Regulatory Stipulations and Scientific Underpinnings for Inhaled Biologics for Local Action in the Respiratory Tract—Part II: A Characterization of Inhaled Biological Proteins ^†

Abstract

Following the discovery of therapeutic molecules and the identification of specific biological targets, preparation of regulatory dossiers entails extensive product development and characterization to support their safety, efficacy, and stability. We have examined the drug development and relevant regulatory considerations related to inhaled biological proteins in the accompanying article. This review focuses on the characterization of locally acting inhaled biological proteins. Drug product characterization is a regulatory requirement, and it ensures drug product safety, efficacy, stability, and usability by the target populations. Together, these two articles provide a comprehensive discussion based on our review and analysis of the available open literature. We have attempted to fill gaps and simulate discussion of challenges following sound scientific pathways. This approach has the prospect of addressing regulatory expectations leading to rapid solutions to unmet medical needs. The robustness of characterization strategies and the development of analytical methods used in the in vitro testing for the evaluation of drug product attributes is assured through application of the Design-of-Experiment (DOE) and Quality-by-Design (QBD) approaches. Drug product characterization entails a variety of in vitro studies evaluating drug products for purity and contamination, and determination of drug delivery by the intended route of administration. Measurement of the proportion of the labeled amount per dose and the form suitable for delivery to the intended target sites is central to this assessment. For respiratory Drug–Device combination products, the testing may vary with the product designs. However, determination of the single-dose content, delivered-dose uniformity, aerodynamic particle size distribution, and device robustness when used by the target populations is common to all combination products. Characterization of aerosol plumes is limited to inhalation aerosols that produce specific aerosol clouds upon actuation. The flow rate dependency of devices is also examined. Product characterization also includes safety-related product attributes such as degradation products and leachables. For inhaled biological proteins, safety-related in vitro testing includes additional testing to assure maintenance of the three-dimensional structural integrity and the sustained biological activity of the drug substance in the formulation, during aerosolization and upon deposition. This article discusses various tests employed for regulatory-compliant product characterization. In addition, the stability testing and handling of possible changes during product development and post-approval are discussed.

1. Introduction

Biological therapeutics have been approved for a variety of diseases, and the majority of approved biotherapeutics are proteins. Of these, there are only two currently marketed inhaled protein products, Pulmozyme^® (Genentech, San Francisco, CA, USA)—an aqueous solution formulation for nebulization of alpha Dornase for management of cystic fibrosis [1]—and the insulin dry powder inhaler Afrezza^® (MannKind Corporation, Danbury, CT, USA)—indicated for improvement in glycemic control in adult patients with diabetes mellitus [2]. Since 1993, Pulmozyme^® has been the only locally acting protein therapeutic approved for administration to the lung, despite approval of numerous small-molecule Drug–Device combination products approved for local delivery for treatment of pulmonary diseases and several parenterally administered proteins approved for allergy and pulmonary conditions, including Xolair^® (Genentech, San Francisco, CA, USA) [3], Nucala^® (GSK, Durham, UK) [4], Fasenra^® (AstraZeneca, Södertälje, Sweden) [5], Cinqair^® (Teva, Frazer, PA, USA) [6], and Dupixent^® (Regeneron, Tarrytown, NY, USA) [7]. With the exception of Pulmozyme^®, the protein biotherapeutics indicated for pulmonary conditions are approved for parenteral delivery, and they reach the site of action through the systemic circulation. Topical delivery to the lung epithelium can deliver drugs in quantities orders of magnitude smaller than those administered systemically to achieve the same effect with potentially reduced toxicity. In addition, absorption of drugs from the blood stream is likely to deposit a greater amount in the endothelium and less in the lung epithelium [8]. The pulmonary delivery of biologicals following inhalation delivers greater and durable exposure in the lungs while minimizing the systemic side effects [9], providing therapy at doses much less than are required for parenteral therapy [10], with rapid onset of action, low cost, and mucosal immunity [11].

The absence of certain drugs and a suitable modality for a given route of administration is usually due to the challenges in drug development and delivery and/or lack of specific regulatory guidelines. The development and approval of hundreds of biological proteins for a variety of diseases is a testament to the status of technology and the ability of the biotechnology industry to manufacture products which are indicated for parenteral administration for which regulatory expectations are also well laid out in the regulatory guidelines.

The development of respiratory drug products of biological origin for local administration is complicated by the complex (combination) nature of the drug products and the lack of guidance specifically for inhaled biologics. We have attempted to conduct an in-depth review to determine such complexities in the development of inhaled products and provide comprehensive treatises that may serve as a source of reference and stimulate discussion to bring into focus multidisciplinary endeavors to provide for the relevant unmet patient needs.

Preparation of drug products for market authorization includes development and characterization portfolios submitted in regulatory dossiers. We have examined both aspects and our views are presented in two review articles. The development and characterization of respiratory drug products for local action include evaluation of drug substances, formulations, and devices, as the product performance is influenced by interactions both between and among these entities (Figure 1).

Characterization of drug products is a regulatory requirement, and it is based on evaluation of the quality attributes that assure their safety and efficacy, as well as their regulatory compliance in terms of their robustness, stability, and efficient handling by the target patient populations. The basic element to support safety is purity, which requires control over the contamination (foreign materials) and unwanted products of manufacturing, as well as impurities resulting from degradation of drug substances. For proteins, purity also assures absence of undesirable physical (aggregation) or chemical modifications which induce toxic immune responses.

Efficacy is assured through delivery of the labeled dose(s) in the manner and form suitable for the route of administration over the approved shelf life, for which the potency and delivery characteristics constitute key considerations. For inhaled proteins, consistent delivery of the labeled doses with sustained biological activity along with the product-specific clinically relevant aerosol attributes throughout the product life determines efficacy. Regulatory compliance warrants fulfilling the expectations/recommendations in terms of robust characterization data in an analytical package to support the drug product safety, efficacy, and stability. Regulatory compliance also includes drug product characterization to monitor consistency in manufacturing to assure within-batch and batch-to-batch consistency in meeting all product-specific specifications at release and throughout the shelf life.

For biological products, “The Process is Product”, and changes in the manufacturing process can affect the biological molecule, impacting the product and its performance, safety, or efficacy [12]. Thus, manufacturers are required to perform process characterization to validate that manufacturing delivers products that are safe and effective. From a regulatory standpoint, process validation entails the collection and evaluation of data, from the process design stage through production, which establishes scientific evidence that a process is capable of consistently delivering quality products [13,14,15,16].

Process characterization strategies are designed to identify and control process parameters and critical quality attributes to ensure that processes deliver drugs with attributes determinant of safety and efficacy [17]. Validated processes are essential for successful technology transfer, which facilitates the transmission of scientific methods and manufacturing processes from development to production. Technology transfer for biological products is more complex than that for small-molecular-weight products. Process understanding and control begins when it is determined that a molecule has clinical significance [18]. The process validation of biological proteins is a multistep process [19] that requires careful consideration of the acceptance criteria, in-process specifications, in-process tests, in-process hold times, operating limits, and validation/evaluation [20].

This paper focuses on product characterization with consideration of the quality attributes and in vitro performance relevant to the safety, efficacy, and stability of inhaled biological proteins subsequent to the product development considerations discussed in Part I [21].

2. Quality Considerations

We have previously described the significance and application of the Quality-by-Design (QbD) methodology to the development of protein therapeutics [21]. However, application of the QbD principles to assure drug product safety and efficacy is relevant to all phases of the development and in vitro testing of the drug product performance [22,23]. It is also relevant to the development and validation of analytical methods used for in vitro testing to assure the safety and efficacy of drug products over their approved shelf lives [24,25,26]. Likewise, evaluation of the in vitro performance of respiratory drug products is crucial to ascertain the reproducible and targeted drug delivery determinant of the safety and efficacy of the drug product at batch release and over the entire labeled shelf life. Though the basic quality considerations for in vitro testing are applicable to all respiratory drug products, certain elements of QbD may vary between nebulizers, dry powder inhalers (DPIs), and soft mist inhalers (SMIs). Pressurized metered dose inhalers (pMDIs) are not included in this discussion because they are deemed unsuitable for delivery of biological proteins.

