Patient Focus and Regulatory Considerations for Inhalation Device Design: Report from the 2015 IPAC-RS/ISAM Workshop

Inhaler use technique

Rates of incorrect inhaler use vary by device type and between studies, but are generally high for all types of inhalers,⁽²⁰⁾ and correlate with poor disease control,⁽²¹⁾ as measured by hospitalizations and exacerbations. The difficulty of using inhalers correctly is due not only to the complexity of administering any inhaled medicine, which typically involves several critical steps, but also to the fact that those steps may differ radically from one device type to another (Table 1).

Furthermore, the diversity of available device types only increases with time, and some patients have to use more than one device type during the course of their treatment. Delivering all inhaler medications through a single device type would be ideal for patients and healthcare providers, but is objectively impossible for a number of reasons, given the broad population of patient situations that need to be considered, the range of physiochemical attributes of the active pharmaceutical ingredients, accepted formulation technologies, pharmacologically active dose, and so on. Hence, the ideal may not be practicable. For example, acute distress/rescue medications must be delivered even when a patient's breathing is compromised; in the most severe situations, this demands pressurized metered dose inhalers (pMDIs) or nebulizers, rather than dry powder inhalers (DPIs), which require energetic inhalation. Similarly, young pediatric patients may be able to inhale medicine when delivered by a pMDI with a spacer, but not by a DPI.

In addition, a number of active therapeutic molecules have extremely low solubility in available propellants, thereby making DPIs a preferred choice as delivery devices (e.g., for some inhaled corticosteroids). Furthermore, patients are unique and have different preferences and capabilities when it comes to using different inhaler technologies. Overall, therefore, availability of multiple options of OIP device types remains essential to ensuring that each patient can find an inhaler that works for them.⁽²²⁾ At the same time, this variety may represent a challenge for physician knowledge as well as patients’ adherence and correct use of inhalers. Strategies to better understand and to simplify patient–device interactions are therefore much needed, ideally resulting in clear recommendations both for device design and patient/physician training and support materials.

(Non) adherence

Adherence in patients using pills or tablets—which could be considered an “ideal delivery system” since it requires minimal manipulation by the patient—is generally higher than for inhalers, but is still <100%,⁽²³⁾ confirming at an intuitive level what rigorous studies have found—device design is important, but it cannot overcome all patient adherence issues.

Adherence is relatively poor in patients with chronic diseases of airways, ranking lower than for other chronic conditions.⁽²⁴⁾ This finding is especially worrying, given that poor technique and poor adherence have been repeatedly shown to lead to decreased effectiveness of treatment and decreased survival rates.^(11,25–27) Determinants of adherence are diverse and complex. Besides device design, they include patient's predispositions and abilities, social factors (relationships with physician, family, community support), disease state, frequency, and complexity of regimen.⁽²⁸⁾ Taking into account the patient's perspective⁽²⁹⁾ and involving the patient in clinical decision-making⁽³⁰⁾ were shown to improve adherence. However, phenomena such as intentional (and, sometimes, “intelligent”) nonadherence^(31,32) further complicate the picture and make easy solutions all but impossible. Nevertheless, improving adherence may help save lives and reduce overall healthcare costs.⁽³³⁾ To reach these goals, a combination of multiple forms of intervention, education, and regularly repeated follow-ups produces best results.^(34–37)

Training and education

Training usually leads to demonstrated improvements in technique and adherence.⁽³⁸⁾ Even a relatively short (an average of 6 minutes) education session was shown to increase correct use of pMDIs from 25% to 80% of patients in one study and this was accompanied by improved asthma control.⁽²⁶⁾ Interestingly, age and educational level of patients have not been consistently associated with device mastery, suggesting that most patients can be trained to use inhalers correctly.⁽³⁹⁾ However, a large proportion of patients never receive proper training,⁽²⁶⁾ and when a prescribed inhaler is switched at the pharmacy without retraining, disease control suffers.⁽⁴⁰⁾

Observation of patient technique as well as providing and receiving feedback to/from patient are critical components of effective training. Without such interaction with a trained professional, and without meaningful feedback, patients often believe they use the device correctly even when they do not.^(41,42)

To enable appropriate patient education, physicians, nurses, and pharmacists also need to be educated so they understand not only different inhalation techniques but also the importance of using the correct technique. These healthcare workers need to be provided with appropriate training materials and follow-up tools.^(43,44) The cost, effectiveness, and availability of such materials and training systems may be an important consideration of healthcare policy, and this may vary from region to region.

