The applications of machine learning to predict the forming of chemically stable amorphous solid dispersions prepared by hot-melt extrusion.

Amorphous solid dispersion (ASD) is one of the most important strategies to improve the solubility and dissolution rate of poorly water-soluble drugs. As a widely used technique to prepare ASDs, hot-melt extrusion (HME) provides various benefits, including a solvent-free process, continuous manufacturing, and efficient mixing compared to solvent-based methods, such as spray drying. Energy input, consisting of thermal and specific mechanical energy, should be carefully controlled during the HME process to prevent chemical degradation and residual crystallinity. However, a conventional ASD development process uses a trial-and-error approach, which is laborious and time-consuming. In this study, we have successfully built multiple machine learning (ML) models to predict the amorphization of crystalline drug formulations and the chemical stability of subsequent ASDs prepared by the HME process. We utilized 760 formulations containing 49 active pharmaceutical ingredients (APIs) and multiple types of excipients. By evaluating the built ML models, we found that ECFP-LightGBM was the best model to predict amorphization with an accuracy of 92.8%. Furthermore, ECFP-XGBoost was the best in estimating chemical stability with an accuracy of 96.0%. In addition, the feature importance analyses based on SHapley Additive exPlanations (SHAP) and information gain (IG) revealed that several processing parameters and material attributes (i.e., drug loading, polymer ratio, drug's Extended-connectivity fingerprints (ECFP) fingerprints, and polymer's properties) are critical for achieving accurate predictions for the selected models. Moreover, important API's substructures related to amorphization and chemical stability were determined, and the results are largely consistent with the literature. In conclusion, we established the ML models to predict formation of chemically stable ASDs and identify the critical attributes during HME processing. Importantly, the developed ML methodology has the potential to facilitate the product development of ASDs manufactured by HME with a much reduced human workload.

Drop-on-powder 3D printing of amorphous high dose oral dosage forms: Process development, opportunities and printing limitations.

Drop-on-powder 3D printing is able to produce highly drug loaded solid oral dosage forms. However, this technique is mainly limited to well soluble drugs. The majority of pipeline compounds is poorly soluble, though, and requires solubility enhancement, e.g., via formation of amorphous solid dispersions. This study presents a detailed and systematic development approach for the production of tablets containing high amounts of a poorly soluble, amorphized drug via drop-on-powder 3D printing (also known as binder jetting). Amorphization of the compound was achieved via hot-melt extrusion using the exemplary system of the model compound ketoconazole and copovidone as matrix polymer at drug loadings of 20% and 40%. The milled extrudate was used as powder for printing and the influence of inks and different ink-to-powder ratios on recrystallization of ketoconazole was investigated in a material-saving small-scale screening. Crystallinity assessment was performed using differential scanning calorimetry and polarized light microscopy to identify even small traces of crystallinity. Printing of tablets showed that the performed small-scale screening was capable to identify printing parameters for the development of amorphous and mechanically stable tablets via drop-on-powder printing. A stability study demonstrated physically stable tablets over twelve weeks at accelerated storage conditions.

Compression of amorphous solid dispersions prepared by hot-melt extrusion, spray drying and vacuum drum drying.

The present study explored vacuum drum drying (VDD) as an alternative technology for amorphous solid dispersions (ASDs) manufacture compared to hot-melt extrusion (HME) and spray drying (SD) focusing on downstream processability (powder properties, compression behavior and tablet performance). Ritonavir (15% w/w) in a copovidone/sorbitan monolaurate matrix was used as ASD model system. The pure ASDs and respective tablet blends (TB) (addition of filler, glidant, lubricant) were investigated. Milled extrudate showed superior powder properties (e.g., flowability, bulk density) compared to VDD and SD, which could be compensated by the addition of 12.9% outer phase. Advantageously, the VDD intermediate was directly compressible, whereas the SD material was not, resulting in tablets with defects based on a high degree of elastic recovery. Compared to HME, the VDD material showed superior tabletability when formulated as TB, resulting in stronger compacts at even lower solid fraction values. Despite the differences in tablet processing, tablets showed similar tablet performance in terms of disintegration and dissolution independent of the ASD origin. In conclusion, VDD is a valid alternative to manufacture ASDs. VDD offered advantageous downstream processability compared to SD: less solvents and process steps required (no second drying), improved powder properties and suitable for direct compression.

Influence of high pressure compaction on solubility and intrinsic dissolution of ibuprofen binary mixtures employing standard excipients.

Enabling formulations often depend on functional excipients. However, the question remains whether excipients regarded as standard establish similar interactions and subsequently improvement of solubility when employed at unusual manufacturing process conditions. In this study, compaction of API under high pressure in the presence of hydrophilic excipients is proposed as a technique to improve the solubility and/or dissolution rate with an acceptable preservation of the supersaturation state. Binary mixtures of ibuprofen (IBU) with hydroxypropyl cellulose, isomalt, mannitol and sorbitol were compacted applying high pressure (500 MPa) with and without a previous co-milling step. Intrinsic dissolution rate (IDR) was selected to characterize and evaluate dissolution performance. The IDR of neat IBU increased from 5 to 88 fold and the aqueous solubility in the range of 3 to 54%. Regarding the polyols isomalt showed the highest impact on solubility and dissolution, without changing the crystallinity of IBU independent of a co-milling step. Even higher impact was achieved in combination with HPC. However, only without a previous co-milling step, ibuprofen remained crystalline, while co-milling induced an amorphous IBU-content of 38%. Based on XRPD and DSC findings, higher IDR and solubility values correlated with crystal modifications as well as IBU/excipient interactions.

In vitro and in vivo assessment of hydroxypropyl cellulose as functional additive for enabling formulations containing itraconazole.

Using polymers as additives to formulate ternary amorphous solid dispersions (ASDs) has successfully been established to increase the bioavailability of poorly soluble drugs, when one polymer is not able to provide both, stabilizing the drug in the matrix and the supersaturated solution. Therefore, we investigated the influence of low-viscosity hydroxypropyl cellulose (HPC) polymers as an additive in HPMC based ternary ASD formulations made by hot-melt extrusion (HME) on the bioavailability of itraconazole (ITZ). The partitioning potential of the different HPC grades was screened in biphasic supersaturation assays. Solid-state analytics were performed using differential scanning calorimetry (DSC), X-ray powder diffraction (XRPD). The addition of HPCs, especially HPC-UL, resulted in a superior partitioned amount of ITZ in biphasic supersaturation assays. Moreover, the approach in using HPCs as an additive in HPMC based ASDs led to an increase in partitioned ITZ compared to Sporanox® in biorelevant biphasic dissolution studies. The results from the biphasic dissolution experiments correlated well with the in vivo studies, which revealed the highest oral bioavailability for the ternary ASD comprising HPC-UL and HPMC.