Applying Clinical Trials to Patient Cases: The Melanoma Landscape

The panel.

As part of a Satellite Sessions program hosted by CancerNetwork®, a group of experts discussed the current landscape of melanoma treatment, ranging from relevant clinical trials to patient cases. Of particular interest were the phase 3 CheckMate 067 (NCT01844505), RELATIVITY-047 (NCT03470922), and NADINA (NCT04949113) trials.

Adil Daud, MD, HS clinical professor in the Department of Medicine (hematology/oncology) at the University of California, San Francisco (UCSF), and director of melanoma clinical research at the UCSF Helen Diller Family Comprehensive Cancer Center, led the panel. His fellow panelists were Bartosz Chmielowski, MD, PhD, a clinical professor of medicine in the Division of Hematology-Oncology at the UCLA Jonsson Comprehensive Cancer Center at UCLA Health in Los Angeles, California; Gregory Daniels, MD, PhD, a medical oncologist and professor of medicine at UC San Diego Health in California; Warren Chow, MD, clinical professor in the Division of Hematology/Oncology in the School of Medicine and associate director for clinical sciences in the UCI Health Chao Family Comprehensive Cancer Center in Orange, California; and Thuy Tran, MD, PhD, assistant professor of medicine (medical oncology) at Yale School of Medicine in New Haven, Connecticut.

CheckMate 067

Daud / The 10-year data from CheckMate 067 still show a persistent advantage for ipilimumab [Yervoy] plus nivolumab [Opdivo]. The first few times, the statistical difference between the ipilimumab plus nivolumab arm and the nivolumab-alone arm hasn’t been there. We’ve all been thinking that this is probably going to be positive at some point, but the most recent data do show that you still have this benefit for the ipilimumab plus nivolumab arm with overall survival [OS] and melanoma-specific survival. There is not a clear signal for a definite type of melanoma that seemingly does not benefit [between] both the mutant and wild type. There was a point where the mutant did look like it was better than the wild type, but this most recent analysis doesn’t show that.1 The PD-L1–positive/negative and [lactate dehydrogenase] LDH-high/low [subgroups] don’t show any differences.

Chmielowski / But I think it was different because nivolumab did not do well in BRAF-mutant [disease] alone. The curves looked like they were separated because nivolumab was doing worse in BRAF-mutant disease than in BRAF wild type.

Daud / Do you make judgments based on BRAF-mutant [status]?

Chmielowski / There was this JAMA paper [that said], “You are showing [these] data [where], if PD-L1 expression is higher, there is not much of a benefit from adding ipilimumab. We are showing that, with BRAF wild type, the difference between nivolumab alone and nivolumab plus ipilimumab is smaller. Should you not be using [these] data and saving patients from toxicities and financial toxicity…?”2 [For now,] I still use the combination.

Daud / Is there a class of patients that you’re treating with single-agent PD-1 [inhibitors]? If there’s an older [man] who has lung metastases, is PD-L1 high, and doesn’t have any liver or brain involvement, if he’s 70 or 80 years old, I often will give those people a single-agent [inhibitor].

Chmielowski / I often do. Although I’m treating an [89-year-old patient] with ipilimumab plus nivolumab. But yes, if the disease is a little too aggressive, either we are going to have a response or he’s going to be dead…. Yes, I do use a single agent.

Ipilimumab is given only for 4 doses. After these 4 doses—so after the first 9 weeks of therapy—both arms are identical, but the curves do not get closer. Even 5 or 7 years later, the same separation remains. The message is if I choose the wrong therapy in the first-line setting, I will never be able to catch up. This is what I think when I am selecting, because if I select the wrong one, I will always be on a worse curve, no matter what I do later. Patients on the nivolumab arm had the chance to get ipilimumab [later] if they were in shape, but they did not catch up.

Daniels / [Patients who received ipilimumab] never caught up to the [patients who received nivolumab]?

Chmielowski / Correct. It’s only 9 weeks of therapy, and we don’t see separation early when this therapy is given. We start seeing separation around 16 months into the trial. It means your first-line treatment had an influence on subsequent lines of treatment. Looking at progression-free survival 2 [PFS2]—like on the nivolumab plus relatlimab [Opdualag] trial—PFS2 was better in patients who got nivolumab plus relatlimab in the first line than in patients who got nivolumab alone.