Patients’ preferences

A patient's preference for a particular device is a crucial determinant of correct actuation, inhalation technique, and adherence. In surveys, patients identify a long list of factors that influence their device preference,⁽⁴⁵⁾ ranging from having a dose counter to the ease of carrying the inhaler in a pocket and simplicity of use (Fig. 1). Taking the preferences of target patient populations into account when designing new inhaler devices should go a long way toward improving patients’ understanding of their devices and adherence to the prescribed therapy.⁽⁴⁶⁾

OPEN IN VIEWER

FIG. 1.

Preference of patients and physicians for various aspects of inhaler design. Source: Respiratory Disease Specific Program 2008. Adelphi Real World, Bollington, UK. Unpublished raw data, cited with permission.

Instructions for use and human factors

“Easy and simple instructions” rank high among preference factors by both patients and physicians (Fig. 1). Currently available Instructions for Use (IFU) often seem to have been written by, or for, engineers and lawyers. Instead, they should be written with a patient (as both the audience and user) in mind. There are currently no standard IFU even for common OIPs such as pMDIs, although many organizations post their own written and image-based instructions online, all differing in approach and terminology. An IPAC-RS Working Group has examined IFU for several commercially available pMDIs, also taking into account other published resources, with the goal of identifying commonalities and developing recommendations for a consistent layout and standardized terminology used in IFU.⁽⁴⁷⁾ General recommendations identified so far include:

• Break up the dense text into parts.

The testing of an IFU should focus on what the patient thinks instructions say, not what engineers think they say. Testing how patients interact with an inhaler after having studied an IFU is the key determinant as to whether the IFU has been designed appropriately for the target patient population.

The testing of an IFU is a component part of the Human Factors (HF) studies, which is a critical part of any OIP development. HF studies are different from clinical trials in several ways. One key distinction is that their focus is to learn what patients do with a device (rather than “what the medicine does to the patient once it's delivered” —which is studied in clinical trials).

HF studies and the application of Human Factors Engineering (HFE) principles to drug–device combination products such as OIPs have been receiving increased attention from the scientific community, standard-setting bodies, and regulators both in the United States and European Union (EU).^{(5–7,48–55)} These guidances and publications are providing significant impetus for further exchange of information and ideas among the stakeholders, which should lead to an increase in everyone's understanding of HF, and to improved designs of future devices and their IFU.

Telehealth: what role it might play

“Telehealth” can be defined as a collection of methods and approaches for monitoring healthcare delivery using telecommunication technologies. Telehealth may provide a tool for some patient groups to help improve poor adherence, for example, by enabling objective monitoring and providing feedback to patients and physicians.⁽⁵⁶⁾ A “health care package” that includes a drug product, an electronic monitoring app, and an ongoing training service might therefore be pursued by some companies to differentiate their product or service. Table 2 shows components of an ongoing patient care program, highlighting those elements related to device design, patient training, and monitoring. Many of these components can be targeted using electronic support, helping address patients’ behavior as well as motivation.

There are an increasing number of software tools that allow researchers and healthcare professionals to better understand motivation, as well as an increasing number of technologies that measure adherence specifically with inhaler-based therapies. The components of a successful telehealth deployment are summarized in Table 3. For a wide adoption, however, a broad-based effort would be required, including a change in how medical software is regulated. The need for a broad-based collaboration and coordination among various communities involved in collecting, analyzing, and using digital data made available by the proliferating electronic tools is being recognized not only in the field of OIPs and drug–device combination products but also in the more general discussions about “big data” in the service of better, patient-centered healthcare delivery.^(57–60)

Taking patient's breathing into account

The breathing pattern of patients is one of the major factors that determine the amount of drug reaching the respiratory tract overall and the distribution of the drug among different regions of the respiratory tract.⁽⁶¹⁾ Patient-independent (i.e., in-vitro) characteristics, or critical quality attributes of pharmaceutical aerosols produced by OIPs, such as their delivered dose, aerodynamic particle size distribution (APSD), mass median aerodynamic diameter, fine particle fraction, aerosol velocity (in pMDIs), device airflow resistance (in DPIs)—all play an important role, but cannot override inappropriate (for a given device) inhalation technique, including inhalation timing, speed, volume, and holding.^(62–67) For example, drug dose reaching the lung can range from 0% to 50% of the expected dose in pMDIs depending on the breathing pattern.⁽⁶⁸⁾ Similarly, most errors in DPI drug delivery relate to the breathing maneuver.⁽⁶⁹⁾

Per regulatory guidances,^(3,4) OIP devices should be characterized over a range of flow rates reflective of the patient populations. Nevertheless, they are designed to control APSD under standardized laboratory conditions rather than under inhalation maneuvers that might be seen in patients, who have very heterogeneous spontaneous inhalation modes that are further influenced by the severity of their airflow obstruction.

