Quality by Design can help determine which inhalation device is best for a drug. John Vennari, West Analytical Services, explains how to obtain the required optimised automated actuation parameters
Quality by Design (QbD), the most recent methodology for improving the safety and efficacy of drug products, is causing pharmaceutical companies to re-evaluate the costs and, more importantly, safety measures associated with achieving high quality overall. Although not yet adopted industry wide, QbD is arguably an essential strategy, especially given the increasingly complex and expensive drug products reaching the market.
One area where pharmaceutical and delivery device companies are successfully applying QbD principles is in manually activated inhalation devices. When applied properly, QbD can help companies determine which inhalation device is best for a specific drug product and how to optimise automated actuation parameters to meet targeted performance specifications that help ensure accurate patient use.
There is a wide variation in the actuation process for nasal spray or metered dose inhalers. Because of this, the FDA recommends performing an exploratory study of hand actuation that simulates in vitro performance of a spray or inhaled drug product (i.e. use a representative sample of trained patients) to determine the relevant actuation parameters for a product. Hand actuation studies have shown that the human hand uses a positional control mechanism when actuating spray pumps. This is as a result of the hand's ability to sense and react to the internal resistance or back pressure of the spray pump mechanism.
Over the years, procedures and equipment have been developed to determine statistically an optimised set of automated actuation parameters to measure velocity, acceleration, hold time and stroke length accurately. By following a proven patented process,1 an optimised set of automated actuation parameters can be developed based on trained patient hand actuation performance - this is the inherent value of QbD.
To establish the patented, proven QbD process, a set of parameters must be established: Design Space, Control Space and Operating Space (Fig. 1). The Operating Space ultimately provides the optimised set of automated actuation parameters, as recommended by the FDA. Unfortunately, despite FDA recommendations, many pump manufacturers do not conduct a study of hand actuation to determine the relevant actuation parameters for a product.
In the document Guidance for Industry: Bioavailability and Bioequivalence Studies for Nasal Aerosols and Nasal Sprays for Local Action, (April 2003), the FDA recommends the use of automated actuation systems to perform in vitro measurements related to nasal aerosol and spray product characterisation. In this paper, the FDA also advises that appropriate settings used by an automated actuation system be determined either by consulting with the pump manufacturer or by performing exploratory studies based on hand actuation.
The challenge is that pump manufacturers rarely, if ever, know what the appropriate settings should be for automated actuation. This is because pump manufacturers are usually not familiar with the proprietary physical properties, such as viscosity and surface tension, of the drug formulation. Since physical formulation properties affect the characteristics of the spray, it is difficult to choose the best pump to fit an application without performing tests that will ensure a set of optimal actuation parameters for a specific dosage form and target indication.
Although the FDA made similar actuation recommendations in the past, the technology available was not able to support accurate automated actuation stations and mimic the in-use actuation performance based on actual patient use. This idea was long considered a commonsense way to decrease variability in drug delivery caused by patient use, provide accurate and repeatable actuation, remove potential analyst bias in actuation and increase the sensitivity for detecting potential differences between products.
However, without technology for automated actuation studies, it was virtually impossible to develop methods that would accurately determine a single set of parameters based on hand actuation that could be used for all comparative in vitro bioequivalence tests - i.e. single actuation content (SAC) through container life (shot weight), droplet size distribution by laser diffraction, drug in small particles/droplets, or particle/droplet size distribution by cascade impactor, spray pattern, plume geometry, and priming and repriming.
The first positional controlled automated actuation station was developed in 2001 by Proveris Scientific Corporation. For the first time, the SprayVIEW NSx station provided the industry with automated actuation parameters that could mimic patient hand actuation performance. However, with the advent of this new technology several new challenges arose:
1. Capturing the actual hand movements of trained patients.
2. Developing a study to determine an optimised set of automated actuation parameters based on trained patient hand actuation performance caused by the variation of the individual's hand strength.
3. Determining the appropriate settings for the spray drug product (i.e. pump/formulation combination) drug formulations with different physical properties (i.e. viscosity, surface tension, etc.) that can affect hand actuation performance.
4. Creating the optimal actuation parameters to actuate the product with an automated system as part of a comprehensive in-vitro testing plan.
hands-on study
Reintroducing reality to the testing of nasal spray devices required bringing back the hand, but returning to live-subject-only, hand-based test procedures was not the answer: automation needed to be part of the test equation. In keeping with the FDA recommendation to develop exploratory studies based on actual hand movements of trained patients, a new test device was needed that would collect the in-use actuation parameters.
Proveris Scientific developed the first hand-held response sensor, Ergo (patent pending), that indicates the mechanical relationship between the movable and stationary parts of a spray device. Ergo is constructed of aluminum or similar lightweight material and allows a person to operate the spray device in an unimpeded manner and simulate typical use of the spray device. The nasal spray device is affixed to Ergo so that a subject's hands fit around built-in flanges on the unit and the thumb is placed in a natural position at the bottom.
A person actuates the spray device in a typical manner and the motion expels a shot.or dose from the spray device. When the spray device is activated, Ergo captures the hand movement and electronically sends raw data to a computer. The data is then processed to remove any noise, such as that from the electrical power line, and is then put through a computer programme that determines the actuation and return velocity, acceleration, hold time and stoke length.
