median (months)2011201120142014BRIM-3break-3COMBI-dCOMBI-v
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
PFS, median, across 4 trials in this lane. BRIM-3 (2011): experimental 5.3 months vs control 1.6 months; break-3 (2011): experimental 5.1 months vs control 2.7 months; COMBI-d (2014): experimental 9.3 months vs control 8.8 months; COMBI-v (2014): experimental 11.4 months vs control 7.3 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
BRIM-32011PFS1.6 months5.3 months
break-32011PFS2.7 months5.1 months
COMBI-d2014PFS8.8 months9.3 months
COMBI-v2014PFS7.3 months11.4 months

Advanced melanoma without earlier checkpoint therapy, immunotherapy

Unresectable stage III or IV melanoma, mostly previously untreated; CheckMate 066 enrolled only BRAF wild-type disease and KEYNOTE-006 allowed up to one earlier systemic therapy (ipilimumab-naive) · Metastatic · endpoint: PFS

Results are not comparable across trials.

Controls differ: dacarbazine (CheckMate 066), dacarbazine plus placebo (CA184-024), ipilimumab (KEYNOTE-006 and CheckMate 067) and nivolumab (RELATIVITY-047). The CA184-024 point is median overall survival, not progression-free survival, and is drawn with a distinct marker; that trial gave no progression-free survival result in this record. KEYNOTE-006 reported 6-month landmark rates, so it is listed but not drawn on the median axis. CheckMate 066 is BRAF wild-type only. MDX010-20 (previously treated, in the ipilimumab milestone) and CheckMate 037 (after ipilimumab) are different populations and are not plotted. CheckMate 067 was not designed to compare the combination with nivolumab alone, and no randomised trial here compares nivolumab plus ipilimumab with nivolumab plus relatlimab. Compare only each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

02468101214PFS, median (months)20102014201520152022ca184-024CheckMate 066CheckMate 067CheckMate 067RELATIVITY-047
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
PFS, median, across 5 trials in this lane. ca184-024 (2010): experimental 11.2 months vs control 9.1 months; CheckMate 066 (2014): experimental 5.1 months vs control 2.2 months; CheckMate 067 (2015): experimental 11.5 months vs control 2.9 months; CheckMate 067 (2015): experimental 6.9 months vs control 2.9 months; RELATIVITY-047 (2022): experimental 10.1 months vs control 4.6 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
ca184-0242010OS (differs from lane)9.1 months11.2 months
CheckMate 0662014PFS2.2 months5.1 months
CheckMate 0672015PFS2.9 months11.5 months
CheckMate 0672015PFS2.9 months6.9 months
RELATIVITY-0472022PFS4.6 months10.1 months
KEYNOTE-0062015PFS (not plotted: different measure or time point)26.5 %47.3 %
KEYNOTE-0062015PFS (not plotted: different measure or time point)26.5 %46.4 %

Resected high-risk melanoma, adjuvant therapy

Completely resected high-risk cutaneous melanoma: T4 or node-positive (E1684), stage III (EORTC 18991, EORTC 18071, KEYNOTE-054), stage IIIB-C or IV (CheckMate 238), and stage III with a BRAF V600E or V600K mutation (COMBI-AD) · Adjuvant · endpoint: RFS

Results are not comparable across trials.

Eligibility, staging editions and controls differ: observation (E1684, EORTC 18991), placebo (EORTC 18071, COMBI-AD, KEYNOTE-054) and ipilimumab 10 mg/kg (CheckMate 238). E1684 and EORTC 18071 report median relapse-free survival (E1684 in years) and set the axis; EORTC 18991, CheckMate 238, COMBI-AD and KEYNOTE-054 report landmark rates at different times and are listed but not drawn. E1609 (ipilimumab 3 mg/kg or 10 mg/kg versus high-dose interferon, resected stage IIIB-C or M1a-b) is listed without a plotted value because its record holds hazard ratios only. CheckMate 238's 9-year values await an erratum check and are not used. Follow-up lengths differ. Compare only each trial's own arms; never compare hazard ratios across trials.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

