The major barrier to both generic competition and on-going product improvements of OINDPs is the cost and requirements for clinical endpoint BE studies. Patient numbers can be significant, sometimes larger than the originator’s efficacy study. Furthermore, considering the high variability, low sensitivity and the inability to detect formulation differences these studies are only confirmatory of local equivalence.
To address the need to establish equivalence in local delivery, the FDA Office of Generic Drugs, through GDUFA I and II, have explored alternative pathways for these complex drug-device combination products, that can reliably ensure equivalence in bioavailability without the need for a clinical endpoint BE study. The general pathway that FDA have adopted for BE testing of locally-acting topical dosage forms, applies Q3 microstructural equivalence where differences between the same components (Q1) in the same concentration (Q2) under a non-equilibrium state can be related to the arrangement of matter and/or its state of aggregation . For transdermal suspensions, creams, gels etc., a road-map for Q3 microstructural equivalence has been determined and implemented. In the case of transdermal a combination of rheology and dissolution/permeability in vitro tools have been validated. The question therefore is that could such an approach be applied to the OINDP space?
For OINDPs, the addition of a device and the variable energy source imparted by the inspiratory flow of the patient can also have a secondary influence on the structural characteristics of an aerosolized dose. This introduces the need to investigate the microstructural relationship between both formulated and aerosolized forms of the product and the influence of device and patient factors in demonstrating the validity of Q3 for bioequivalence. Some of the techniques currently under assessment for characterising Q3 of OINDPs include the use of morphologically directed and surface mapping Raman spectroscopy (MDRS) and an integrated measurement of structure using in vitro dissolution and permeability kinetics.
MDRS is a novel in vitro technology for elucidation of morphological and chemical features of physical blends of drugs and excipients. For example, FDA has recognized the potential of MDRS to provide supportive information on equivalence in drug particle size between test and reference nasal spray suspension products. We have utilised MDRS to measure the particle size of nasal suspension formulation containing mometasone furoate of different particle size and compared to Nasonex. In addition, we attempted to correlate the particle size of the API in the finished products with their dissolution behaviour using our UniDose technology. The in vitro dissolution profiles and the relationship between the first order half-life and the % of particles by volume less than 5µm determined by MDRS of the formulated API batches and the Nasonex RLD product are shown in Figure 1. These data suggest that there is a direct relationship between particle size of the API and the release behaviour of API, when tested under sink conditions. Batch 1 has the largest particle size and slowest rate of dissolution, whilst batch 2 has the smallest particle size and therefore fastest rate of dissolution. These data support the consideration made by the regulatory agencies that the rate and extent of release for local absorption of nasal spray suspensions may be indicated by particle-size distribution measurements of the API within these formulated products.
Figure 1: Relationship between the first order dissolution half-life and % of particles <5µm for commercial Nasonex RLD and the prepared formulations with four different batches of API.
We have attempted to use the combination of dissolution testing using our UniDose system and MDRS to determine the microstructure of dry powder inhaler (DPI) formulations. Figure 2 compares the in vitro dissolution release profiles of the FP component of the aerosolized impactor stage mass (ISM) dose of US Advair DPI Diskus 100/50µg, 250/50µg and 500/50µg FP/SX. These data suggest that the dissolution rate of the FP dose collected was inversely proportional to drug loading, in which the low strength exhibited the fastest rate of dissolution and the high strength product the slowest rate of dissolution. For a fixed concentration of SX and a constant fill weight (12.5mg) for these unit dose blister formulations, increasing the surface coverage of FP appears to lead to a concomitant decrease in the rate of dissolution of the aerosolized dose. Previous studies that have shown that the dissolution rate of a poorly soluble compound in an interactive mixture is inversely proportional to the degree of surface coverage and more specifically to the surface area ratio between drug and excipient. There is a greater likelihood of agglomerate formation over discrete drug particle-excipient interactions as drug concentration increases.
Figure 2: USP Apparatus V Paddle-over-Disk (POD) dissolution release profiles of aerosolized ISM dose of FP from US Advair Diskus 100/50µg, 250/50µg, 500/50µg DPIs (n=5).
The collected ISM dose using the UniDose apparatus was also characterized using MDRS. Upon collection, the filter substrate was mounted directly on to the sample stage of the Morphologi G3-ID. The microstructural differences following MDRS analysis for the ISM dose collected from Advair Diskus 100/50µg, 250/50 µg and 500/50µg FP/SX DPI products are shown in Figure 6. These data show the presence of “free-standing” or discrete API together with mixed agglomerates within the aerosolized dose as a percentage of the total particles analysed in each case. While there are constraints associated with the spatial resolution of the G3-ID apparatus and the acquisition times needed for analysis, the Q3 map is an elegant way to compare the aerosol that would be delivered to the cascade impactor following APSD testing in each case. As shown, the percentage of discrete FP increased with increased concentration of FP in the powder blend. Moreover, the presence of FP-lactose-SX agglomerates decreased as the dose strength of FP in Advair increased. A comparison of these data with the dissolution data presented implies that the faster dissolution of FP from the US Advair Diskus 100/50µg low dose product may be related to the smaller amount of discrete FP alongside the increased FP agglomeration with more soluble components, lactose and SX, that could accelerate FP dissolution. The slower rate of dissolution of FP from both the mid and high-strength formulations was consistent with the increased amount of free-standing FP (Figure 3) and lower volume of mixed agglomerate structures, resulting in poor wettability of FP and a reduction in dissolution rate. These data support the role that both in vitro dissolution testing of sparingly soluble compounds and microstructural characterization by MDRS may play in enabling scientifically valid measurements of the state of aggregation of APIs within a representative lung dose.
Figure 3: MDRS Structural analyses of the state of agglomeration as a percentage of the total number of particles analysed as either free-standing FP or agglomerated with lactose, SX and SX-lactose from US Advair Diskus 100/50µg, 250/50µg, 500/50µg DPIs
At Nanopharm we have pioneered the concept of Q3 structural equivalence for OINDPs. We believe this alternative approach to seek generic product approval is vital for our industry. The utility of dissolution testing and MDRS, together with validated in silico mechanistic modelling, may help provide alternative approaches for inter-product comparisons that may enable clinical biowaivers for OINPD development programs.