The Lancet Oncology APRIL 12, 2021

Adjuvant pembrolizumab versus placebo in resected stage III melanoma (EORTC 1325-MG/KEYNOTE-054): distant metastasis-free survival results from a double-blind, randomised, controlled, phase 3 trial

Alexander M. M. Eggermont, Christian U. Blank, Mario Mandala, et al.

Bottom Line

In patients with completely resected, high-risk stage III melanoma, adjuvant therapy with pembrolizumab significantly improves recurrence-free and distant metastasis-free survival compared to placebo.

Key Findings

1. At a median follow-up of 3.5 years, adjuvant pembrolizumab significantly reduced the risk of distant metastasis or death by 40% compared to placebo (hazard ratio [HR], 0.60; 95% CI, 0.49–0.73; P < .0001).
2. The 3.5-year distant metastasis-free survival rate was 65.3% in the pembrolizumab arm versus 49.4% in the placebo arm.
3. Recurrence-free survival (RFS) benefit was sustained at 3.5 years, with an HR of 0.59 (95% CI, 0.49–0.70; P < .0001) and 3.5-year RFS rates of 59.8% for pembrolizumab compared to 41.4% for placebo.
4. The survival benefit was consistent across key clinical subgroups, including stage IIIA-IIIC, PD-L1 status, and BRAF-V600 mutation status.

Study Design

Design
RCT
Double-Blind
Sample
1,019
Patients
Duration
3.5 yr
Median
Setting
Multicenter, international
Population Adults with resected high-risk stage IIIA, IIIB, or IIIC cutaneous melanoma
Intervention Pembrolizumab 200 mg intravenously every 3 weeks for up to 18 doses
Comparator Placebo every 3 weeks for up to 18 doses
Outcome Recurrence-free survival (RFS)

Study Limitations

Overall survival data remain immature, precluding definitive conclusions on whether adjuvant pembrolizumab confers a survival advantage.
The study did not compare pembrolizumab to other systemic adjuvant options such as targeted therapy (BRAF/MEK inhibitors) for BRAF-mutated patients.
The risk of immune-related adverse events associated with prolonged PD-1 blockade requires careful consideration in a curative-intent setting.

Clinical Significance

This trial established pembrolizumab as a standard-of-care adjuvant treatment for high-risk resected stage III melanoma, demonstrating that immune checkpoint inhibition can effectively delay disease recurrence and distant metastatic progression.

Historical Context

Prior to the success of checkpoint inhibitors like pembrolizumab, adjuvant treatment options for high-risk melanoma were limited, with interferon-alpha showing only modest benefits at the cost of substantial toxicity. The KEYNOTE-054 trial was pivotal in shifting the paradigm toward targeted immunotherapy in the adjuvant setting.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

How does the mechanism of action of pembrolizumab differ from traditional cytotoxic chemotherapy in the management of resected stage III melanoma?

Key Response

Pembrolizumab is a humanized monoclonal antibody that blocks the PD-1 receptor on T-cells, preventing it from binding to PD-L1/PD-L2 on tumor cells. This 'releases the brakes' on the immune system to allow T-cell mediated killing of residual micrometastatic disease. In contrast, chemotherapy targets rapidly dividing cells via DNA damage or microtubule disruption. In melanoma, immunotherapy has proven far more effective than traditional chemotherapy because melanoma is highly immunogenic due to a high tumor mutational burden.

Resident
Resident

Based on the KEYNOTE-054 inclusion criteria, which specific subgroups of stage III melanoma patients should be considered for adjuvant pembrolizumab, and how does the AJCC 7th vs 8th edition staging impact this decision?

Key Response

The trial enrolled patients with completely resected stage IIIA (at least one nodal metastasis >1mm), IIIB, and IIIC disease based on AJCC-7. Residents must recognize that AJCC-8 introduced a stage IIID and redefined IIIA to include nodes <1mm. Clinical reasoning is required because the benefit in the 'low-risk' IIIA group (nodes <1mm) is less certain, as they were excluded from the primary trial population, meaning clinicians must weigh the 14% risk of grade 3-5 adverse events against a lower absolute risk of recurrence.

Fellow
Fellow

The KEYNOTE-054 trial demonstrated significant improvement in Distant Metastasis-Free Survival (DMFS). How should this influence your choice between adjuvant pembrolizumab and BRAF/MEK inhibitors (e.g., dabrafenib/trametinib) for a patient with a BRAF V600E mutation?

