The clinical development of targeted agents and immunotherapy for use in melanoma has occurred on parallel tracks, amassing a wealth of positive data and expanding treatment choices for patients. However, no head-to-head trials have compared the two approaches — so for a patient with BRAF-mutated metastatic melanoma, which treatment should be tried first?
Some answers come from a network meta-analysis — a statistical modeling method — that allows comparisons between therapeutic options across clinical studies.
The analysis found no difference in overall survival between targeted therapy with BRAF/MEK inhibitors and immunotherapy with programmed cell death 1 (PD-1) inhibitors, but both of these treatment approaches were superior to chemotherapy.
However, patient outcomes differed.
BRAF/MEK inhibitors provided better progression-free survival (PFS) and objective response rates compared with immunotherapy with PD-1 or cytotoxic T-lymphocyte–associated antigen 4 (CTLA-4) inhibitors.
In addition, the PD-1 inhibitors were associated with the lowest risk for serious adverse events (SAEs).
"This is the first analysis to draw comparison between targeted and immune therapies for BRAF-mutated melanoma," corresponding author Feng Xie, PhD, from the Department of Clinical Epidemiology & Biostatistics at McMaster University, Hamilton, Ontario, Canada, said in a press release.
"Although overall survival [OS] was similar with both classes of drugs, the lower risk for adverse events with PD-1 inhibitors made them the drugs of choice in the first-line setting for BRAF metastatic melanoma," Dr Xie told Medscape Medical News.
"In situations where rapid response is a priority, targeted therapy with BRAF/MEK inhibitors may be considered first," he added.
The report was published online October 27 in JAMA Oncology.
Melanoma Experts Weigh In
Medscape Medical News reached out to several melanoma experts — some of whom were involved in clinical trials included in the analysis — for comments on how the results of this analysis relate to clinical practice.
"While the analysis reflects current clinical practice, there are some nuances of clinical practice that a meta-analysis cannot address," Marc S. Ernstoff, MD, professor and chair of the Department of Medicine and senior vice president of clinical investigation at Roswell Park Cancer Institute, Buffalo, New York, told Medscape Medical News. "This analysis is useful because it provides some security for practicing physicians in choosing one class of agents over another," he added
However, some nuances of using these agents in clinical practice revolves around an overestimating or underestimating of toxicities, depending on one's bias, Dr Ernstoff explained. This is true of both targeted agents and immunotherapies, he indicated.
It is true that toxicities with single-agent immunotherapy are lower than with combination immunotherapy, he said. However, most toxicity is reversible with high-dose steroids, he pointed out. The percentage of patients who fully recover from toxicities may be available from a QWIST (quality time without symptoms from treating) analysis, he pointed out.
Another nuance of clinical practice that a meta-analysis cannot distinguish is how long a patient may require treatment. He advocates enrollment in a clinical trial to answer this question: one that compares an immunotherapy combination (PD-1 + CTLA-4 inhibition) vs BRAF/MEK inhibition. Such a trial is ongoing and will address an issue that the meta-analysis is trying to tease out, he said.
"This is an interesting study that tries to answer an important and relevant clinical question," said Michael Postow, MD, medical oncologist at the Memorial Sloan Kettering Cancer Center, New York, New York.
"The conclusions are consistent with clinical practice," he added, indicating that immunotherapy is the go-to choice for patients unless symptom relief is a priority. He explained that on the basis of data from randomized trials, there is equipoise between BRAF/MEK inhibitors and PD-1 inhibitors, but safety tilts in favor of PD-1 inhibitors.
Dr Postow also indicated that in some countries, physicians are required to use BRAF/MEK inhibitors before PD-1 inhibitors.
However, he indicated that there is still an elephant in the room: the combination of PD-1 and CTLA4 inhibitors, for which the study drew no substantive conclusions.
Dr Xie concurred. "While the data in our study represents the best available evidence, using more than one kind of immunotherapy shows promise in clinical trials and may change the treatment landscape once longer-term results are published," he commented in a press release.
Dr Postow indicated that CheckMate 067, which compared a combination of PD-1/CTLA-4 inhibitors vs a PD-1 inhibitor, has yet to yield long-term overall survival information.
Georgina Long, PhD, MBBS, professor of melanoma medical oncology and translational research at the Melanoma Institute Australia at the University of Sydney, also indicated that this study is consistent with clinical experience. "But enough follow-up and maturity to be sure at this stage is desired," she told Medscape Medical News.
"It highlights issues with primary resistance with anti-PD1 agents and hence the shorter and poorer PFS for these agents compared with targeted therapy," she explained. She further expanded that acquired resistance is more of a problem with targeted therapy; hence, the PFS advantage is lost and the OS is equalized with PD-1. "We need more mature data, however," she said.
The Network Meta-Analysis
For the network meta-analysis, the researchers included phase 2 or 3 trials that included treatment-naive patients with unresectable BRAF metastatic melanoma. Interventions had to include targeted therapy (BRAF or MEK) or an immune checkpoint (PD-1 or CTLA4) inhibitor.
Of 2546 publications, 2 reviewers whittled the list to 16 studies eligible for the meta-analysis, which included 15 randomized controlled studies with 6662 patients; all were multicenter (13 were global) and published between 2011 and 2015.
Dr Xie said, "We incorporated all the evidence identified from published randomized clinical trials using robust analytical tools into a single network." Using Bayesian network meta-analysis, the authors compared the drugs across the studies with "a Markov Chain Monte Carlo…simulation technique with 100 000 iterations in each of the 4 chains."
For the OS network, 13 trials (5361 patients) compared 9 treatment strategies. The analysis indicated that BRAF/MEK and PD-1 inhibition were associated with improved OS compared with all other treatments.
For the PFS network, 14 trials (6738 patients) compared 10 treatment strategies. BRAF/MEK inhibition showed a significant advantage over all other treatment modalities, followed by PD-1/CTLA-4 inhibition compared with PD-1 inhibition alone; MEK inhibition alone was likely the worst treatment option.
For the SAE network, 8 trials (4395 patients) compared 8 treatment strategies reporting grade 3 or higher SAEs. With an odds ratio (OR) of 1.0, risk for SAEs between chemotherapy and PD-1 inhibitors was similar.
Treatment with PD-1/CTLA4 inhibitors was associated with the highest risk for SAEs compared with either CTLA4 (odds ratio [OR], 1.63; 95% credible interval [CrI], 1.19 - 2.26) or PD-1 inhibition alone (OR, 2.99; 95% CrI, 2.18 - 4.12).
Dr Xie indicated that as evidence accumulates and longer-term data become available, such an analysis might be undertaken for individual drugs rather than be restricted to drug classes. In future studies, it will also be important to determine the cost-effectiveness of each therapy, he added.
So far, there has been a relatively purist approach in treatment strategies — keeping immunotherapies apart from targeted therapies. However, the horizon is slowly expanding to include the combination. For example, Roche/Genentech recently announced that it is exploring the combination of its PD-L1 inhibitor together with its BRAF/MEK inhibitors for patients with BRAF-mutant melanoma.
The study authors have disclosed no relevant financial relationships. Dr Postow and Dr Long were investigators on several industry-sponsored trials used in the network meta-analysis. Dr Ernstoff was an investigator of early studies with immunotherapies. JAMA Oncol. Published online October 27, 2016.