Subgroup findings suggest a PFS benefit with the addition of encorafenib/binimetinib in patients with liver metastases.
A progression-free survival (PFS) improvement did not occur when adding encorafenib (Braftovi) and binimetinib (Mektovi) to nivolumab (Opdivo)/ipilimumab (Yervoy) before subsequent nivolumab monotherapy in those with metastatic or unresectable BRAF V600E/K–mutated melanoma, according to data from the phase 2 EBIN trial (NCT03235245) presented at the 2024 American Society of Clinical Oncology (ASCO) Annual Meeting.1
Findings showed that patients treated in the encorafenib/binimetinib arm (n = 136) experienced a median PFS of 9 months (95% CI, 7-13) and 9 months (95% CI, 5-14) for those given nivolumab/ipilimumab alone (n = 135; HR, 0.87; 90% CI, 0.67-1.12; stratified log-rank P = .360). The 12-week, 6-month, and 24-month PFS rates in the encorafenib/binimetinib arm were 99% (95% CI, 95%-100%), 62% (95% CI, 53%-70%), and 29% (95% CI, 20%-38%), respectively. Those respective rates in the control arm were 73% (95% CI, 64%-80%), 56% (95% CI, 47%-64%), and 35% (95% CI, 26%-44%).
“If we take into consideration the intention-to-treat [ITT] population, there was no difference in PFS, which was the main objective of the trial. However, findings do support the hypothesis that patients with high lactate dehydrogenase [(LDH) at or above] twice the upper limit of normal [ULN] and patients with liver metastases—which are a frequent occurrence in metastatic melanoma—could benefit from this approach [using encorafenib/binimetinib induction],” lead study author Caroline Robert, MD, PhD, of the Department of Cancer Medicine at Gustave Roussy Cancer Campus at the University of Paris-Saclay in France, said in a presentation of the data.
Immunotherapy-based regimens consisting of a PD-1 inhibitor with or without a CTLA-4 inhibitor or a LAG-3 inhibitor, as well as targeted therapy consisting of the combination of BRAF and MEK inhibitors, are approved first-line treatments for patients with advanced BRAF V600E/K–mutated melanoma. Although using a targeted therapy regimen following disease progression on an immunotherapy-based regimen is considered the optimal sequencing for this patient population, the potential benefit of targeted therapy followed by immunotherapy before disease progression is unknown.
EBIN evaluated this approach in patients with unresectable or metastatic melanoma harboring BRAF V600E/K mutations per investigator assessment. Patients were required to have an ECOG performance status of 0 or 1. No prior treatment for advanced melanoma was allowed; however, adjuvant therapy completed at least 6 months prior to enrollment was permitted. Patients were not allowed to have uveal melanoma, untreated or symptomatic brain metastases, or leptomeningeal involvement.
During the international, multicenter, superiority, open-label trial, patients were randomly assigned 1:1. In the experimental arm, patients received 450 mg of encorafenib once per day plus 45 mg of binimetinib twice per day for 12 weeks as induction therapy. After 1 week off therapy, they received 3 mg/kg of nivolumab plus 1 mg/kg of ipilimumab once every 3 weeks for 4 total doses, followed by 480 mg of nivolumab once every 4 weeks. In the control arm, patients received the same regimen of nivolumab plus ipilimumab, followed by nivolumab monotherapy.1,2
Notably, after disease progression, patients in the experimental arm went on to receive encorafenib plus binimetinib. Those in the control arm received investigator’s choice of therapy after progression.1
Stratification factors included stage/LDH level (unresectable stage III/M1a with LDH ≤ ULN vs M1b/M1c with LDH ≤ ULN vs ULN < LDH ≤2 ULN vs LDH > 2ULN).