An “optimal” inhalation flow rate (which would maximize drug delivery to the relevant portion of the airways) is device dependent. In general, for pMDIs and soft mist inhalers, slower controlled inhalation leads to higher lung deposition,⁽⁷⁰⁾ while inhaling too fast may increase oropharyngeal impaction of the drug and reduce the amount available for lung deposition. Accordingly, medium-to-slow inhalation is recommended for pMDIs, also because evaporation of the propellant (leading to smaller particle size and consequently deeper lung deposition) takes time. Conversely, DPIs need a sufficiently high inhalation flow rate to achieve deaggregation of the powder. For this reason, slow breathing is not recommended for DPIs. Optimum flow rate for DPIs depends on the design of the DPI and the formulation.

The importance of characterizing in-vitro OIP performance under more realistic conditions has been raised in the past, with the important distinction drawn between tests conducted for product development versus those intended for routine quality control.^(71,72) Anatomically correct induction ports (“throats”) for cascade impactors have been developed,^(73,74) but remain a research tool rather than a standard approach in OIP development. Finally, an alternative way to test APSDs, which involves a breathing simulator, has been described previously,^(75,76) but so far has not received wide adoption or regulatory recognition.

Food and Drug Administration Center for Devices and Radiological Health perspective

In the United States, most OIPs are considered drug–device combination products. The primary Food and Drug Administration (FDA) center responsible for their review and approval is the Center for Drug Evaluation and Research (CDER) because the primary mode of action in these OIPs is the drug. By contrast, those devices that are not integrated with the drug formulation are reviewed by the FDA Center for Devices and Radiological Health (CDRH). For example, general indication (not drug specific) nebulizers⁽⁷⁷⁾ are usually approved by the “510(k) application”^(78,79) through CDRH, whereas drug-specific nebulizers would typically be reviewed as a New Drug Application (NDA) or Abbreviated NDA (ANDA) through CDER, with consultation from CDRH. FDA's Center for Biologics Evaluation and Research (CBER) typically is the lead Center for review when the formulation is a vaccine or includes proteins, peptides, or other biological components (e.g., if a nebulizer is developed for the delivery of one specific biologic product).

FDA encourages sponsors to meet with the agency early in the development of a novel device or a combination product so that the sponsor would have correct expectations for the regulatory pathway and corresponding requirements.⁽⁸⁰⁾ Where the primary mode of action is unclear, a sponsor may submit a request for designation.⁽⁸¹⁾ When a sponsor meets with its primary review division in the lead center within FDA, they should also request attendance by an expert from other appropriate centers/offices for the combination product. When reviewing the device component of an OIP, CDRH should be consulted. Some of the key factors considered by CDRH are summarized in Table 4. Further advice can be found in published guidance documents and consensus standards.^(82–85)

FDA CDER Office of Generic Drugs perspective

In the United States, innovator OIPs are reviewed through an NDA. Generic products are reviewed through an ANDA. A generic product is one that is therapeutically equivalent to the innovator product, which is designated as the reference listed drug (RLD). Therapeutically equivalent products are those that are both pharmaceutically equivalent (same active ingredient(s), same dosage form, same route of administration, identical in strength or concentration, meet compendial or other applicable standards of strength, quality, purity, and identity) and bioequivalent (as can be demonstrated through comparative in-vitro and in-vivo studies) to the RLD. Generic drug products are considered to have the same efficacy and safety profiles as the RLD under labeled conditions of use and are, therefore, substitutable without any dose adjustment or additional monitoring. In a generic drug development program, there are no new clinical trials conducted for demonstration of efficacy and/or safety.

Generic DPI devices present a special challenge. DPIs can differ greatly with respect to the external design of the delivery device and operating principles, and these differences may affect patient handling. Therefore, when designing a generic DPI, it is important to consider, among other things, the design and operating principles of the device component. For instance, a switch from one DPI to another without additional retraining (e.g., from a premetered multiunit dose device to a drug reservoir multidose device) may cause patient confusion and incorrect use of the DPI device, resulting in inability to achieve proper dosing and treatment. Thus, the assurance of substitutability of the generic and reference DPIs needs to account for patient–device interactions. Therefore, to ensure substitutability in the patient's hands, generic DPI devices are recommended to have similar or same characteristics, relative to the RLD device (Table 5).

