(T1530-08-51) Effect of Morphologies and Process Parameters on Aerodynamic Performance of Fluticasone Propionate in Dry Powder Inhaler

Purpose: Dry powder inhalers (DPIs), one of pulmonary drug delivery systems, have several advantages, including higher therapeutic efficacy and reduced systemic exposure. These advantages have contributed to the popularity of their use for treating asthma and chronic obstructive pulmonary disease (COPD) over the last decade. Fluticasone propionate (FP), a steroidal drug used to treat asthma and COPD, has long been used in DPIs. In the DPI system, drugs are loaded onto carriers for handling purposes and delivered to the targeted area by detached from carriers. It is known that there are many factors that influence the aerodynamic performance of DPI systems, such as drug polymorphism, the interaction between drugs and carriers, morphology, size, etc. The purpose of this study was to produce different morphologies of FP crystals and assess FP's aerodynamic performance by examining the FPF values and the proportion of FP deposited on each stage, considering both morphology and blending time.

Methods: Different morphologies and sizes of FP particles were prepared via anti-solvent recrystallization, slurrying, or shock cooling crystallization. Short fiber-like crystals were prepared by pouring 18 mL distilled water into 6 mL of polyethylene glycol 400 (PEG400) solution containing 20 mg of FP and stirring at 1200 rpm for 5 minutes. Long fiber-like crystals were obtained by heating a 0.5 mL acetonitrile (ACN) solution containing 16.95 mg of FP and pouring it into a mixture of 13 mL PEG 400 and 39 mL of distilled water, followed by stirring it at 500 rpm for 5 minutes. Small square-1 crystals were also prepared by slurrying 33.9 mg of FP in 1mL ACN at 200 rpm for 30 min. Rod-shape crystals were obtained by heating a 0.5 mL ACN solution containing 25.3 mg of FP at 85 °C and shock-cooling the solution in ice-water at 900 rpm for 30 minutes. Small square-2 crystals were obtained in the same manner as Rod-shape crystals, except for an additional 10 seconds of ball milling. Solid state characterization of the FP particles was performed using PXRD, DSC, TGA, and polarized microscope. Lactohale-200 (LH-200) was used as a carrier. Blending time was varied (15, 30, 60, 90, and 120 min.).
Next generation impactor (NGI) was used to evaluate the aerodynamic performance of FP crystals such as fine particle fraction (FPF), recovered dose, emitted dose (ED), and fine particle dose (FPD) as well as the percentage of FP on each stage. The amounts of FP on each stage of NGI were analyzed by HPLC.

Results: Crystals of FP with five different morphologies were produced by different crystallization methods. Size of short fiber-like crystals was ranged from 2.04−5.79 mm, that of long fiber-like crystals from 6.5−18.06 mm, that of small square-1 crystals with static electricity from 1.77−3 mm, that of small square-2 without static electricity from 1.40−2.43 mm, and that of rod-shaped crystals from 6.45−11.74 mm. The resulted FP crystals were all Form I. The effect of morphology and blending time on the aerodynamic performance of FP crystals was analyzed by measuring FPF, FPD, ED, and the proportion of FP crystals on each stage. The maximum FPF values, ranked in descending order, are as follows: long fiber-like (23.77%, 60 min), Seretide 500 Diskus (20.82%), short fiber-like (17.89%, 30 min), rod-shape (16.77%, 60 min), small square-2 (5.87%, 30 min), and small square-1 (1.39%, 90 min). Interestingly, long fiber-like FP crystals (23.77%) show higher FPF value than Seretide 500 Diskus (20.82%). It seems that the blending time required for the maximum FPF values depends on morphology of FP crystals. Specifically, long fiber-like and rod shows the maximum FPF value at 60 min. while, short fiber-like FP crystals shows the maximum FPF value at 30 min. In the case of small squares, the blending time does not seem to affect the FPF values. The impact of morphology on the aerodynamic performance of FP crystals was also evaluated by comparing the portion of FP crystals deposited on each stage during NGI test. Short fiber-like crystals tend to deposit on stage 5, long fiber-like crystals on stage 4, rod-shape crystals on stage 3~4, small square crystals on stage 4, and Seretide 500 Diskus on stage 5.

Conclusion: The interaction between the FP crystals and the carrier appeared to depend on the morphology of the FP crystals as well as the blending time. As a result, the trends of FPF values as a function of blending time were similar to FP crystals with similar morphologies. In addition, the stage deposited with the maximum portion varied depending on morphology meaning that the depth where inhaled drugs are deposited can be controlled by modifying morphology of drugs. Since the required depth for drug particles to be deposited varies depending on the disease, this study can significantly improve the efficacy of pulmonary drug delivery system by controlling the depth for drug particle to be deposited via morphology modifications.

References: Crystallization and crystallinity of fluticasone propionate

Figure 1. FPF values of FP crystals with five distinct morphologies at various blending time

Figure 2. FPD values of FP crystals with five distinct morphologies at various blending time

Figure 3. The amount fraction of FPD at the blending time that has the maximum FPF value.