Fianlimab/Cemiplimab Demonstrates Persistent Activity in Advanced Melanoma, Independent of LAG-3 and PD-L1 Status

Fianlimab combined with cemiplimab-rwlc (Libtayo) demonstrated persistent and significant clinical activity in patients with advanced melanoma, irrespective of LAG-3 or PD-L1 expression, according to results from a combined post hoc analysis of three independent cohorts from a phase 1 trial (NCT03005782) presented at the 2024 ESMO Congress.

Tumor responses assessed by blinded independent central review (BICR) in a combined cohort of 98 patients showed that at a median follow-up of 23 months (interquartile range, 15-31), the overall response rate (ORR) was 57% (95% CI 47%-67%), including a complete response (CR) rate of 25% and a partial response (PR) rate of 33%. The rates of stable disease (SD) and progressive disease (PD) were 17% and 15%, respectively. Seven percent of patients were not evaluable (NE), and 3% of patients did not reach CR or PD. The disease control rate (DCR) was 78% (95% CI, 68%-85%). The median time to response was 1.5 months, and the median time to CR was 4.1 months.

Notably, 70% of patients experienced some level of tumor reduction. The median duration of response (DOR) was not reached (NR; 95% CI, 23–not estimable [NE]). The median progression-free survival (PFS) was 24 months (95% CI, 12-NE); the 12- and 24-month PFS rates were 60% (95% CI, 49%-69%) and 49% (95% CI, 36%-62%), respectively. The median overall survival was NR (95% CI, 42-NE).

"With a longer follow-up of 23 months, fianlimab plus cemiplimab in patients with advanced melanoma showed persistent high clinical activity by BICR assessment regardless of PD-L1 or LAG-3 status and across high-risk subgroups with an acceptable safety profile,” lead study author Meredith McKean, MD, of Sarah Cannon Research Institute in Nashville, Tennessee, and colleagues, wrote in a poster presentation of the data.

Study Overview

The open-label, nonrandomized, multi-cohort expansion study enrolled patients at least 18 years of age with metastatic or locally advanced, inoperable non-uveal melanoma. Eligible participants were required to have an ECOG performance status of 0 or 1 and possess at least one measurable lesion by RECIST 1.1 criteria. Key exclusion criteria included prior treatment with LAG-3 targeting agents, prior anti–PD-(L)1 therapy for advanced melanoma, and radiation therapy within 2 weeks prior to enrollment.

Patients were divided into three distinct cohorts: the initial cohort (MM1) included 40 patients with first- or second-line advanced melanoma who had never received anti–PD-(L)1 therapy. The confirmatory cohort (MM2) consisted of 40 patients with first-line advanced melanoma who were also naive to anti–PD-(L)1 agents. The neoadjuvant/adjuvant experienced cohort (MM3) enrolled 18 patients who received prior perioperative systemic therapy, including 13 who had received anti–PD-(L)1 therapy.

All patients received fianlimab at 1600 mg combined with cemiplimab at 350 mg once every 3 weeks for up to 24 months. The primary end point was ORR per RECIST 1.1 criteria. Secondary end points included PFS, DOR, DCR, safety, and pharmacokinetics. The data cutoff for this analysis was October 31, 2023.

Patient Characteristics and ORR by Cohort

In the combined population of all three cohorts, the median age was 68 years (range, 24-88). Sixty percent of patients were male, and 90% were White.

In the MM1 cohort, the ORR was 60% (95% CI, 43%-75%) with a median DOR that was NR (95% CI, 23-NE). Patients in the MM2 cohort achieved an ORR of 63% (95% CI, 46%-77%) with a median DOR of NR (95% CI, 14-NE). In the MM3 cohort, the ORR was 39% (95% CI, 17%-64%) with a median DOR of NR (95% CI, 6-NE).

CR rates were 23% in MM1, 25% in MM2, and 28% in MM3. The median PFS was NR (95% CI, 8-NE), 19 months (95% CI, 8-NE), and 12 months (95% CI, 1-NE), respectively.

Clinical Activity by BICR in Subgroups With Poor Prognosis

Among patients with prior anti–PD-(L)1 treatment in the neoadjuvant or adjuvant setting (n = 13), the ORR was 46% (95% CI, 19%-75%) with a CR rate of 31% and a DCR of 69% (95% CI, 39%-91%). The median DOR was NR (95% CI, 6-NE).

In patients with liver metastases (n = 20), the ORR was 35% (95% CI, 15%-59%) with a CR rate of 0% and a DCR of 55% (95% CO, 32%-77%). The median DOR was NR (95% CI, 7-NE).

The subgroup with elevated lactate dehydrogenase above the upper limit of normal (n = 31) achieved an ORR of 55% (95% CI, 36%-73%) with a CR rate of 13% and a DCR of 71% (95% CI, 52-86). The median DOR was NR (95% CI, 13-NE).

Efficacy By LAG-3 and PD-L1 Expression

Fianlimab and cemiplimab demonstrated an ORR of 50% in patients with PD-L1 expression of less than 1% and 71% in those with a PD-L1 expression of at least 1%. Similarly, the ORRs were 50% and 61% in patients with LAG-3 expression of less than 1% and at least 1%, respectively.

In patients evaluable for circulating tumor DNA (ctDNA; n = 31), 15 patients cleared ctDNA by cycle 4, day 1; 2 patients had no detectable ctDNA at baseline.

The mean time to progression without ctDNA clearance was 3.2 months. Conversely, among the 15 ctDNA-negative patients, only one experienced disease progression at the time of the analysis.

Safety Analysis

The safety profile of the fianlimab plus cemiplimab was largely consistent with that observed for cemiplimab monotherapy and other anti–PD-(L)1 agents. However, study authors noted that treatment-related adrenal insufficiency of any grade occurred in 12% of patients, with 5% experiencing grade 3 or higher effects. Among patients who experienced any-grade adrenal insufficiency (n = 12), the ORR was 92%, (95% CI 62%-100%).

Adverse effects (AEs) of any grade occurred in 95% of patients. Grade 3 or higher AEs were reported in 47% of patients, serious AEs occurred in 36%, and immune-mediated AEs were noted in 13% of patients.

Reference

McKean M, Nye S, Papadopoulos K, et al. Long-term follow-up of advanced melanoma (unresectable/metastatic - aMel) patients treated with fianlimab (FIAN) + cemiplimab (CEMI): Results from blinded independent central review (BICR) efficacy assessment. Presented at: 2024 ESMO Congress; September 13-17, 2024; Barcelona, Spain. Abstract 1097P.