Biomarker-Driven Computational PK/PD Modeling and Dosing Optimization of the Targeted Protein Degrader Fulvestrant in Estrogen Receptor–Positive Breast Cancer

Computational models that help establish dose, exposure and response relationship are essential for maximizing a drug’s clinical efficacy. This paper describes a novel biomarker-driven computational model that simulates the pharmacological effects of an important targeted protein degradation drug, fulvestrant (Faslodex), used to treat Estrogen Receptor (ER)-positive breast cancer. Fulvestrant promotes ER degradation in tumors, and this slows the disease progression. Using PK/PD modeling to optimize dosing to achieve maximal intratumoral ER degradation hence is key to improving fulvestrant efficacy. This PK/PD model development effort utilized fulvestrant’s pharmacokinetics (half-life: 14 h; clearance: 11 mL/kg/min), potency in tamoxifen-resistant cells (IC₅₀: 2.4 nM), tumor penetration (cell/plasma ratio: 0.7), plasma protein binding (~99%), and intracellular ER turnover rate (0.003 h⁻¹) data. Tumor growth inhibition in mice was simulated (doses: 25 and 200 mg/kg), and the model refined using in vivo xenograft tumor growth inhibition (TGI) data. This approach was translated to develop a human computational model and fine-tuned by aligning predicted ER degradation rates with reported patient biomarker data. Our model indicates that the current monthly 500 mg IM dose is unlikely to achieve sufficient ER degradation in patients, while a 250 mg weekly IM schedule could reduce ER levels by >90% in six months and >99% in ten months, potentially improving treatment outcomes.