FDA Office of Generic Drugs (OGD) has also conducted research and published findings to illustrate the importance of enhanced product understanding in the fabrication and refinement of test DPI devices that provide a closer match to the aerosolization performance of the reference DPI device at multiple flow rates.^(86,87) Computational fluid dynamics and in-vitro characterization by a multistage cascade impactor were utilized to provide engineering assistance in identifying and understanding the key performance attributes, which would influence the criteria for designing and modifying a test DPI device. Similarly, the material property attributes of micronized fluticasone propionate and carrier lactose were measured to design formulation that enabled in-vitro equivalence of test and reference product at three flow rates.

European regulatory framework

In contrast to the United States, Europe has no legal definition of a drug–device combination product, and such products are currently regulated either as medical devices or as medicinal products. Differences between the two routes in the EU are summarized in Table 6. Corresponding regulation is provided in the Medical Devices Directive (MDD),⁽⁸⁸⁾ Medicinal Product Directive,^(89,90) and Regulation.⁽⁹¹⁾ The general legislative basis for pharmaceuticals in the EU is presented in Eudralex volumes.⁽⁹²⁾ The European Medicines Agency (EMA) has also published a series of specific guidelines on clinical⁽⁹³⁾ and quality⁽⁴⁾ aspects of OIPs.

The concept of a “generic” is also defined differently in Europe compared to the United States, and the requirements for their approval differ in many respects between the regions. For example, in Europe (but not in the United States), if a proposed “follow-on” product has the same active ingredient(s) as the reference product(s), and if it meets a number of other criteria, it could be approved based on in-vitro data only—provided that the proposed product meets strictly delineated requirements relative to its reference product,⁽⁹⁴⁾ some of which are summarized in Table 7.

In general, European regulators recommend that OIP performance be investigated under conditions simulating use by patients. This includes activating the delivery device at the frequency indicated in the IFU, considering effects of cleaning and of carrying the delivery device between use. The robustness of the device should also be assessed by dropping the device from various heights onto various surfaces and determining the effects of possible drug accumulation on parts of the device over its lifetime, the effectiveness of any lockout mechanism, the vibrational stability (for powder mixtures), and potentially other aspects of robustness.⁽⁹⁵⁾ Similar device characterizations may be expected by FDA regulators, and an international standard with recommended tests for design verification of aerosol delivery devices was issued by ISO.⁽⁹⁶⁾

Although the European system generally works well, there is limited European consensus on the requirements for OIP devices regulated as European Conformity (CE)-marked medical devices.⁽⁹⁷⁾ There are no definitions of Drug–Device Combination medicinal products (DDCs) in the EU legislation. The device may be an integral component of the medicinal product (e.g., a pMDI), copackaged (e.g., a refillable device), comarketed (e.g., named nebulizers for use with specific nebulizer solutions), or independently marketed (in cases where the device meets the requirements for the necessary delivery system stated in the Summary of Product Characteristics of the drug product). Challenges arise due to the fundamental differences between the operation of legislation for medical devices and for medicines. Integral DDCs should comply with the Medicinal Products Directive (MPD) 2001/83/EC and should also provide evidence of compliance with the MDD 93/42/EC-Annex 1 as far as safety and performance-related device features are concerned. In the nonintegrated DDCs, the medicinal product component should comply with the MPD, while the medical device component should comply with the MDD 93/42/EC and should be CE marked.

This situation can result in a lack of clarity and inconsistent approaches for device components. To further complicate the planning of a regulatory strategy, at the time of this writing, the MDDs are being revised into a new Regulation; the number of Notified Bodies is decreasing (from the current number of more than 60); and the need for usability engineering and HF testing is being increasingly recognized. A guidance on this last topic is expected to be released by the UK's Medicines and Healthcare Products Regulatory Agency (MHRA) in mid-2016.⁽⁹⁷⁾ For the new Medical Device Regulations, there are proposals to update the medicines’ legislation (2001/83/EEC) to require input of Notified Bodies into the assessment of integral medical device components of medicinal products for conformity with Annex I of Essential Requirements. The final text of the new Regulation is currently being discussed between the European Commission, European Council, and European Parliament and is expected to be released in 2016.

Other tools for device development

ICH guidelines Q8–11^(98,99) set out a framework for pharmaceutical product development using QbD principles. The application of QbD principles specifically to OIPs was illustrated by IPAC-RS using several case studies of real-world products.⁽¹⁰⁰⁾ QbD is a risk-based approach that provides the sponsoring company a more systematic understanding of the “boundaries” of product and process parameters, within which in-vitro performance remains acceptable.⁽¹⁰¹⁾ Such a deepened scientific knowledge base provides the sponsoring companies and regulators with an increased assurance as to the quality, safety, and efficacy of the drug product through its product life cycle.

