(T1030-03-18) Drug and Colloidal Characteristics Impact the Unstirred Water Layer Thickness

Purpose: The use of colloidal systems, such as surfactant micelles and nanosized drug particles, is gaining renewed interests in the pharmaceutical industry as an enabling strategy to enhance the solubility and bioavailability of poorly soluble drugs. Colloidal formulations not only increase drug’s aqueous solubility, but also generate colloidal particles that move into the unstirred water layer (UWL) and deliver a high concentration of drug close to the surface of a membrane or the intestinal mucosa, thus promoting permeability and drug absorption. This is termed the particle drifting effect. However, key formulation properties affecting the particle drifting effect remain poorly understood. Currently, developments of these formulations remains highly empirical, and bioavailability predictions are often unsatisfactory. Therefore, to better understand the particle drifting effect and to achieve better predictions of drug absorption, it is critical to understand how formulation properties such as drug and colloidal characteristics affect the particle drifting effect.

Methods: Free drug, bile micelle, and amorphous drug particle were selected as three representative species involved in membrane diffusion and the particle drifting effect. Sodium taurocholate (STC), an abundant bile salt present in the human body, was used as a model bile salt at 15 mM (fed-state condition). Amorphous drug particles of 300 nm were generated for different drugs. Model drugs were chosen based on their UV absorption, amorphous solubility, crystallization propensity, bile salt solubilization window, and particle stability. A biphasic diffusion apparatus was used to determine the thickness of UWL (h_UWL) at the interface between organic and aqueous phases. Amorphous drug particles were generated via antisolvent titration and stabilized using either 100 ug/mL HPMCAS or 15 mM STC pre-dissolved in pH 6.5 phosphate buffer. Drug concentration in organic phase was monitored using a UV dip probe. UWL thickness was then calculated using a mechanistic mass transport model¹. For free drugs, aqueous diffusion coefficients (D) were calculated using their molecular weight using an empirical equation derived from 160 compounds². The diffusion coefficient of drug particles was measured using dynamic light scattering (DLS). For STC micelles, the diffusion coefficient was calculated using the Stokes-Einstein equation with a diameter of 7 nm³. The amorphous solubility of model drugs was determined using three different methods: UV spectrophotometry, DLS, and ultracentrifugation.

Results: According to the Levich equation, the h_UWL is proportional to the one-third power of the diffusion coefficient of diffusing species. If multiple species simultaneously diffuse, each of them has its own UWL with its corresponding thickness. In this study, h_UWL ranging from around 38 to 84 µm were observed for systems containing free drugs only. The presence of colloidal species, such as bile micelles and amorphous drug nanoparticles, significantly reduced the h_UWL to 3.0 to 6.4 µm and 0.1 to 5.7 µm, respectively. The extent of h_UWL reduction appeared to be largely dependent on diffusion coefficient of species, where the presence of large species, such as micelles and amorphous particles, resulted in significant UWL reduction.
For free drug, the relationship between UWL thickness and diffusion coefficient is shown in Figure 1, using atazanavir as a reference drug. Experimental errors from this study appeared to be more significant than variations in diffusion coefficients for small molecule drugs with a molecular weight ranging from 206 to 721. This molecular weight range covers most poorly soluble small molecule drugs. Therefore, it would be reasonable to use an average h_UWL of around 52 µm for free drugs in particle drifting effect predictions.
For STC micelles, similarly, it appears that h_UWL remained similar for micelles containing different drugs as shown in Figure 2, with a value of 4.8 ± 0.8 µm.
For amorphous drug particles, although particle size was kept the same, diffusion rates of extremely poorly soluble drugs were significantly lower than drugs with higher amorphous solubilities. As mass transfer from the UWL to the membrane occurs only through the molecularly dissolved drug rather than intact particles, these drug particles act as reservoirs and release the free drug near membrane surface. Hence, drugs with higher amorphous solubility dissolve faster, leading to higher flux and reduced UWL thickness. A linear relationship was established between amorphous solubility of the drug and normalized reciprocal of h_UWL as shown in Figure 3.

Conclusion: Drug and colloidal properties, including solubility and diffusion coefficients, were found to be important factors affecting h_UWL. Overall, h_UWL was dependent on the diffusion coefficient of the moving species, with larger species showing thinner UWLs and vice versa. For free drugs, the diffusion coefficient dependent h_UWL was not significant. For bile micelles, their h_UWL were similar and appeared to be less dependent on the model drug used. For drug particles, UWL thickness was not only dependent on particle size, but also amorphous solubility of the drug composing the particle. Quantitative understanding of UWL thickness variations obtained from this study will lead to improved prediction of the particle drifting effect for colloid-containing formulations.

References: (1) Mudie, D. M.; Shi, Y.; Ping, H.; Gao, P.; Amidon, G. L.; Amidon, G. E. Mechanistic analysis of solute transport in an in vitro physiological two-phase dissolution apparatus. Biopharm Drug Dispos 2012, 33 (7), 378-402. DOI: 10.1002/bdd.1803 From NLM Medline.
(2) Avdeef, A. Leakiness and size exclusion of paracellular channels in cultured epithelial cell monolayers-interlaboratory comparison. Pharm Res 2010, 27 (3), 480-489. DOI: 10.1007/s11095-009-0036-7 From NLM Medline.
(3) Stewart, A. M.; Grass, M. E.; Brodeur, T. J.; Goodwin, A. K.; Morgen, M. M.; Friesen, D. T.; Vodak, D. T. Impact of Drug-Rich Colloids of Itraconazole and HPMCAS on Membrane Flux in Vitro and Oral Bioavailability in Rats. Mol Pharm 2017, 14 (7), 2437-2449. DOI: 10.1021/acs.molpharmaceut.7b00338 From NLM Medline.

Figure 1. Relationship between UWL thickness and diffusion coefficient for the free drug (dashed line showing theoretical values according to the Levich equation)



Figure 2. The thickness of UWLs for bile micelle-bound drugs



Figure 3. The dependence of the UWL on the solubility of the drug