Elements of QbD to assure quality are based on Quality Target Product Profiles (QTPPs) that ascertain safety and efficacy, which constitute the fundamental attributes for establishing the clinical performance of all therapeutics. The QbD approach establishes a relationship between the product performance-related QTPPs and the corresponding critical quality attributes (CQAs) that are used for evaluation of the in vitro performance of drug products. For respiratory drug products, the readers would find useful the discussion in the public domain, including, but not limited to, the relevant regulatory documents [27,28,29] and elsewhere [30,31,32,33,34,35]. Our understanding of the various QTPPs and CQAs relevant to the in vitro performance testing of inhaled proteins is reflected in

Table 1. Elements of Quality-by-Design Approach Applicable to the In Vitro Performance Testing of Inhaled Biologics.

CLINICAL ATTRIBUTE	PERFORMANCE TESTING (IN VITRO)		NEBULIZERS	DRY POWDER INHALERS	SOFT MIST INHALERS
	QTPP	CQA
EFFICACY	REPRODUCIBLE TARGETED DRUG DELIVERY	DELIVERED DOSE (DD)	DELIVERED AMOUNT	SINGLE
				ACTUATION CONTENT (SAC)
		IMPACTOR SIZED MASS (ISM)	FRACTION OF DD DEPOSITED IN THE IMPACTOR FROM STAGE 1 TO FILTER
		FINE PARTICLE DOSE (FPD)	FRACTION OF DD < 5μ
		FINE PARTICLE FRACTION	FPD EXPRESSED AS PERCENT OF DD
		PULMONARY DELIVERY	TOTAL LUNG DEPOSITION AND WITHIN-LUNG DISTRIBUTION
		DELIVERY RATE	AMOUNT/MIN	NA
		NEBULIZATION TIME	YES	NA
		RESIDUAL VOLUME/AMOUNT	% SINGLE DOSE a	NA
		FORMULATION PSD	YES	NA
		APSD b	MMAD c, GSD d
		DRUG RELEASE/DISSOLUTION	FOR SUSPENSIONS	YES	NA
		BIOLOGICAL ACTIVITY	YES
		AGGREGATION/STRUCTURAL INTEGRITY	YES	YES	YES
		PLUME SPEED	NA		YES
		PLUME DURATION	NA		YES
	DEVICE PERFORMANCE	FLOW RATE, BREATHING PATTERNS, FLOW PROFILE, INSPIRATORY VOLUME	YES
		ROBUSTNESS	YES
		RESISTANCE	NA	YES
		PATIENT HANDLING	YES
		PATIENT IN-USE	YES
SAFETY	STABILITY	TESTING	DD, FPD, IMPURITIES/DEGRADANTS, LEACHABLES
			AGGREGATION, IMMUNOGENCITY
		BATCH RELEASE	YES
		SHELF LIFE	YES
		SHIPPING AND STORAGE	YES

^a Aerodynamic Particle Size Distribution. ^b Amount of Drug Remaining After Nebulization. ^c Mass Median Diameter. ^d Geometric Standard Deviation.

. The same QbD elements would be applicable to in vitro testing to support post-approval changes, where applicable.

The quality of in vitro performance testing relies heavily on the robustness and sensitivity of the analytical techniques that are used for the quantitative determination of CQAs. Evaluation of certain CQAs may require more than one analytical method. Advances in the instrumentation and techniques for CQA characterization provide greater opportunities for incorporation of QbD [36,37,38,39,40,41]. Modern method development and validation incorporate Analytical QbD (AQbD) [42,43]. This approach integrates principles of QbD into analytical procedures, aiming to ensure the quality and robustness of analytical methods. AQbD provides a systematic framework for method development that emphasizes understanding the relationship between critical process parameters (CPPs) and CQAs [44]. Regulatory guidelines also emphasize the use of QbD and DoE in method development and validation [45,46,47,48].

For registration and market authorization, drug products must be safe, effective, and stable over their approved shelf lives. These drug product attributes provide the basis for the selection of QTPPs and CQAs used for QbD-aligned in vitro performance testing. Thus, for inhalation products, the QTPPs included essentially fall into three main categories, i.e., reproducible targeted drug delivery, device performance, and stability. These QTPPs are achieved through testing for a variety of CQAs, outlined in Table 1.

CQAs can also be affected by critical material attributes (CMAs) and critical process parameters (CPPs), in addition to the other parameters used in testing. For example, the CMAs for DPIs may include drug substance properties such as particle size, surface properties, shape, and cohesiveness. The CMAs relevant to the widely used excipient lactose may include grade, size, shape, and surface texture. If the DPI uses capsules, the type of material (gelatin or HPMC), capsule dimensions, lubricant, and moisture content are monitored. The relevant CPPs include mixing/blending (speed, duration, and intensity), fill weight uniformity, and in vitro parameters such as flow rates. Inhalation airflow profiles may also influence both DD and APSD.

The impact of relevant variables is generally evaluated in a Design of Experiment (DoE) during method development and validation to establish “Design Spaces”, which are multidimensional combinations and interactions of input variables (CMAs and CPPs). They represent robust operational ranges and form the basis for establishing specifications that ensure product quality. The QbD approach also warrants a control strategy to assure reproducibility. Strategic planning for controlling the quality of in vitro testing includes controlling the input material through specifications for CMAs for raw materials, in-process control by monitoring CPPs during in vitro testing, verification during the finished product testing that all CQAs meet their target profiles, and lifecycle management by the continual refinement of methods based on the acquired data. The control strategy may also include patient considerations, such as optimal delivery at low (weak) airflow rates that can be achieved by compromised patients.

This section does not include discussion on the dissolution testing of aerosolized proteins, principally because most proteins are readily soluble. Nonetheless, it is relevant to mention that the lung absorption of inhaled compounds is dictated, and the local action is influenced, by the dissolution of the drug deposited in the fluids of the epithelial lining, the composition of which may differ within the lung [49]. Drugs that exhibit rapid dissolution are predisposed to ready absorption. Therefore, for locally acting drugs, development of formulations that exhibit prolonged dissolution is desirable [50]. Development of methods for testing the dissolution of aerosolized drugs has been deliberated [51], and the dissolution testing of inhaled drugs is an important consideration from the product development and regulatory viewpoints [29,52,53,54]. In reference to the development of dissolution methods for aerosolized drugs, the reader may benefit from the readily available literature [55,56,57]. It should also be noted that in addition to the intrinsic solubility of the drug substance, the presence of additives and the total amount of powder administered may influence the dissolution rate.

Like any method designed to support product development for regulatory submission, the development and validation of dissolution testing methods are also expected to follow a DoE approach based on application of QbD. For development of dissolution procedures, QbD establishes the target method profile, assesses risks associated with critical method parameters, defines analytical method design spaces, and develops the analytical method control strategy to establish a discriminating and robust dissolution method that reflects the drug product quality [58,59,60,61]. The QbD application also ascertains the robustness of the analytical methods developed to support dissolution testing [62,63,64].

3. In Vitro Drug Delivery Testing

Regulatory approval of most therapeutics mandates in vitro testing that characterizes the formulations and delivery systems to ascertain the amount of, and the form in which, the active is delivered to the target population. For inhaled therapeutics, the in vitro data necessary to support product approval includes determination of the delivered-dose uniformity and characterization of the aerosolized drug for its aerodynamic particle size distribution, indicative of the portion of the inhaled dose reaching the lung, as well as subfractions of the respirable dose that are determined within the lung distribution of the inhaled therapeutic. The nature of in vitro testing includes determination of the emitted dose (ED) and aerodynamic particle size distribution (APSD) applicable to all respiratory Drug–Device combinations, in addition to certain device-specific tests.

Data from in vitro studies are used to provide information for inclusion on the approved labels related to the dose recommended for the therapeutic use of the approved products. For nebulizers, in vitro data are submitted to support the labeled dose, reported as the volume of formulation and the concentration/amount of drug in a given respule/unit. Drug delivery is characterized in terms of the unit dose content of drug in the ampules, mean nebulization time, and mean delivered dose, comparative drug particle and agglomerate particle size distribution (for suspensions in ampoules), and nebulized aerosol. The recommended method for the latter is the aerodynamic APSD of the nebulized aerosol [65].

Drug delivery from nebulizers may vary with the methods used for measurement [66], the physicochemical properties of the formulations [67,68,69], and the breathing patterns representative of different target populations [70,71,72,73]. Indeed, the US FDA approval of nebulizers includes evaluation of the influence of breathing patterns on drug delivery [74] and inclusion of the observed effect on the approved product labels [75].

The US FDA’s key in vitro BE testing requirements applicable to all Drug–Device combination products are evaluations of the Single-Actuation Content (SAC) and APSD. In addition, the agency requires device resistance data and determination of the flow rate dependency of the SAC and APSD for DPIs and SMIs. Furthermore, comparative priming/repriming, spray pattern, and plume geometry data are required for pMDIs. Likewise, spray characterization for plume velocity and plume duration are included in the in vitro testing to support approval of SMIs.

