乌鸦传媒

Expert Critique

FROM THE ASCO Reading Room

Keith Michael Giles, PhD Postdoctoral Research Fellow New York University School of Medicine

Small molecule drugs targeting the MAPK pathway and antibodies blocking immune checkpoints can prolong the survival of patients with advanced melanoma. While some patients with Stage III melanoma 鈥� with lymph node involvement and the absence of distant metastases 鈥� can be cured by complete surgical resection, others are at significant risk of disease relapse and could benefit from adjuvant treatment. This article explores the debate around the best approach to adjuvant treatment for resected Stage III melanoma patients, drawing on recent clinical trials of combined BRAF/MEK inhibition or immune checkpoint inhibitors in this patient population.

While the BRIM-8 study showed that adjuvant BRAF inhibitor monotherapy does not improve recurrence-free survival (RFS), adjuvant treatment of resected Stage III, BRAF-mutated melanoma with combined BRAF/MEK inhibition can reduce the risk of relapse and extends overall survival (OS), with side effects that usually resolve when the therapy is stopped. Other studies have shown that adjuvant immune checkpoint inhibitor therapy improves RFS in patients with resected Stage III melanoma, and that anti-PD-1 agents (nivolumab or pembrolizumab) produce superior RFS to anti-CTLA-4 (ipilimumab). However, it is not yet clear whether adjuvant anti-PD-1 improves OS in these patients.

Toxicity is an important issue for adjuvant treatment, and severe toxicity is even less acceptable in this setting as many resected Stage III melanoma patients will be cured by surgery alone. Toxicity from checkpoint inhibitors can often persist beyond cessation of treatment and can require multi-disciplinary management. The ongoing CheckMate 915 study is comparing ipilimumab and nivolumab against nivolumab alone in the adjuvant setting; in unresectable melanoma the combination of these drugs is generally thought to be more efficacious but also tends to be more toxic. As treatments previously reserved for unresectable Stage III or IV melanoma enter the adjuvant setting, the total number of patients receiving these agents and hence experiencing toxicity will increase.

Ultimately, the optimal choice for adjuvant treatment depends on the benefit that can be achieved against the risks of relapse and toxicity. At the present time, combined BRAF/MEK inhibitor or anti-PD-1 therapies are likely to be good options for adjuvant treatment, although BRAF/MEK inhibitors are obviously not appropriate for BRAF wild-type melanoma. Randomized comparator studies that assess RFS and OS are needed where combined BRAF/MEK inhibition is pitted directly against immune checkpoint inhibitor therapy in patients with resected Stage III melanoma. Development and validation of pre-treatment or early on-treatment biomarkers to help predict the risk of onset of severe toxicity from targeted therapy or immunotherapy could help clinicians better understand the risk-benefit equation for these patients. Similarly, biomarkers that can predict the risk of relapse might spare some patients from unnecessary adjuvant treatment, or prioritize others for this therapy. In other words, can we predict who is likely to have been cured by surgery alone? Identifying patients at significant risk of relapse and who are unlikely to experience severe treatment-related side effects will be important for future applications of adjuvant treatment for melanoma, for example, in high-risk, node-negative Stage IIB or IIC disease.

Full Critique

The landscape of adjuvant treatment for patients with stage III melanoma has changed since the introduction of targeted therapies and immunotherapy. The most recent ASCO annual meeting featured a debate about the optimal choice for first-line adjuvant therapy in this patient population and the criteria for treatment selection.

The Case for Targeted Therapy

Randomized phase III data have demonstrated that BRAF/MEK inhibition improves both overall survival (OS) and relapse-free survival (RFS) in patients with resected stage III melanoma. In addition, multiple trials in stage III and stage IV disease show that permanent toxicities from BRAF/MEK inhibition occur only rarely. "I mention that because as we look into using more and more treatments in the adjuvant setting, quality of life becomes more of an issue," said Ragini Reiny Kudchadkar, MD, of Winship Cancer Institute of Emory University in Atlanta.

Given those realities, "it seems that everyone should be given targeted therapy in the adjuvant setting," she said. "However, I think adjuvant BRAF/MEK inhibition has challenged our traditional dogma. There are many melanoma oncologists that I've spoken to that don't believe targeted therapy should be used at all. There's a dogma that only immunotherapy leads to durable survival and that immune therapy is ideal, because you can stop treatment."

The other dogma, she said, is that BRAF/MEK inhibition ultimately leads to resistance and disease progression and that patients must be on the drug to actively suppress cancer growth.

Kudchadkar set out to debunk these theories and convince the audience that targeted therapy in the adjuvant setting can lead to durable survival.

The case for durability with BRAF/MEK inhibition (in stage IV disease) comes from a recent study by , in which a tail to the OS curve was observed about 3 years out. The 5-year OS rate in the overall study population was 28%, which increased to 45% in patients with normal lactate dehydrogenase (LDH) and to 51% in those with normal serum LDH levels and ≤3 organ sites with metastases.

These data were corroborated by from the COMBI-d and COMBI-v studies using dabrafenib and trametinib. in 2017 found that adjuvant dabrafenib and trametinib was associated with a 3-year OS rate of 86% and 3-year RFS rate of 58% -- both highly statistically significant, Kudchadkar said. The risk of death was improved significantly regardless of the disease stage.

In the stage IV setting, immunotherapy, by comparison, led to a 5-year OS rate of 41% in the treatment-naïve population and 34% in the treatment-refractory plus treatment-naïve patients in the KEYNOTE-001 study. demonstrated a 2-year OS rate of 55% with pembrolizumab or ipilimumab.

