Welcome to the CCO Site

Thank you for your interest in CCO content. As a guest, please complete the following information fields. These data help ensure our continued delivery of impactful education. 

Become a member (or login)? Member benefits include accreditation certificates, downloadable slides, and decision support tools.


Key Studies in Skin Cancer: Independent Conference Coverage of the 2021 ASCO Scientific Meeting

Michael A. Davies, MD, PhD
Evan J. Lipson, MD
Released: August 4, 2021

Key Studies in Melanoma

CheckMate 067: Long-term Follow-up of Nivolumab/Ipilimumab vs Nivolumab vs Ipilimumab

Michael A. Davies, MD, PhD:
One of the highlights of ASCO 2021 for melanoma was the follow-up data from the phase III CheckMate 067 trial of immunotherapy in patients with previously untreated, unresectable stage III or IV melanoma (N = 945).1 Patients were randomized to the control arm of ipilimumab 3 mg/kg plus placebo that was compared with either nivolumab 3 mg/kg plus placebo or ipilimumab 3 kg/mg plus nivolumab 1 mg/kg followed by nivolumab 3 mg/kg. The coprimary endpoints were PFS and OS.

CheckMate 067 6.5-Year Follow-up: OS

Michael A. Davies, MD, PhD:
This study’s results have been closely watched since they were initially posted in 2013. I think the most exciting and clinically impactful data presented at ASCO 2021 are the OS results.

For patients treated with ipilimumab plus nivolumab, the median OS was just over 6 years at 72.1 months (HR: 0.52 vs ipilimumab alone). Median OS with nivolumab alone was 36.9 months, and with ipilimumab alone, it was 19.9 months. At the 6-year benchmark, OS rates were 50%, 43%, and 23%, respectively.

These results are especially noteworthy, given that approximately a decade ago, the median survival for stage IV patients was 6 months and long-term survival was <10%.2 This remarkable improvement sets the benchmark for future advances.

CheckMate 067 6.5-Year Follow-up: PFS

Michael A. Davies, MD, PhD:
The median PFS was 11.5 months with the combined immunotherapy regimen vs 2.9 months with ipilimumab alone. By the third year, the curve is flattening, emphasizing the durability of therapy. Nevertheless, long-term PFS with either the combination of ipilimumab plus nivolumab or nivolumab alone is in the 30% to 35% range, meaning that approximately 70% of patients treated either with nivolumab or ipilimumab plus nivolumab will still develop disease progression at some point, a reminder of the continuing need for improvement.

A key question remains: Since combination therapy does have much higher rates of toxicity, and many patients are doing well with single-agent PD-1 inhibition, which patients benefit the most from combination therapy?

CheckMate 067 6.5-Year Follow-up: OS Outcomes by BRAF Mutation Status

Michael A. Davies, MD, PhD:
The BRAF V600 mutation is present in 40% to 45% of patients with of cutaneous melanoma.3 The graph on the left shows an absolute difference of almost 15% in OS at 5 and 6 years for patients with the BRAF mutation who were treated with ipilimumab plus nivolumab compared with nivolumab alone and 30% compared with ipilimumab alone.

By contrast, the graph for patients with wild-type BRAF shows very similar OS curves for ipilimumab plus nivolumab vs nivolumab alone. This suggests that most of the benefit from the combined therapy is in patients with a BRAF V600 mutation. 

Evan J. Lipson, MD:
It is important to note that the trial was not statistically designed to compare these groups, and caution should be used when interpreting these data. The numbers of patients in these subgroups are relatively small, and the results should be considered descriptive, but they do paint a picture that may help guide clinical decision making.

Michael A. Davies, MD, PhD:
Keep in mind, too, that this is a univariate analysis. Other markers may also predict differential outcomes from these treatments. An example is PD-L1–negative vs PD-L1–positive tumors, with the former showing more PFS and OS benefits from the combination treatment vs nivolumab alone, and the latter showing no apparent difference between regimens4—again, in a univariate analysis.

Many different factors are considered when deciding among therapeutic options. Multivariate analyses that consider these factors together would help healthcare professionals understand which treatments or combinations of treatments would be the most helpful for different subgroups of patients.

CheckMate 067 6.5-Year Follow-up: Patients Alive and Treatment Free

Michael A. Davies, MD, PhD:
These pie charts show how many patients in all 3 treatment arms are not only alive, but also treatment free. This extends previous results and illustrates that a high percentage of patients who received ipilimumab plus nivolumab who are alive are also off therapy. These results underscore the potential for long-term disease control and OS with this regimen, even with very short-term therapy. This trend is also seen with nivolumab alone, but a larger number of patients received subsequent therapy in this arm. 

These greatly improved outcomes can be achieved with a relatively short period of treatment with ipilimumab plus nivolumab. At the same time, many patients have a short term of treatment because of the toxicity associated with this regimen. 

Evan J. Lipson, MD:
This is a testament to the power of immunotherapy in general and the immune system’s memory and capacity to fight cancer, specifically. Over a relatively brief period of time, these patients received immune-modulating therapy; once their immune systems were “empowered to fight cancer,” they continued to do so for long periods of time.

This powerful characteristic of the immune system was evident in one of the large, early studies of nivolumab monotherapy, dating back to 2012.5 Some of those patients are still alive after stopping therapy almost a decade ago.

CheckMate 067 6.5-Year Follow-up: Safety

Michael A. Davies, MD, PhD:
This table shows that no new safety issues have emerged in extended follow-up of CheckMate 067. With ipilimumab plus nivolumab, approximately 60% of patients experienced grade 3/4 treatment-related adverse events (TRAEs), and 31% discontinued due to those toxicities. This is much higher than is seen with nivolumab alone: 24% of patients experienced grade 3/4 TRAEs and 8% discontinued therapy.