Demonstration of equivalence of the in vitro performance data is necessary to support post-approval changes or approval of generic respiratory drug products. Documentation of equivalence is based on statistical analyses of data from three lots each of test and reference products. European regulators use the Average BE, whereas the Population BE methodology (PBE) is employed for documentation of in vitro equivalences in US applications. Criteria for acceptance of equivalence also differ between the two geographies. Compared with the EU acceptance criteria of 85–115%, the US FDA exercises tighter acceptance limits (90–111%). However, the PBE methodology used by the US FDA allows for scaling based on the relative variability of the test and reference products.

The two agencies also differ in their evaluations of the equivalence of APSD data. The APSD of OIDPs is determined by fractionation of the aerosolized dose using the Andersen Cascade Impactor or Next-Generation Impactors—both work on the same principle [76]. The aerosolized dose is distributed onto the impactor stages based on the cut-off diameter, which decreases from stages 1–7 [77]. The mass of the drug within the impactor is used for comparison of the APSDs between the test and RLDs. The US FDA recommends the “Impactor Size Mass (ISM)”, which represents drug deposition on stages 1 to filter, whereas the EU employs the fine particle mass (FPM), representing drug < 5 µM. While the US determines the in vitro BE based on PBE analyses of the ISM, the EU examines the drug deposition within the impactor by splitting into several groups, which may vary among drug products based on the clinical relevance of the groupings. For determination of the BE, all groupings must be within the recommended ±15%. Thus, it is possible that two products that have the same ISMs/FPMs but qualitatively different impactor deposition profiles may meet the US FDA criteria of the equivalence of the APSD but fail the EU criteria. Both agencies also review the Mass Median Aerodynamic Diameter (MMAD) and the associated Geometric Standard Deviation (GSD)—both derived from the impactor data.

The quality and relevance of the in vitro testing of the performance of inhaled drugs have evolved over time, with valuable advances in the relevant scientific disciplines and regulatory arenas. Though the standard compendial methods are still commonly used in batch release and stability testing, clinically relevant in vitro testing is now preferred to support product approval as well as demonstrations of equivalences [78], as it enhances the likelihood of achieving good in vitro–in vivo relationships [79,80] and provides valuable input for the in silico determination of the total lung deposition and within-lung distribution [81,82]. Clinically relevant in vitro testing uses anatomically relevant mouth throat (MT) models and employs flow rates representative of the target populations and/or disease severity [83], instead of using the compendial right-angle MT models and flow rates used in the conventional testing. Small, medium, and large MTs may represent MT variations among age groups, whereas the weak, medium, and strong flow rate profiles may be representative of mild, moderate, and severe levels of airway disease.

Submission of acceptable in vitro testing data is necessary for approval of new and generic drugs. In vitro data are relevant to the characterization of new drug products and are a mandatory component of the “Weight-of-Evidence” approach used by the US FDA for approval of generic respiratory drug products. Similarly, the European regulators require in vitro data to support approval of new and follow-on (hybrid/generic) respiratory products. However, for approval of generic inhalation products in the EU, bioequivalence may be documented based on comparative in vitro performance data without any clinical studies—whereas it constitutes one of the several elements of the “Weight-of-Evidence approach used by the US FDA.

Besides supporting product characterization and bioequivalence (where applicable) in drug application, in vitro tests are determinant of the product quality during manufacturing, at batch release, and over the approved shelf life. Quality is ascertained and sustained by enforcement of specifications for the applicable in vitro tests defined during product development and established upon the regulatory review of the release and stability data from commercial batches [84].

Of the two approved inhaled biologics, determination of the APSD of Pulmozyme^® is based on laser diffraction, as the drug product is formulated as an aqueous solution for nebulization, whereas the New-Generation Impactor (NGI) is used for characterization of insulin aerosol from Afrezza^® formulated as dry powder. The in vitro tests for characterization of nebulized aerosols have been discussed above. DPIs are characterized by the SAC and APSD. Though the same methods are used for determination of the in vitro performance of inhalation products made up of small molecules and biologics, extra caution is warranted for the latter with respect to the quality of the analytical methods and conduct of the experiments due to the labile nature of biologics. For both classes of products, the analytical methods are required to be stability-indicating, which means that the assays have the ability to account for the chemical degradation of the drug. However, for biologics, the analytical methods should also be able to determine the biological activity of the molecules and physical stability assessment in the form of monitoring for aggregate formation, in addition to chemical/biochemical degradation. Furthermore, during method development and validation, particular attention is necessary to monitor the biological molecular integrity and functional activity at various steps in aerosol generation, and the testing conditions and procedures [85,86,87,88,89], which can be achieved through case-by-case technical feasibility studies of drug-specific aerosolization and analytical methods [90,91,92,93] with capability of monitoring for biological activity.

The in vitro testing of inhaled biologics becomes more complicated, as the formulation of products for inhalation and the process of drug administration may cause, or add to, the immunogenicity potential of the drug. Complexity in ascertaining such potential in vitro is further enhanced by changes that may happen during aerosolization and is also affected by the aerosol sampling technique [87]. Furthermore, droplet deposition at 37 °C with 100% relative humidity and multidirectional air movements on the fluid-covered epithelial surface is principally governed by the combination of inertial impaction, gravitational sedimentation, and Brownian diffusion [94]. These conditions are different from current in vitro aerosol collection processes. Thus, analytical testing for protein aggregation should be designed to monitor/prevent the creation or reduction of protein aggregation during sample collection or handling. Furthermore, because administration involves the reconstitution of powder in aqueous media, assessment of protein aggregation post-reconstitution is required to determine the aggregation related to the reconstitution. Furthermore, it is important to consider that the reconstitution of dry powder for analytical testing may not be representative of the rehydration and dissolution of dry powder in the epithelial lining fluid with its unique, different chemical composition in the lungs. The in vivo aggregation may be differentiated from aggregation detected in vitro due to the powder disposition and dispersion in the limited volume of the lung fluid [95]. The risk of post-inhalation protein aggregation in the lungs may depend upon the inhaled dose, dosing frequency, and site of deposition [96]. Protein aggregates have been reported in preclinical studies [97,98], producing a dose-dependent induction of immune response [99].

Thus, it is advisable to evaluate the potential for protein aggregation in the lungs given the correlation between protein aggregation and immunogenicity from parenteral products [100]. However, caution should be exercised in selecting the in vitro testing conditions to mimic the in vivo conditions encountered by the protein post-administration. The alteration of proteins in in vitro dissolution studies for inhaled dry powders submerged in a large volume of media in a dissolution cup or in culture media of lung cells may not correlate with the in vivo disposition of the drug deposited in the lung [86]. Furthermore, the in vitro determination of protein aggregation is also affected by stress triggered by air–liquid interface stress [101]. This issue can potentially be addressed by using in vitro air–liquid interface (ALI) cell culture models [102,103,104].

In these models, the apical surface of cells is exposed to air, while the basal side is nourished by contact with the liquid cell culture medium [103]. ALI models allow for generation of data relevant to the respiratory tract [105], and the availability of such culture models for different inflammatory lung diseases should provide opportunity for better correlation to the in vivo environment. However, the complex structure of the lungs and the particle size-dependent regional deposition of the inhaled drug add a layer of complexity in determination of aggregation, which may vary with the deposited fraction of the inhaled dose., because the resulting concentration in the volume of lung fluid available for dissolution in different parts of the lungs may also vary.

In the absence of regulatory guidelines specifically tailored to aerosolized protein therapeutics, development considerations related to protein aggregation—including its formation during manufacturing, storage, and patient use, the importance of its measurement, and the applicability of established analytical approaches—have largely been extrapolated from regulatory frameworks and the extensive scientific knowledge base developed for parenterally administered protein products. However, inhalation delivery introduces distinct formulation-, device-, and use-related stresses, including aerosolization, exposure to air–liquid and solid–air interfaces, shear forces, and surface adsorption, which may fundamentally alter aggregation pathways and profiles. Accordingly, this work examines differences in regulatory considerations between protein therapeutics designed for parenteral administration and those intended for inhalation (see

Table 2. Comparison of Regulatory Consideration for Aggregation Testing of Parenteral and Inhaled Proteins.

