Development of mannitol-based composite particles by spray drying: uncovering the critical process parameters

Eunice Costa, Filipa M. Maia, Susana Campos, Filipe Neves

Mannitol has been extensively explored in dry powder inhalation (DPI), namely i) as an alternative to lactose monohydrate in carrier-based formulations, being a non-reducing sugar alcohol, ii) for the treatment of cystic fibrosis as an airway hydrating agent and iii) for bronchial provocation testing. Moreover, mannitol holds great potential for engineering DPI composite particle formulations by spray drying (SD), since it is highly crystalline after the process, ensuring enhanced physical stability.
Although it is known that mannitol particle properties can be manipulated by adjusting SD parameters, few works can be found in the literature, especially on the field of respirable composite DPI formulations. For this reason, the current study reports the impact of SD parameters on mannitol particle size, morphology and solid state properties, aiming at optimal aerodynamic performance, using a design of experiments (DoE) approach. The best performing system was co-spray dried with nanoparticles of a model active pharmaceutical ingredient (API).
The aerodynamic performance of the mannitol particles were significantly impacted by the drying temperature, since particle morphology changed substantially within the explored range, due to differences in crystallization kinetics. The feed solution concentration, contrarily to expectation, was not a very significant parameter and the best performing system considered an intermediate temperature/feed concentration. Moreover, inclusion of the nanoparticles did not change significantly the deposition profile, regardless of the API relative concentration.
The illustrated development strategy allows a priori optimization of complex crystallizing composite particles, while balancing performance with process throughput. The resulting particles should be particularly suitable for dose-ranging studies, given its negligible dependence on the API load.

Join today to view and download the full abstract/presentation