Regarding disease stage at randomization, patients had unresectable stage III (experimental arm, 19%; control arm, 18%), stage IV M1a (9%; 9%), stage IV M1b (13%; 13%), and stage IV M1c (59%; 60%) disease. LDH level at randomization included no higher than ULN (52%; 53%), between ULN and two times ULN (36%; 37%), and greater than two times ULN (12%; 10%). Thirteen percent of patients in the experimental arm and 21% of patients in the control arm had an ECOG performance status of 1. The rates of liver metastases were 30% in the experimental arm and 24% in the control arm. Seven percent of patients in both arms received prior adjuvant therapy.
All 136 patients randomly assigned to the experimental arm started treatment with encorafenib/binimetinib, and 126 started nivolumab/ipilimumab. In the control arm, 4 of the 135 patients randomly assigned did not start protocol treatment due to withdrawn consent (n = 2) or the initiation of targeted therapy (n = 2).
Additional data from a prespecified subgroup analysis showed varying PFS outcomes for the experimental vs control arm, including in patients with stage M0/M1a disease with LDH no higher than ULN (HR, 2.09; 95% CI, 0.97-4.54), those with stage M1b/M1c disease with LDH no higher than ULN (HR, 0.74; 95% CI, 0.43-1.29), patients with LDH above ULN but no greater than two times ULN (HR, 0.86; 95% CI, 0.54-1.37), and those with LDH above two times ULN (HR, 0.46; 95% CI, 0.21-1.03).
Notably, a post-hoc analysis of PFS in other subgroups showed that patients with liver metastases experienced a benefit when treated with encorafenib/binimetinib induction (HR, 0.49; 95% CI, 0.29-0.84). Other subgroups examined in the post-hoc analysis included those without liver metastases (HR, 1.10; 95% CI, 0.76-1.58), patients with less than 3 involved sites (HR, 0.72; 95% CI, 0.48-1.08), those with 3 or more involved sites (HR, 1.45; 95% CI, 0.92-2.26), patients with target lesions with a sum of less than 5 cm (HR, 1.01; 95% CI, 0.62-1.63), those with target lesions with a sum of at least 5 but less than 10 cm (HR, 1.04; 95% CI, 0.62-1.74), and patients with target lesions with a sum of at least 10 cm (HR, 0.83; 95% CI, 0.45-1.51).
PFS events in the ITT population included metastasis in the lymph node, skin, or other soft tissue only (experimental arm, n = 10; control arm, n = 10); metastasis in the lung and not in the visceral area or central nervous system (CNS; n = 9; n = 3); metastasis in the visceral area but not the CNS (n = 17; n = 24); CNS metastasis (n = 34; n = 11), progression without new reported metastases (n = 18; n = 33); and death without progression (n = 2; n = 2).
Patients treated with encorafenib/binimetinib experienced an objective response rate (ORR) of 53%, including a complete response (CR) rate of 12% and a partial response (PR) rate of 41%. In the control arm, the ORR was 45% with CR and PR rates of 10% and 36%, respectively.
An exploratory analysis showed that the 2-year overall survival rates were 68% (95% CI, 57%-77%) for patients treated with encorafenib/binimetinib induction compared with 74% (95% CI, 64%-82%) for those who started with nivolumab/ipilimumab.
Regarding safety, grade 3 or higher treatment-related adverse effects (TRAEs) occurred in 43% of patients in the encorafenib/binimetinib arm vs 32% of patients in the control arm. Notably, instances of grade 3 or higher myocarditis and grade 3 or higher adrenal insufficiency were each reported in 1 patient in the experimental arm; neither of these events occurred in the control arm. The rates of grade 5 TRAEs were 1% in each arm, which included heart failure (n = 1) and arrhythmia (n = 1) in the experimental arm and meningitis (n = 1) in the control arm.
Disclosures: Robert reported stock ownership of RiboNexus; serving in a consulting or advisory role with AstraZeneca, Bristol Myers Squibb, MSD, Novartis, Pfizer, Pierre Fabre, Roche, Sanofi, and Sun Pharma; and receiving institutional research funding from Novartis and Phio Pharmaceuticals.