(M1530-03-15) Biopredictive Dissolution of Budesonide Enteric Pellets under Simulated Fasted Intake Conditions

Purpose: For patients diagnosed with ulcerative colitis (UC), budesonide therapy stands as a well-recognized approach for addressing bowel inflammation. This treatment modality typically utilizes multiparticulate drug delivery systems, such as enteric pellets, wherein drug release is triggered upon pH levels, time (beyond the swelling threshold of the polymer), or a combination of both. Given the inherent pH variations throughout the gastrointestinal tract and the buffering capacity of physiological fluids, it is likely that the drug release kinetics of enteric pellets may be subject to physiological influences. Herein, budesonide granules were tested under various physiologically relevant pH conditions in the fasted state simulating edge conditions of the human gastrointestinal (GI) passage to check on the dissolution performance under conditions simulating the variability range of pH that is specific for fasted intake.

Methods: Budenofalk® Uno 9 mg gastro-resistant granules (Dr. Falk Pharma, Germany) were purchased locally. Dissolution was carried out in advanced modular platform (AMP) dissolution device in a two-stage experiment comprising 30 min incubation in gastric environment, followed by addition of modified Fasted State Intestinal Fluid (mFaSSIF) buffer concentrate containing sodium hydrogen carbonate and bio-relevant surfactants (final concentration of 4.2-12.0 mM and 2.24 g/L, respectively). The pH in the gastric phase was static (1.6, 1.8 or 4.0) and the pH during simulated small intestine and colon passage was dynamically adjusted with the in-built pHysio-grad® functionality [2], thus enabling simulation of a high-resolution pH profile. Edge conditions of physiological relevance were simulated according to Koziolek et al. [1, Fig. 1]. Three distinct setups were conducted, simulating mean GI transit conditions (SI pH range: 5.81-7.69, colon pH range: 5.43-7.3) as well as minimum (SI pH range: 5.43-6.64, colon pH range: 5.2-7.0) and maximum intestinal and colonic pH profiles (SI pH range: 6.09-8.0, colon pH range: 5.62-7.79), respectively. Dissolution samples were withdrawn through PE filters (1 µm pore size) every 30 min via an automated sampling system. All experiments were conducted with n=6 samples. Released budesonide in bio-relevant media was determined by RP-HPLC, using following equipment: Merck/Hitachi Chromaster equipped with in-line filter (stainless steel; 0.5 µM), Aquity UPLC BEH C18 column (2.1 x 50 mm; 1.7 µm) and according guard column. Detection was carried out at 244 nm using UV-Vis detector. Isocratic elution was carried out with 72% water + 0.05% formic acid (mob. phase A) and 28% acetonitrile + 0.05% formic acid (mob. phase B) at a flow rate of 0.4 mL/min. Elution time amounted to ~8-9 min. The method was characterized according to ICH guidelines.

Results: Enteric coated pellets containing 9 mg budesonide displayed distinct differences when subjected to variable pH conditions representative for the fasted state [Fig. 2]. In two-stage experiments with high pH (gastric pH = 4.0, intestinal pH profile = maximum) the dissolution started around 2 h and release levels of 71.4% (+/- 5.8%) of the label claim were reached after 3.5 h. Clogging of the filters prevented further sampling until 6 h and beyond 7.5 h. After 12 h, 95.2% (+/-1.8%) of the label claim were determined in solution. Under mean pH conditions (gastric pH = 1.8, intestinal profile = mean), the released API at the time of simulated colonic arrival at 4.5 h amounted to 4.0% (+/- 0.6%) of the label claim and the dissolution rate accelerated subsequently. The major release of the API occurred within the following 4-5 h, reaching the plateau at 9.5 h with 77.1% (+/- 4.0%) drug dissolved. Noteworthy, after 12 h 80.7% (+/- 3.8%) were dissolved, due to generally lower colonic pH levels compared to the small intestine. In the other extreme scenario simulating low gastric and intestinal pH values, a distinct onset of dissolution cannot be clearly defined due to the slow gradual release from enteric pellets. After 6 h of incubation (which corresponds to the proximal colon), only 5.2% (+/- 2.1%) were dissolved, increasing to 17.3 (+/- 11.3%) after 12 h. In all conducted experiments, levels below 2.5% were determined during the first 2 h.

Conclusion: The results indicate that relying only on pH as the sole triggering factor may result in high variability of drug exposure. Considering the physiology of GI tract it becomes likely that the higher pH levels in the small intestine, full release of budesonide can be expected in the mid-section of the jejunum or ileum, allowing effective treatment in the terminal ileum or proximal colon. At mean transit conditions, major release can be expected in broader sections of the ascending colon whereas insufficient release in rare cases of massively decreased intestinal pH levels can be expected.

References: [1] Investigation of pH and Temperature Profiles in the GI Tract of Fasted Human Subjects Using the Intellicap® System; M. Koziolek, M. Grimm, D. Becker, V. Iordanov, H. Zou, J. Shimizu, C. Wanke, G. Garbacz, W. Weitschies; Pharmaceutics; 2014.
[2] A dynamic system for the simulation of fasting luminal pH-gradients using hydrogen carbonate buffers for dissolution testing of ionisable compounds; G. Garbacz, B. Kołodziej, M. Koziolek, W. Weitschies, S. Klein; Eur. J. Pharm. Sci.; 2014.


Fig. 1: Physiologically relevant intestinal and colonic pH profiles during human GI passage up to 12 h. Mean and edge conditions according to Koziolek et al. [1]. The pH during simulated gastric residence was static and set to 1.6 (minimum pH), 1.8 (mean pH) and 4.0 (maximum pH), respectively.


Fig. 2: Dissolution of budesonide from enteric pellets in scenarios representative for fasted intake conditions with mean (blue), maximum (green) and minimum (red) physiologic pH ranges during human GI passage. Mean dissolution profiles (n=6) are displayed as distinct graphs. Simulated pH profiles of each scenario are presented as dashed graphs in according colors. Standard deviation is indicated by the error bars. Data points missing due to clogged filters during automated sampling.