Abstract
Purpose: The broader application of self-nanoemulsifying drug delivery systems (SNEDDS) has been hampered by the limited solubility of lipophilic, hydrophobic drugs, often requiring the inconvenient dosing of multiple units. In addition, the drug’s solubility in SNEDDS has not been fully utilised due to the current paradigm that crystalline drug precipitation during the digestion of the delivery system should be avoided. However, recently it has been reported that drugs might precipitate in an amorphous form during in vitro digestion. The purpose of this thesis was to investigate the feasibility of SNEDDS containing drug above the saturation solubility (super-SNEDDS) as a means to increase the drug load in SNEDDS.
Methods: Using a partial phase diagram approach, mixtures of lipids, surfactant, and cosolvent were screened for their ability to generate fine dispersions in aqueous medium while keeping the lipid/surfactant/cosolvent ratio constant. Pre-concentrates were loaded with increasing amounts of the poorly water-soluble drug simvastatin (SIM) and were evaluated by dynamic light scattering with regard to their dispersion characteristics in water and simulated gastric fluid. The effect of increasing drug loads on the ability to keep SIM in solution was determined using the dynamic in vitro lipolysis. Following ultracentrifugation, the aqueous phase and the pellet were quantified for SIM by HPLC. Super-SNEDDS containing SIM up to 200% of the drug’s equilibrium solubility (Seq) were produced by subjecting the pre-concentrates to a heating and cooling cycle, while the drug load of halofantrine (HAL) super-SNEDDS was 150%. All super-SNEDDS were regularly investigated by polarising light microscopy to determine the physical stability during storage at room temperature. The super-SNEDDS were characterised during in vitro lipolysis and compared with single units or multiple units of conventional SNEDDS (75% drug load). The solid state properties of precipitates obtained after in vitro lipolysis were characterised by XRPD and dissolution studies of the precipitates were carried out in lipolysis medium. The in vivo performance after oral administration of super-SNEDDS (one capsule 150% drug load) was compared with one capsule SNEDDS (75% drug load) and two capsules SNEDDS (75% drug load), containing the same dose as super-SNEDDS.
Results: Increasing the drug loads from 25% to 100% Seq affected the dispersion characteristics of SNEDDS depending on the type of lipid and surfactant used. During in vitro lipolysis of SIM (super-) SNEDDS the concentration of SIM in the aqueous phase increased proportionally to the drug loads and amount of SNEDDS employed, generating an initial supersaturation of the lipolysis medium followed by rapid precipi-tation of SIM. In contrast, rapid precipitation was observed for HAL super-SNEDDS from the beginning of in vitro lipolysis and the concentration of HAL in the aqueous phase could only be increased when multiple units of super-SNEDDS were subjected to in vitro lipolysis. Both HAL and SIM precipitated in an amorphous form. The precipi-tated drugs demonstrated a faster dissolution rate compared to blank pellets spiked with crystalline drug. Super-SNEDDS were physically stable for at least six (HAL super-SNEDDS) and ten months (SIM super-SNEDDS). In vivo, the oral bioavailability of SIM super-SNEDDS (one capsule 150%) significantly increased to 185% relative to two capsules conventional SNEDDS of the same dose, possibly due to both saturation effects of the effects on pre-systemic metabolising enzymes and prolonged absorption along the small intestine, where differences are observed in the CYP 3A4 expression. After administration of HAL super-SNEDDS the maximum plasma concentrations and areas under the plasma-time curves were similar to those observed for two capsules conventional SNEDDS containing the same dose.
Conclusion: Super-SNEDDS are a viable formulation option to enhance the bioavaila-bility of poorly water-soluble drugs while reducing the pill burden by an increased drug load. The paradigm to avoid drug precipitation needs to be put into perspective since drugs can precipitate in an amorphous form. Whilst the in vitro lipolysis model alone was found to be useful to predict the in vivo performance of some SNEDDS, it failed to predict the in vivo performance of super-SNEDDS. Notwithstanding this, the solid state characterisation of the precipitate should always accompany the in vitro assessment of SNEDDS and super-SNEDDS to improve the predictability of the in vitro model for in vivo studies.