Aspect	Route of Administration
	Parenteral	Inhalation
Size limits for particles/aggregates	Pharmacopeial limits for particulate matter apply (e.g., USP <788>: particles ≥ 10 μm and ≥ 25μm per container) to control safety risks such as embolism and inflammation.	No inhalation-specific numeric limits. Aggregate or particulate thresholds for inhaled biologics are not defined. Regulatory expectations are risk-based and product-specific.
Regulatory framework	ICH Q5C (stability), ICH Q6B (specifications), FDA/EMA biologics quality guidance; USP <788> for injectables.	General biologics guidance (ICH Q5C, Q6B) combined with inhalation product quality guidance (FDA/EMA); no dedicated aggregate limits for lungs.
Analytical expectations	Routine, validated testing for soluble and insoluble aggregates (e.g., SEC, DLS, MFI, light obscuration) as part of release and stability programs.	Comparable analytical tools expected, plus assessment before and after aerosolization/ to demonstrate molecular integrity through the device and process.
Stress (Device and process)	Minimal (direct injection); aggregation primarily driven by formulation, storage, and handling	Nebulization, spray drying, shear, air-liquid interfaces, and, device materials can induce unfolding and aggregation; regulators expect stress and compatibility studies.
Immunogenicity	Aggregates recognized as a key contributor to systemic immunogenicity; evaluated per FDA Immunogenicity Guidance.	Uncertainty regarding local lung and systemic immune responses; regulators may expect a risk-based justification linking aggregate data to local tolerance and systemic exposure.
Specification setting	Defined specifications justified against pharmacopeial limits and clinical experience.	Sponsor-defined specifications supported by scientific rationale, device performance data, and nonclinical/clinical risk assessment.
Regulatory maturity	Well established regulatory history and harmonized expectations.	Evolving—case-by-case -assessment with reliance on totality of evidence.

).

The foregoing discussion highlights the need to critically identify potential gaps in the in vivo relevance of currently employed in vitro testing approaches for protein therapeutics, particularly when extrapolated across different routes of administration. While in vitro assays are essential for characterizing product quality, stability, and aggregation behavior, their ability to reliably predict in vivo performance, safety, and clinical outcomes remains uncertain in several contexts. This is especially relevant for inhaled protein products, where unique physiological, aerodynamic, and immunological factors may influence the clinical response.

Table 3. Gaps in In Vivo Predictability of In Vitro Testing of Inhaled Biologic Proteins.

Test/Parameter	In Vitro Test	Objective	Predictability Gaps	Clinical Relevance
Aerodynamic Particle Size	Laser Diffraction, Cascade Impactor, Next Generation	Determine Emitted Dose and Aerodynamic Particle Size Distribution.	Utilizes Pharmacopeia listed flow rates. Does not account for biological activity and ignores patient variability (breathing patterns, airway geometry).	Limited clinical reference as the testing ignores patient variability (breathing patterns, airway geometry).
Pulmonary Deposition	Impactor	Determine Fine Particle Dose/Fraction.
Protein Stability & Aggregation	SEC, DLS, MFI, stress tests	Detect aggregates, fragments, chemical degradation.	Conducted in simplified buffer systems that may not reflect lung microenvironment.	Aggregates may behave differently in vivo, affecting immunogenicity and efficacy.
Dissolution/Solubility	Artificial lung fluid or saline dissolution	Predict release of protein in lungs.	Simplified media do not mimic surfactant composition, pH gradients, and mucus interactions.	Biopharmaceutics and bioavailability predictions may be inaccurate.
Cellular Uptake/Barrier Models	Calu-3, 16HBE, A549 and primary alveolar epithelial cells, Trans well monolayers	Study epithelial permeability, transport, or immune activation.	Overly simplified; lack alveolar macrophages, surfactant, dynamic airflow.	Limits prediction of systemic exposure, local clearance, and immune response.
Device–Protein Compatibility	Aerosolization tests, shear stress simulation	Assess protein integrity post-device.	Uses fixed flow rates and temperature; patient inhalation varies.	May misrepresent delivered dose, deposition, and protein integrity in humans.
Immunogenicity Screening	PBMC cytokine release, aggregation-induced response	Early prediction of immune response.	Lacks lung-specific immune cells and microenvironment.	Poor correlation with local lung immune activation or systemic immunogenicity risk.
Microbial/Particulate Safety	Filter-based particle counting, sterility tests	Detect sub-visible particles or contaminants.	In vitro conditions do not mimic particle clearance or local host defenses.	Particles may aggregate or be cleared differently in vivo; risk assessment may be inaccurate.

SEC: Size Exclusion Chromatography; DLS: Dynamic Light Scattering; MFI: Micro-Flow Imaging; PBMC: Peripheral Blood Mononuclear Cells.

summarizes the outcomes of our assessment, identifying potential predictability gaps in existing in vitro tests and evaluating their anticipated clinical relevance.

4. Specifications

Drug products require a combination of science and technology, where basic science related to applicable biology, chemistry, and other disciplines once established does not fail. However, technology applicable to processes from the manufacture of drug substances to product release is dynamic with a finite risk of failure. The risk is managed during product development, and specifications provide valuable tools to successfully control the risk. For the manufacturer, specifications are an important means of insuring that product design features and consumer expectations are consistently fulfilled, and that the marketed products remain safe and effective throughout the product life. For regulatory agencies, specifications represent the principal means by which they exercise their mandate of protecting and improving public health [106]. Manufacturer compliance with acceptance criteria for defined quality attributes applicable at the various steps between the qualification of the raw materials to the production-scale manufacturing of the finished products and their release and stability ensures their safety and efficacy.

From a regulatory standpoint, specifications are defined as the quality standards (i.e., tests, analytical procedures, and associated acceptance criteria) used to confirm the quality of drug substances, drug products, intermediates, raw materials, reagents, components, in-process materials, container closure systems, and other materials used in the production of drug substances and drug products [107]. Generally, the drug product specifications are based on data from multiple batches of the drug product used in critical clinical and primary stability studies. The associated analytical procedures for the test attributes need to be properly validated and documented [108,109]. These quality standards constitute a set of criteria that DSs and DPs should conform to during their manufacturing, release, and over their approved shelf lives [110].

Quality control by application of specifications is a regulatory mandate with specific testing that is “all or none” because DSs and DPs must pass all evaluations relevant to guarantee quality. Specifications are proposed by sponsors based on the relevant product development data and are finalized based on the regulatory review of drug applications. In some cases, limited-time data may be included at the time of the initial dossier submission, which, however, is updated during the evaluation of applications [111]. Specifications apply to DSs, excipients, DP manufacturing, batch release, and both accelerated and long-term stability testing. Release specifications determine the quality at release, and the acceptance criteria for drug products are set to control the performance at batch release and thereafter throughout the shelf life. Specifications for batch release are generally tighter than the criteria set for monitoring quality over the shelf life to accommodate changes ascertained and qualified by QbD during product development [112,113]. ICH Q6B provides an outline of testing procedures and acceptance criteria for biotechnological/biological products, and ICHM10 [114] enlists key performance characteristics for analytical method validation. Based on ICH Q6B, quality attributes relevant to DSs and DPs include appearance and description, identity, purity/impurities, and quantity, as general tests. Specific tests may include potency, DP-specific tests, and testing relevant to the dosage form. Of these, appearance is generally based on the visual inspection of the physical state of the formulation, color, and clarity of the solution. For protein therapeutics, changes in color may be indicative of degradation products, the presence of other impurities/contaminants, or aggregation [115,116,117]. Appearance of therapeutic proteins manufactured as solutions where aggregation occurs may also result in cloudiness. Drug products intended for parenteral administration are also inspected for visible particulate matter [118,119]. Appearance is the first among the various tests selected for quality control. Failure at this stage generally indicates product failure. Testing for identity is applicable to all drugs, which is both complex and specific to each molecule due to the primary and higher-order structures as well as molecular folding. Thus, from both the safety and efficacy viewpoints, testing for the sequence and folding of protein therapeutics is recommended [120].

Entities that are not chemically similar to raw materials or formulation components are described as impurities. They have no therapeutic benefit; instead, impurities may pose a risk of toxicity [121,122]. For protein therapeutics, impurities may represent size variants, charge variants, and post-translational modification/variants [123,124,125]. ICH Q6B categorizes substances related to the therapeutic molecule that are product-related variants of the active substance with no detrimental effects and product-related impurities whose therapeutic profiles are not comparable to those qualified for the desired therapeutic [117]. Impurities also originate from manufacturing processes. Regulatory authorities require monitoring of impurities that are formed during manufacturing and storage. Analytical methods for characterization of impurities must be stability-indicating to separate desired protein species from impurities, and the selected techniques should be sensitive enough for detection of trace impurities.

