New England Journal of Medicine May 10, 2018

Adjuvant Pembrolizumab versus Placebo in Resected Stage III Melanoma

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

Bottom Line

In patients with completely resected high-risk stage III melanoma, adjuvant pembrolizumab significantly prolonged recurrence-free survival compared to placebo with a manageable safety profile.

Key Findings

1. At a median follow-up of 15 months, pembrolizumab significantly improved the 1-year recurrence-free survival (RFS) rate to 75.4% compared with 61.0% for placebo (HR 0.57; 98.4% CI, 0.43-0.74; P<0.001) [1.1.3].
2. The RFS benefit was maintained in the PD-L1-positive subgroup (comprising 83.5% of the population), with a 1-year RFS of 77.1% for pembrolizumab vs 62.6% for placebo (HR 0.54; 95% CI, 0.42-0.69; P<0.001).
3. Treatment-related adverse events of grade 3 to 5 occurred in 14.7% of the pembrolizumab group versus 3.4% of the placebo group, including one treatment-related death due to myositis in the pembrolizumab arm.
4. Grade 3 or 4 immune-related adverse events occurred in 7.1% of patients receiving pembrolizumab versus 0.6% receiving placebo.

Study Design

Design
Phase 3 RCT
Double-Blind
Sample
1,019
Patients
Duration
15 mo
Median
Setting
Multicenter, Global
Population Patients 18 years or older with completely resected, high-risk stage III melanoma (stage IIIA with at least one lymph node metastasis >1 mm, IIIB, or IIIC without in-transit metastasis).
Intervention Pembrolizumab 200 mg intravenously every 3 weeks for up to 1 year (18 doses).
Comparator Placebo intravenously every 3 weeks for up to 1 year (18 doses).
Outcome Recurrence-free survival (RFS) in the overall intention-to-treat population and in the PD-L1-positive subgroup.

Study Limitations

The median follow-up of 15 months at the time of the primary publication was relatively short, leaving long-term outcomes and overall survival (OS) data immature.
The trial restricted enrollment to stage IIIA patients with lymph node metastases >1 mm, excluding those with lower-burden stage IIIA or high-risk stage II melanoma.
A placebo comparator was used rather than an active control like ipilimumab or interferon, reflecting the lack of a universally accepted, well-tolerated standard of care at the time of study design.

Clinical Significance

Pembrolizumab reduced the risk of disease recurrence or death by 43% compared to placebo, establishing adjuvant anti-PD-1 therapy as a highly effective, standard-of-care option for patients with resected high-risk stage III melanoma.

Historical Context

Before the advent of modern immune checkpoint inhibitors, adjuvant options for high-risk melanoma, such as high-dose interferon alfa and ipilimumab, were limited by severe toxicities and varying efficacy. While the CheckMate 238 trial established nivolumab as superior to ipilimumab, KEYNOTE-054 parallelly compared pembrolizumab against a placebo control. Together, these milestone trials decisively shifted the adjuvant landscape for melanoma to anti-PD-1 monotherapy, demonstrating that early systemic intervention significantly delays disease recurrence.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

How does the mechanism of action of pembrolizumab differ from targeted therapies like BRAF inhibitors, and why is an immunotherapy theoretically advantageous in the adjuvant setting for a patient with resected stage III melanoma?

Key Response

Pembrolizumab is an anti-PD-1 monoclonal antibody that blocks the inhibitory interaction between PD-1 on T-cells and PD-L1/2 on tumor cells, reinvigorating exhausted T-cells to attack micrometastatic disease. Unlike BRAF inhibitors, which directly target a specific kinase mutation in the tumor but often face acquired resistance, immunotherapy can induce a durable, long-term immunologic memory, which is the primary goal of adjuvant therapy to prevent distant recurrence.

Resident
Resident

Based on the KEYNOTE-054 trial, what are the most common immune-related adverse events (irAEs) you must monitor for in a patient receiving adjuvant pembrolizumab, and what is the general management strategy if a patient develops grade 3 colitis?

Key Response

Residents must recognize that pembrolizumab can cause irAEs such as thyroiditis, pneumonitis, colitis, and hepatitis. For a grade 3 toxicity like severe colitis, the management involves immediately holding the pembrolizumab, ruling out infectious causes, and initiating high-dose systemic corticosteroids (e.g., 1-2 mg/kg/day of prednisone or equivalent), with careful tapering over several weeks.

Fellow
Fellow

The KEYNOTE-054 trial included both BRAF-mutated and BRAF-wild-type patients. How did the efficacy of adjuvant pembrolizumab compare between these subgroups, and in the absence of head-to-head data, how do you choose between adjuvant anti-PD-1 therapy and targeted therapy (e.g., dabrafenib/trametinib) for a BRAF-mutated stage III melanoma patient?

Key Response

Subgroup analysis in KEYNOTE-054 showed significant recurrence-free survival (RFS) benefit regardless of BRAF mutation status. For a BRAF-mutated patient, fellows must weigh the choice between targeted therapy (rapid onset, finite toxicity but high rates of pyrexia) and immunotherapy (potential for permanent autoimmune toxicities like endocrinopathies but potentially deeper durable memory). Decisions often hinge on patient preference, comorbidities, and toxicity tolerance.

Attending
Attending

While KEYNOTE-054 demonstrated a significant improvement in RFS, overall survival (OS) data in adjuvant melanoma trials often takes years to mature. How do you counsel patients on the distinction between RFS and OS, and how do you justify the risk of lifelong, permanent endocrinopathies for an RFS benefit that might not translate to an OS advantage due to highly effective salvage therapies at relapse?

Key Response

Attendings must navigate the difficult risk-benefit discussion of adjuvant therapy. RFS means delaying recurrence, which provides profound psychological relief and prevents morbid surgeries. However, because metastatic melanoma can now be salvaged effectively with dual immunotherapy, some patients cured by surgery alone might be unnecessarily exposed to permanent toxicities (like type 1 diabetes or hypothyroidism). Shared decision-making is critical.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The trial protocol allowed patients in the placebo arm to cross over to receive pembrolizumab upon disease recurrence. How does this crossover design statistically dilute the ability to detect an Overall Survival (OS) difference, and what statistical modeling techniques are most appropriate to estimate the true treatment effect on OS?

Key Response

Crossover is ethically sound but complicates long-term endpoint analysis, as the control arm effectively receives the active drug, narrowing the OS gap. PhDs and methodologists must employ techniques like the Rank Preserving Structural Failure Time (RPSFT) model or Inverse Probability of Censoring Weighting (IPCW) to adjust for crossover and estimate the true OS benefit of early versus delayed pembrolizumab.

Journal Editor
Journal Editor

Given that high-dose interferon and adjuvant ipilimumab were already approved for high-risk stage III melanoma when KEYNOTE-054 was initiated, how would you evaluate the ethical and methodological justification of using a placebo control rather than an active comparator?

Key Response

A critical reviewer would scrutinize the choice of the control arm. The justification relies on the fact that high-dose interferon was highly toxic with marginal survival benefit, and adjuvant ipilimumab (from EORTC 18071) had severe, often fatal toxicity profiles, making them poorly adopted globally. Therefore, placebo remained an ethically acceptable and methodologically clean standard for many regulatory agencies at the time.

Guideline Committee
Guideline Committee

Integrating the robust RFS benefit from KEYNOTE-054 and the comparable results from CheckMate 238 (nivolumab), how should current NCCN and ESMO guidelines classify the strength of recommendation for adjuvant anti-PD-1 therapy, and what happens to the prior standard of adjuvant ipilimumab?

Key Response

Guideline committees elevated adjuvant anti-PD-1 (pembrolizumab and nivolumab) to Category 1 recommendations for resected high-risk stage III melanoma based on these trials. Because of the superior efficacy-to-toxicity ratio compared to ipilimumab, anti-PD-1 therapy effectively replaced ipilimumab and interferon in both NCCN and ESMO guidelines, rendering older therapies obsolete in the adjuvant setting.

Clinical Landscape

Noteworthy Related Trials

2015

EORTC 18071

n = 951 · Lancet Oncol

Tested

Ipilimumab 10 mg/kg

Population

Resected stage III melanoma

Comparator

Placebo

Endpoint

Recurrence-free survival

Key result: Adjuvant ipilimumab significantly improved recurrence-free and overall survival versus placebo but was associated with severe immune-related adverse events.
2017

CheckMate 238

n = 906 · NEJM

Tested

Nivolumab

Population

Resected stage III or IV melanoma

Comparator

Ipilimumab

Endpoint

Recurrence-free survival

Key result: Nivolumab significantly improved recurrence-free survival compared to ipilimumab and had a much better safety profile.
2017

COMBI-AD Trial

n = 870 · NEJM

Tested

Dabrafenib plus Trametinib

Population

Resected stage III melanoma with BRAF V600 mutations

Comparator

Placebo

Endpoint

Relapse-free survival

Key result: The combination of dabrafenib and trametinib significantly lowered the risk of disease recurrence or death compared to 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