The use of modeling tools to drive efficient oral product design

Abstract

Modeling and simulation of drug dissolution and oral absorption has been increasingly used over the last decade to understand drug behavior in vivo based on the physicochemical properties of Active Pharmaceutical Ingredients (API) and dosage forms. As in silico and in vitro tools become more sophisticated and our knowledge of physiological processes has grown, model simulations can provide a valuable confluence, tying-in in vitro data with in vivo data while offering mechanistic insights into clinical performance. To a formulation scientist, this unveils not just the parameters that are predicted to significantly impact dissolution/absorption, but helps probe explanations around drug product performance and address specific in vivo mechanisms. In formulation, development, in silico dissolution-absorption modeling can be effectively used to guide: API selection (form comparison and particle size properties), influence clinical study design, assess dosage form performance, guide strategy for dosage form design, and breakdown clinically relevant conditions on dosage form performance (pH effect for patients on pH-elevating treatments, and food effect). This minireview describes examples of these applications in guiding product development including those with strategies to mitigate observed clinical exposure liability or mechanistically probe product in vivo performance attributes.

Fig. 1

Case study for modeling with BMS compound A. a In vivo observed (symbols) and predicted (line) profiles in the fasted state, b in the fed state, c in famotidine treated subjects; points are means ± standard deviation. d Surface response plot of simulated Cmax change with respect to mean particle diameter and pH change. e Surface response plot with simulated AUC change with respect to mean particle diameter and pH change

Fig. 2

a Modeling for BMS compound B. Observed (symbols) and simulated (line) profile in subjects in the fasted state (start) and the fed state (12 h). Points are means ± standard deviation. b Dog PK results for BMS compound B, solution and tablet in the fasted and fed state. Data is mean AUC levels, n = 4

Fig. 3

Case study with BMS compound C. a In vitro dissolution profile in 0.01 N HCl in USP-II apparatus; b simulated plasma concentration–time profile when Tpptn is set at 10, 60, and 360 min. c Gastroplus plot of amount dissolved with Tpptn set at 10 min; d gastroplus plot of the amount dissolved with Tpptn set at 360 min

Fig. 4

Case study 4 for Metoproplol and Ranitidine-modified release formulations. Surface response plot of Cmax change with respect to gastric emptying time and drug release time from MR dosage form for metoprolol a and ranitidine b; AUC change with respect to gastric emptying and drug release time for metoprolol c and ranitidine d

Fig. 5

Caste study 5 Metformin release from FDC and bioequivalence to Metformin product. a In vitro release profile between Metformin from FDC and reference Metformin product, b in vivo model prediction in virtual trial population for change in the Cmax at various drug release rates, and impact on AUC c. Values are geometric means ± standard deviations