Oral Delivery of Semaglutide to the Colon: Preliminary Formulation Study

INTRODUCTION Millions of individuals globally are diagnosed with type 2 diabetes and are subject to the many well-known diabetes related complications1. Glucagon-like peptide-1 (GLP-1) receptor agonists offer an innovative treatment option in the management of this chronic condition2. Beyond their glucose-lowering efficacy, GLP-1 analogs may help preserve pancreatic β-cell function and exert favorable effects on common comorbidities, such as hypertension, dyslipidemia and obesity3. However, the need for subcutaneous injections represents an important barrier to long-term adherence to the therapy. GLP-1 receptor agonist semaglutide is currently available as both a subcutaneous and an oral formulation (Rybelsus®). Rybelsus® employs absorption enhancer sodium N - (8 - [2 - hydroxylbenzoyl] amino) caprylate (SNAC), which protects the drug from enzyme degradation by locally increasing pH, promotes drug dissolution and transiently modulates gut epithelial permeability2. Despite these advances, the oral bioavailability of the drug remains poor: in dogs, it was found to be 1.2 ±0.25 with a 10 mg dose, and human values were slightly lower4,5. This may be ascribed to enzymatic activity, variable pH and low responsiveness to promoters in the upper gastrointestinal tract. Furthermore, side effects such as nausea, vomiting and abdominal discomfort are frequently observed and constitute a leading cause of treatment discontinuation. Considerable benefits could thus arise from colon targeting, an alternative delivery strategy that may offer better chances of absorption while ensuring high patient compliance6. To achieve colon delivery, a double coating technology has recently been developed for improved site selectivity of release. The outer layer comprises Eudragit® S, an enteric polymer with a pH dependent dissolution threshold of 7, and guar gum, a natural polysaccharide degraded by the colon microbiota7. This layer protects the dosage form in the stomach and upper intestine while allowing release to take place in the colon. An inner low-viscosity hydroxypropyl methylcellulose (HPMC) coating introduces a lag phase, further delaying exposure of the drug-loaded core to intestinal fluids after dissolution of the enteric outer coating. This would be of major importance in the event the latter coating is dissolved before colon arrival. In the present study, an oral semaglutide formulation intended for colon-targeted release was proposed based on the above-described oral delivery technology and evaluated for physico-technological properties and in vitro release performance. MATERIALS AND METHODS Oblong placebo cores were manufactured from a blend of 48.8% Emcompress®, 48.7% Avicel® PH102, 2.0% Explotab® CLV and 0.5% magnesium stearate, tableted by a rotary press (AM-8S) equipped with 14.5 × 7.5 mm punches at 23 kN and characterized. Coating of placebo and of 3 mg Rybelsus® tablets (Novo Nordisk) was performed using an 8% w/w low-viscosity HPMC (Methocel™ E50) solution plasticized with PEG 400 (0.8% w/w on dry polymer) in a tangential-spray fluid bed (Glatt GPCG 1.1), followed by a hydroalcoholic Eudragit® S and guar gum dispersion; coated tablets were cured at 40 °C for 24 h. Photomicrographs were acquired using a scanning electron microscope (SEM, LEO1430, Carl Zeiss, IT). Release testing (n=3) was run in a USP paddle apparatus employing 800 mL of media, i.e. 0.1 N HCl for 2 h and then phosphate buffer pH 7.4, at 37.0 ± 0.5 °C, 100 rpm and UV quantification at 230 nm. Testing was repeated after 3 months of storage at 40 ± 2 °C and 75 ± 5% relative humidity (RH). Lag time was defined as t10% in pH 7.4 buffer by linear interpolation of the two bracketing time points. RESULTS AND DISCUSSION The two sequential coating processes, based on low-viscosity HPMC and Eudragit® S with guar gum, respectively, were initially set up with in-house manufactured tablets, which were also used as a bulking substrate. The size, shape and compression forces of such tablets were selected to reflect the physico-technological properties of the commercial semaglutide tablets as closely as possible. The coating processes were then successfully run with the peptide containing cores under the operating conditions reported in Table 1. Methocel™ E50 Eudragit® S/ Guar gum Batch size (g) 700 700 Nozzle diameter 1.2 1.2 Nebulization air pressure (bar) 2 2 Air volume (m3/h) 100 100 Disk rotation (rpm) 400 450 Inlet air temperature (°C) 60 40 Product temperature (°C) 42-44 42-44 Spray rate (g/min) 4.4 4.4 Following each coating step, the thickness, weight gain and amount of Eudragit® S applied per surface area were recorded (Table 2). The resulting double-coated systems exhibited reproducible coating levels and continuous overlapping layers of consistent thickness (Table 2, Figure 1). The two polymer structures appear homogeneous despite the dispersed nature of the Eudragit® S/guar gum formula, pointing out proper deposition of the outer coating on the inner one. Inner coating Outer coating Weight gain (%) 9.02 10.75* Total amount applied (mg/cm2) 10.26 15.44 t10% (min) 19 140 Table 2. Coating level and in vitro lag time of double-coated delivery systems (*on single-coated tablets). In vitro gastric resistance of the double-coated systems was demonstrated according to compendial requirements (Figure 2). After transition to intestinal pH fluid and dissolution of the outer coating, a lag phase was observed as needed to protect the drug-containing core in case of early in vivo failure of the enteric coating. After 3 months of storage under accelerated stability testing conditions, the release patterns showed no major changes, aside from a slight decrease in the release rate after the lag phase. Such a decrease, however, was proved not to be statistically significant (p > 0.05, Student’s t-test). CONCLUSION The present study features a first effort toward distal intestinal release of semaglutide. The results obtained showed that the previously developed oral colon delivery technology could successfully be applied to this GLP-1 analog.