CheckMate 067 6.5-Year Follow-up: Conclusions

Michael A. Davies, MD, PhD:
We are excited about the tremendous long-term outcomes seen with these treatments, particularly with ipilimumab plus nivolumab. It also may be possible to achieve similar outcomes, either by using alternative therapeutic approaches with less toxicity or by employing biomarkers that help healthcare professionals identify which patients would benefit most.

The CheckMate 067 results set the benchmark for improvements in therapeutic regimens for advanced melanoma. The major milestone in this study is the determination of median OS for all 3 treatment arms. After 6.5 years, the data show a median OS of 72.1 months in patients treated with the combination of nivolumab and ipilimumab. With fewer than one half of these patients receiving subsequent therapy, the median time to subsequent therapy has not yet been reached. In comparison, patients in the single-agent arms saw a median OS of 36.9 months with nivolumab and 19.9 months with ipilimumab. The median time to subsequent therapy for these arms was 25.2 months and 8.0 months, respectively. These OS data set the benchmark for new therapeutic regimens.

After 6.5 years, 77% of patients who received nivolumab plus ipilimumab and 69% of patients who received nivolumab alone remained alive and off treatment. In addition, safety findings remained consistent with previous reports, and all subgroups, including patients with baseline liver metastases and BRAF-mutant tumors, saw durable clinical benefit.

ABC Study: Nivolumab With or Without Ipilimumab in Patients With Melanoma Brain Metastases

Michael A. Davies, MD, PhD:
The ABC Study was conducted in Australia between 2014 and 2017.6,7 This was a multicenter, open-label, randomized phase II trial of nivolumab alone or with ipilimumab in immunotherapy-naive adult patients with melanoma brain metastases (N = 76).

Patients with asymptomatic metastases and no previous local brain therapy were randomly assigned to cohort A (nivolumab 1 mg/kg with ipilimumab 3 mg/kg, then nivolumab 3 mg/kg) or cohort B (nivolumab 3 mg/kg). Patients with symptomatic metastases, who had received previous therapy, or who experienced leptomeningeal disease with MRI progression were enrolled in the nonrandomized cohort C (nivolumab 3 mg/kg).

The primary endpoint was intracranial responses at Week 12 and beyond, with secondary endpoints including extracranial response rate, ORR, PFS, and OS.

There were 35 and 25 patients in the asymptomatic cohorts A and B, respectively, and 16 in the symptomatic, previously treated group. These are relatively small numbers, but still informative. 

ABC Study: Baseline Characteristics

Michael A. Davies, MD, PhD:
The updated results7 reported at ASCO 2021 reinforce earlier evidence that patients with asymptomatic brain metastases benefit most from the combination of nivolumab and ipilimumab.8

Baseline characteristics included a median age of 51-63 years, all but 2 patients had an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0/1, and approximately one half of the patients in the asymptomatic cohorts had the BRAF V600 mutation. Most patients in this study had more than 1 brain metastasis, with approximately one third of the asymptomatic patients having more than 4 metastases.

ABC Study: Response by Modified RECIST v1.1

Michael A. Davies, MD, PhD:
The intracranial ORR with nivolumab plus ipilimumab was 51% vs 20% with single-agent nivolumab.

Referring back to the CheckMate 067 data on patients with extracranial disease only, there was a relatively subtle difference between combination immunotherapy and single-agent nivolumab.1 By contrast, results from the ABC study showed a response rate of 51% with nivolumab plus ipilimumab in patients with brain metastases.8 The response rate with single-agent nivolumab was much lower at 20% compared with the combination, which is very similar to the pattern with single-agent pembrolizumab reported in previous clinical trials for patients with brain metastases.9 

The lower section of this table shows that this difference in intracranial response rate is even more pronounced in patients without previous treatment: 59% with combination therapy vs 21% with single-agent nivolumab.

ABC Study: Intracranial PFS

Michael J. Davies, MD, PhD:
The curves for intracranial, 5-year PFS illustrate that responses are quite durable. This durability was also seen in CheckMate 067.1 This stands in contrast to intracranial responses with targeted therapy in BRAF-mutant melanoma that have high response rates but short duration.10

This figure also shows an absolute improvement of 31% for treatment with nivolumab plus ipilimumab (46%) vs nivolumab alone in patients with previous treatment (15%) at 5 years and underscores the marked difference in the outcomes with combination vs single-agent immunotherapy for patients with brain metastases.

Evan J. Lipson, MD:
Few trial results are clearly “a game changer” for clinical decision-making, but I think this is one of them. For a patient with brain metastases, this finding would influence the decision to use nivolumab with ipilimumab over an anti–PD-1 antibody alone.

Michael A. Davies, MD, PhD:
In addition, for patients with BRAF-mutant disease, the OS data also argue in favor of combination immunotherapy vs targeted therapy or any other systemic therapy regimen. 

ABC Study: Overall Survival

Michael A. Davies, MD, PhD:
The 5-year OS rate is 51% for nivolumab plus ipilimumab vs 34% for nivolumab in patients with no previous treatment, and 13% for nivolumab in patients with previous treatment. This absolute improvement of 17% for patients with asymptomatic brain metastases is an influential finding for clinical decision-making.

ABC Study: Safety

Michael A. Davies, MD, PhD:
The toxicity data for this high-risk population are similar to those for patients without brain metastases. Only 5% of patients in this trial had serious neurologic adverse events (AEs). In this high-risk population, the frequency of new AEs, particularly neurologic AEs, was consistent with previous data, supporting more aggressive combination therapy as an option for patients with brain metastases.

ABC Study: Conclusions

Michael A. Davies, MD, PhD:
In summary, for a tumor type with historically poor outcomes, the results of the ABC study set new benchmarks for patients with asymptomatic brain metastases. After 5 years, the intracranial response rate for the combination therapy is in the 50% to 60% range, with no additional safety issues. Looking at the different subgroups with ongoing responses, there is a clear improvement in both PFS and OS for the nivolumab plus ipilimumab therapy vs nivolumab alone in patients with untreated brain metastases.

