Using cascade impaction data
Once the raw data are gathered, the question arises of how best to use them to meet testing requirements. The industry has
a number of routinely applied metrics, including MMAD, fine particle dose or fraction (FPD or FPF), and geometric standard
deviation (GSD). Of these, MMAD, the particle size below which 50% of the particle population lies on the basis of mass, is
probably the most widespread, although acceptance criteria are typically based on mass-per-stage or fine particle dose (FPD)
applicable to the active. Together, MMAD and GSD locate the central point of the APSD and describe its spread, thereby summarising
the key features of what is most often a monomodal particle-size distribution.
During drug development, the aim is to formulate towards a predetermined delivery of the active. The particle-size distribution
of the delivered dose is closely connected with the delivery efficiency because it influences deposition within the lungs.
In general, particles above 5 μm in diameter are considered too large for pulmonary deposition; for the deep lung, even finer
particles may be preferred, perhaps 2–3 μm (3). The lower size limits of OINDP formulations are usually of much less concern,
although research suggests that a high proportion of sub-micron particles may be an effective means of achieving bronchodilation,
thereby enhancing the efficiency of drug delivery (4).
APSD data support the manipulation of a device or formulation towards desirable performance. FPD or fine particle fraction
(FPF), the amount or fraction of the delivered dose that lies below the 5 μm range, is a useful metric; achieving an MMAD
of 2–3 μm is a frequent goal. A relatively low MMAD coupled with a low GSD is indicative of a tight size distribution centred
on a fine particle size, a potentially beneficial combination for efficient delivery.
Once a product transitions into manufacture, detecting any differences between manufactured material and the defined specification
becomes critical for quality assurance. The same metrics may be used to summarize APSD data because they focus attention on
those features of the distribution that are especially crucial in terms of the delivery to the patient and, by inference,
clinical efficacy. It is important to recognize, however, that regulators currently request full resolution data rather than
single number metrics to verify product quality, just as they do to support a new drug application.
Most testing, therefore, is comparative, against a predetermined target (i.e., development) or a specification (i.e., QC).
Overlaying two distributions shows up differences but not necessarily in a particularly informative way. However, the metrics
identified are indicative not only of the difference but also of relative performance, which explains their popularity.