Introduction LBL-034 is a humanized IgG1 subtype asymmetric bispecific antibody targeting GPRC5D and CD3 with a 2:1 format. Its unique design features affinity optimization and steric hindrance within the anti-CD3 arm, effectively reducing off-target engagement with bystander T cells and minimizing non-specific T cell activation. Additionally, the bivalent, high-affinity anti-GPRC5D Fab domains enable potent, conditional T cell activation exclusively in the presence of GPRC5D-expressing tumor cells. Here, we report the robust clinical efficacy and favorable safety data from the phase I/II study (NCT06049290) in patients with relapsed/refractory multiple myeloma (RRMM).

Methods This ongoing phase I/II, open-label study includes a dose escalation stage followed by dose expansion (phase I). Eligible RRMM patients had received ≥3 prior lines of therapy, including a proteasome inhibitor and an immunomodulatory drug, with or without an anti-CD38 monoclonal antibody. LBL-034 was administered intravenously at escalating dose levels of 10, 30, 80, 200, 400, 800, and 1200 μg/kg. Dosing occurred on Days 1 and 15 of each 4-week cycle. Starting from the 80 μg/kg cohort, a step-up dosing strategy (10 to 80 μg/kg) was implemented to mitigate CRS risk. Tumor responses were assessed by investigators per IMWG criteria. AEs were graded per CTCAE v5.0, and CRS and ICANS were graded according to the ASTCT consensus. Minimal residual disease (MRD) negativity in bone marrow aspirates was evaluated by next-generation sequencing (Seq-MRD®).

Results As of July 15, 2025, 56 patients were enrolled in phase I dose escalation/expansion. The dose distribution was as follows: 10 μg/kg (n=1), 30 μg/kg (n=1), 80 μg/kg (n=6), 200 μg/kg (n=7), 400 μg/kg (n=19), 800 μg/kg (n=11), and 1200 μg/kg (n=11).

Median age was 64 years (range 34-79), and patients had received with a median of 5 prior lines of therapy (range 3-11). Most patients were triple-class exposed (n=39, 69.6%) and penta-drug exposed (n=32, 57.1%). Prior therapies included ASCT (n=16, 28.6%) and BCMA-directed CAR-T (n=9, 16.1%). Extramedullary disease (EMD) was present in 13 patients (23.2%).

Efficacy: Among 55 efficacy evaluable patients (≥1 post-baseline tumor assessment), ORR was 70.9% (39/55), with dose-dependent trends observed:

200 μg/kg (mFU 11.6 m): ORR 57.1%; ≥VGPR 57.1%; ≥CR 14.3%

400 μg/kg (mFU 10.4 m): ORR 77.8%; ≥VGPR 61.1%; ≥CR 55.6%

800 μg/kg (mFU 5.1 m): ORR 90.9%; ≥VGPR 81.8%; ≥CR 54.5%

1200 μg/kg (mFU 2.7 m): ORR 81.8%; ≥VGPR 63.6%; ≥CR 27.3%

MRD negativity was achieved in 84.2% (16/19) at the 10-5 threshold among MRD-evaluable and ≥CR patients. Median time to first CR shortened with higher doses: 6.0 months (m) at 400 μg/kg, 3.4m at 800 μg/kg, and 2.7m at 1200 μg/kg. Notably, strong clinical benefit was observed in difficult-to-treat subgroups at doses ≥400 μg/kg:

Penta-drug exposed: ORR 78.1%

Post-BCMA-targeted therapy: ORR 85.8%

EMD

ORR 75.0%

The mPFS at doses ≥200 μg/kg (n=47) was 11.7m (8.1, NE) with a median follow-up (mFU) of 6.8m.

Safety: All patients experienced TEAEs, with ≥ grade 3 (G3) TEAEs reported in 83.9%. The most frequent ≥ G3 hematologic TEAEs included:

Lymphopenia (51.8%)

Neutropenia (28.6%)

Leukopenia (25.0%)

Thrombocytopenia (17.9%)

Anemia (16.1%)

CRS occurred in 73.2% of patients, predominantly G1 (51.8%) and G2 (19.6%), and was primarily observed during the first few days of Cycle 1. Only one patient experienced G3 (1.8%) CRS, with no ≥ G4 event reported.

Quality-of-life related TEAEs included dysgeusia (G1 44.6%, G2 5.4%, ≥G3 none), nail disorder (G1 42.9%, G2 5.4%, ≥G3 none), skin disorder (G1 39.3%, G2 3.6%, ≥G3 none) and decreased appetite (G1 3.6%, G2 7.1%, ≥G3 none), all of which were low grades and manageable. Only one patient discontinued treatment due to a TEAE. No DLTs or dose reduction were reported.

Conclusions LBL-034 demonstrated highly encouraging anti-tumor activity in patients with RRMM, including those in difficult-to-treat subgroups with high-risk features. Deep and durable response—including ≥CR and MRD negativity—were achieved across multiple dose levels (200-1200 μg/kg). Its conditionally activated design contributed to a favorable safety profile, particularly at higher dose levels. These compelling efficacy and safety results strongly support the continued clinical development of LBL-034 in RRMM. Updated data, including longer PFS follow-up and extended safety analyses, will be presented at the upcoming ASH Annual Meeting.

This content is only available as a PDF.
2025
Sign in via your Institution