Unfortunately, patients who require steroids to control central nervous system symptoms show much lower response rates, and those are patients for whom we continue to need new and more effective therapy regimens.

CheckMate 511 3-Year Follow-up: Comparison of Different Nivolumab/Ipilimumab Regimens in Untreated Melanoma

Evan J. Lipson, MD:
CheckMate 511 is a randomized, double-blind phase IIIb/IV trial of 2 different nivolumab/ipilimumab regimens, and the 3-year follow-up was reported by Lebbe and colleagues.11 This study was designed to determine whether “flipped dose” nivolumab at 3 mg/kg plus ipilimumab at 1 mg/kg (nivo3 + ipi1) might have an improved safety profile compared with the current US standard of nivolumab 1 mg/kg plus ipilimumab at 3 mg/kg (nivo1 + ipi3) for patients with unresectable or metastatic melanoma (N = 360). It was not designed or powered to test efficacy or noninferiority. However, the descriptive analyses of the results are interesting.

Following primary treatment, both groups entered a maintenance phase with standard nivolumab 480 mg monotherapy given every 4 weeks until progressive disease or unacceptable toxicity for up to 2 years.

The primary endpoint was the incidence of grade 3-5 TRAEs. The secondary endpoints included investigator-assessed ORR, PFS, and OS.

CheckMate 511-3-Year Follow-up: Patient Characteristics

Evan J. Lipson, MD:
The 2 groups of patients were relatively well balanced in terms of age (median: 58.5 years), ECOG PS, BRAF mutation status, and PD-L1 expression. Baseline lactate dehydrogenase (LDH) was above the upper limit of normal (ULN) in approximately 40% of patients. Of note, only approximately 5 patients in each group had brain metastases.

CheckMate 511 3-Year Follow-up: Safety Summary

Evan J. Lipson, MD:
Grade ≥3 TRAEs were reported in 34% of the patients in the nivo3 + ipi1 group vs 48% in the nivo1 + ipi3 group. This is not surprising. Nivolumab has been tested at various dose levels, as low as 0.1 mg/kg up to approximately 10 mg/kg (approximately 2 orders of magnitude). Historical studies did not show a difference in toxicity across those dose ranges.

But the same cannot be said for ipilimumab. There does seem to be a dose-response relationship in terms of toxicity. Certainly, at lower doses, the drug is better tolerated. At the higher doses, it is quite toxic in some patients. For example, it has been used in the adjuvant setting at 10 mg/kg, although some healthcare professionals shy away from that dose because of its toxicity profile.12

TRAEs led to discontinuation of therapy for 24% of patients in the nivo3 + ipi1 group vs 34% in the nivo1 + ipi3 group.

CheckMate 511 3-Year Follow-up: TRAEs Occurring in ≥10% of Patients

Evan J. Lipson, MD:
Toxicities observed in ≥10% of the patients reflect previous observations. For each, there is a slight increase associated with the higher dose of ipilimumab.

CheckMate 511 3-Year Follow-up: Response

Evan J. Lipson, MD:
The response data are only descriptive because the study was not powered to assess noninferiority or efficacy. However, the 2 regimens appear to be approximately similar in terms of efficacy. There is a 22% complete response rate in patients in the low-dose ipilimumab group vs 23% for patients in the higher-dose group; partial response rates were 25% vs 30%, respectively. Both groups had 9% rates of stable disease.

CheckMate 511 3-Year Follow-up: Survival

Evan J. Lipson, MD:
The same trend is seen in survival rates, with the median PFS being nearly equivalent at approximately 10 months. The median OS for both groups had not been reached at the time of reporting.

CheckMate 511 3-Year Follow-up: Treatment-Free Analyses

Evan J. Lipson, MD:
This was an interesting analysis of treatment-free intervals. After 3 years, there were similarities between the nivo3 + ipi1 and nivo1 + ipi3 groups. In the low-dose ipilimumab group (nivo3 + ipi1), 74% of patients remained treatment free vs 77% in the high-dose ipilimumab group; similar percentages were seen for patients receiving subsequent therapy, 43% and 38%, respectively.

CheckMate 511 3-Year Follow-up: Conclusions

Evan J. Lipson, MD:
The 3-year follow-up data from CheckMate 511 demonstrate that there is a difference between the rate of grade 3-5 TRAEs in the nivo3 + ipi1 group vs the nivo1 + ipi3 group at 34% vs 48%, respectively, meeting the trial's primary endpoint. Although the study was not designed or powered to demonstrate noninferiority between the 2 regimens, the descriptive analyses illustrate similar response rates across the subgroups, both in terms of survival and treatment-free intervals.

The authors of the study conclude that the information here is important when talking about the benefit–risk profile. In practice, I think this has supported the decision of many melanoma investigators and healthcare professionals to use the lower-dose ipilimumab/higher-dose nivolumab regimen, particularly in patients who may do poorly when exposed to severe toxicity.

Michael A. Davies, MD, PhD:
This information is reassuring in terms of clinical decision-making, particularly for PFS and OS. A key limitation of the study, however, is the absence of certain groups, for example, patients with new or progressing brain metastases. That group has seen such remarkable benefit with the nivo1 + ipi3 regimen, and it would be good to know if the flipped dosing regimen would result in equivalent outcomes in patients with central nervous system involvement or other melanoma subtypes, particularly mucosal melanoma. The MD Anderson Cancer Center is currently conducting a phase II clinical trial of low-dose ipilimumab with pembrolizumab in patients with melanoma brain metastases, and they hope to present the results in the near future.13

Evan J. Lipson, MD:
Excellent points, and in fact, I err on the side of higher-dose ipilimumab for many of those patient populations.