Product-related impurities also include variants of size, charge, and post-translational modifications (PTMs). Size variants may include low-molecular-weight species due to protein truncation by chemical degradation or proteolysis [126,127,128] and high-molecular-weight species that are the outcome of misfolding or certain PTMs [122,129,130]. The majority of the charge variants result from PTMs of therapeutic proteins, and this CQA can overlap with a PTM quality attribute. For the characterization of PTM variants by efficient chromatography, analyses based on modified peptides may be more informative than the characterization using more proteins or their subunits [131,132,133,134].

For biological proteins, the quantity relates to the amount of drug, and in most cases, the potency that is measured in terms of the biological activity. Based on ICH Q6B, quantity determinations are based on the protein content (mass) and should be determined using an appropriate assay. As required for all pharmaceuticals, analytical methods for determination of protein quantity must be specific, stability-indicative, and sensitive. The quantity of highly heterogeneous proteins is generally measured as the total protein concentration [135,136,137,138].

Potency reflects the quantitative determination of protein biologics based on the functional activity and/or extent of binding. The quantitative assessment of potency in units is achieved using a validated bioassay and based on ICH Q6B, and the validated potency assay should be part of the specifications. Bioassay methods should be specific, stability-indicating, and sensitive for capturing minor changes. If an appropriate potency assay is used for the drug product, an alternative method (physicochemical and/or biological) may be sufficient for determination of potency at the drug substance stage. A variety of in vitro and in vivo assays can be used to determine the potency as binding or functional activities [123]. The US Pharmacopeia includes a variety of tests relevant to potency determination, including, but not limited to, Design and analysis of biological assays <121> Insulin assays, <124> Erythropoietin bioidentity tests, <1030> Biological assay chapters—overview and glossary, <1032> Design and development of biological assays—(1033), Biological assay validation, <1034> Analysis of biological assays <1102> Immunological Test Methods—General Considerations, <1103>—ELISA, Surface Plasm resonance, and <1108> Assays to Evaluate Fragment Crystallizable (Fc)-Mediated Effector Function.

Biological assays encounter greater variability than the chemical tests employed for determination of the potency of pharmaceuticals. Therefore, establishing the quantitative reliability of the assay requires cognizance of the inherent variability due to the very biological nature of the DS [139,140]. Consistent with the regulatory requirements for all bioanalytical methods, validation/verification of biological assays is crucial to determining their suitability for the potency test [USP <1033>, <1225>, <1286>, ICH Q2(R1)]. Furthermore, if quantitation reliability is eclipsed by biological variation, physicochemical testing may be necessary to ascertain the higher-order structure correlation with biological activity.

In addition to the above-mentioned specific evaluations, certain tests, such as those for pH, microbial contaminants, total microbial counts, bacterial endotoxins, and osmolarity, are relevant to protein drug substances and drug products. Additionally, tests specific to drug products may include those for particulate matter, microbial limits, sterility, uniformity of dosage units, moisture content, and amount of excipient(s). Where applicable, pharmacopeial tests may be indicated for these quality attributes [141].

Specifications represent lists of acceptance criteria and the associated ranges for the tests identified as relevant to assuring the drug product quality. In QbD-based product development, specifications comprise a crucial element of the overall control strategy designed to maintain consistent quality over the various phases of product evaluation. For instance, in early clinical development, acceptance criteria furnish the sufficient quality control of DSs and drug products, whereas product-specific acceptance criteria are essential for subsequent clinical development [142,143].

Though the foregoing exposition on specification principally deals with the DSs and products that represent the mainstay of the currently marketed biological parenterals, it should be applicable to a wide range of products. However, inhalation products consisting of Drug–Device combinations require development of formulations, modes of drug delivery, and mechanisms of aerosolization that add to both complexity and additional stress on the drug products, requiring specific considerations for the drug substances, formulation, excipients, device selection, determination of parameters that control conversion of the formulation to aerosol, characteristics of the aerosols, lung deposition, and within-lung distribution.

Combination inhalation products are delivery systems containing substance(s) to furnish pharmacological action(s) or other direct physiological effects in humans. These products aerosolize defined amount(s) of substance(s) into aerosols that contain solid particles or liquid droplets for transport to the lower respiratory tract. The rate of drug delivery to the lung and its regional distribution are controlled by design to achieve specific deposition and temporal patterns [144]. Thus, a multitude of factors are considered in managing the risks that arise from drug, device, formulation, and formulation–device interactions and patient factors that determine drug delivery to the lung and the within-lung distribution. The significance of the latter attribute may vary with the drug and its intended target, mechanism of action, and DS/DP-specific biopharmaceutics. The appropriate distribution of the aerosolized active(s) between the central and peripheral lung is ascertained during product development. Acceptance criteria and specifications are established to control characteristics of the aerosolized drug and its pulmonary deposition and within lung distribution. Any change in these parameters would alter the pulmonary deposition and regional distribution. In turn, these outcomes would influence not only the amount of drug for therapy but also the drug quantity delivered to the lung region crucial for the therapeutic effect. Alteration in drug deposition between the central and peripheral regions may lead to changes in drug release and biopharmaceutic processes related to drug availability at the site of action, absorption, and elimination. Mechanical stress exerted during aerosolization may negatively affect susceptible molecules, distorting the defined three-dimensional structure in the formulation. Furthermore, higher-order structures that are often maintained by ionic and hydrophobic interactions can be perturbed when exposed to pharmaceutically relevant stresses, such as shear or thermal excursions. This can cause protein aggregation resulting in deactivation of the biological molecules that may lead to loss of biological activity and enhancement of immunogenicity [87].

No regulatory guidance has been published regarding target specifications for inhaled biologics. The US FDA 2018 Draft guidance discusses specifications and acceptance criteria for Drug–Device combination inhalers (MDIs and DPIs) using small molecules. It outlines specifications for drug substances, excipients, devices, and drug products per the ICH M4Q guidelines, along with lists and descriptions of the relevant quality attributes, drug delivery characteristics, container closure system quality, and product release testing. Additional discussion on QC test attributes and relevant development/characterization studies can be found elsewhere [84,145,146,147,148].

Drug development uses both short-term (accelerated) and long-term stability data. Regulatory authorities require submission of all stability data in product dossiers [113,149]. Accelerated stability data helps promote the pace of product development, as it can be used to predict long-term stability at an abbreviated time scale [150,151,152]. To establish the nature of the quality variation over time, specifications for accelerated and long-term stability determination are applicable to both the drug substance and drug product (ICH Q5C). The early-phase specifications are crucial to ensure that materials used in clinical studies are of the desired quality and pose no risk to the study volunteers or patients. Therefore, it is common in the pharmaceutical industry to develop platform approaches to support specification setting at the first in-human clinical trial in product development [153]. Of particular relevance in this regard are specifications related to impurities for which both concentrations and patient safety are key considerations [154].

Drug delivery from inhalation drug products is generally measured in terms of the labeled dose and APSD of the emitted aerosols. These are the key measures of drug delivery, and they are subject to the release and long-term specifications based on real-time data from the testing of exhibit/clinical batches of the to-be-marketed products to assure product safety and efficacy over the shelf life. The measures of labeled doses may vary with geographies. Thus, in the US, the product-labeled dose is expressed as the amount emitted ex-actuator (MDIs and SMIs) and upon inhalation (DPIs), which is different from the EU where, for example, the ex-valve amount may be the labeled dose. The APSD is a determinant of the fraction of the emitted dose that is delivered to the lung, as well as the within-lung distribution. Determination of the APSDs from nebulizers uses essentially the same techniques that are employed to determine the APSDs of emitted doses from MDIs, DPIs, and SMIs; thus, the scientific bases for setting the APSD specification are similar for all these delivery devices. However, aqueous aerosols are also evaluated for droplet size distribution by laser diffraction.

Nebulizer products are labeled for the amount of drug in the respules, and the delivered dose is controlled by the nebulization time, which may vary with the type of nebulizer. Thus, the specification is generally applicable to the amount in the respules, the nebulization time, which may be specified for drug products approved with dedicated nebulizers, and the fine particle mass or respirable fraction, if such data are included in the product label [75,155,156].

