Approaches for estimating the clinical starting dose of new dosage forms: An example of a long-acting injectable formulation of finasteride

In our previous study, a long-acting injectable (LAI) formulation of finasteride was prepared as a new dosage form of PROPECIA®, and in vivo pharmacokinetics (PKs)-pharmacodynamics (PDs) was evaluated in beagle dogs. The resulting PK-PD profiles of the formulation showed pharmacological effects and achievability for monthly delivery. In this study, a first-in-human (FIH) dose of the LAI formulation loaded with finasteride was predicted. The three approaches were used for estimating a FIH dose of the LAI formulation: (1) No observed adverse effect level (NOAEL)-based approach; (2) Pharmacokinetically-guided approach; (3) Pharmacokinetic/pharmacodynamic model-based approach. The advantage, assumptions, limitations, and estimated FIH dose from each approach was discussed and compared since there is no consensus on the best approach. For the prediction of clinical exposures and estimation of FIH doses, the clinical PK-PD parameters were allometrically scaled from the nonclinical data, extracted from reported clinical studies, or fixed from published literature. The starting dose range of the LAI formulation (as finasteride) was estimated to be 16.80–81.06 mg from the three approaches, and the PK/PD model-based approach suggests the most optimal starting dose (16.80 mg) of the LAI formulation. The approaches for estimating starting doses presented in the study could be used as a basis for an Investigational New Drug (IND) application of new dosage forms.