RELATIVITY-047: First-line Relatlimab Plus Nivolumab vs Nivolumab in Advanced Melanoma

Evan J. Lipson, MD:
My colleagues and I conducted RELATIVITY-047, a global, randomized, registrational phase II/III study of relatlimab plus nivolumab vs nivolumab monotherapy in advanced melanoma (N = 714).14 In preclinical results, this combination therapy shows synergistic activity and is tolerated well.15,16 At ASCO 2021, we reported initial efficacy and safety results from RELATIVITY-047.

Relatlimab is a human monoclonal antibody that targets LAG-3, an immune checkpoint pathway that restores the effector function of exhausted T-cells.17 Previous research showed that some patients whose melanoma was relapsed or refractory to anti–PD-1 monotherapy responded when relatlimab was added.16

In RELATIVITY-047, patients with previously untreated unresectable or metastatic melanoma were randomized to either relatlimab plus nivolumab or nivolumab alone. Relatlimab and nivolumab were administered as a fixed-dose preparation.

The study’s primary endpoint was PFS. Secondary endpoints included OS and ORR. The PFS and ORR were determined by blinded independent central review (BICR).

RELATIVITY-047: Baseline Characteristics

Evan J. Lipson, MD:
The groups were well balanced with a relatively high percentage of patients with M1c or M1d disease. A number of patients had serum LDH levels that were either above the ULN or 2 times the ULN. Fewer than 10% of patients had adjuvant or neoadjuvant treatment before enrolling. Because this was a global study, most of that treatment would have been interferon.

Patients were stratified by LAG-3 expression, PD-L1 expression, BRAF mutation status, and American Joint Committee on Cancer (AJCC) M stage.

RELATIVITY-047: PFS by BICR (Primary Endpoint)

Evan J. Lipson, MD:
The study met its primary endpoint: median PFS for the combination therapy was 10.1 months, which is more than double the 4.6 months with nivolumab alone (HR: 0.75; P = .0055).

It is noteworthy that the PFS benefit appeared relatively early, at 3 months, which was approximately the time of the first on-treatment imaging. The separation of PFS curves extends throughout the study. The median follow-up was 13.2 months and the 1-year OS rate was approximately 48% with combination therapy vs 36% for nivolumab.

RELATIVITY-047: PFS by LAG-3 Expression Level

Evan J. Lipson, MD:
We also looked at whether LAG-3 expression in the tumor microenvironment was a marker for the likelihood of increased response to relatlimab, and it does not appear to be. LAG-3 expression in the tumor microenvironment is probably just a marker for better outcomes generally. In this setting, LAG-3 may just be a marker for infiltrating T-cells in a tumor and increases in infiltrating T-cells (CD8 positive in particular) might correlate with increased efficacy of immunotherapy across the board.

RELATIVITY-047: PFS Across Subgroups

Evan J. Lipson, MD:
This forest plot illustrates the relative benefit of relatlimab with nivolumab and single-agent nivolumab in different patient subgroups. The subgroups, in almost all cases, favored the combination therapy; this was true for BRAF mutation status, AJCC stage, and PD-L1 and LAG-3 expression, as well as factors such as sex and LDH level.

Michael A. Davies, MD, PhD:
It is reassuring that the benefit was seen in so many subgroups. In patients who were PD-L1 positive (≥1%) the HR is 0.95. However, there also was a marked improvement in PFS for patients who are PD-L1 negative (<1%), with an HR of 0.66. This resembles the improvement in PFS for patients with PD-L1–negative disease treated with ipilimumab plus nivolumab vs nivolumab alone in CheckMate 067.1 Consequently, this begs the question: Is the relatlimab and nivolumab combination rescuing the same group of patients who are rescued by the ipilimumab and nivolumab combination, or is this a distinct group? Using these treatments in sequence in the future may shed some light on clinical benefits.

RELATIVITY-047: Safety

Evan J. Lipson, MD:
Regarding safety, the TRAEs that were associated with relatlimab plus nivolumab were generally manageable and reflected the safety profile typically seen with immune checkpoint inhibitors. Although the incidence of grade 3/4 TRAEs was higher with the combined therapy compared with nivolumab alone, the AEs in this study occurred at a lower rate than those with other immunotherapy combinations.

In RELATIVITY-047, TRAEs that led to discontinuation of trial therapy occurred in approximately 15% of patients in the relatlimab with nivolumab group and approximately 7% of patients who received nivolumab alone.

RELATIVITY-047: Immune-Mediated AEs

Evan J. Lipson, MD:
Based on preclinical data that suggested that blocking LAG-3 might cause immune-mediated myocarditis,18 investigators prospectively monitored troponin for the first 2 months of the protocol. They saw myocarditis only rarely: in 1.7% of patients treated with relatlimab plus nivolumab and 0.6% for nivolumab alone.

RELATIVITY-047: Conclusions

Evan J. Lipson, MD:
In conclusion, in RELATIVITY-047, the addition of relatlimab to nivolumab to patients with previously untreated and unresectable or metastatic melanoma led to a significant improvement in PFS compared with nivolumab alone. This combination demonstrated a manageable safety profile with a lower rate of grade ≥3 TRAEs compared with other immune checkpoint inhibitor combinations.

These findings demonstrate that relatlimab with nivolumab is a potential novel treatment option for this patient population. They also validate LAG-3 as the third checkpoint pathway to be identified (after PD-1 and CTLA-4) for which blockade has clinical benefit.

Neoadjuvant Relatlimab Plus Nivolumab for Stage III Melanoma: Study Design

Evan J. Lipson, MD:
Amaria and colleagues19 also evaluated relatlimab with nivolumab but in the neoadjuvant setting. The combination was assessed in a single-arm phase II trial of patients with resectable stage IIIB/C or oligometastatic stage IV melanoma prior to surgery (N = 30). The primary endpoint was pCR and secondary endpoints included ORR, relapse-free survival (RFS), event-free survival, OS, and safety.

