Phase 1b dose extension study of a next-generation anti-CD47 monoclonal antibody IMC-002 combined with lenvatinib in patients with advanced hepatocellular carcinoma (HCC)

Background: IMC-002 has shown significant preclinical efficacy and safety, which are attributed to its unique binding site and distinct mechanism of action. These preclinical findings strongly support its clinical development as a cancer therapeutic. Phase 1a trial confirmed its superior safety and tolerability. Here we present initial results from the phase Ib trial, focusing on safety, efficacy, PK, and biomarker. Methods: Eligible pts had advanced HCC that progressed following at least 1 prior systemic therapy and ECOG PS #1. IMC-002 was administered at 20 mg/kg Q3W in combination with Lenvatinib, continuing until disease progression. Tumor assessments were conducted every 6 weeks using RECIST 1.1 and iRECIST. A target-mediated drug disposition (TMDD) PK model incorporating FcRn recycling was developed to predict PK values for Q3W dosing and evaluated for consistency with observed data. Immunohistochemistry (IHC) images of CD47 expression were analyzed using Lunit SCOPE uIHC, an AI-based platform capable of distinguishing staining positivity and cell types at the single cell level. Results: A total of 13 pts with refractory HCC received IMC-002 in combination with Lenvatinib. Most patients had received prior anti-PD-(L)1 therapy (11 pts) and had an ECOG PS of 1 (9 pts). Among the 10 pts evaluable for efficacy, the ORR was 30%, and the DCR was 70%. The median TTP was 8.3 months. AI-driven analysis of CD47 membrane specificity, using a subcellular model, revealed that samples with a high proportion of non-membrane-specific cells were associated with poor clinical outcomes (ORR 0%, DCR 33%). In contrast, samples with a low proportion demonstrated improved responses (ORR 60%, DCR 80%). We confirmed that 96.3% of the observed concentrations of IMC-002 in Phase 1b not only fell within the 90% prediction percentiles of PK model developed for Q3W dosing schedule but also demonstrated steady-state achievement (after cycle 2 of Q3W) and consistent C_trough exposure above the MEC (. 24 mg/mL). All TRAEs were grade 1-2 (100%), with 92% occurring during cycle 1. TRAEs reported in more than one patient included skin rash and transient vitreous floaters. Anemia was observed in only one patient, while no cases of neutropenia, thrombocytopenia, or treatment-related SAEs were reported. Conclusions: IMC-002, when combined with Lenvatinib at a dose of 20 mg/kg Q3W, demonstrated a promising efficacy and safety profile. AI-driven biomarker analysis identified potential predictive value, supporting the need for further investigation in larger clinical trials. Clinical trial information: NCT05276310. Research Sponsor: ImmuneOncia Therapeutics Inc.