Although cross-study comparisons are fraught with peril, "I do think it gives you an idea of how long patients, even with stage IV, are living with both PD-1 antibody as well as targeted therapy," said Kudchadkar. "Overall, I think this tells us that there are patients with long-term durable responses with BRAF/MEK inhibition, that there are some predictors of who will do well from BRAF/MEK inhibition with normal LDH and low-volume disease."

In patients with resected stage III melanoma, it makes complete sense to use these drugs in the adjuvant setting, she argued. "Most of these patients have normal LDH and low-volume micrometastatic disease that hopefully can be eradicated."

A total of 41% of patients receiving targeted therapy have grade 3 or 4 toxicity, primarily fever, representing one challenge to the use of targeted therapy. Although the rate of toxicity may be high, most is fully reversible with drug discontinuation, Kudchadkar said. In contrast, permanent toxicities, such as the development of type I diabetes, are a risk with immunotherapy.

The rate of secondary cancers including secondary melanomas is about 2.3% in those treated with BRAF/MEK inhibition. Finally, testing for BRAF mutation can be challenging in patients with very low-volume disease, she said.

"Targeted therapy should be front line for adjuvant treatment of stage III melanoma in the BRAF-mutated setting," Kudchadkar concluded. "The reasons you can be confident in that is that studies [of targeted therapy] have longer follow-up than any other modern adjuvant studies."

Immune Therapy the Choice in BRAF Wild Type

Taking the other side in the debate was Olivier Michielin, MD, PhD, head of the Personalized Medicine Analytical Oncology and Melanoma Clinic of Lausanne University Hospital in Switzerland.

In BRAF wild-type patients, PD-1 blockade is definitely the first choice for adjuvant treatment for stage IIIA, B, and C melanoma, and in BRAF-mutated melanoma, both PD-1 blockade and dabrafenib plus trametinib "are excellent options," but he said he leans toward immunotherapy with an eye on OS data to come.

Once the immune system is mobilized, T cells can go into adaptation, said Olivier, in arguing for immunotherapy as a first choice in the adjuvant setting. "There is a memory effect and clearly a well-demonstrated long-term benefit [with immunotherapy] in the metastatic setting. The key question is whether or not we already have enough data now to see the same trend in the adjuvant setting," he said.

Although the nature of residual disease in the adjuvant setting is difficult to assess, PD-1/PD-L1 can be found in the lymph node in microscopic disease (adjuvant setting). "We still need to have a broader view on what's happening at the systemic level on the subclinical metastasis," he admitted.

The evidence to support immunotherapy in the adjuvant setting in resected stage III melanoma is plentiful, and includes the trial showing the superiority of ipilimumab over placebo, the trial demonstrating superiority of pembrolizumab over placebo, and Checkmate 238 showing superiority of nivolumab over ipilimumab.

With a plethora of treatment options, including targeted therapy (dabrafenib plus trametinib), in the adjuvant setting, the choice of agent is becoming more complex, said Michielin. Hazard ratios for OS are missing for many of the immunotherapy trials, although the improvement in distant metastasis-free survival serves as an interesting surrogate in that it is usually associated with OS in the adjuvant setting.

Patients with stage III disease entered into the immunotherapy trials were required to have complete lymph node dissection, "which is no longer the population that is going to walk through our doors," Michielin said. "How these numbers are going to be affected by this new population resulting from the absence of complete lymph node dissection remains to be validated."

An overview of progression-free survival outcomes shows consistent improvement across disease stage in the more recent immunotherapy trials, he noted, although confidence intervals are wide given the low number of events.

The data available so far show three independent prospective randomized trials (two placebo-controlled) with immunotherapy in advanced melanoma that were all positive for the primary endpoint of RFS. One placebo-controlled randomized trial of targeted therapy in this setting was positive for the primary endpoint of RFS and OS as a secondary endpoint. One other prospective placebo-controlled randomized trial of single-agent BRAF inhibition was negative for RFS as the primary endpoint in patients with stage IIIC melanoma, but RFS was numerically improved in patients with stage IIC-IIIB disease.

"Very important, we have no head-to-head comparison, and that's clearly what we need to move forward," said Michielin.

The rate of treatment-related toxicity, including grade 3-4 events, is about twice as high with targeted therapy versus immunotherapy, with a higher rate of adverse events leading to treatment discontinuation in patients receiving targeted therapy. Michielin admitted that toxicities from immunotherapy tend to be long term, and require experienced medical teams to manage, as treatment cessation will not resolve the toxicity in most cases, unlike with tyrosine kinase inhibitors and spontaneous resolution of toxicities upon cessation.

An important question to guide decision-making is whether the PD-1-blocking antibodies are similar, he said, noting that a network meta-analysis predicts, based on direct and indirect treatment comparisons, that the hazard ratio for nivolumab is similar to that of pembrolizumab for RFS. Although an OS benefit to immunotherapy can be inferred from an improvement in RFS, a definitive answer on OS benefit will have to await analyses of this endpoint in the EORTC 1325 and Checkmate 238 trials.

Similarly, longer follow-up from COMBI-AD is needed to determine whether dabrafenib-trametinib improves OS or simply delays relapse, Michielin said. After 1 year of adjuvant treatment with targeted therapies in placebo-controlled trials, the rate of progression-free survival declines to that of the placebo arm.

The development of predictive biomarkers should be a priority to best select patients for targeted therapy or immunotherapy in the years to come, he concluded.