Population pharmacokinetics (popPK) and exposure-response (ER) analyses of lbl-034, a potent GPRC5D T cell engager in patients with relapsed/refractory multiple myeloma

Abstract Introduction LBL-034 is a humanized IgG1 subtype asymmetric bispecific antibody targeting GPRC5D and CD3. Its affinity-optimized anti-CD3 arm, engineered with steric hindrance, allows for conditional T cell activation and proliferation. LBL-034 has shown promising activity following intravenous infusion every two weeks (Q2W) as monotherapy in patients with relapsed/refractory multiple myeloma (RRMM) in an ongoing Phase Ⅰ/Ⅱ study (NCT06049290). The objectives of the present analyses were to characterize the pharmacokinetics (PK) of LBL-034 in this population, assess the exposure response relationships for clinical efficacy and safety, and inform dose selection in the upcoming clinical trials. Methods As of June 30, 2025, a total of 945 PK concentration records from 55 patients at 10 (1), 30 (1), 80 (5), 200 (7), 400 (19), 800 (11), and 1200 (11) μg/kg were used in popPK analysis. Dosing was initiated on Days 1 and 15 of each 4-week cycle. Beginning with the 80 μg/kg cohort, a step-up dosing strategy (10 to 80 μg/kg) was employed. The potential impacts of selected potential covariates, including weight, disease type (IgG or non-IgG), baseline albumin, and the baseline total T-cell number, on key PK parameters were explored. Exposure response analyses on clinical efficacy and safety signals were conducted using logistic regression methodology (R version 4.3.2). Clinical response data analysis included ORR, ≥VGPR, and ≥CR, while safety signals evaluation included CRS, leukopenia, neutropenia, thrombocytopenia, dysgeusia, oral pain, nail disorder, bacterial infection, upper respiratory infection, pruritus, and rash. Simulations were conducted to explore optimal doses and regimens, including body weight normalized dose or flat dose of Q2W, Q3W and even Q4W. Results The PK of LBL-034 was best described by a 2-compartment model with both a constant clearance (CL1) and a time-dependent clearance (CL2) that faded over time. The estimated central volume (Vc) and peripheral volume (Vp) were approximately 3700 mL and 5980 mL, respectively. The model predicted CL1 and CL2 were about 14.5 mL/hr and 654 mL/hr, respectively. Notably, CL2 is a surrogate parameter mimicking target-mediated drug disposition, which was estimated to disappear within 7-day after the first dose. Covariates including weight and disease type had no significant effect on clearance. At steady state, the terminal half-life was estimated to be approximately 23 days. In general, LBL-034 exhibits typical antibody PK characteristics. ER analyses demonstrated statistically significant correlations between higher LBL-034 exposure and improved clinical response in Phase Ⅰ patients. In contrast, no significant ER relationship was observed for safety outcomes, aligning with clinical findings. Simulations with the popPK model suggest a shift from weight-based dosing to flat dosing is feasible. These results support evaluating more convenient dosing regimen in future clinical development of LBL-034. Conclusions LBL-034 exhibited an excellent PK profile with a long half-life. ER analysis indicated a statistically significant positive correlation between drug exposure and efficacy. Additionally, higher doses were associated with improved clinical responses without increased safety risk. These findings provide important insights for optimizing dose and dosing frequency in future clinical development of LBL-034.