(T0930-09-57) Dissolution-Membrane Permeation Combined with Dissolution Imaging for Bioperformance Prediction of Amorphous Solid Dispersion Manufactured by a Novel Co-Precipitation Process

Purpose: Enabling formulation are often needed in oral drug delivery to obtain desired in-vivo bioavailability of poorly water-soluble drugs. Amorphous solid dispersion (ASD) is one of the leading strategies to enhance oral bioavailability by increasing drug solubility and dissolution rate, improving the driving force for oral absorption. However, the rate and extend of drug release, drug supersaturation precipitation and drug speciation along the gastrointestinal tract on oral absorption is not well understood and make it difficult to predict the in-vivo performance of ASDs. The purpose of this study was to develop an in-vitro absorption model to understand how amorphous solubility, colloidal species impact oral drug absorption and to assess the impact of drug load and polymer grade on the performance of ASDs manufactured using a novel co-precipitation process.

Methods: ASDs of a BCS class II drug with HPMCAS grade L, M, H with drug load DL 20-50% were manufactured using a spray-dried and a novel co-precipitate based process. Amorphous solubility was predicted from crystalline material using the solvent shift method with UV extinction combined with the Zero Intercept Method (ZIM) analysis in AuPro. In-vitro dissolution-permeation studies were carried out using µFLUX apparatus (Pion Inc., Billerica MA, USA) which consist of a donor and a receiver chamber separated by an artificial membrane (polyvinylidenfluoride filter impregnated with GIT Lipid, Pion Inc. solution). Measurements were performed using canine fasted state simulated intestinal fluid (FaSSIFc) in the donor side and Acceptor Sink Buffer (ASB pH 7.4) in the receiver compartments.  ASDs powder or dissolved drug in DMSO stock were added into the donor compartments of the µFLUX and concentration time profiles in both chambers were monitored for 2 hours using in-situ monitoring technique implemented with Rainbow fiber optic UV-Vis spectrometer (Rainbow R2D8, Pion Inc). Flux across the membrane representing the amount of drug substance crossing a unit area of the membrane per unit time, was calculated from the concentration-time profiles in the receiver compartments using Predictor software. Dissolution imaging was conducted using an Olympus confocal fluorescence microscope with a 405 nm excitation to stimulate the intrinsic drug fluorescence.

Results: Amorphous solubility was determined to be ~ 250 µg/mL in FaSSIFc, while crystalline drug solubility evaluated by shake flask method was 13 µg/mL. In-situ concentration time profiles when DMSO stock was added to the donor chambers were collected below and above amorphous solubility. It was found that diffusive flux increased linearly with drug concentration to reach a maximum at amorphous solubility and then stabilized in the liquid-liquid phase separation up to 350 µg/mL. Flux at the amorphous solubility was significantly higher than flux of crystalline API, 0.6 compared to 0.01 µg/(min*cm²). Flux measured from ASDs powder was similar to the flux evaluated using the DMSO stock method. In-situ concentration time profiles from donor and receiver chambers of ASDs powder with HPMCAS grade L-M-H were collected. Drug release rate was found to be higher for L, M while grade H showed slower release. Calculated flux from the receiver compartments were also able to rank order the impact of the polymer grades on drug release from ASDs with L >M >H (0.5872, 0.4675, 0.2870 µg/(min*cm²)). When comparing performance of ASDs at various DL, higher flux was found for ASDs with lowest DL. At DL (43-50%) flux had the lowest values. Also flux values from ASDs with 20-33% DL were similar to flux measured at amorphous solubility suggesting that higher bioavailability will be reached for ASDs with 20-33% DL. Confocal fluorescence microscope videos have been collected to reveal critical physical properties influencing the dissolution behavior of the ASD formulations.

Conclusion: This study shows that: The maximum flux was reached at the amorphous solubility and was maintained in the colloidal drug-rich-phase well above amorphous solubility. DMSO stock method developed was predictive of ASDs bioperformance and can be used in preformulation stage with crystalline material for formulation development. Confocal fluorescence microscopy was able to predict drug release from ASDs with drug loading. Overall, the flux-based approach could be used as a material sparing screening tool to assess the impact of the critical quality attributes on ASDs bioperformance and could be a useful in guiding formulation design for optimal product performance.