Lead optimization of a CHI3L1 inhibitor for Glioblastoma: Enhanced target engagement, pharmacokinetics, and efficacy in 3D spheroid models

Abstract

Glioblastoma (GBM) remains one of the most aggressive and treatment-resistant brain tumors, necessitating the development of novel therapeutics with improved efficacy and pharmacokinetic (PK) profiles. CHI3L1 (chitinase-3-like protein 1) is a secreted glycoprotein overexpressed in GBM, where it promotes tumor progression and immune evasion; however, no small molecule CHI3L1 inhibitors with demonstrated in vivo efficacy in GBM models are currently available. Herein, we present the lead optimization of K284 - a previously reported CHI3L1 inhibitor - toward its application in glioblastoma (GBM) therapy. We employed microscale thermophoresis (MST), surface plasmon resonance (SPR), and molecular docking to design and evaluate novel K284 derivatives targeting GBM. Among the synthesized compounds, 11g emerged as a promising candidate, exhibiting superior CHI3L1 binding affinity and improved PK parameters compared to K284. Notably, 11g demonstrated extended plasma half-lives (t₁/₂ = 2.5 ± 0.2 h in human and 2.0 ± 0.1 h in mouse) and microsomal stability (t₁/₂ = 2.8 ± 0.5 h in human and 2.2 ± 0.4 h in mouse), along with reduced intrinsic clearance (Cl_int = 15 ± 1.2 mL/min/mg in mouse and 19 ± 1.1 mL/min/mg in human). The 11g compound also showed improved solubility, favorable permeability, and decreased plasma protein binding. Safety profiling revealed lower cardiotoxic potential and reduced cytotoxicity toward normal human astrocytes relative to K284. Furthermore, in a 3D multicellular GBM spheroid model, 11g induced dose-dependent cytotoxicity, reduced spheroid mass by over 50 %, and inhibited migration by approximately 60 %, whereas K284 exhibited minimal activity. These findings underscore compound 11g as a potent and multifaceted anti-GBM agent with favorable pharmacological and safety characteristics, warranting further in vivo investigation.