(249d) Modelling Amorphous Solid Dispersions: The Role of Supersaturation, Nucleation and Crystal Growth

Amorphous solid dispersion (ASDs) are a rising strategy to overcome the poor solubility of drugs in water, since they can maintain supersaturation concentrations of the drug and thus increase its oral bioavailability. Many approaches are reported to predict the oral absorption of drugs, namely the combined use of biorelevant dissolution testing with in silico modeling.¹ There are not many publications regarding the prediction of in vivo ASDs behavior² and the current mathematical models are simplistic on describing supersaturation as the ratio between the bulk concentration and the concentration at saturation of the drug. This approach seems inadequate to describe supersaturation and can be a roadblock to accurate physiologically based pharmacokinetic (PBPK) modelling of ASDs.

This work aims at describing the behavior of ASDs when dissolving into a biorelevant medium taking into consideration a combination of four events: dissolution, supersaturation, nucleation and crystal growth. These events were described by differential equations and constants with physical meaning.

ASDs were produced by spray-drying following a full-factorial design of experiments, varying the drug load, the ratio of atomization and the outlet temperature of spray-drying. The behavior of these ASDs was evaluated by dissolution in biorelevant medium including a pH shift from FaSSGF medium (pH 1.2) to FaSSIF medium (pH 6.5) to mimic the conditions in the gastrointestinal tract.

The first attempt combined models describing dissolution and precipitation by using differential equations but describing supersaturation as a simple ratio between the concentration of the drug and its concentration at saturation. However, this attempt failed to be a good descriptor of supersaturation, as the model did not reflect the experimental data from ASDs presenting significant burst release.

In order to have the drug’s concentration at saturation described by differential equations, rather than by the ratio between concentrations, the Noyes-Whitney equation was modified taking into consideration the maximum concentration for each ASD, rather than being restricted by the concentration at saturation of the drug. Then, the behavior of ASDs was modeled by combining four differential equations: the Noyes-Whitney equation for dissolution, the modified Noyes-Whitney equation for supersaturation, and the nucleation and crystal growth equations for precipitation. This approach was able to describe ASDs with different behaviors (error between experimental and simulated data < 2%). This outcome was possible because the dynamic process of dissolution was based on dissolution, supersaturation and precipitation equations of the drug.

Figure 1 shows the experimental and simulated data for ASDs produced by spray-drying with different behaviors: spring and parachute, absence of precipitation, absence of supersaturation, and delayed and incomplete drug release (produced by hot melt extrusion, HME).

This work contributes for a better understanding of an ASD behavior describing the supersaturation mechanism, which directly impacts on the drug’s bioavailability and plasmatic concentration.

References:

1. Kaur, N., Narang, A. & Bansal, A. K. Use of biorelevant dissolution and PBPK modeling to predict oral drug absorption. Eur. J. Pharm. Biopharm. 129, 222–246 (2018).

2. Mitra, A., Zhu, W. & Kesisoglou, F. Physiologically based absorption modeling for amorphous solid dispersion formulations. Mol. Pharm. 13, 3206–3215 (2016).