Specifications for measures of drug delivery are a regulatory requirement. Delivered drugs, related substances, and particle/droplet size distributions are determined at batch release and from stability samples taken over the shelf lives. Quantitation of drugs in such testing is based on stability-indicating assays, which detect degradation of the active(s), which, for small molecules, generally occurs in the drug product, with little possibility of degradation causing drug substance deactivation during aerosolization. However, protein therapeutics may undergo alteration during aerosolization, resulting in loss of biological activity. Thus, the analytical tests and methods for quantitation of protein therapeutics should be capable of monitoring molecule-specific biological activity, in addition to changes in protein structures and the presence of related substances [157]. Aggregates in therapeutic protein products as subvisible particles can promote immunogenicity, an important risk factor to consider when assessing product quality [158,159,160,161]. Thus, subvisible protein particles have the potential to negatively impact clinical performance to a similar or greater degree than other degradation products, such as soluble aggregates and chemically modified species that are evaluated and quantified as part of product characterization and quality assurance programs [162]. Furthermore, protein particles (visible and subvisible) can be generated from protein alone or from heterogeneous nucleation on foreign micro- and nanoparticles that are shed, for example, from filling pumps or product containers [163,164,165].

Subvisible particles are usually defined as particles that are too large for analysis by size exclusion chromatography (SEC) (e.g., ~ >0.1 μM) but too small to be visible to the unaided eye (e.g., <100 μM). Subvisible protein particles are thus relatively large assemblies (e.g., 0.1–10 μM) that contain thousands to millions of protein molecules. There are no guidelines for the quantitative analysis or reporting of subvisible particles in combination respiratory drug products. For liquid formulations intended for parenteral administration, the US FDA controls subvisible particles through United States Pharmacopeia (USP) <788>. The compendial limits for subvisible particles per 100 mL container are “No more than 6000 particles ≥10 µm, and No more than 600 particles ≥25 µm”. It also recommends monitoring particles ≥2 µM. Though applicable to all parenteral products, USP <788> was introduced for small-molecule therapies. Nonetheless, all protein therapies may contain some subvisible protein aggregate particles, with potential for toxicity/immunogenicity. Aggregates are often challenging to analyze via membrane microscopy due to their fragile, “soft” structure.

The methods in USP <788> may not be suitable for analysis of protein solutions that come in much smaller volumes. Thus, USP <787> provides an alternative test to USP <788> to analyze subvisible particles in protein therapeutics. It recommends the light obscuration method as in USP <788>. However, unlike USP <788>, it accommodates small volumes over 1 and 25 mL. Also, laser light obscuration is preferable for protein therapies due to the difficulties of processing protein aggregates with membrane microscopy.

The USP recommendation for parenteral formulations is intended to derisk vascular capillary occlusion in vivo through the infusion of foreign particulate matter, and it is not applicable for controlling the risk of in vivo protein aggregation or immunogenicity concerns. The inherent proteinaceous particulates are generally different from the traditional “foreign” particulates. Generally, they are amorphous and soft and have irregular morphologies, with a refractive index resulting in low contrast against an aqueous background, making them more difficult to detect and count compared to “foreign” particulates. Thus, for inhaled proteins, analytical methods that can assess particulate characteristics (including the composition, amount, and reversibility of the protein aggregate) are critical for developing scientifically sound approaches for evaluating and mitigating risks to product quality caused by large protein aggregates [166]. Regulatory considerations include Raman spectroscopy for characterization of subvisible particles [167].

Although ICH quality guidance Q6B states that the requirements set forth by pharmacopoeias pertaining to analytical procedures and acceptance criteria for particulate matter are applicable to biotechnological products, the risks associated with the administration of large, aggregated protein particles were not considered in the establishment of the USP light obscuration test <788>. However, ICH Q6B clearly states that specifications “should focus on those molecular characteristics found to be useful in ensuring the safety and efficacy of the product” and, logically, that recommendations should potentially be applicable to the control of large protein aggregates.

The EMA has not published specific guidelines for determination of subvisible particles in protein therapeutics. However, like the US FDA, the EMA also relies on the recommendation made in the European Pharmacopoeia (Ph. Eur.), which aligns with the methods published in the USP. For parenteral products, light obscuration and Raman spectroscopy are the preferred methods [168,169,170]. However, the EMA does recommend determination of visible and subvisible particles in its guidance for determination of the immunogenicity of therapeutic proteins [171], as does the US FDA [172].

5. Stability Testing

Documentation of the stability of drug products is a regulatory mandate worldwide. Stability is indicative of the quality of being stable, whereas instability may represent chemical and physical aberrations, microbiological contaminations, loss of therapeutic potential, and toxicological manifestations [173,174]. Stability studies provide evidence that the quality of a drug product remains acceptable during clinical testing and the stipulated shelf life [175,176]. Thus, stability studies are considered an essential element of product development, and they are typically conducted under long-term storage, accelerated, and stressed conditions to determine the product shelf life. Stability study results are also used to support evaluation of shipping and handling excursions [177].

Regulatory guidelines for determination of the stability of pharmaceutical drug substances and drug products have been well established [178,179]. While the EMA guidelines indicate its applicability to the EU, the scope of the US FDA guidance is tripartite, as it defines that a “stability data package for a new drug substance or drug product is sufficient for a registration application within the three regions of the European Union (EU), Japan, and the United States”. The US FDA guidelines indicate of the document that “This guidance was developed within the Expert Working Group (Quality) of the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) and has been subject to consultation by the regulatory parties, in accordance with the ICH process. This document was endorsed by the ICH Steering Committee at Step 4 of the ICH process, February 2003. At Step 4 of the process, the final draft is recommended for adoption to the regulatory bodies of the European Union, Japan, and the United States.” Furthermore, the recommended specifications relevant to analytical procedures and proposed acceptance criteria are addressed in the relevant ICH guidelines [116,180,181].

Though the US FDA guidance includes certain recommendations, such as conditions for storage under refrigeration, that are also applicable to biologics, stability testing for biological drug substances and drug products warrants special considerations owing to the unique characteristics of biologic drugs which are relevant to instability, which can be an issue in the process, storage, and delivery of any therapeutic product [182]. Deviations in structural integrity have ramifications for biological activity. The pharmacokinetics may be affected even if the therapeutic activity is retained [183], and instability can be implicated in immunological reactions [184]. Furthermore, biologic drugs are highly complex molecules and are more susceptible to environmental factors than small molecules. In comparison with small molecules, biologics are more heat- and light-sensitive, tend to denature at surfaces, are subject to enzymatic degradation, and aggregate under various conditions to form immunoreactive species (a potential safety issue). Therefore, stability testing studies are performed to evaluate biologics under a wider range of environmental conditions.

Manifestation of the instability of biologics may be physical, thermal, mechanical, chemical, and immunologic. The impacts of these processes are not mutually exclusive; the outcome may result from the interaction of more than one factor. Changes in temperature during manufacturing, storage, and shipping can also contribute to protein denaturation. Therefore, selecting applicable temperature and humidity conditions is an important consideration because exposure to both low and elevated temperatures can influence stability. Denaturation can occur at low temperatures when formulations are stored frozen or lyophilized. There is the risk of ice crystallization at very low temperatures and denaturation leading to aggregation at high temperatures. Elevated temperatures may accelerate chemical reactions which lead to aggregation. Therefore, selecting applicable temperature and humidity conditions is an important consideration. Aggregation also results from mechanical processes which induce agitation stress.

The identification of stressors that compromise the stability of biologic proteins, the characterization of the resulting degradation mechanisms, and the informed selection of analytical methodologies are well recognized as critical elements in the development of parenteral protein therapeutics. Extending this framework to inhaled biologic proteins,

Table 4. Stress Conditions, Degradation Mechanisms, and Analytical Readouts Specifically for Inhaled Protein Therapeutics.

Stress Condition	Degradation Mechanism	Analytical Readout
Aerosolization/Shear	Aggregation, particle formation, unfolding	Size-exclusion chromatography (SEC), Dynamic light scattering (DLS), Microscopic imaging (MFI), FTIR
Air–Liquid/Solid–Air Interfaces	Surface-induced aggregation, adsorption	Surface plasmon resonance (SPR), SEC, MFI, Quartz crystal microbalance (QCM)
Temperature (Refrigerated/Accelerated/Stress)	Denaturation, chemical modifications (deamidation, oxidation)	Circular dichroism (CD), Mass spectroscopy (MS), HPLC, ELISA potency
Humidity/Residual Moisture (for lyophilized powders)	Hydrolysis, crystallization, aggregation	Karl Fischer titration, DSC, SEC, Visual inspection
Light/Photo stress	Photo-oxidation, degradation of sensitive residues	UV–Vis spectroscopy, HPLC, MS, Fluorescence spectroscopy
Agitation/Vibration (during filling, shipping, or device actuation)	Aggregation, fragmentation, cavitation	SEC, DLS, MFI, SDS-PAGE

presents a structured overview linking stress conditions, degradation pathways, and analytical readouts while highlighting the unique physicochemical and mechanical challenges inherent to aerosol delivery. Inhaled proteins encounter stresses such as aerosolization, air–liquid and solid–air interfaces, thermal fluctuations, humidity, photo stress, and mechanical agitation, each of which can promote aggregation, unfolding, chemical modification, or surface adsorption. Analytical approaches including SEC, DLS, MFI, CD, HPLC, mass spectrometry, and potency assays are critical for the sensitive and quantitative assessment of these degradation events. Collectively, this framework establishes a robust scientific basis for systematic stability evaluation, informed formulation and device design, and the assurance of quality, safety, and efficacy in inhaled protein therapeutics.

