Issue link: https://www.e-digitaleditions.com/i/1276475
Inhalation August 2020 11 intended might impact only the mass dimension of a metered dose inhaler's APSD. Alternatively, a batch of the same product with less co-solvent than intended might deliver the same total mass of formulation, but with the distribution of the API mass shifted toward a finer aerodynamic diameter. Of course, there may also be changes in both dimensions of a product's APSD simultaneously, from either a single root cause or a combination of multiple causes. Regardless of the cause, though, successful control of an APSD requires a system of metrics which is sensitive to changes in each of these two orthogonal dimensions. Returning to your specification setting task, you might then choose impactor-sized mass (ISM) and mass median aerodynamic diameter (MMAD) as orthogonal metrics to independently control the mass and size dimensions, respectively, of your product's APSD. ISM is the size-fractionated portion of the API mass captured by an impactor, and MMAD is the median aerodynamic diameter of the ISM. 5, 6 ese two metrics are widely accepted measures of the mass and size dimensions of an APSD. Note that these met- rics include only the size-fractionated portion of the aerosol that is considered most directly related to pul- monary deposition. Let's now look at how you might use these two met- rics to control the product's APSD, as summarized in Figure 2. e left-hand plot presents the product's APSD data in terms of the two chosen metrics, with each point reporting the ISM (on the y-axis) and the MMAD (on the x-axis) of an individual APSD deter- mination. Note that there is no correlation between the two variables, confirming the expected orthogonality of the two dimensions of the APSD (i.e., the amount of the API mass and its aerodynamic size are independent of each other). Attempting to control your APSD To illustrate the shortcomings of stage groupings, let's walk through an example comparing different treat- ments of CI data and the information they offer. Note that this example focuses solely on the APSD as a mea- sure of an OIP's performance. Imagine that you are a development scientist working on an OIP. e 201 distributions in Figure 1 summarize the recent histor- ical performance of the product, including release and stability data from the batches used in pivotal clinical trials. e APSDs provided for this example are, in fact, development data from an actual OIP, drawn from the IPAC-RS blinded product database (see the side bar at the end of this article for more information). Looking ahead to commercial production, you need to define specifications that the quality control (QC) lab will eventually use to make batch disposition deci- sions. To do so, the QC lab will need to assess whether the performance of a newly manufactured batch is sufficiently similar to that of the clinical batches. Now, imagine for a minute that you do not have any pre- conceived notions regarding health authorities' expec- tations for reporting APSD data. Imagine that your concern is simply to establish metrics, the bounds of which define a reasonable range around your target APSD, enabling batch disposition decisions. So how does one place appropriate limits on a prod- uct's APSD? A wide variety of metrics are available for describing cascade impaction data, however their utility for making batch disposition decisions varies widely. Moreover, the utility of a given metric depends heavily upon how it is applied and, in particular, with which other metrics it is used. e key is to recognize that APSDs have two orthogonal dimensions, mass and size, and that APSDs can differ in either one, or both, of these dimensions. For example, an irregular meter- ing valve that delivers slightly more formulation than Figure 1 APSD determinations (201 total) for a solution metered dose inhaler (MDI) from the IPAC-RS blinded product database (see sidebar for details). API mass (expressed in terms of percent label claim (%LC)) is plotted for each impactor component (left), and the size-fractionated mass is plotted vs. aerodynamic diameter (right). Typical stage groupings, as defined in Table 1, are shown on each plot. Group 1 Group 4 70 60 50 40 30 20 10 0 35 30 25 20 15 10 5 0 Group 2 Group 3 IP Stage 0 Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Stage 6 Stage 7 Filter 0 1 2 3 4 5 6 7 8 9 API mass (%LC) Recovery (%LC) Group 3 Group 2 Group 4 Aerodynamic Diameter (microns)