← All summaries

Melanoma ๐Ÿ›ก๏ธ Preventing recurrence

CheckMate 238 at 9 years: nivolumab beats ipilimumab after melanoma surgery

This nine-year update of a landmark trial confirms that for people with high-risk melanoma removed by surgery, one year of nivolumab kept the cancer away for longer than one year of ipilimumab โ€” and was better tolerated. The overall survival difference remained small and was not statistically significant.

~9 min readPhase IIIMultiple countriesNEJM, 2026

Who is this trial for?

This trial was for adults with high-risk melanoma that had been completely removed by surgery.

People in the trial had:

  • Stage IIIB, IIIC, or IV melanoma that was surgically removed with clear margins (complete resection).
  • Good general health and ability to perform normal daily activities.
  • No prior systemic treatment for melanoma.
What is an immune checkpoint inhibitor? The immune system has built-in “checkpoints” — proteins that stop it from attacking the body’s own cells. Cancer cells can exploit these checkpoints to hide from immune attack. A checkpoint inhibitor is a drug that blocks one of these hiding mechanisms, freeing immune cells to recognise and kill the cancer. Two different checkpoints appear in this trial: PD-1 (targeted by nivolumab) and CTLA-4 (targeted by ipilimumab). Both unleash the immune system, but through different pathways — and with different side-effect profiles. The high dose of ipilimumab used here (10 mg/kg) is associated with substantially more immune-related side effects than nivolumab.

What kind of trial is this?

This trial tested which immunotherapy drug better prevented melanoma from returning after surgery.

Understanding the disease
Finding cancer earlier
Preventing recurrence
Treating the cancer
Feeling better during treatment
First steps in humans
How we make decisions

Background: why researchers asked this question

Melanoma at stages IIIB–C and IV has a high risk of coming back after surgery, even after all visible tumour has been removed. Microscopic cells can remain and later cause the cancer to return.

Ipilimumab (anti-CTLA-4 immunotherapy) was the first approved adjuvant treatment for high-risk resected melanoma. It showed a meaningful benefit over placebo, but at a high dose (10 mg/kg) it caused significant immune-related side effects in many patients, and treatment was often discontinued early.

Nivolumab — a newer PD-1 inhibitor — offered a more targeted way to release the immune system’s brakes. It was already approved for advanced melanoma. CheckMate 238 asked: is nivolumab better than ipilimumab when given for one year after surgery to prevent recurrence?

How nivolumab releases the immune brake Left panel: without nivolumab, the PD-L1 protein on melanoma cells binds PD-1 on a T-cell, switching off the immune attack. Right panel: nivolumab blocks PD-1, breaking that connection and freeing the T-cell to attack remaining melanoma cells. WITHOUT NIVOLUMAB Melanoma cell PD-L1 PD-1 T-cell (blocked) Immune attack: OFF WITH NIVOLUMAB Melanoma cell PD-L1 PD-1 nivolumab T-cell (active) ATTACKS Immune attack: ON
Nivolumab works by blocking PD-1 on immune T-cells. Without it, the PD-L1 protein on melanoma cells uses PD-1 to switch off immune attack. Nivolumab breaks this “immune brake,” allowing T-cells to seek out and destroy any remaining melanoma cells after surgery.

The trial: what was tested and how

CheckMate 238 enrolled 906 people with resected stage IIIB–C or IV melanoma. They were randomly assigned 1:1 to one year of:

  • Nivolumab: 3 mg/kg intravenously every two weeks for up to one year.
  • Ipilimumab: 10 mg/kg intravenously every three weeks for four doses, then every 12 weeks, for up to one year.

The primary measure was recurrence-free survival (RFS) — time until cancer came back or until death. This report represents nine-year follow-up (minimum 107 months) — among the longest follow-up periods for any adjuvant immunotherapy trial in melanoma.

A note before the results

Melanoma was the first cancer type where these two immunotherapies were tested after surgery, and this nine-year update comes from a trial that began more than a decade ago — when we were still discovering the power of checkpoint inhibitors. Before them, we gave a drug called interferon for a year after surgery, with no clear evidence of benefit, and patients literally felt like they had the flu for the entire year. This trial showed us which immunotherapy actually works in the adjuvant setting — nivolumab — and set the stage for its use in many other cancer types since.

Results: what they found

At nine years, nivolumab continued to outperform ipilimumab on recurrence-free survival. The overall survival difference was numerically in favour of nivolumab, but did not reach statistical significance.

Nivolumab vs ipilimumab โ€” 9-year follow-up

Median time without recurrence (RFS)
61.1 moNivolumab vs 24.2 moIpilimumab

On average, people in the nivolumab group went more than five years without cancer returning, compared to two years with ipilimumab. That is a striking difference in a high-risk disease.

Still recurrence-free at 9 years
44%Nivolumab vs 37%Ipilimumab

About 7 percentage points more people remained recurrence-free at 9 years with nivolumab (HR 0.76 — a 24% lower risk of recurrence or death).

Still alive at 9 years (overall survival)
69%Nivolumab vs 65%Ipilimumab

A trend toward better survival with nivolumab, but the 4-percentage-point difference did not reach statistical significance (HR 0.88; 95% CI 0.69–1.11). No new late-onset adverse events were reported at nine years.

Why might the overall survival difference be small if recurrence-free survival is so different?

When cancer recurred in ipilimumab-treated patients, many went on to receive nivolumab as salvage therapy. This “crossover” effect means that some of the survival benefit from nivolumab was “transferred” to the ipilimumab group after recurrence, making the two survival curves look more similar than the recurrence curves. This is common in trials where effective subsequent treatments are available. The recurrence-free survival difference better reflects the direct benefit of one drug over the other.

Look up this trial ClinicalTrials.gov ID NCT02388906
View on ClinicalTrials.gov →

The bottom line

Nine years after the trial began, nivolumab remains clearly superior to ipilimumab for preventing recurrence after high-risk melanoma surgery: 44% vs 37% cancer-free at nine years, with a median recurrence-free survival of more than five years vs two years. The overall survival advantage numerically favours nivolumab but was not statistically significant, likely influenced by subsequent treatments. Nivolumab is now the preferred adjuvant immunotherapy for this population and is better tolerated than high-dose ipilimumab.

What this could mean for you

  • Both drugs are active — this trial compared two immunotherapies, not immunotherapy vs no treatment. The question was which was better.
  • One year of treatment. Adjuvant nivolumab is given for up to 12 months after surgery, then stopped. Long-term protection extends far beyond the treatment period.
  • Side effects differ between drugs. High-dose ipilimumab is more toxic; nivolumab at the dose used here has a more manageable immune-related side-effect profile. Your oncologist will discuss what to watch for.
  • BRAF testing also matters. If your melanoma has a BRAF mutation, targeted therapy (dabrafenib + trametinib) is another option โ€” your team will discuss which approach best suits you.

An oncologist’s perspective

Matt had his cancer diagnosed at age 45, just when his eldest child graduated college, after a mole on his leg grew and spread to some lymph nodes in his groin. He got through surgery just fine, but the risk of spread to distant organs loomed over him like a dark cloud. He jumped on the chance to have a year of adjuvant nivolumab, and although the nivolumab caused an itchy rash and affected his thyroid function, he got through a year of treatment. Every few months he had a scan, with a few sleepless nights afterwards waiting for the all clear. After a couple of years he started to relax as scan after scan was normal. We started chatting about our local sports teams more than we did about his melanoma, and finally, at 10 years, when I discharged him, it felt like saying goodbye to a friend. This is the result we always want to see in cancer — a wave goodbye out of the cancer clinic, because everything will be ok. This is the outcome that drugs like nivolumab have helped make a reality.

For information purposes only. This summary explains published research in plain language. It is not medical advice and is not a substitute for care from your own doctors. Trial results describe what happened in a study group and may not apply to your situation. Always discuss your diagnosis, treatment options, and any clinical trial with your own oncology team before making any decisions.

Questions & comments

Have a question about this trial? Ask below. Questions are read and answered by the site. We can’t give personal medical advice, but we’re glad to explain the research more clearly.

Please don’t include personal health details. Comments are moderated before they appear.