While a QbD approach may be familiar to many OIP manufacturers, its assessment may benefit from a clear communication with medicine regulators on the development work that has been conducted for a device product. This is especially important in the EU, where reviewers of drug–device combination products are likely to be pharmacists and clinicians by training, rather than engineers or physicists, and thus currently are less experienced in the design and validation strategies routinely used in those more “mechanical” disciplines. Sponsoring companies should be mindful that the regulatory reviewers may not be experts in mechanical engineering, and therefore, the QbD approach should be presented in such a format that a nonexpert could understand the scientific approach adopted.

Another useful international standard that describes a process for change management through the product life cycle is being developed by ISO.⁽¹⁰²⁾ All companies that develop and have marketed devices recognize that the design will evolve and change through the product life cycle. The drivers of these changes can vary from, but are not limited to the following:

• Increase in manufacturing scale (e.g., design features incorporated into part design to support high-volume manufacture)

The proposed ISO standard, when implemented, will provide the following:

• To sponsoring companies—guidance as to what data are required to support any change after pivotal clinical studies (since the safety/efficacy of any drug product for which market approval is sought or granted is dependent upon maintenance of a link with these pivotal data)

The number and diversity of changes that OIP sponsors are facing through a product's development cycle and postapproval are enormous, especially because regulators usually advise that device be “locked” (finalized) early on, before the start of pivotal clinical trials. However, this is not always ideal or possible because the need for device refinements may emerge during the clinical experience.

At the moment, the approach to justifying device changes is highly inconsistent,⁽¹⁰³⁾ as well as lengthy—with most of the changes taking more than a year to implement, including the collection of justification/qualification data and the regulatory approval.⁽¹⁰⁴⁾ Accordingly, in a previous ISAM/IPAC-RS conference, it was recommended that a standard be developed for approaches to change management specifically in drug–device combination products⁽¹⁰⁵⁾; the work within ISO commenced shortly thereafter. The finished ISO standard is expected to be available in a few years. In the meantime, OIP developers can rely on the more general US quality management systems regulations,⁽¹⁰⁶⁾ and ISO risk management and quality management standards for devices.^(107,108)

Key messages and recommendations from the IPAC-RS/ISAM 2015 workshop

In asthma and COPD, disease control remains suboptimal for many patients, despite the availability of effective treatments. This situation is due in part to poor inhaler use technique and poor patient adherence. Improving adherence and correct use is a multidimensional problem. Its solution requires new thinking on the part of the following:

• industry developing devices and IFU;

Educating and training each patient are important but not sufficient to achieve sustained adherence. Patients themselves differ widely in their motivation, attitudes to treatment, capacity for correct device handling, and for correct inhalation technique. In light of the strong and varied contribution of patient-related factors, there is no single “ideal” device that would be used perfectly by any and every patient. The availability of various device and drug types and the individualization of treatment choice by the physician in collaboration with the patient are therefore of paramount importance.

Training on correct use of devices should be provided on a regular basis and not just once, with the first prescription for an inhaler. Training should extend not only to patients but to physicians and other healthcare providers who interact with patients. Switching of a prescribed device at the point of sale should be discouraged unless the physician is consulted and the patient is retrained on the use of the new device. IFU should use efficient visual layouts and standardized unambiguous terminology that are readily understandable by the patients.

Technological innovations such as electronic monitoring apps could be used to improve training or adherence, but by themselves cannot guarantee adherence, and might even hinder compliance in some situations (e.g., some elderly patients may not be familiar or comfortable with new technologies). The implications of precision (personalized) medicine and patient input would need to be carefully considered, since the variability in motivation and ability to use different tools are likely to be great. Furthermore, the use of innovations such as apps raises new challenges related to regulatory requirements, healthcare delivery, cybersecurity, and privacy.

Device developers would do well to aim for devices that are intuitive to use, easy to activate and robust, have a dose counter, indicate correct inhalation/dose administration, and do not require patients to inhale strongly and rapidly or to perform too many maneuvers to get their therapy. OIP developers should study—and optimize—drug deposition and inhaler handling in target patient populations, paying special attention to children, the elderly, and patients with severely compromised respiratory function. In-vitro analytical tests for characterizing inhalers during development should reflect characteristics of target patient populations (e.g., with respect to flow rate and device resistance).

All of the above will require substantial commitments and sustained efforts by the United States, European, and other government agencies, medical professionals, and industry.