First, we measured the particle size of excipients as powder, since the particle size was related the flowability (Fig. 1). The particle size of SP was smaller than that of the lactose. This result suggested that this implies that the flowability of starch is low and should be improved as it would likely cause packaging errors and result in residual powder on packaging paper. Subsequently, we attempted to prepare the SGs, and commercial HPC-30 and HPC-80 of different molecular weights were used as binders in the process, since the granule can be not provided by the granulation without binding agent (Fig. 1). Granulation increased the particle size of the SP, and the particle size of the HPC-30-SG and HPC-80-SG were larger than that of lactose (Fig. 1). The content of binder is also important in the production of SGs, since the binder strongly agglomerates between particles. In addition, it was known that the granule stability decreases when the amount added is too large [10], and an optimal concentration is required to prepare SGs. We showed that the 4.5% concentration was determined to be too viscous for use following preliminary investigations. Therefore, 0.75%, 1.5%, and 3% binder concentrations were used in this study. On the other hand, the particle size of the HPC-30-SG and HPC-80-SG was difference, and the largest particle size of granules at a 1.5% addition rate. The molecular weights of HPC-80 and HPC-30 are approximately 80,000 and 30,000, respectively, and the viscosity of HPC-80 is higher than that of HPC-30 (the viscosity of 1.5% HPC-30 and HPC-80 in water are 5 mPa·s, 18 mPa·s, respectively, 25℃). Taken together, these differences in the degree of polymerization, molecular weight, and viscosity may affect the particle size of SGs in this study.

Next, we measured the adhesion and recovery rate, respectively, during heat sealing of the ODP process of HPC-30-SG and HPC-80-SG (Fig. 2). The recovery of HPC-30-SG and HPC-80-SG was significantly higher than that of the SP, although, the recovery of both 0.75% HPC-30-SG and HPC-80-SG were significantly lower in comparison with 1.5% and 3% HPC-30-SG and HPC-80-SG. Moreover, the recoveries of 1.5% and 3% of HPC-30-SG were comparable to those of lactose. These results indicated that the SGs using HPC-30 and HPC-80 improved the issue observed with SP during ODP [11], given that the investigated excipients with binding agents are not inferior to lactose. On the other hand, it is important to clarify the lot-to-lot reproducibility. Therefore, we measured the particle size distribution and recovery rate in lot-to-lot of the corresponding 1.5% HPC-30-SG and HPC-80-SG (lot-to-lot reproducibility), and the no lot-to-lot differences were observed.

Moreover, evaluation of granule durability is of practical importance. Therefore, we demonstrated the changes in storage and transport properties of SGs when they were subjected to physical impact (Fig. 3). Following physical impact, 1.5% and 3% HPC-80-SG were not inferior to lactose, while other SGs were inferior with significant differences. From this result, 1.5% and 3% HPC-80-SG can be considered appropriate excipients to substitute lactose. In addition, the packaging error in 1.5% and 3% HPC-80-SG tend to be lower in comparison with other SP and corresponding SGs (Table 1). These results suggested that 1.5% and 3% HPC-80-SG provided fewer mistakes during packaging result in an accurate dose.

The results of this basic experiment demonstrated the formulation's usefulness, and the possible advantages of using this formulation in the clinical setting are following; (I) hospital preparation allows for tailoring prescriptions to meet individual patient needs or specific cases, providing optimal treatment. (II) it enables tighter quality control, especially when specific requirements for preparation and compounding are necessary. (III) a smaller scale can reduce formulation costs compared to mass production. On the contrary, demerits is following; (I) it is time-consuming and labor-intensive. Individual compounding for each prescription may restrict time available for other tasks or patient care. (II) it is a higher risk of errors or variability, Errors could affect efficacy or safety, requiring careful attention. These are the research limitations of this experiment, and it requires careful consideration and appropriate quality management.

In Japan, commercially available corn starch granules is able to provide from Japan corn starch co., ltd. (Japan). However, the D50 is 81.58 µm, and larger particle size is needed to attenuate the adhesion of corn starch granules during one dose packaging. In fact, the recovery rate of 0.75% HPC-30-SG, which is comparable particle size, is not enough (Figs. 2 and 3). Therefore, the preparation of SG using binding agents (HPC-30 or HPC-80) would be useful. On the other hand, it is possible to show similar effects with other HPC-30 derivatives by setting optimal concentrations. Further studies are needed to determine the optimal concentration settings for other HPC-30 derivatives. In conclusion, we designed SGs using HPC-30 or HPC-80 as binding agents, and found that drug loss during the ODP process was improved when between 1.5%–3% HPC-80 was used.