RELATIVITY-047

Daud / [There’s] the RELATIVITY study, which we all know is nivolumab alone vs the nivolumab plus relatlimab combination…. PFS is 10.2 months vs 4.6 months; OS, according to the most recent analysis, is 51 months vs 34 months.3 It’s interesting how the response rate and the OS for nivolumab alone in the nivolumab plus relatlimab study seemed to be underperforming the nivolumab-alone arm of the CheckMate 067 study. Maybe that’s just an artifact of the different populations, although there wasn’t an obvious worsening of the population that I could tell.

Chmielowski / I was paying attention to the same thing. Also, nivolumab in this trial was less toxic than nivolumab in the CheckMate 067 trial. When you look at the frequency of adverse events [AEs], there were fewer AEs with nivolumab alone in this trial, which maybe reflects that people are more familiar with the management [of them]. But there was a follow-up [Journal of Clinical Oncology] paper where they weighted the trials and compared them with each other, and when they selected the patients based on all the characteristics and matched them, they got identical results.4

Daud / This makes sense because it would be hard to imagine that nivolumab is just losing its effectiveness as you do more studies….

The most recent OS data we have are 51 months for the nivolumab plus relatlimab arm and 34 months for the nivolumab- alone arm. Melanoma-specific survival also looks like it’s superior with nivolumab plus relatlimab, although it hasn’t been reached yet. Most interestingly, the response data are about 10% better for nivolumab plus relatlimab.

Safety

Daud / Looking at the safety summary…for grade 3 or 4 AEs, it’s 62% for nivolumab plus relatlimab vs 46% for nivolumab plus relatlimab, and it is relatively similar for nivolumab alone…. What are your impressions of the updated RELATIVITY data, and what practice do you consider when choosing dual therapy? What are you focusing on?

Chow / In the reanalysis published in NEJM Evidence [in 2023], they looked at the subset, and the forest plots favored nivolumab plus relatlimab if you had a PD-L1 [expression] of less than 1%.5 That’s an important factor to consider if you go with single-agent nivolumab vs the combination of nivolumab plus relatlimab.

Daud / If you had a PD-L1 [expression] that was high, [would you give nivolumab alone]?

Chow / Yes. If this is an older patient with other comorbid conditions, you don’t want to give ipilimumab. If you have a high PD-L1 [expression], the benefit of adding the relatlimab is not so great. You could consider just giving single-agent nivolumab to those patients with a high PD-L1 expression.

Daud / Is your default regimen at this point nivolumab plus relatlimab or nivolumab plus ipilimumab, or is it one of the different single-agent combinations? What are you using to decide?

Chow / ECOG [performance status] is a big differentiator in comorbid conditions. If they’re older and they have a lot of comorbidities, I don’t like to give nivolumab plus ipilimumab. For the older patients—over 70 [years]—the decision point for me is the PD-L1 expression.

A Patient Case

Daud / Our last case is a 56-year-old [patient]. The patient had COVID-19—which may or may not be connected—and had a left submandibular lymph node and NRAS mutation with a high [tumor mutation burden], so it was skin to begin with. Would you give him pembrolizumab [Keytruda], nivolumab plus ipilimumab, or send him to surgery?

Daniels / I’d give him nivolumab plus ipilimumab.

Chmielowski / I would give [the NADINA regimen].

Chow / NADINA.

Daud / NADINA, you would do that? It’s funny, I saw 2 patients of the same type on the same day. For whatever reason, I gave one of them nivolumab plus ipilimumab, and I put [the other] on a clinical trial with relatlimab—it was the [phase 3 Regeneron study (NCT06246916)] with the [fianlimab and cemiplimab (Libtayo)] combination. The [patient who received] ipilimumab plus nivolumab had 7% residual tumor…. Would you stick with the NADINA data?

Daniels / Yes, that’s done.

Chow / The data are the data.

Daud / I haven’t decided what to do. I’m thinking about whether to give him a little bit more ipilimumab plus nivolumab. I might have led him to think that the results were going to be good because his scans were so great. But he still had that little bit of residual tumor left.