Key Response

Both PD-1 inhibitors and BRAF/MEK inhibitors (COMBI-AD trial) show adjuvant benefit. Fellows must integrate the evidence: PD-1 inhibitors like pembrolizumab offer the potential for 'treatment-free' long-term survival due to immune memory and a durable plateau in the survival curve, whereas BRAF/MEK inhibitors have a high early efficacy but the risk of relapse may increase after the 12-month treatment course ends. DMFS is a key metric because it is a strong surrogate for overall survival in melanoma.

Attending
Attending

A unique feature of KEYNOTE-054 was the provision for crossover or re-challenge with pembrolizumab upon recurrence. How does this trial design feature impact the interpretation of Overall Survival (OS) data and the clinical debate regarding 'upfront' adjuvant therapy versus 'salvage' therapy at the time of relapse?

Key Response

The crossover design is critical for determining if adjuvant therapy actually saves lives or if it simply delays recurrence that could have been treated just as effectively at the time of relapse. If the OS curves eventually merge, it suggests that the timing of PD-1 inhibition (adjuvant vs. therapeutic) may be less critical than the exposure to the drug itself. However, the DMFS benefit suggests that preventing the first distant metastasis is a superior clinical strategy to waiting for systemic spread.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

Critique the use of Distant Metastasis-Free Survival (DMFS) as the primary endpoint in KEYNOTE-054. What statistical validation is required to accept DMFS as a true surrogate for Overall Survival (OS) in the era of modern immunotherapy?

Key Response

To validate a surrogate endpoint, one must demonstrate a high correlation coefficient (R-squared) between the treatment effect on the surrogate and the treatment effect on the true endpoint (OS). In melanoma, adjuvant OS is difficult to reach due to effective subsequent lines of therapy (confounding). PhD researchers look for 'Prentice criteria' satisfaction and evaluate whether the DMFS benefit captures the full 'net clinical benefit' of the drug, including the avoidance of the morbidity associated with treating visceral metastases.

Journal Editor
Journal Editor

The control arm of KEYNOTE-054 used a placebo despite ipilimumab being FDA-approved for adjuvant melanoma at the time of the study. Does this 'historical' control strategy compromise the study's internal validity or its relevance to the current standard of care?

Key Response

Editors look for the 'appropriate comparator.' While ipilimumab was approved, its severe toxicity (54% grade 3-4 AEs at 10mg/kg) meant it was not universally adopted as a standard. While a head-to-head trial (like CheckMate-238) provides higher comparative evidence, the use of a placebo in KEYNOTE-054 allowed for a clear determination of the absolute benefit and safety profile of pembrolizumab, which was necessary given the context of the high toxicity of the only other approved agent at the time.

Guideline Committee
Guideline Committee

Considering the results of KEYNOTE-054, should adjuvant pembrolizumab be recommended for stage IIIA patients with a sentinel lymph node metastasis of 0.5mm? How does this align with current NCCN or ESMO guidelines?

Key Response

Current guidelines (NCCN Category 1) recommend pembrolizumab for stage III disease. However, KEYNOTE-054 specifically required a nodal deposit >1mm for stage IIIA. The committee must decide whether to extrapolate the benefit to the 'ultra-low-risk' IIIA group (AJCC-8 <1mm) or to advise observation. Most guidelines currently suggest that for patients with very low-volume N1a disease (<1mm), the risk-benefit ratio of immunotherapy is less clear, and shared decision-making is essential since their 10-year melanoma-specific survival with surgery alone exceeds 90%.

Clinical Landscape

Noteworthy Related Trials

2010

EORTC 18071 Trial

n = 951 · NEJM

Tested

Ipilimumab 10 mg/kg every 3 weeks for 4 doses, then every 3 months

Population

Patients with stage III melanoma who had undergone complete resection

Comparator

Placebo

Endpoint

Recurrence-free survival

Key result: Adjuvant ipilimumab therapy significantly improved recurrence-free survival and overall survival compared with placebo in patients with resected stage III melanoma.
2017

CheckMate 238 Trial

n = 906 · NEJM

Tested

Nivolumab 3 mg/kg every 2 weeks

Population

Resected stage IIIb, IIIc, or IV melanoma

Comparator

Ipilimumab 10 mg/kg every 3 weeks

Endpoint

Recurrence-free survival

Key result: Nivolumab resulted in significantly longer recurrence-free survival than ipilimumab in patients with resected stage III or IV melanoma.
2017

COMBI-AD Trial

n = 870 · NEJM

Tested

Dabrafenib 150 mg twice daily plus trametinib 2 mg once daily

Population

Patients with resected stage III melanoma with BRAF V600E or V600K mutations

Comparator

Placebo

Endpoint

Relapse-free survival

Key result: Adjuvant treatment with dabrafenib plus trametinib led to significantly higher rates of relapse-free survival and overall survival than placebo.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis