Researchers are seeking to determine whether the plethora of new therapies that have been introduced in recent years in the advanced and metastatic settings can benefit patients at risk of recurrence earlier in the treatment timeline for melanoma.
Jeffrey Weber, MD, PhD
For decades, physicians treating patients with melanoma had 1 option with demonstrated efficacy in the adjuvant setting: interferon alfa-2B.
That changed in 2015 with the approval of ipilimumab (Yervoy), an anti—CTLA-4 checkpoint blockade immunotherapy, as adjuvant therapy for patients with high-risk stage III melanoma. Now researchers are seeking to determine whether the plethora of new therapies that have been introduced in recent years in the advanced and metastatic settings can benefit patients at risk of recurrence earlier in the treatment timeline.
The potential for the expansion of therapeutic options was evident at the 2017 European Society for Medical Oncology (ESMO) Annual Congress, held in September in Madrid, Spain. Findings reported at the conference and simultaneously published in the New England Journal of Medicine demonstrate a role for 2 modalities: nivolumab (Opdivo) monotherapy, a checkpoint blockade immunotherapy, for patients of any mutation status and combination therapy with dabrafenib (Tafinlar) and trametinib (Mekinist), small molecules that target kinases in the MAPK cell-signaling network, for patients with BRAF-mutant disease.
In the CheckMate 238 trial, nivolumab resulted in a 35% improvement in relapse-free survival (RFS) compared with high-dose ipilimumab after a minimum 18-month follow-up, with a lower rate of adverse events (AEs).1,2 Nivolumab, an anti—PD-1 monoclonal antibody, was also superior on 1-year RFS in patients with BRAF-mutant disease. In the COMBI-AD trial, adjuvant treatment with dabrafenib and trametinib reduced the risk of relapse or death by 53% compared with placebo for patients with BRAF-mutant stage III melanoma after a median follow-up of 2.8 years.3,4 Dabrafenib inhibits BRAF while trametinib is aimed at MEK. The findings sparked a lively debate among key researchers at the ESMO conference about how best to administer adjuvant treatment.
Jeffrey S. Weber, MD, PhD, who presented the CheckMate 238 data, said nivolumab “looks like a superior adjuvant melanoma regimen compared with ipilimumab from every angle” and may also be a better choice than targeted therapy in this setting for patients with BRAF-mutations.
“It comes down to whether you believe there will be a tail on the curve for immunotherapy with RFS in the adjuvant mode as you see in metastatic patients, compared with a lesser tail on the curve for BRAF-MEK modes and metastatic modes,” said Weber, who is deputy director of the Perlmutter Cancer Center at New York University Langone Health in New York City. “It’s going to be a tough decision. I would speculate that a lot of doctors will like immunotherapy and use nivolumab even in mutated patients, and some considerable number of patients will get BRAF/MEK in the adjuvant setting.”
Axel Hauschild, MD, lead investigator on the COMBI-AD study, noted that patients might have a different perspective. Immunotherapy must be infused twice weekly while dabrafenib/trametinib are taken orally. He also noted that the treatments have distinct toxicity profiles.
“There will be patients who are not good candidates for targeted therapy,” said Hauschild, who is a professor of dermatology at the University of Kiel in Germany. “There will be patients who are not good candidates for immunotherapy. That is just a practical consideration.”The phase III CheckMate 238 trial included 906 patients with stages IIIB, IIIC, and IV completely resected melanoma who had ≥50% risk of relapse over 5 years. Patients were randomly assigned 1:1 to 3 mg/kg intravenous (IV) nivolumab given every 2 weeks or 10 mg/kg IV ipilimumab every 3 weeks for 4 doses, followed by 10 mg/kg every 12 weeks. The primary endpoint was RFS.
The FDA approved ipilimumab for adjuvant therapy of resected stage III melanoma based on an improvement in RFS versus placebo in a randomized phase III trial, but more than 50% of patients treated with ipilimumab experienced a grade 3/4 AE.5,6
“There is a clear need to improve the risk-benefit ratio of adjuvant treatment given the toxicity observed with the approved ipilimumab adjuvant regimen of 10 mg/kg, particularly for patients with high-risk resected stage IIIB/C and IV melanoma,” said Weber.
Most patients (85%) in the trial had cutaneous melanoma. Four of 5 patients in each arm had stage IIIB or IIIC disease, about 42% had a BRAF mutation, and nearly 34% had PD-L1 expression ≥5% (Figure). The median number of doses administered was 24 in the nivolumab group and 4 in the ipilimumab group. None of the participants remained on treatment at the time results were reported.
The Data Safety Monitoring Committee halted the study early due to clear evidence of benefit for nivolumab. In the planned interim analysis at a minimum follow-up of 18 months, the rate of RFS was significantly improved with nivolumab compared with ipilimumab (66.4% vs 52.7%), corresponding to a 35% reduction in risk (HR, 0.65; P <.0001).
Weber noted key milestones demonstrating superiority of nivolumab, “a 10% difference at 12 months, slightly widening to a 13% difference at 18 months, with 66% relapse-free at 18 months.” When the primary endpoint was analyzed by PD-L1 status, 1-year RFS was superior with nivolumab versus ipilimumab in the subset with PD-L1 expression <5% (64% vs 54%; HR, 0.71) and in those with PD-L1 expression ≥5% (82% vs 74%; HR, 0.50). The median progression-free survival was not reached in the nivolumab arm and was 15.9 months in the ipilimumab arm in the former subset; it was not reached in either arm in the latter subset.
Nivolumab was also superior for 1-year RFS in patients with BRAF-mutant disease (68% vs 63%; HR, 0.72) and BRAF wild-type disease (72% vs 57%; HR, 0.58).
Results were similar across prespecified patient subgroups. Distant metastases-free survival among stage III patients, an exploratory endpoint, was 80% at 1 year in the nivolumab arm and 73% in the ipilimumab arm (HR, 0.73; P = .0204). Postprotocol treatment with immunotherapy was twice as frequent in the ipilimumab arm versus the nivolumab arm (23% vs 11%). Weber noted that roughly 15% of patients in each arm had surgery as postprotocol treatment.
“I was struck by the fact that a fair number of patients could have surgery, suggesting that some of these patients were rendered free of disease yet again,” he said. He added that ipilimumab might retain a role for patients when nivolumab fails and another resection is possible.
There were significantly fewer treatmentrelated, clinically relevant AEs (grade 3/4) in the nivolumab arm compared with the ipilimumab arm (14% vs 46%). Ten percent of patients in the nivolumab group discontinued treatment due to AEs compared with 43% in the ipilimumab group.
There were no treatment-related deaths in the nivolumab group. One patient in the ipilimumab group died of marrow aplasia and another died of colitis. The percentage of patients with low-grade thyroid disorders, primarily hypothyroidism, was greater with nivolumab compared with ipilimumab (20.4% vs 12.6%).
“The results of CheckMate 238 are very exciting,” said John Haanen, MD, head of the Division of Medical Oncology at the Netherlands Cancer Institute in Amsterdam. “They show for the first time that an anti—PD-1 drug is superior in the adjuvant setting and, because of its lower toxicity, nivolumab is much easier to give than ipilimumab. The same occurs in the metastatic setting where anti-PD-1 treatment is more efficacious and has a much better safety profile and has replaced ipilimumab as first-line treatment.”
The FDA initially approved nivolumab for the treatment of patients with unresectable or metastatic melanoma and disease progression following ipilimumab and, if BRAF V600—mutation positive, a BRAF inhibitor. Nivolumab is now approved for patients with unresectable or metastatic disease as a single agent regardless of BRAF mutation status and in combination with ipilimumab.Results from COMBI-AD adjuvant treatment show that dabrafenib plus trametinib reduced the risk of relapse or death by 53% compared with placebo for patients with BRAF-mutant stage III melanoma, suggesting that the combination is likely to play an important role for these patients going forward.
After a median follow-up of 2.8 years, the 3-year RFS rate with dabrafenib and trametinib was 58% compared with 39% for placebo (HR, 0.47; 95% CI, 0.39-0.58; P <.001). “This is a very remarkable hazard ratio and if you look at the P value for the primary endpoint of RFS there are 13 zeroes, and therefore it is very impressive,” Hauschild said. Early data for overall survival (OS) showed that 86% of patients in the combination arm were alive at 3 years versus 77% with placebo (HR, 0.57; 95% CI, 0.42-0.79; P = .0006). At the interim analysis, the OS advantage was not yet deemed statistically significant, according to predefined criteria requiring a P value of .000019.
“For overall survival, there was a 43% reduction in the risk of dying from metastatic melanoma, noted Hauschild. “A very impressive result that is maintained over time.”
Patients with BRAF V600E/K stage III melanoma were randomly assigned to dabrafenib plus trametinib (n = 438) or placebo (n = 432). All patients were within 12 weeks of complete surgical resection and had stage IIIA (18%), IIIB (41%), or IIIC (40%) disease (Figure). Dabrafenib was administered at 150 mg twice daily with 2 mg trametinib once daily for 12 months. The salvage therapies received following the study were similar in each arm, and, in some cases, included a rechallenge with BRAF/MEK inhibition, Hauschild noted.
“Both the placebo and the combination group received the same amounts of salvage therapy. Seventy-four percent of the patients received systemic treatment for progressive disease,” he said. “It is unlikely that overall survival was impacted by the salvage therapies, since it was so balanced between the 2 groups.”
The baseline characteristics were similar between groups. In the combination arm, the median age of patients was 50 years and 91% of tumors had the BRAF V600E mutation with the remainder having the V600K alteration. Most patients (92%) had an ECOG performance status of 0. Twelve percent of patients in the combination group had in-transit metastases versus 8% with the placebo. Seventeen percent of patients had ≥4 positive lymph nodes, with the remainder having <4.
The median RFS was not reached with the combination versus 16.6 months for placebo. RFS was improved with dabrafenib/trametinib across all subgroups, Hauschild noted. “There was not a single outlier,” he noted. “All 3 stages benefited from the combination in the same way.” HRs across all subgroups ranged from 0.33 to 0.55 in favor of dabrafenib and trametinib versus placebo.
The 1-year OS rates were 97% versus 94% and the 2-year OS rates were 91% and 83% for the combination and placebo groups, respectively. The 1-year RFS rates were 67% versus 56% and the 2-year rates were 67% versus 44% for dabrafenib and trametinib versus placebo, respectively. The most common locations of recurrence, for the combination and placebo, respectively, were loco-regional (12% vs 25%), distant (22% vs 29%), and both local and distant (2% vs 2%).
The risk of distant metastases or death was reduced by 49% with the combination versus placebo (HR, 0.51; 95% CI, 0.40-0.65). Additionally, there was a 53% improvement in freedom from recurrence with the combination (HR, 0.47; 95% CI, 0.39-0.57).
Almost all patients (97%) assigned to dabrafenib and trametinib experienced AEs compared with 88% for placebo. The rates of grade 3/4 AEs were 41% for the combination and 14% for placebo. Overall, AEs led to discontinuation for 26% of those in the combination arm versus 3% with placebo. The most common all-grade AEs in the combination arm, which were mostly grade 1/2, were pyrexia (63%), fatigue (47%), and nausea (40%).
“There were fortunately no fatal adverse events with the combination of dabrafenib and trametinib. There were no new toxicities that were not described for stage IV melanoma,” said Hauschild. “There were AEs leading to discontinuation in 26% of the patients, which is a bit more than we expected in advance but this is the adjuvant setting so the pressure on the patient to continue getting therapy is lower than in the stage IV setting.”
The FDA initially granted an accelerated approval for the combination of dabrafenib and trametinib for patients with BRAF-mutant metastatic melanoma in January 2014. The combination received a full approval in November 2015, and is now also indicated for the treatment of patients with BRAF-mutant non—small cell lung cancer.Additional research presented at ESMO showed how elusive effective treatment for patients with completely resected, high-risk melanoma can be. In the BRIM8 trial, vemurafenib (Zelboraf), a BRAF inhibitor, induced mixed results for patients with high-risk disease.7 Vemurafenib is approved as single-agent therapy and in combination with cobimetinib (Cotellic), a MEK inhibitor, for BRAF-mutant unresectable or metastatic melanoma.
Patients with stage IIIC disease experienced a median disease-free survival (DFS) of 23.1 months in the experimental arm versus 15.4 months in the placebo arm. However, the difference did not achieve statistical significance. Patients with stage IIC-IIIB disease had a nonstatistically significant 46% reduction in the HR for DFS, but the trial design defined a positive outcome on the basis of results in later-stage disease.
In all, 498 patients were enrolled: cohort 1 was composed of 314 patients with stage IIC-IIIB disease and cohort 2 was composed of 184 patients with stage IIIC disease. Patients in each cohort were randomly assigned to vemurafenib or placebo (surgery only), and treatment continued for a year. The trial design prespecified DFS testing in the later-stage cohort and required a P value of ≤.05.
This DFS analysis yielded a 20% reduction in the HR in favor of vemurafenib, but the difference did not meet the prespecified definition of statistical significance (95% CI, 0.54-1.18; P = .2598).
The later-stage vemurafenib cohort had a 12-month DFS of 78.9% versus 58.0% with placebo. However, the 24-month DFS showed no difference between treatment groups; it was 46.3% with vemurafenib and 47.5% with placebo.
In patients with resected stage IIC-IIIB melanoma, median DFS had yet to be reached in the vemurafenib arm, whereas the placebo group had a median DFS of 36.9 months. The difference was associated with an HR of 0.54 (95% CI, 0.37- 0.78; P = .0010).
The 12-month DFS was 84.3% with vemurafenib and 66.2% with placebo. At 24 months, 72.3% of vemurafenib-treated patients remained alive without disease recurrence versus 56.5% in the placebo group.
“The magnitude of the DFS benefit [in earlier-stage disease] is of a proportion not previously demonstrated in this clinical setting,” said lead author Karl Lewis, MD, a medical oncologist at the University of Colorado Comprehensive Cancer Center in Aurora. “However, due to the prespecified statistical design of the study, these data cannot be described as statistically significant.”
Lewis said it is possible there may be different disease biology in later-stage disease as compared with stage IIC-IIIB. Furthermore, the 12-month duration of adjuvant therapy might not have been sufficient for the stage IIIC-IV cohort.
While the work to pair the ideal patient with the ideal treatment continues, physicians treating melanoma now have more tools in their armamentarium than ever before.
“As I tell patients, ‘Cancer is a bad disease. The good news is you have options,’” Weber said. “Now our patients have clearly beneficial options and that’s great news.”