05101520253035RFS, median (months)19962015ECOG E1684EORTC 18071
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
RFS, median, across 2 trials in this lane. ECOG E1684 (1996): experimental 20.4 months vs control 12 months; EORTC 18071 (2015): experimental 26.1 months vs control 17.1 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
ECOG E16841996RFS12 months20.4 months
EORTC 180712015RFS17.1 months26.1 months
EORTC 189911996RFS (not plotted: different measure or time point)38.9 %45.6 %
CheckMate 2382017RFS (not plotted: different measure or time point)60.8 %70.5 %
COMBI-AD2017RFS (not plotted: different measure or time point)39 %58 %
EORTC 1325/KEYNOTE-0542018RFS (not plotted: different measure or time point)61 %75.4 %

Resectable stage III melanoma, neoadjuvant immunotherapy

Resectable macroscopic stage III melanoma (NADINA); SWOG S1801 also enrolled stage IIID and oligometastatic resectable stage IV disease · Neoadjuvant · endpoint: EFS

Results are not comparable across trials.

Only NADINA is drawn. SWOG S1801 (phase 2) is listed but not plotted against it: it used a 24-month event-free survival landmark, pembrolizumab given before and after surgery versus after surgery only, and a broader population. The OpACIN, OpACIN-neo and PRADO studies are phase 2 with pathological-response endpoints. NADINA's control was surgery followed by adjuvant nivolumab, and its follow-up was short (median 9.9 months). Compare only each trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100EFS at 12 months (%)2024NADINA
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
EFS at 12 months, rate, across 1 trial in this lane. NADINA (2024): experimental 83.7 % vs control 57.2 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
NADINA2024EFS57.2 %83.7 %

Positive sentinel node: completion dissection or observation

Cutaneous melanoma with sentinel-node metastasis randomised to immediate completion lymph-node dissection or observation (MSLT-II, DeCOG-SLT) · Localised · endpoint: MSS

Results are not comparable across trials.

Both trials asked the same question, but MSLT-II reports 3-year melanoma-specific survival (per-protocol analysis) while DeCOG-SLT reports 3-year distant metastasis-free survival, drawn with a distinct marker. DeCOG-SLT closed early and its authors describe it as underpowered. In both, the experimental arm is observation and the control arm is completion dissection. Compare only each trial's own two arms. MSLT-I (whether to perform sentinel-node biopsy) is a different question and is not in this lane.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100MSS at 36 months (%)20162016MSLT-IIDeCOG-SLT
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
MSS at 36 months, rate, across 2 trials in this lane. MSLT-II (2016): experimental 86 % vs control 86 %; DeCOG-SLT (2016): experimental 77 % vs control 74.9 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
MSLT-II2016MSS86 %86 %
DeCOG-SLT2016DMFS (differs from lane)74.9 %77 %

Standard of care, then and now

How the usual sequence of treatment changed for one setting. Choose a point in time; each step links to the milestone behind it.

Advanced melanoma: systemic treatment over time

Adults with unresectable stage III or stage IV melanoma · Metastatic

  1. Targeted1BRAF V600: dabrafenib + trametinibDabrafenib plus trametinib: COMBI-d and COMBI-v (2014) (2014)
  2. Immunotherapy2Nivolumab (untreated, BRAF wild-type)Nivolumab monotherapy: CheckMate 066 and CheckMate 037 (2014-2015) (2014)
  3. Immunotherapy3Pembrolizumab compared with ipilimumabKEYNOTE-006: pembrolizumab versus ipilimumab (2015, 10-year follow-up) (2015)
  4. Immunotherapy4Nivolumab + ipilimumabNivolumab plus ipilimumab: CheckMate 069/067 and the 10-year follow-up (2015)
  5. Immunotherapy5Nivolumab + relatlimabRelatlimab plus nivolumab: RELATIVITY-047 and LAG-3 blockade (2022) (2022)
  6. Cell therapy & transplant6After anti-PD-1: lifileucel (TIL)Lifileucel: from phase 2 to FDA accelerated approval (2024) (2021)
Today. Based only on the trials cited in each step; options are not ranked. After earlier anti-PD-1 treatment, lifileucel (tumour-infiltrating lymphocytes) has FDA accelerated approval, based on a single-arm study.
Text version
  • 2010: Dacarbazine (reference chemotherapy) → Ipilimumab (anti-CTLA-4)
  • 2015: BRAF V600: dabrafenib + trametinib → Nivolumab (untreated, BRAF wild-type) → Pembrolizumab compared with ipilimumab → Nivolumab + ipilimumab
  • 2022: BRAF V600: dabrafenib + trametinib → Nivolumab (untreated, BRAF wild-type) → Pembrolizumab compared with ipilimumab → Nivolumab + ipilimumab → Nivolumab + relatlimab
  • Today: BRAF V600: dabrafenib + trametinib → Nivolumab (untreated, BRAF wild-type) → Pembrolizumab compared with ipilimumab → Nivolumab + ipilimumab → Nivolumab + relatlimab → After anti-PD-1: lifileucel (TIL)