This process is repeated, typically using three subjects, until an acceptable set of actuation parameter ranges have been recorded for each device under consideration. The collected parameters are then examined by a trained analyst using QbD principles to first determine the design space of the device under test.
The Design Space of a selected multidose nasal spray device is defined by the minimum and maximum range of the raw data collected from a trained patient sampling. During the testing process, subjects must produce enough sprays to determine accurately spray parameters to program the automated actuation equipment.
For this article, data was collected using a multidose Pfeiffer nasal spray pump filled with water and the analysis data was blinded to protect the identity of the trained testers. The pump's delivery volume was 70 microlitres (~70mg of water) per actuation. Test subjects were chosen to span the targeted age group of the specific drug product and three trained testers were used: two men and one woman, ages 42, 27 and 49, respectively. Since a multidose nasal spray device was used as part of this study, the first step was to determine the product's label claim for the number of priming shots required.
The graph in Figure 2 illustrates raw data collected from the trained testers actuating three devices from a single product lot. Each device was initially actuated 10 times by hand into a spray collection vessel to determine the label claim for the number of shots required to prime the devices. Each spray was weighed as delivered spray weight using a 4-place analytical balance (Mettler-Toledo AX204). The hand actuation data was collected from each trained tester and analysed. As shown in Fig. 2, the delivered spray weight reaches a consistent value (near its target of 70mg) after five shots. Although based on a small sample size, this test clearly demonstrates the number of priming strokes needed to confirm a label claim value.
Once the priming specification is confirmed, the Design Space is determined (without using the priming shot data) for the delivered shot weight, velocity, acceleration, hold time and stoke length, as shown in Table 1 and Fig. 3.
The variation in hold time, stroke length, actuation velocity, return velocity, actuation acceleration and return acceleration are also plotted and analysed similarly to the above Delivered Shot Weight by the same 30 shots per device for a total of 90 shots per device.
Once the raw data is collected for each actuation control parameter, an analyst can then determine the Design Space, as shown in Table 1.
control space
The Control Space is determined by statistical analysis of the Design Space results (e.g. average ±2s [standard deviation] for each design space actuation parameter) as shown in Table 2 (Control Space values are shown in dark purple). To maintain data integrity, if the calculated Control Space minimum (maximum) value was found to be less (greater) than the Design Space minimum (maximum), then the Design Space value was used in the Control Space. This safeguard ensures that the Control Space stays within the Design Space parameter values.
The final step in the QbD process is ascertaining the Operating Space. This is determined by using a combination of Control Space settings that produce the delivered shot weight performance closest to the target dosage, e.g. minimum % difference from target shot weight (70mg).
A technique of adjusting only one parameter at a time for each Control Space setting is used - along with measurements of the delivered shot weight - to determine the Control Space results, from which the Operating Space was selected as described above. Once the adjustment levels are known they can be incorporated into an automated actuation system and shot weight values can then be recorded for multiple actuations at each adjustment level. Typically, five adjustment levels are collected for each measured parameter. The actuation parameters are then selected individually based on those parameters that produce the shot weight with the lowest percentage difference from the target dosage.
Delivered shot weight values are collected with an automated actuation station and 4-place analytical balance using the Control Space actuation settings from Table 2. The results are shown in Table 3 along with the selected Operating Space values (shown in dark purple).
The variation as a function of stroke length, actuation velocity, return velocity, actuation acceleration and return acceleration are also plotted and analysed after five priming shots (Fig. 4). Each point on this plot represents the average of two consecutive shots taken with identical settings, for a total of 10 actuations and five plotted points.
Once all the parameters are tested in the Control Space, an analyst can determine the Optimised Set of Actuation Parameters Space as part of the Operating Space, as shown in Table 4.
Applying QbD principles to determine actuation parameters can greatly reduce product development risks and potentially provide pharmaceutical companies with significant cost savings. Furthermore, the QbD principles used to determine actuation parameters in a multi-dose nasal spray device can be slightly modified to simulate in vitro performance for other dosage forms such as oral sprays, metered dose inhalers, and unit- and bi-dose nasal/oral spray devices.
Together, global pharmaceutical organisations and delivery device companies can gain a competitive edge by working with analytical laboratories, such as West, an expert in applying QbD principles. Working with an outside laboratory reduces development time as well as the risk of selecting an inhalation spray device that may not be appropriate for a specific target indication. In addition, the overall quality of the regulatory submission is improved and promotes sound QbD principles, as advocated by industry2 and the FDA.3,4
Some drug makers remain unconvinced that studies should be performed, despite the fact that the FDA may ask for evidence that a hand-actuation study was used to produce test results for nasal spray devices. The FDA draft guidance is currently only a recommendation, and in cases where an inappropriate delivery device was chosen, a drug manufacturer may have to start the selection process anew, a process that is certain to be costly. Developing an appropriate Operating Space early in the drug development process is an investment in success that can help ensure that a multi-dose inhalation drug travels the shortest path to approval.
acknowledgement
The author would like to thank Dino J. Farina, Proveris Scientific Corporation, for his help in collecting the data used in this article.