Abstract
OBJECTIVE: This study aimed to determine the optimal dosing strategy for ribociclib (RIB) in patients with breast cancer and brain metastasis (BCBM) by using a developed and validated physiologically based pharmacokinetic (PBPK) model. METHODS: A population PBPK model for RIB was established by integrating its physicochemical properties, pharmacokinetic (PK) data, cyclin-dependent kinase 4/6 (CDK4/6) occupancy profiles, and the physiological characteristics of the BCBM population. The predictive performance of the PBPK model was verified against data from two clinical plasma PK studies and two cerebrospinal fluid (CSF) PK studies. RESULTS: The PBPK model demonstrated concordance with observed clinical data. All predicted-to-observed ratios for plasma AUC, C(max), C(min), and RIB concentrations in CSF fell within the acceptable 2-fold range (0.5–2.0), with only one exception of 0.43. Furthermore, the model predicted the CSF-to-plasma unbound concentration ratio (K(p, uu)), with values closely aligning with clinical observations (400 mg: predicted 0.99 vs. observed 1.22; 600 mg: predicted 1.03 vs. observed 1.29; 900 mg: predicted 1.10 vs. mean observed 1.63). The PBPK model identified CYP3A4/FMO3 (Flavin-containing monooxygenase 3) concentrations, ABCB1 expression, and plasma albumin levels as key physiological parameters governing intracranial C(min) and CDK4/6 occupancy. Based on these simulations, a regimen of 300 mg twice daily (BID) was identified as the optimal dosing strategy for RIB in BCBM patients, with > 90% mean intracranial CDK4/6 occupancy and safety mean C(min) threshold. CONCLUSION: This study successfully developed and applied a population PBPK model to simulate the PK and pharmacodynamic profiles of RIB. The model proved to be a robust tool for identifying a rational, optimized dosing regimen specifically for BCBM patients. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12885-026-15561-x.