Resected stage III melanoma: adjuvant treatment over time

Adults after complete resection of stage III melanoma (some trials also included stage IV) · Adjuvant

  1. Immunotherapy1Nivolumab (compared with ipilimumab)Adjuvant nivolumab versus ipilimumab: CheckMate 238 (2017) (2017)
  2. Immunotherapy2Pembrolizumab (compared with placebo)Adjuvant pembrolizumab: EORTC 1325/KEYNOTE-054 (2018) (2018)
  3. Targeted3BRAF V600: dabrafenib + trametinibAdjuvant dabrafenib plus trametinib: COMBI-AD (2017) (2017)
  4. Immunotherapy4Stage IIB-C: pembrolizumab or nivolumabAdjuvant therapy in stage IIB/IIC: KEYNOTE-716 and CheckMate 76K (2021)
Today. Based only on the cited trials; options are not ranked. Interferon and ipilimumab 10 mg/kg are not shown (the guideline read says the latter is no longer used because of toxicity). Stage IIB-C trials added.
Text version
  • 1996: Interferon alfa-2b
  • 2015: Interferon alfa-2b → Ipilimumab 10 mg/kg
  • 2018: Interferon alfa-2b → Ipilimumab 10 mg/kg → Nivolumab (compared with ipilimumab) → Pembrolizumab (compared with placebo) → BRAF V600: dabrafenib + trametinib
  • Today: Nivolumab (compared with ipilimumab) → Pembrolizumab (compared with placebo) → BRAF V600: dabrafenib + trametinib → Stage IIB-C: pembrolizumab or nivolumab

Today and tomorrow

Standard of care today

Screening
The US Preventive Services Task Force (2023 recommendation statement; abstract read) concludes that current evidence is insufficient to assess the balance of benefits and harms of a clinician's visual skin examination to screen for skin cancer in adolescents and adults (I statement). The statement applies to asymptomatic people with no history of premalignant or malignant skin lesions and covers skin cancer as a whole, not melanoma alone. Screening programmes in other countries were not researched for this page.
US Preventive Services Task Force 2023
Localised

Guideline text read:

  • ESMO Clinical Practice Guideline for cutaneous melanoma (Ann Oncol 2025;36(1):10-30
  • approved by the ESMO Guidelines Committee, last update November 2024
  • open-access author manuscript). Recommendations are restated with the guideline's own evidence grades and are not ranked by this site. Staging is recommended according to the AJCC 8th edition [II, A]. Full-depth wide local excision is recommended with safety margins of 0.5 cm for in situ melanoma, 1 cm for tumours up to 2 mm thick and 2 cm for tumours thicker than 2 mm [III, B]
  • reduced margins or micrographic surgery are acceptable to preserve function in acral and facial melanomas. Sentinel lymph node biopsy is not routinely recommended for pT1a [II, E], should be discussed for pT1b [III, B] and is recommended for clinically node-negative T2a or higher tumours (thickness over 1.0 mm) [I, A]. Complete lymph node dissection is not recommended after a positive sentinel node biopsy [I, E]. The NCCN guideline (Melanoma: Cutaneous, Version 2.2026, J Natl Compr Canc Netw 2026;24(6):272-277) is cited by identifier only
  • its text was not read. MelMarT-II, a trial of 1 cm versus 2 cm margins in stage II melanoma, is listed under Horizon. No FDA or EMA action specific to localized disease was read for this page
Amaral T, Ottaviano M, Arance A, Blank C, Chiarion-Sileni V, Donia M, et al 2025Swetter SM, Johnson D, Albertini MR, Barker CA, Bateni S, Baumgartner J, et al 2026ClinicalTrials 2026
Adjuvant