Chmielowski / It’s difficult to make it better. That’s why I would stop. Also, the argument is to stop because I sense that if he recurs, he will respond again.

Daud / True. As long as you’re keeping close tabs on him, [and monitor him for any adverse reactions].

Daniels / For a [patient] like this, are you putting in clips and taking out a single node, or are you doing the full neck?

Daud / With this [patient], I did put a clip in, but I haven’t been consistently putting a clip in. There was another [man] who was a pathologist, and that [patient] had a forehead melanoma and bilateral lymph nodes. He’s had 4 cycles of pembrolizumab, and now, just when I had the bright idea of putting clips in, the lymph nodes are undetectable. I sent him to get a Magseed, and they [said], “What do you want us to put a Magseed in? There’s nothing.” What are you doing? Are you putting clips or a Magseed in?

Daniels / I’m starting to put clips in. If there’s hesitation, I’ll wait, and if I start seeing a response, I’ll get them in quick…. NADINA didn’t clip them. They did a full dissection.

Tran / I usually try to get them clipped before starting neoadjuvant immunotherapy. But in spite of that, in the NADINA trial, of the 198 patients who underwent surgery, 197 of them—which is almost 99.9% of the patients—had a total lymph node dissection. They didn’t get a limited dissection. Only 1 had an index lymph node removed. Technically, if you’re going to follow the NADINA trial, we should be doing the total therapeutic lymph node dissection….

Tran / In my practice, I discuss this with my patients. I tell them that yes, even though in the purest form, we technically do the total therapeutic lymph node dissection, the data are evolving. We can clip the lymph nodes. We could do limited. [There’s] less morbidity. I go through the options with the patients and tell them, “Yes, the standard of care is still technically a total therapeutic dissection, but limited [therapeutic dissection] is an option.” Especially since I always rescan the patients after neoadjuvant immunotherapy. If I see the lymph nodes completely disappearing, and I have already clipped the lymph node, it isn’t unreasonable to offer a limited dissection.

Daud / What do you do for a [patient] who started with bilateral lymph nodes and has had a great clinical response? Would you do a limited sampling of both necks? Or would you go full bilateral neck if the patient is willing to do that? With neck dissections, you do get these seromas and some serious morbidity if you do a multilevel neck dissection.

Tran / It’s hard. If a patient started with a high disease burden, then I might be more inclined to do a bigger dissection vs someone with very limited disease from the get-go. [If] they have a great response with immunotherapy, I would lean toward considering doing limited [therapeutic dissection]. But I don’t think the data are out there yet.

References

1. Wolchok JD, Chiarion-Sileni V, Rutkowski P, et al. Final, 10-year outcomes with nivolumab plus ipilimumab in advanced melanoma. N Engl J Med. 2025;392(1):11-22. doi:10.1056/NEJMoa2407417
2. Courtney PT, Yip AT, Cherry DR, Salans MA, Kumar A, Murphy JD. Cost-effectiveness of nivolumab-ipilimumab combination therapy for the treatment of advanced non–small cell lung cancer. JAMA Netw Open. 2021;4(5):e218787. doi:10.1001/jamanetworkopen.2021.8787
3. Tawbi HA, Stephen Hodi F, Lipson EJ, et al. Nivolumab (NIVO) plus relatlimab (RELA) vs NIVO in previously untreated metastatic or unresectable melanoma (RELATIVITY-047): overall survival (OS) and melanoma-specific survival (MSS) outcomes at 3 years. J Clin Oncol. 2024;42(suppl 16):9524. doi:10.1200/JCO.2024.42.16_suppl.9524
4. Long GV, Lipson EJ, Stephen Hodi F, et al. First-line nivolumab plus relatlimab versus nivolumab plus ipilimumab in advanced melanoma: an indirect treatment comparison using RELATIVITY-047 and CheckMate 067 trial data. J Clin Oncol. 2024;42(33):3926-3934. doi:10.1200/JCO.24.01125
5. Long GV, Stephen Hodi F, Lipson EJ, et al. Overall survival and response with nivolumab and relatlimab in advanced melanoma. NEJM Evid. 2023;2(4):EVIDoa2200239. doi:10.1056/EVIDoa2200239