Patients received relatlimab 160 mg plus nivolumab 480 mg every 4 weeks, for 2 doses, then underwent CT scans for response and surgical resection of disease for pCR assessment. Afterward, patients received adjuvant relatlimab plus nivolumab every 4 weeks for 10 doses, with follow-up for 2 years.

Neoadjuvant Relatlimab Plus Nivolumab for Stage III Melanoma: Baseline Characteristics

Evan J. Lipson, MD:
The median age was 60 years (range: 35-79). Of note, nearly all patients had an ECOG PS of 0, 60% had stage IIIB disease, 67% had previous surgery, and only 5 patients (17%) had the BRAF V600E/K mutation.

Neoadjuvant Relatlimab Plus Nivolumab for Stage III Melanoma: ORR and pCR

Evan J. Lipson, MD:
Data showed that 17 patients (59%) achieved a pCR, and another 2 (7%) had a near pCR. These data mirror not only some of the results from prior neoadjuvant melanoma trials20 but also results from neoadjuvant trials in other tumor types, for example, the CheckMate 358 study of neoadjuvant nivolumab in Merkel cell carcinoma.21 It seems that, in the neoadjuvant setting, the use of immune checkpoint blocking therapy can result in a relatively high rate of pCR, near 60%.

Michael A. Davies, MD, PhD:
The previous RELATIVITY-047 study provided excellent data but not response data. Although this neoadjuvant trial was not randomized, it is one of the first quantifications of the response rate in treatment-naive patients with melanoma. The pCR rate of 59% with a 73% pathologic response rate overall is very impressive.

These numbers are reminiscent of data seen in neoadjuvant trials of nivolumab plus ipilimumab in melanoma and appear to be much higher than the pathologic response rates seen with single-agent anti–PD-1 therapy in the neoadjuvant setting.

Neoadjuvant Relatlimab Plus Nivolumab for Stage III Melanoma: Radiographic Response

Evan J. Lipson, MD:
Of the 30 patients enrolled, only 1 had a delay in surgery. This is an important consideration for a patient receiving neoadjuvant therapy whose surgical window may close in the near term. This waterfall plot illustrates radiographic response and shows that approximately two thirds of patients did see some regression of their tumors.

Neoadjuvant Relatlimab Plus Nivolumab for Stage III Melanoma: RFS by MPR Status

Evan J. Lipson, MD:
I think the most interesting finding from this study is the RFS data. The blue line represents patients who had a major pathologic response with 100% probability of RFS during the 2-year follow-up period after surgery. The orange line represents patients who did not experience a major pathologic response, and their RFS curve dips over time.

Michael A. Davies, MD, PhD:
These data are reminiscent of a pooled analysis published by the International Neoadjuvant Melanoma Consortium.20 It examined studies of neoadjuvant immunotherapy and targeted therapy that were conducted by several groups. It showed that the achievement of either a pCR or near-pCR (or even a partial pathologic response) in patients treated with neoadjuvant, single-agent PD-1 inhibition or nivolumab with ipilimumab correlates with outstanding RFS (2-year RFS: 96%). These results are interesting when compared with neoadjuvant targeted therapy, where pCR correlated with only a 79% 2-year RFS.

Neoadjuvant Relatlimab Plus Nivolumab for Stage III Melanoma: Safety

Evan J. Lipson, MD:
The balance of safety and efficacy is especially important in the neoadjuvant setting. One issue is the potentially negative influence of a TRAE on scheduling of surgery. The relatlimab with nivolumab combination might hit the sweet spot in terms of efficacy when compared with neoadjuvant nivolumab or pembrolizumab vs neoadjuvant ipilimumab with nivolumab. Larger comparative clinical trials are needed.

The data in this study certainly are promising. There is no evidence of myocarditis, and although the grade 3 toxicity occurring in 26% of patients is a little higher than reported in RELATIVITY-047,14 it may be considered acceptable.

Neoadjuvant Relatlimab Plus Nivolumab for Stage III Melanoma: Conclusions

Evan J. Lipson, MD:
In summary, the neoadjuvant use of relatlimab with nivolumab demonstrated encouraging activity, despite the small size of this phase II study (N = 30). We saw high rates of pCR and major pathologic response in approximately two thirds of patients. Of note, the RFS rates were improved in patients who achieved a major pathologic response after surgery.

But what do you do for patients who have not achieved or experienced a major pathologic response after surgery? Do these patients need more therapy? Do they need different therapy? These questions still remain to be answered.

Title: COLUMBUS: Encorafenib/Binimetinib vs Encorafenib vs Vemurafenib in BRAF-Positive Advanced Melanoma

Michael A. Davies, MD, PhD:
COLUMBUS is a multicenter, randomized, open-label phase III trial of encorafenib plus binimetinib vs encorafenib alone vs vemurafenib in patients with advanced melanoma and a BRAF V600E/K mutation (N = 577).22 The primary endpoint was PFS by BICR for encorafenib plus binimetinib vs vemurafenib with secondary endpoints including additional PFS analysis, responses, and survival.

Immunotherapy continues to yield remarkable outcomes in patients with stage IV melanoma without biomarker restrictions (ie, other than autoimmune disease limitations). In addition, there are targeted therapy regimens approved for patients with stage IV melanoma who have a BRAF V600 mutation.23 These patients comprise 40%-45% of those with cutaneous melanoma. Encorafenib plus binimetinib is the third BRAF/MEK inhibitor combination that has been approved for patients with BRAF V600–positive stage IV melanoma.24

In this 3-arm trial, one arm received the combination of encorafenib 450 mg/day and the MEK inhibitor binimetinib 45 mg twice daily, and another arm received encorafenib at the single-agent, maximum-tolerated dose of 300 mg/day. When encorafenib is given with binimetinib, patients can tolerate a higher dose of encorafenib at 450 mg/day (vs the single-agent dose of 300 mg). Both of these arms were compared with single-agent vemurafenib at the FDA-approved dose of 960 mg twice daily.

At ASCO 2021, Dummer and colleagues22 presented 5-year landmark results from COLUMBUS.

COLUMBUS: Baseline Characteristics

Michael A. Davies, MD, PhD:
The baseline characteristics of the patient population are typical compared with other phase III studies of patients with stage IV melanoma. Approximately 24% to 28% of the patients in this study have high LDH, which is a consistent and poor prognostic factor across virtually all clinical trials conducted.


Michael A. Davies, MD, PhD:
The median PFS was 14.9 months for encorafenib with binimetinib vs 9.6 months for encorafenib alone and 7.3 months for vemurafenib alone. Although targeted therapy regimens have a reputation for achieving high rates of initial response and initial clinical benefit, they are not known for durable benefit.

Looking at registration studies for each of the BRAF/MEK inhibitor combinations—encorafenib and binimetinib,22 dabrafenib and trametinib,25 and vemurafenib and cobimetinib26—there is a subset of patients who seem to do well for a very long period of time. In COLUMBUS at 5 years, 22.9% of patients treated with encorafenib and binimetinib were progression free. A 5-year PFS rate in the 20% range has consistently been seen with each FDA-approved BRAF/MEK inhibitor regimen.


Michael A. Davies, MD, PhD:
The 5-year OS rate for encorafenib with binimetinib was 34.7%. Median survival with encorafenib and binimetinib was 33.6 months vs 16.9 months with vemurafenib. At later time points, long-term survival rates plateau and show similar rates for single-agent encorafenib vs encorafenib plus binimetinib. However, it should be noted that encorafenib with binimetinib is generally better tolerated than single-agent encorafenib. This graph shows significant OS separation between single-agent encorafenib and single-agent vemurafenib, which is consistent with the PFS difference.

COLUMBUS: 5-Year Survival With Encorafenib Plus Binimetinib by Subgroup

Michael A. Davies, MD, PhD:
Baseline LDH levels continue to correlate with outcomes in clinical trials, and this trend was seen in COLUMBUS as well. The 5-year PFS rate for patients who had a baseline LDH less than the ULN was 30.9%. However, essentially none of the patients who had a baseline LDH greater than the ULN were progression free at 5 years. Similarly, marked differences in long-term OS are seen, depending on baseline LDH. At 5 years, 51.7% of those with baseline LDH at or below ULN remained alive vs 9.1% of those with LDH above ULN.

I think many treatments are very active in patients with normal LDH, which generally correlates with lower tumor burden. But the high-LDH population remains a challenging subgroup in melanoma, particularly for achieving long-term survival.

COLUMBUS: Subsequent Therapy

Michael A. Davies, MD, PhD
The most common subsequent therapy was a checkpoint inhibitor, and the prevalence was similar across the 3 arms: 34.4% for encorafenib and binimetinib, 37.1% for encorafenib, and 42.4% for vemurafenib.


Michael A. Davies, MD, PhD:
There were no new safety events, demonstrating that encorafenib and binimetinib continue to be tolerated well over longer periods in most patients. This combination has a much lower rate of pyrexia than is seen with dabrafenib and trametinib and does not have the photosensitivity issue seen with vemurafenib and cobimetinib.27 Encorafenib and binimetinib does have other particular AEs, such as laboratory-based toxicities including increase of liver transaminases.24

Evan J. Lipson, MD:
An important safety issue is the potential for long-term, possibly permanent toxicity with cancer therapy. This is not seen to any great degree with the encorafenib and binimetinib combination, but there is a possibility with any of the immune-based agents. If long-term safety is a consideration for some patients, they may choose to pursue therapy with one of the targeted combinations instead of an immunotherapy.

Michael A. Davies, MD, PhD:
These first-line trials cannot inform patients of the potential impact of any future sequencing of therapies. However, in retrospective studies, some patients treated with targeted therapy after progressing on immune therapies have seen a higher incidence of toxicities than in the frontline setting.28

Evan J. Lipson, MD:
That has been our experience as well.

COLUMBUS: Conclusions

Michael Davies, MD
To summarize, COLUMBUS provides 5-year landmark data for encorafenib and binimetinib with a 5-year PFS of 22.9% and OS of 34.7%. These data are very consistent with the long-term outcomes seen with the 2 other FDA-approved BRAF/MEK inhibitor regimens for melanoma (cobimetinib plus vemurafenib and dabrafenib plus trametinib). These results also suggest there is a subgroup of patients who do very well long-term with targeted therapy, even as it remains challenging to identify those patients.

Lifileucel in Previously Treated Advanced Melanoma: Study Design

Evan J. Lipson, MD:
There are no therapies that are currently approved for patients who progress on immune checkpoint inhibitors and BRAF or MEK inhibitors. So, this population is in great need of effective therapy. Larkin and colleagues29 presented an update from a global, multicenter, open-label phase II trial of lifileucel in patients with previously treated advanced melanoma (N = 171). Lifileucel is a cryopreserved adoptive cell therapy based on autologous tumor-infiltrating lymphocytes (TILs). Of note, although investigation of various TIL regimens has a long history,30 they have not been commercialized to any degree.

There were 4 cohorts on this trial, but this discussion will focus only on cohort 2, which received cryopreserved lifileucel. The primary endpoint is investigator-assessed ORR. Secondary endpoints include efficacy and safety.

Lifileucel in Previously Treated Advanced Melanoma: Baseline Characteristics

Evan J. Lipson, MD:
The median age of cohort 2 is 55 years (range: 20-79). This cohort skews a little younger than some of the other melanoma patient cohorts in the literature.

Generally, patients had a good ECOG PS (44% ECOG PS 1). All patients were pretreated, with a mean number of previous therapies of 3.3. All patients had received an anti–PD-1/PD-L1 antibody, other previous therapies included an anti–CTLA4 antibody, the combination of an anti–PD-1 plus anti–CTLA4, and a BRAF inhibitor with or without a MEK inhibitor. Of note, approximately 40% of patients had LDH levels greater than the ULN. In this cohort, 42% of patients had liver and/or brain lesions. Overall, this is a patient population with difficult-to-treat disease.