Resected stage III (and, for some agents, stage IV) melanoma. FDA actions read in approval letters:

  • ipilimumab, 28 October 2015, for cutaneous melanoma with pathologic involvement of regional lymph nodes of more than 1 mm after complete resection including total lymphadenectomy
  • nivolumab, 20 December 2017, for melanoma with involvement of lymph nodes or metastatic disease after complete resection
  • dabrafenib with trametinib, 30 April 2018, for BRAF V600E or V600K-mutated melanoma with lymph node involvement after complete resection
  • pembrolizumab, 15 February 2019, for melanoma with lymph node involvement after complete resection. Current openFDA label records list adjuvant nivolumab for completely resected stage IIB, IIC, III or IV melanoma (Opdivo, effective 12 August 2026) and adjuvant pembrolizumab for stage IIB, IIC or III melanoma (Keytruda, effective 10 July 2026). Guideline text read: ESMO Clinical Practice Guideline for cutaneous melanoma (Ann Oncol 2025;36(1):10-30
  • approved by the ESMO Guidelines Committee, last update November 2024
  • open-access author manuscript). Recommendations are restated with the guideline's own evidence grades and are not ranked by this site. It lists anti-PD-1 therapy (nivolumab for resected stage IIIB to IV with no evidence of disease, pembrolizumab for resected stage III) or dabrafenib with trametinib for resected stage III BRAF V600E-mutated melanoma as adjuvant options [I, A], to be started within 12 weeks of resection
  • it advises against targeted therapy for BRAF V600K-mutated melanoma [II, D], against routine adjuvant treatment for AJCC8 stage IIIA with a sentinel node tumour burden under 1 mm [I, D], and against adjuvant nivolumab with ipilimumab for resected stage III [I, D]. It states that ipilimumab 10 mg/kg is no longer used because of toxicity, and attributes EMA approval of adjuvant nivolumab and pembrolizumab to August and December 2018 (EMA documents not read). The NCCN guideline (Melanoma: Cutaneous, Version 2.2026, J Natl Compr Canc Netw 2026;24(6):272-277) is cited by identifier only
  • its text was not read
FDA 2015FDA 2017FDA 2018FDA 2019U 2026U 2026Amaral T, Ottaviano M, Arance A, Blank C, Chiarion-Sileni V, Donia M, et al 2025Swetter SM, Johnson D, Albertini MR, Barker CA, Bateni S, Baumgartner J, et al 2026
Adjuvant

Resected stage IIB or IIC melanoma. FDA actions read in approval letters:

  • pembrolizumab, 3 December 2021, for adult and pediatric patients 12 years and older with stage IIB or IIC melanoma after complete resection
  • nivolumab, 13 October 2023, extending its adjuvant indication to completely resected stage IIB and IIC melanoma for adult and pediatric patients 12 years and older. Guideline text read: ESMO Clinical Practice Guideline for cutaneous melanoma (Ann Oncol 2025;36(1):10-30
  • approved by the ESMO Guidelines Committee, last update November 2024
  • open-access author manuscript). Recommendations are restated with the guideline's own evidence grades and are not ranked by this site. It states that adjuvant pembrolizumab or nivolumab for 12 months should be considered for stage IIB to IIC disease, with the discussion including the recurrence-free survival benefit and the lack of mature overall survival data [I, A]
  • it reports that no overall survival data were available from KEYNOTE-716 or CheckMate 76K at the time of writing, and attributes EMA approval of adjuvant pembrolizumab for stage IIB to IIC to June 2022 (EMA documents not read). Mutation testing is to be considered in clinical stage IIB to IIC [V, C]. The NCCN guideline (Melanoma: Cutaneous, Version 2.2026, J Natl Compr Canc Netw 2026;24(6):272-277) is cited by identifier only
  • its text was not read
FDA 2021FDA 2023Amaral T, Ottaviano M, Arance A, Blank C, Chiarion-Sileni V, Donia M, et al 2025Swetter SM, Johnson D, Albertini MR, Barker CA, Bateni S, Baumgartner J, et al 2026
Neoadjuvant

