A machine learning model (Sol_ME) assisted development of a unit-dose lipid formulation for apalutamide: formulation, pharmacokinetics, and pharmacodynamics evaluation.

PURPOSE: The development of a novel lipid-based formulation for apalutamide, a potent androgen receptor inhibitor for non-metastatic castration-resistant prostate cancer (nmCRPC), is explored in this study.

METHOD: To address its poor solubility and high dose requirement, a single-dose lipid-based soft gelatin capsule delivering 240 mg of apalutamide was developed using machine learning (ML)-assisted excipient selection. The ML model, Sol_ME, predicted cinnamon oil as the optimal solubilizer, enhanced further by vanillin. The optimized formulation showed improved dispersion, stability, and bioavailability.

RESULT: In pharmacokinetic studies, the test formulation achieved a Cmax of 7860.66 ng/mL at 8.34 h compared to 4850.19 ng/mL at 12.82 h for the standard formulation. AUC0-∞ values were 190,500.75 and 123,879.63 ng·h/mL, respectively, indicating significantly enhanced absorption. Pharmacodynamic assessments confirmed superior androgen suppression and tumor inhibition.

CONCLUSION

This AI-guided, lipid-based system offers an efficient, patient-friendly alternative for apalutamide administration.