Considerations for the stability testing of biological proteins [185] are based on ICH Q5C. The stability determinations consider the quality combination of physicochemical as well as biological testing, together with the production process and its control, as biological drug substances are deemed complex based on the primary, secondary, tertiary, and quaternary structural lability during manufacturing, shipping, storage, and usage. The need for the special consideration of stability testing for biologics is to impart maintenance of biological activity dependent on non-covalent and covalent interactions, as well as sensitivity to environmental factors, including temperature, oxidation, light, ionic content, and sheer [186]. Thus, the stability testing of biologic products includes physicochemical characterization, biochemical evaluations, immunological assessments, and determination of biological activity.

The regulatory requirements include the type of stability studies to be provided in support of marketing authorization applications for various aspects of biological medicinal products. Documentation of the stability of drug substances requires at least three batches representative of the manufacturing scale of production representative of the quality of batches used in the preclinical and clinical studies, manufacturing process, and storage conditions and containers. Data from pilot plant-scale batches may be submitted—if the pilot-scale batches are produced and stored in conditions representative of the commercial scale and use the same container/closure system—with the commitment to place the first three manufacturing-scale commercial batches in a stability program after approval. The applicants must provide stability data for a minimum of 6 months at the time of submission; applications providing data for less than 6 months may be evaluated on a case-by-case basis. Similarly, for the drug products, at least three batches of the final container product representative of the manufacture scale are required. The drug product batches should be prepared from different batches of drug substances and excipients (where applicable).

Establishment of stability requires analytical determinations based on “Stability Indicating” assays. These assays are drug- and product-specific and validated to detect any changes in purity, identity, and potency. Assays for potency are based on the quantitative measurement of attributes reflective of the clinical effect. In potency determination, the functional similarity of the reference standard and sample is a fundamental consideration in assay validation. The assay system should affect the test and reference preparations to the same extent.

Information on purity and molecular characterization is an integral part of stability data in the dossiers. Determination of purity generally requires more than one method. It includes determination of degradation products, and acceptable degradation should be derived from the analytical profiles of batches of the drug substance used in preclinical and clinical studies. The methods should allow for the comprehensive characterization determinant of molecular size, charge, hydrophobicity, and degradation, including deamidation, oxidation, sulfoxidation, aggregation, and fragmentation during storage.

Shelf life is established using real-time temperature and humidity data. Testing for the effect of light (photosensitivity) may be on a case-by-case basis. Determination of shelf life is based on the applicable long-term data. However, accelerated studies can provide information on post-development changes, validate stability-indicating tests, and generate the degradation profile. Accelerated testing is usually conducted at a level higher than real storage conditions (

Table 5. Stability testing conditions for biologics ^a.

TESTING CONDITION
Accelerated	Long Term	Stress
+5 ± 3 °C and/or +25 ± 2 °C/60% RH	≤−20 ± 5 °C	Temperature, pH, light, oxidation, shaking, freeze–thaw, device robustness…
+25 ± 2 °C/60% RH	+5 ± 3 °C
+40 ± 2 °C/75% RH	+25 ± 2 °C/60% RH or +30 ± 2 °C/65% RH

^a Source: Jimenez AG, Brake B (2011) [187].

).

Stability testing conditions for biological protein drug products also warrant consideration of the possible influence of formulation-dependent degradation pathways between lyophilized powders and liquid formulations. Lyophilized proteins generally show improved stability and are evaluated under long-term (2–8 °C), accelerated (25 °C/60% RH), and stress conditions, with particular focus on the residual moisture, glass transition temperature, cake appearance, and reconstitution behavior, as outlined in the ICH Q5C and ICH Q1A(R2) guidelines [186,188]. In contrast, liquid formulations require tighter control under refrigerated and accelerated conditions, with extensive monitoring of the aggregation, oxidation, deamidation, subvisible particles, and microbial integrity, reflecting their higher physicochemical vulnerability, as in the EMA stability guideline [189]. Photostability and agitation stress testing are especially critical for liquid proteins, consistent with WHO [190] and EMA regulatory expectations for biological products.

The regulatory agencies recommend a “Full study design” with samples for every combination, where all factors included in the design of the stability program are evaluated at all time points. A reduced study design [191,192] may be considered if supported by scientific justification. However, it may introduce the risk of establishing a shorter shelf life. Therefore, shelf life is based on data representative of real time and conditions. Data can be provided during the review and evaluation process. However, it should be representative of the quality of batches used in the preclinical and clinical studies and the manufacturing process and storage conditions, and with the final containers, devices, or container closure systems. Recommendations for long-term stability are summarized in

Table 6. Long-term stability testing intervals ^a.

Shelf Life	Testing Frequency	Time Points
≤One Year	Every month for up to 3 months then at 3-month intervals	1, 2, 3 6, 9, and 12
≥One Year	Every 3 months during the first year, then every 6 months	0, 3, 6, 9, 12, 18, 24, 36, 48…

^a Source: Jimenez AG, Brake B (2011) [187].

.

Stability data to support the claimed shelf life should include all conditions affecting the potency, purity, and quality. Primary data to support the requested shelf life should be based on long-term, real-time, and real-condition stability studies. The design of the long-term stability program is critical. For pharmaceutical drug products, regulatory agencies acknowledge requests for extension of the shelf life beyond the approved shelf life. The US FDA guidance [193], which is endorsed by the ICH Committee and is also applicable in Europe and Japan, provides the agency recommendations on how to use the available stability data to support the extension of shelf life beyond the period covered by available data from the stability study under the long-term storage condition (long-term data). Extrapolation to extend the retest period or shelf life beyond the period covered by long-term data can be proposed in the application if no significant change is observed under the accelerated condition. An extrapolation of stability data assumes that the same change pattern will continue to apply beyond the period covered by long-term data. The correctness of the assumed change pattern is critical when extrapolation is considered. A retest period or shelf life granted based on extrapolation is always verified by additional long-term stability data as soon as these data become available. The maximum shelf life after the extension may not be more than double, or more than twelve months longer than the period covered by the real-time stability data obtained with representative batch(es).

The consideration of the impact of the formulation format stated above is also applicable to the long-term stability evaluation of lyophilized and liquid formulations. Lyophilized products are typically stored at 2–8 °C, or at controlled room temperature where justified, with stability programs focused on preserving the solid-state matrix, including the residual moisture content, glass transition temperature, cake appearance, reconstitution time, and maintenance of physicochemical and biological activity after reconstitution, as described in the ICH Q5C and EMA guidelines [186,189]. In contrast, liquid protein formulations usually require long-term storage at 2–8 °C, with stability assessments emphasizing aggregation, oxidation, deamidation, subvisible particles, potency, and container–closure interactions, reflecting their higher susceptibility to temperature- and interface-driven degradation [186,189,190].

Where extensions of the shelf life are planned, the applicant should commit to performing the proposed stability program according to the executed protocol(s). As is true for all situations warranting deviation from the established norms, applicants planning to obtain shelf-life extensions should seek the agency’s advice based on all available data, including computational studies. Though extrapolation of the available stability data using linear regression for biologics can be performed to provide some estimation of the anticipated shelf life, the statistical approach described in ICH Q1E [194] to evaluate batch data for poolability and support extrapolation to extend a retest date or shelf life is not recognized as suitable for supporting extrapolations of shelf lives for biological products [195]. Furthermore, extension of the shelf life beyond the intended duration of the long-term stability studies is not acceptable for biologics [196].