Guideline text read:

  • ESMO Clinical Practice Guideline for cutaneous melanoma (Ann Oncol 2025;36(1):10-30
  • approved by the ESMO Guidelines Committee, last update November 2024
  • open-access author manuscript). Recommendations are restated with the guideline's own evidence grades and are not ranked by this site. For resectable stage III melanoma with pathologically proven, clinically or radiologically detectable lymph node metastasis, it recommends that neoadjuvant nivolumab with ipilimumab followed by surgery should be offered [I, A], with adjuvant treatment omitted after a major pathological response and further treatment discussed otherwise, and it also recommends neoadjuvant plus adjuvant pembrolizumab [II, A]
  • both are labelled in the guideline as not EMA or FDA approved. The guideline text discusses NADINA (described as phase 3) and SWOG S1801 (described as randomised phase 2) before these recommendations
  • the trial publications were not read for this entry. No neoadjuvant melanoma indication appears in the melanoma indications of the current openFDA label records read for Opdivo (effective 12 August 2026) and Keytruda (effective 10 July 2026), and no FDA or EMA document for a neoadjuvant melanoma regimen was found. The NCCN guideline (Melanoma: Cutaneous, Version 2.2026, J Natl Compr Canc Netw 2026;24(6):272-277) is cited by identifier only
  • its text was not read
Amaral T, Ottaviano M, Arance A, Blank C, Chiarion-Sileni V, Donia M, et al 2025Swetter SM, Johnson D, Albertini MR, Barker CA, Bateni S, Baumgartner J, et al 2026U 2026U 2026
Metastatic

First line, unresectable or metastatic disease, any BRAF status. FDA actions read in approval letters:

  • pembrolizumab, 4 September 2014, accelerated approval after progression on ipilimumab (and a BRAF inhibitor if BRAF V600-mutant)
  • the letter of 18 December 2015 (supplements 4 and 6) removed the progression-after-ipilimumab restriction and the accelerated-approval statement and states that the approval fulfils the postmarketing requirement (the letter does not use the word 'converted'). Nivolumab, 22 December 2014, accelerated approval after progression on ipilimumab (and a BRAF inhibitor if BRAF V600-mutant)
  • 23 November 2015, regular approval for BRAF wild-type disease (the same letter states that the postmarketing requirement from the 2014 letter was not met, because clinical benefit was verified only for BRAF wild-type disease)
  • nivolumab with ipilimumab, 30 September 2015, accelerated approval for BRAF V600 wild-type disease, and 23 January 2016, expansion removing the BRAF wild-type restriction for the combination and, for single-agent nivolumab, the restriction to BRAF V600-mutant disease after progression on ipilimumab and a BRAF inhibitor, with the letter listing both under the accelerated-approval postmarketing requirement. The FDA letter of 7 March 2019 (supplement 42
  • 'Subpart E fulfilled') removed the accelerated-approval statement ('based on progression-free survival') from the single-agent BRAF V600 mutation-positive and the nivolumab-plus-ipilimumab indications, states that the approval fulfils postmarketing requirements 2838-1 and 2959-1, and merges the melanoma indications into one (single agent or with ipilimumab)
  • the letter does not use the word 'converted'. The current openFDA Opdivo label record (effective 12 August 2026) lists unresectable or metastatic melanoma as a single agent or in combination with ipilimumab without any accelerated-approval statement in its indications text. Nivolumab with relatlimab (Opdualag): FDA, 18 March 2022, for adult and pediatric patients 12 years and older with unresectable or metastatic melanoma, with no PD-L1 restriction in the indication wording
  • EMA: CHMP opinion adopted 21 July 2022 and marketing authorisation issued 15 September 2022 (EMA page), for first-line use from 12 years of age in tumours with tumour cell PD-L1 expression below 1%. Guideline text read: ESMO Clinical Practice Guideline for cutaneous melanoma (Ann Oncol 2025;36(1):10-30
  • approved by the ESMO Guidelines Committee, last update November 2024
  • open-access author manuscript). Recommendations are restated with the guideline's own evidence grades and are not ranked by this site. For unresectable disease it describes ipilimumab with nivolumab as a preferred first-line option for all patients regardless of BRAF status when this can be safely delivered for the first few months, that is when a rapid response is not required because of aggressive or symptomatic disease [I, A]
  • nivolumab or pembrolizumab alone is also recommended [I, A]
  • nivolumab with relatlimab can be offered [I, B]. The NCCN guideline (Melanoma: Cutaneous, Version 2.2026, J Natl Compr Canc Netw 2026;24(6):272-277) is cited by identifier only
  • its text was not read
U 2014U 2015U 2014U 2015U 2015U 2016U 2019U 2026U 2022European Medicines Agency 2026Amaral T, Ottaviano M, Arance A, Blank C, Chiarion-Sileni V, Donia M, et al 2025Swetter SM, Johnson D, Albertini MR, Barker CA, Bateni S, Baumgartner J, et al 2026
Metastatic

