(281e) Assessment of Vacuum Drum Filling Technology Highly Cohesive Dry-Powder Inhaler (DPI) Formulation

Inhaled therapeutics are routinely used to treat pulmonary diseases such as asthma, COPD, and cystic fibrosis. In recent years, the need to deliver high therapeutic doses into the lungs is becoming increasingly more commonplace. In the case of recently approved anti-infectives (e.g. colobreathe, tobramycin DPI, aerovanc) and Parkinson’s therapy (inbrija), for example, the total lung doses are thousands of times greater than those of conventional asthma medications (ICS, LABA, LAMA, etc.). Inhaled drug products, which are either a blend comprising the active pharmaceutical ingredient (API) and excipients, or engineered particles possessing attributes suitable for aerosolization (e.g. pulmospheres), are often packaged in blisters or standard-sized capsules and subsequently dispersed and delivered to the lungs by a specially designed inhaler. To promote patient compliance, it is desirable to administer the required lung dose in as few inhalations as possible. This, combined with the fill volume constraint of standard capsules, often leads to the need to maximize the drug loading in each dose.

This work presents the challenges associated with the process of filling a model high drug loading formulation into capsules. These challenges arise due to the cohesive and/or the low-density nature of the drug product, which is a consequence of the size and other attributes of the API payload: to reach the lung regions relevant to drug absorption, the API is typically micronized to a specific size range, or, in the case of engineered particles, their density is intentionally lowered by making the particles porous. While certain excipients can be blended with the API to improve filling capability, the fact remains that the amount that can be added is limited due to the reasons explained above. Such drug products, then, typically require specialized equipment to be filled into the dosage units.

Vacuum drum filling is one such equipment commonly used to manufacture dry powder inhalers (DPIs). Drum filling involves charging small amounts of formulation into a chamber wherein a drum into which small cylindrical cavities are drilled is placed on the bottom. When processing cohesive formulations, the charging of the formulation is often performed intermittently via a vibrating sieve placed on top of the chamber. Vacuum is then drawn from within the drum, causing some formulation to fill the cavities. The drum is then rotated by 180 degrees and air is blown out of cavities to eject the slightly compressed formulation out and into empty capsule bodies laid out under the drum. Throughout the drum dosing process, a stirrer placed inside the chamber is also intermittently rotating to prevent powder bridging with a rotational frequency that is in sync with the drum motion.

Despite its widespread use in this field, the drum filling process is not well studied. A literature search returns only two handfuls of articles, with most published in the last four years. To address such a lack, we performed a DoE study to investigate the effect of the process parameters hypothesized to affect the capsule filling process, namely the vacuum pressure, the blow-out pressure, the air pressure controlling the sieve vibration, and the rotational speed of the in-chamber stirrer. A ModuC-CS drum filler (Harro Hoefliger, Allmersbach, Germany) was used for the experiments. The following responses were measured: filling yield, fill weight consistency, capsule content uniformity (CU), and the aerosol performance of the filled capsules In addition, the amount of active ingredient in the formulation placed on the vibrating sieve is also tracked over time to assess potential vibration-induced segregations.

The results show high acceptance rates for all batches, ranging from 88-99%. The capability index, a measure of the capability of a process to provide output within the process specification limits, was between 0.56 and 0.99 for all runs, with the batch filled using a shorter filling duration and higher vacuum pressure/blowout pressure giving the highest dose consistency. Composite blend assays determined from samples taken from the vibrating sieve hopper at the beginning, middle, and end of each capsule filling run were within 97-103%, indicating no segregation risk for the formulation tested in this study. Similar trends were observed from the CU data assessed from capsule samples taken at the beginning, middle, and end of each run. All of the above results were also found to be relatively unimpacted by the in-chamber wire stirring speed, which directly affected the filling speed (i.e. faster stirring speed increases the drum rotation frequency, which in turn increases the overall filling speed). Finally, aerosol data assessed through the next generation impactor (NGI) and dosage unit sampling apparatus (DUSA) showed little to no influence of drum filling parameters on the aerosol statistics (e.g. emitted dose and fine particle fraction). All results confirm the feasibility and the robustness of a drum filling system for high-dose DPI manufacture.