Lifileucel in Previously Treated Advanced Melanoma: ORR (Primary Endpoint)

Evan J. Lipson, MD:
Looking at the primary endpoint in cohort 2, the ORR of the 66 patients was 36.4%. In my opinion, this is the most impressive data in this presentation. Three patients had a complete response (4.5%) and 21 patients had a partial response (31.8%). Another 44% of patients experienced stable disease.

The disease control rate is 80.3%, which is an impressive result in a patient population that has lacked effective therapy.

Michael A. Davies, MD, PhD:
To me, the striking result is that after 33 months of follow-up, the median duration of response (DoR) has yet to be reached. Not only does this show an impressive ORR, but responses appear to be very durable, which is uncommon in patients refractory to anti–PD-1 antibodies.

Evan J. Lipson, MD:
I agree, and results from this study of lifileucel are yet another testament to the power of the immune system and the way that it uses memory against invaders like viral and bacterial pathogens and cancers.

Lifileucel in Previously Treated Melanoma: Predictors of DoR and ORR

Evan J. Lipson, MD:
In a multivariate analysis of the data, investigators determined that baseline LDH and cumulative duration of previous anti–PD-1/PD-L1 treatment independently predict DoR. The results suggest that DoR nearly doubled with each 6-month decrease in previous treatment with an anti–PD-1/PD-L1 antibody. Of note, none of the factors examined in a univariate analysis predicted ORR.

Lifileucel in Previously Treated Melanoma: Safety

Evan J. Lipson, MD:
When discussing safety in this study, it is important to understand the treatment regimen. First, the tumor is resected and the TILs collected and expanded. After the patient receives cytotoxic chemotherapy, generally in the inpatient setting, the expanded TILs are infused along with interleukin-2.

All the associated therapies carry attendant toxicities such as the ones listed here (eg, thrombocytopenia, anemia, neutropenia, and in some cases, febrile neutropenia). So, none of these are unexpected. Almost all patients (97%) experience a grade 3/4 toxicity. This can be a difficult therapy for many patients, partly because of underlying comorbidities or worse PS.

Lifileucel in Previously Treated Melanoma: Conclusions

Evan J. Lipson, MD:
In conclusion, lifileucel TIL therapy demonstrated encouraging activity in heavily pretreated patients with advanced melanoma, specifically those who had progressed on previous anti–PD-1/PD-L1 therapy as well as BRAF and MEK inhibitors. The ORR of 36.4% and the disease control rate of approximately 80% are impressive. In addition, the median DoR was not reached at 33 months of median follow-up. These responses deepen over time, which is another testament to the immune system’s ability to continue working even after the therapy itself has been discontinued.

The investigators on this study concluded that newly diagnosed patients receiving initial melanoma therapy should be closely monitored for progression with the possibility of an early switch to TIL therapy if progressive disease is seen.

LEAP-004: Pembrolizumab Plus Lenvatinib After Progression on Anti–PD-1/PD-L1 in Advanced Melanoma

Michael A. Davies, MD, PhD:
A second major study in patients with melanoma who are refractory to anti–PD-1/PD-L1 antibodies was presented at ASCO 2021 by Arance and colleagues.31 This was the LEAP-004 trial of lenvatinib plus pembrolizumab in stage III/IV melanoma. Healthcare professionals have limited treatment options for patients who have progressed on anti–PD-1 antibodies, with or without anti–CTLA-4 agents. Lenvatinib is a tyrosine kinase inhibitor that targets multiple kinases, including both the VEGF receptor and the FGF receptor; both may have positive immunomodulatory effects.32-34

LEAP-004 is an ongoing multicenter, open-label, single-arm phase II study that is assessing the safety and efficacy of lenvatinib 20 mg/day plus pembrolizumab 200 mg every 3 weeks in 103 patients with advanced melanoma who had previously progressed on anti–PD-1/PD-L1 therapy. The primary endpoint was ORR with secondary endpoints including DoR, PFS, OS, and safety. The results presented at ASCO 2021 were from a median follow-up of 15.3 months (range: 12.1-19.0).

LEAP-004: Baseline Characteristics

Michael A. Davies, MD, PhD:
This is a high-risk population with multiple previous lines of therapy, including at least 3 lines in 33% of patients. Recalling the prognostic significance of LDH mentioned earlier, the high levels of LDH in these patients are noteworthy: above the ULN in 55% of patients, including 20% with LDH twice the ULN. Unfortunately, we are increasingly seeing patients like this in our clinics.

LEAP-004: BICR-Confirmed Response (Primary Endpoint)

Michael A. Davies, MD, PhD:
Even with longer follow-up, the BICR-confirmed ORR remained at 21.4%, and the disease control rate was 66%. This update revealed 1 additional patient with complete response and another with stable disease.

LEAP-004: Tumor Response by RECIST v1.1 by BICR

Michael A. Davies, MD, PhD:
This waterfall plot of tumor size changes illustrates that approximately 70% of patients achieved disease control.

LEAP-004: BICR-Confirmed Response by Prior Therapy

Michael A. Davies, MD, PhD:
These are the most interesting and impressive data from this study. As discussed, the overall population had an ORR of 21% and a disease control rate of 66%. Those results were consistent across several challenging patient subgroups, including those who had progressed on combination immunotherapy (anti–PD-1/PD-L1 and anti–CTLA-4 antibodies), and patients with BRAF mutations who progressed on targeted therapy.

LEAP-004: Response by Resistance to Prior PD-1/PD-L1 Therapy

Michael A. Davies, MD, PhD:
To the credit of the investigators, they looked carefully at resistance to anti–PD-1/PD-L1 antibodies, including primary resistance in the adjuvant setting, primary resistance in the metastatic setting, and secondary resistance, that is, patients who had an initial response and then progressed. Results demonstrated very similar response and disease control rates, regardless of type of resistance to previous anti–PD-1/PD-L1 therapy.