BRAF V600-mutant disease. FDA actions read in approval letters:

  • vemurafenib, 17 August 2011 (BRAF V600E, unresectable or metastatic)
  • dabrafenib with trametinib, accelerated approval on 8 and 9 January 2014 (Mekinist and Tafinlar letters
  • basis stated as durable response rate), with the 20 November 2015 Tafinlar letter granting regular approval for the combination and stating that the Subpart H postmarketing requirement was fulfilled
  • cobimetinib with vemurafenib, 10 November 2015
  • encorafenib with binimetinib, 27 June 2018 (Braftovi and Mektovi letters). Guideline text read: ESMO Clinical Practice Guideline for cutaneous melanoma (Ann Oncol 2025;36(1):10-30
  • approved by the ESMO Guidelines Committee, last update November 2024
  • open-access author manuscript). Recommendations are restated with the guideline's own evidence grades and are not ranked by this site. BRAF V600 testing is mandatory [I, A]. Where anti-PD-1-based therapy is not available or the patient is ineligible, BRAF plus MEK inhibitor combination therapy is a first-line option [I, A]
  • the guideline states that first-line nivolumab with ipilimumab is the preferred treatment when it can be safely delivered, with targeted therapy reserved for later lines, citing the DREAMSEQ, SECOMBIT and EBIN sequencing trials
  • BRAF/MEK inhibitors for 8 to 12 weeks followed by ipilimumab with nivolumab is listed as an option, especially with high LDH or liver metastases [II, C], and is labelled as not EMA or FDA approved. The NCCN guideline (Melanoma: Cutaneous, Version 2.2026, J Natl Compr Canc Netw 2026;24(6):272-277) is cited by identifier only
  • its text was not read
U 2011U 2014U 2014U 2015U 2015U 2018U 2018Amaral T, Ottaviano M, Arance A, Blank C, Chiarion-Sileni V, Donia M, et al 2025Swetter SM, Johnson D, Albertini MR, Barker CA, Bateni S, Baumgartner J, et al 2026
Relapsed / refractory

After anti-PD-1 therapy. FDA actions read:

  • lifileucel (Amtagvi), 16 February 2024, accelerated approval for adults with unresectable or metastatic melanoma previously treated with a PD-1 blocking antibody and, if BRAF V600-mutant, a BRAF inhibitor with or without a MEK inhibitor, on the basis of objective response rate and duration of response in a multicohort study (objective response rate 31.5% among 73 patients treated at the recommended dose, per the FDA announcement)
  • a confirmatory trial is described as ongoing. The company withdrew its EMA application on 22 July 2025 (EMA page read). Vusolimogene oderparepvec (RP1
  • Tudriqev) with nivolumab, 6 August 2026, accelerated approval for adults with unresectable advanced cutaneous melanoma that progressed on a PD-1 blocking antibody-based regimen, based on a single-arm trial (IGNYTE, 140 patients, 91 in the efficacy-evaluable population
  • objective response rate 24.2% (interval 15.8 to 34.3) and median duration of response 14.1 months (interval 10.7 to not reached), per the FDA notice, which does not label the intervals)
  • the FDA convened an advisory committee on 30 July 2026, and post-approval trials are required. Guideline text read: ESMO Clinical Practice Guideline for cutaneous melanoma (Ann Oncol 2025;36(1):10-30
  • approved by the ESMO Guidelines Committee, last update November 2024
  • open-access author manuscript). Recommendations are restated with the guideline's own evidence grades and are not ranked by this site. For second-line treatment it lists ipilimumab with nivolumab [II, B], pembrolizumab [I, A], nivolumab [II, B], ipilimumab [II, B] and, for BRAF-mutant disease, BRAF plus MEK inhibitors [II, B], and lists tumour-infiltrating lymphocyte (TIL) therapy for selected patients [II, B] with the label 'not EMA or FDA approved', a wording that sits alongside the FDA action above and is flagged for the reviewer. The guideline predates the RP1 approval
US Food and Drug Administration 2024European Medicines Agency 2025US Food and Drug Administration 2026Amaral T, Ottaviano M, Arance A, Blank C, Chiarion-Sileni V, Donia M, et al 2025
Locally advanced

Unresectable cutaneous, subcutaneous and nodal lesions. FDA, 27 October 2015 (approval letter, archived copy read):

  • talimogene laherparepvec (Imlygic) for local treatment of unresectable cutaneous, subcutaneous and nodal lesions in patients with melanoma recurrent after initial surgery. Guideline text read: ESMO Clinical Practice Guideline for cutaneous melanoma (Ann Oncol 2025;36(1):10-30
  • approved by the ESMO Guidelines Committee, last update November 2024
  • open-access author manuscript). Recommendations are restated with the guideline's own evidence grades and are not ranked by this site. T-VEC is listed as a local-therapy option for unresectable in-transit metastases [I, B] alongside isolated limb infusion or perfusion, radiotherapy, electrochemotherapy and limited palliative excision, which are not expected to change overall survival
  • the guideline text states that combining T-VEC with PD-1 blockade does not provide additional clinical benefit. EMA action for talimogene laherparepvec was not read in this pass
US Food and Drug Administration, CBER 2015Amaral T, Ottaviano M, Arance A, Blank C, Chiarion-Sileni V, Donia M, et al 2025
Metastatic

Uveal melanoma, which the ESMO cutaneous guideline excludes (it refers to a separate ESMO guideline that was not read). Tebentafusp (Kimmtrak):

  • FDA, 25 January 2022, BLA licence for HLA-A*02:01-positive adults with unresectable or metastatic uveal melanoma
  • EMA marketing authorisation issued 1 April 2022 (CHMP opinion 27 January 2022, per the EMA page). Hepzato Kit (melphalan hepatic delivery system): FDA, 14 August 2023, liver-directed treatment of adults with uveal melanoma with unresectable hepatic metastases affecting less than 50% of the liver and no extrahepatic disease, or extrahepatic disease limited to bone, lymph nodes, subcutaneous tissues or lung that is amenable to resection or radiation. Guideline recommendations for uveal melanoma were not read
  • NCCN and ESMO uveal documents are not cited
US Food and Drug Administration 2022European Medicines Agency 2022U 2023
Any setting

Subcutaneous formulations. Opdivo Qvantig (nivolumab and hyaluronidase-nvhy):

  • FDA letter effective 27 December 2024 (corrected letter e-signed 2 January 2025, effective date unchanged), listing for melanoma unresectable or metastatic disease as monotherapy, monotherapy after combination treatment with intravenous nivolumab and ipilimumab, and adjuvant treatment of completely resected stage IIB, IIC, III or IV melanoma in adults
  • the letter states that it is not indicated in combination with ipilimumab for melanoma. Keytruda Qlex (pembrolizumab and berahyaluronidase alfa-pmph): FDA letter of 19 September 2025 (BLA 761467/Original 1), listing unresectable or metastatic melanoma in adults and adjuvant treatment of stage IIB, IIC or III melanoma after complete resection in adults and pediatric patients 12 years and older
  • the companion Original 2 letter describes Original 1 as covering indications under traditional approval. Neither product's EMA status was read
U 2024U 2025U 2025

Trials to watch

  • INTerpath-001 NCT05933577

    In participants with high-risk stage II to IV melanoma after resection, does adjuvant intismeran autogene (V940, an individualised mRNA neoantigen therapy, formerly mRNA-4157) plus pembrolizumab improve recurrence-free survival compared with placebo plus pembrolizumab? Registry record: phase 3, randomised, double-blind; status active, not recruiting; no results posted.

  • Fianlimab plus cemiplimab versus pembrolizumab NCT05352672

    In adolescents and adults with previously untreated unresectable locally advanced or metastatic melanoma, does fianlimab (an anti-LAG-3 antibody) plus cemiplimab improve progression-free survival compared with pembrolizumab? Registry record: phase 3 with two fianlimab-plus-cemiplimab dose arms, a pembrolizumab-plus-placebo arm and a cemiplimab-plus-placebo arm; status active, not recruiting; no results posted.

  • TILVANCE-301 NCT05727904

    In people with untreated unresectable or metastatic melanoma, does the lifileucel regimen (tumour-infiltrating lymphocytes) plus pembrolizumab improve objective response rate and progression-free survival compared with pembrolizumab alone? Registry record: phase 3, open-label, with optional lifileucel crossover after confirmed progression on pembrolizumab; status recruiting; no results posted.

  • IGNYTE-3 NCT06264180

    In people aged 12 years and older with unresectable stage IIIB to IV cutaneous melanoma that progressed on an anti-PD-1 and an anti-CTLA-4 regimen (or who are not candidates for anti-CTLA-4 therapy), does vusolimogene oderparepvec (RP1) plus nivolumab improve overall survival compared with physician's choice of treatment? Registry record: phase 3, open-label; status recruiting; no results posted.

  • PRISM-MEL-301 NCT06112314

    In HLA-A*02:01-positive adults with previously untreated advanced melanoma, does brenetafusp (IMC-F106C) plus nivolumab improve progression-free survival compared with nivolumab or nivolumab plus relatlimab? Registry record: phase 3, randomised, with low-dose and high-dose brenetafusp arms; status recruiting; no results posted.

  • MelMarT-II NCT03860883

    In AJCC stage II primary cutaneous melanoma, does wide local excision with a 1 cm margin match 2 cm for disease-free survival? Registry record: official title 'a phase III, multi-centre randomised controlled trial' (the registry's phase field reads not applicable); status recruiting; no results posted.

References

Show all 156 references

Loading references… Open the list as a page

Every figure was checked against its primary source to the extent possible. Items that could not be fully confirmed are marked “needs review”. Educational history, not medical advice.

Report an error

Links open the primary source. A flag marked ⚑ needs review means a value or claim has not yet been independently confirmed.

2018TreatmentImmunotherapy

majorAdjuvant

Adjuvant pembrolizumab: EORTC 1325/KEYNOTE-054 (2018)

After surgery for high-risk stage III melanoma, a year of pembrolizumab kept more people free of recurrence than placebo, and the benefit was still seen after a median follow-up of almost seven years. Serious side effects were more common with pembrolizumab. The reports summarised here give no overall survival result.

Phase 3, double-blind; resected stage III melanoma (N=1,019): pembrolizumab 200 mg every 3 weeks (n=514) vs placebo (n=505), up to 18 doses. 12-month RFS 75.4% vs 61.0% (HR 0.57; 98.4% CI 0.43-0.74; P<0.001). Distant metastasis-free survival at 3.5 years 65.3% vs 49.4% (HR 0.60; 95% CI 0.49-0.73). At median 6.9 years, 7-year RFS 50% vs 36% (HR 0.63; 95% CI 0.53-0.74).

Standard before

The trial's control arm was placebo after complete resection.

What changed

FDA approved pembrolizumab on 15 February 2019 for the adjuvant treatment of patients with melanoma with involvement of lymph node(s) following complete resection.

Why it mattered

A PD-1 inhibitor was associated with longer recurrence-free survival than placebo after resection of stage III melanoma, with the benefit sustained in follow-up reports to a median of 6.9 years.

Caveats

  • Placebo-controlled: the trial did not compare pembrolizumab with nivolumab or with targeted therapy, and its hazard ratios are not comparable with those of other trials.
  • Grade 3 to 5 treatment-related adverse events were 14.7% with pembrolizumab and 3.4% with placebo in the primary report, including one treatment-related death (myositis).
  • The primary hazard ratio has a 98.4% confidence interval; eligibility used the 7th edition AJCC staging.
  • No overall survival result appears in the abstracts of the papers cited here.