LEAP-004: Duration of BICR-Confirmed Response

Michael A. Davies, MD, PhD:
A noteworthy difference between LEAP-004 and the lifileucel study is the DoR. The lifileucel trial had a median follow-up of 33 months, and the median DoR had not been reached.29 In LEAP-004, the median DoR was 8.3 months.31 Clearly, this therapy achieves good response rates and disease control in a significant subset of patients. The question is are we truly reinvigorating the immune response or just slowing disease down? Is this primarily a targeted therapy effect? Those answers will need more follow-up, but it does seem that most responses are relatively short duration. Keep in mind, this is a very high-risk and treatment-refractory population. I think there is evidence of clinical benefit, but not the full reinvigoration of durable immune responses that we hoped to achieve.

LEAP-004: PFS and OS

Michael A. Davies, MD, PhD:
The median PFS was 4.2 months and the 12-month PFS rate was 18.5%. The median OS was 14.0 months with a 1-year survival rate of 54.5%. A longer follow-up period may determine the extent of the durable benefit.


Michael A. Davies, MD, PhD:
This regimen has toxicities, and TRAEs need to be considered. Approximately one half of the patients (45.6%) developed grade 3-5 TRAEs, with 59.2% having treatment interruptions and 56.3% needing lenvatinib dose reductions. The most common TRAE was hypertension, a known toxicity of lenvatinib.

LEAP-004: AEs of Interest

Michael A. Davies, MD, PhD:
The toxicity profile is dominated by the hypertension. Of interest, among the autoimmune toxicities, there was a significant rate of hypothyroidism, but it is relatively low compared with most other immunotherapy-related toxicities.

LEAP-004: Conclusions

Michael A. Davies, MD, PhD:
In summary, these follow-up data show that the combination of pembrolizumab and lenvatinib achieved good disease control in patients with unresectable, stage III/IV melanoma. What remains unclear is how many patients are able to achieve true, long-term benefit. In terms of the toxicities, the profile appears to be very different from other immunotherapy regimens or, likely, lifileucel.

KEYNOTE-054 Part 2: Adjuvant Pembrolizumab vs Placebo in High-Risk Resected Melanoma

Michael A. Davies, MD, PhD:
KEYNOTE-054 is an ongoing, international, double-blind, randomized phase III study of adjuvant pembrolizumab in adult patients with high-risk, stage III melanoma and crossover or rechallenge with pembrolizumab after recurrence.35 In Part 1, 1019 patients were randomly assigned to receive either pembrolizumab or a placebo as post-surgery therapy. At ASCO 2021, Eggermont and colleagues36 presented results from part 2 (n = 175), comprising patients in the placebo arm with disease recurrence who crossed over to receive pembrolizumab; and patients who recurred at least 6 months after completing treatment with adjuvant pembrolizumab who were rechallenged with pembrolizumab. The primary objectives for part 2 are safety, responses, DoR, and PFS.

Although the KEYNOTE-054 study overall was designed to assess OS benefit with adjuvant pembrolizumab, the new results also shed light on whether waiting until patients have recurrences before administering adjuvant pembrolizumab could achieve similar OS.

KEYNOTE-054 Part 2: Baseline Characteristics

Michael A. Davies, MD, PhD:
In total, 155 of the patients whose disease progressed on the original placebo arm elected to participate in the crossover arm. Twenty patients who relapsed on the adjuvant pembrolizumab arm continued in the rechallenge group. The median time on treatment in part 2 among the crossover group was 8.0 months and was shorter at 3.2 months for the rechallenge group.

KEYNOTE-054 Part 2: PFS

Michael A. Davies, MD, PhD:
Patients in the crossover arm had a 3-year PFS rate of 32% with a median PFS of 8.5 months. Of interest, in the rechallenge group, which was very small (n = 20), the median PFS was just 4.1 months.

KEYNOTE-054 Part 2: Responses

Michael A. Davies, MD, PhD:
The ORR in the patients who crossed over to pembrolizumab was nearly 40%, which is approximately the rate seen in other studies of pembrolizumab in treatment-naive patients with melanoma.37 Of interest, in the patients who were rechallenged, the response rate was only 11% (n = 1), which is lower than in previous studies. The caveat is that only 9 patients in the rechallenge group were assessable for response.

KEYNOTE-054 Part 2: Safety

Michael A. Davies, MD, PhD:
No unexpected toxicities were seen with pembrolizumab, either in the crossover or the rechallenge arm. In the crossover arm, 7.1% experienced grade 3/4 immune-related AEs.

KEYNOTE-054 Part 2: Conclusions

Michael A. Davies, MD, PhD:
In conclusion, the crossover patients in KEYNOTE-054 experienced a median PFS of 8.5 months and an ORR of almost 40%. In practice, almost 90% of patients with resected stage III disease receive adjuvant therapy, but we still do not know if it is improving OS. The strength of the KEYNOTE-054 study is that it is designed to answer that question.

We also know that we are overtreating many patients with stage III melanoma who would not have relapsed. I think the data that are emerging from KEYNOTE-054 and other studies support the need to develop true predictive biomarkers to identify those patients who benefit the most from receiving adjuvant therapy. In particular, we need longer follow-up data from this and other studies to better determine the OS benefit.

Provided by Clinical Care Options, LLC

Contact Clinical Care Options

For customer support please email: customersupport@cealliance.com

Mailing Address
Clinical Care Options, LLC
12001 Sunrise Valley Drive
Suite 300
Reston, VA 20191

Supported by educational grants from
Bristol-Myers Squibb
Pfizer, Inc.

Leaving the CCO site

You are now leaving the CCO site. The new destination site may have different terms of use and privacy policy.


Cookie Settings