Stability data requirements for regulatory submissions may also vary with product development, but this aspect is not elaborated herein because this review does not cover safety and efficacy evaluations. Briefly, however, for US applications, limited stability data may be sufficient for the INDs for initiation of Phase I clinical investigations [197]; additional information would be required to support the INDs for Phase 2 and 3 clinical studies [198]. In addition, during product development, the phase-appropriate approach for GMP considerations is also recommended [199]. In Europe, Regulation 536/20147 provides the legislation related to investigational medicinal products [200]. The EMA provides guidelines for documentation of the quality of biological investigational medicinal products in clinical trials [196]. Like the US FDA guidance, the EMA guideline provides recommendations regarding stability protocols covering the proposed storage period of the product, including specifications; analytical methods and test intervals; and quality of the batches of the product placed into the stability program—with inclusion of stability data for at least one batch prepared by a process representative of that used to manufacture material for use in the clinical trial. Supportive stability data on relevant development batches or batches manufactured using previous manufacturing processes should be provided, if available, along with application of progressive requirements reflecting the amount of available data and emerging knowledge about the stability of the product during the different phases of clinical development.

6. Alterations During Product Development and Post-Approval

Drug product manufacturers frequently make changes to therapeutic products both pre- and post-approval for a variety of reasons, including, but not limited to, scaling up, improving product quality, and enhancing manufacturing efficiency and cost-effectiveness, facilitated by scientific/technological advancements, or necessitated by changes in regulatory considerations. Implementation of any change requires evidence to show that the change does not adversely affect the safety or efficacy of the product, and an evaluation to indicate whether confirmatory nonclinical or clinical studies are necessary to support the change. From a regulatory standpoint, this exercise entails a comparison between the post-change product and pre-change products, and determination of the significance of the impact of the development or manufacturing change on the quality attributes relevant to the safety and efficacy of the product.

Comparisons between the pre- and post-change products are warranted even if the changes occur in the manufacture of the drug substance alone or the starting materials and intermediates used in the manufacturing of the active(s). Studies to support regulatory acceptance may vary with the quality of the evidence. Acceptability of the change may be granted from comprehensive quality studies alone or may need to be supported by bridging studies determinant of comparability. Assurance of comparability through analytical studies alone may obviate the need for comparative nonclinical or clinical studies with the presence of an acceptable relationship between specific quality attributes and safety and efficacy. Essentials in supporting changes in drug substances/products include the appropriate quality (specificity and sensitivity) of analytical techniques and characterizations determinant of the physicochemical properties, quantity, purity, impurities, contaminants, biological activity, and immunochemical properties [110], and their impact on the specifications and stability evaluated under appropriate conditions [201,202]. However, if the analytical procedures are insufficient to detect differences relevant to the safety and efficacy of the product, it becomes necessary to include more elaborate quality testing, and perhaps nonclinical and/or clinical studies to support unequivocal comparability. The regulators might also solicit the available historical data that provides insights into potential change the quality attributes following any change(s). Thus, it is important to establish the overall impact of changes over time on the efficacy and safety of the product.

Besides making alterations in the drug substance(s) and product formulations, the sponsors may implement changes in the manufacturing process. The level of effort to support change in the manufacturing process may vary with the product, the process, the extent of the available development data, and the manufacturer’s knowledge/experience with the process. The multifactorial nature of the manufacturing process adds complexity to assessing the impact of change in one or more processes because it warrants consideration of the potential effects of the planned change(s) on steps downstream and the quality parameters related to these steps, such as acceptance criteria or specifications for in-process evaluations, tests, hold times, and operating limits.

The extent of reevaluation generally depends upon the complexity of the change. A simple change might require evaluation limited to the affected process step if the performance of the downstream process steps or the quality of the resulting intermediates is not affected. However, the manufacturing change that influences more than one step would need to be supported by more extensive analyses/studies.

For management of product updates and regulatory approval, the regulatory agencies periodically publish their expectations for the evaluation and reporting of changes to approved products. The US FDA provides guidance for management of CMC, manufacturing, and control changes in certain biological products with considerations for assessment and implementation by the manufacturers in preparation of application, as well as its regulatory deliberations in their evaluation [203,204]. The assessment and implementation of product change(s) by the sponsors warrants multidisciplinary coordination to determine the potential for a change to influence the product quality relevant to the efficacy or safety. Thus, in addition to the case-specific combination of testing, validation studies, and nonclinical or clinical studies, if necessary [21 CFR 601.12(f)(3)], the data to demonstrate comparability of the product pre- and post-change may also include historic information obtained over the product development and its commercial manufacturing regarding applicability to the product updating of the relationship between the material attributes and product parameters.

Consistent with the established format, the US FDA recommendations for submission changes [205,206,207,208], the regulatory pathways depending upon the ascending order of the complexity of the change and the extent of the information required to support approval, include the Prior Approval Supplement (PAS), or Changes Being Effected in 30 Days/Changes Being Effected Supplements (CBE30/CBE), or in annual reports [21 CFR 601.12(a)(2)]. The regulation also provides for the expedited review of the PAS in response to unexpected circumstances causing interruption of the product supply or product shortages [21 CFR 601.12(b)(4)]. Both the US FDA [209] and EMA [210] also recommend taking into consideration recommendations in the ICH Q12 guidance that pertain to Risk-Based Reporting Categories and Post-Approval Change Management Protocols [211].

Changes in formulation or device may happen during product development or post-approval. The alterations may happen in the drug substance, formulation components and manufacturing, or material used to construct and manufacture the device(s) [212]. Furthermore, the impact of even a change in single components of the product may be complex due to the concerted functioning of the integral components necessary for the sustained safety and efficacy of the product. Additionally, changes in the components may occur at various levels/steps in the supply chain, which may determine the impact of the change on the drug product component, product performance, and responsibility of reporting such changes by the sponsor in application or by holders of the Drug Master File (DMF), where applicable

Although, hitherto, there has been no regulatory guidance to address developmental or post-approval changes to respiratory drug products, a recent review provides information that may be useful in planning to address product updates that may be introduced by change(s) in single or multiple components of the products [213]. The scientific and regulatory considerations in this article should also be applicable to inhaled proteins, with the exception of complexities due to the biologic nature of the drug substances as compared to small molecules. Based on the foregoing, the biological activity of a protein therapeutic may be affected by any changes(s) in the manufacturing of the drug substance and product formulation and the aerosolization. Thus, drug substance-specific characterization and testing and in vitro and in vivo studies may be required to establish the comparability of pre- and post-change products with respect to the integrity of the active(s) in terms of the quantity (assay) and biological activity of the active(s) over the shelf life of the product. The information discussed herein may serve as a source in planning for investigations to support the regulatory approval of changes. However, such plans should be concluded only after endorsement(s) from the relevant regulatory agencies.

7. Regulatory Gaps

Regulatory oversight for the characterization and release testing of biological protein therapeutics has evolved substantially over the past three decades. However, it exhibits scientific and policy gaps that become increasingly evident as product complexity grows and novel delivery routes (such as inhalation) are pursued alongside established parenteral administration. Biological proteins are intrinsically heterogeneous macromolecules. The safety, efficacy, and clinical performance of these complex molecules depend on a complex interplay of primary sequences, higher-order structures, post-translational modifications, and aggregation states—all influencing the functional integrity. Regulatory guidance such as ICH Q6B [110] has established a foundational, risk-based approach for defining specifications based on physicochemical properties, biological activity, purity, and safety. The significance of this document remains limited, as it does not provide prescriptive direction on the depth, choice, validation, or interpretation of modern analytical methods required to adequately characterize these attributes in a clinically meaningful way. As a result, significant gaps persist between the analytical capabilities now routinely available in industry and academia and the clarity of regulatory expectations for how these data should be generated, justified, and translated into release specifications and lifecycle controls.

One of the most noticeable gaps relates to the characterization of higher-order structures, which is critical to protein function, stability, and immunogenicity and yet remains challenging to define quantitatively. While regulators expect the use of orthogonal biophysical techniques, there is no harmonized regulatory guidance on the acceptable variability, sensitivity requirements, or quantitative acceptance criteria for these methods, nor on how deviations detected by highly sensitive techniques should be contextualized in terms of clinical relevance [113,213]. Such lack of clarity complicates both initial development and approval, as well as post-approval comparability assessments—particularly as analytical technologies evolve and reveal differences that were previously undetectable. Similar gaps exist for post-translational modifications that can influence pharmacokinetics, effector function, stability, and immunogenicity. Although regulators expect detailed molecular/particle profiling and control, the current guidance does not define standardized thresholds for acceptable microheterogeneity or site-specific variability, leaving sponsors to justify limits on a case-by-case basis.

Protein aggregation represents another area where regulatory expectations are conceptually clear but operationally ambiguous. Aggregates are recognized as critical quality attributes due to their potential impact on safety (immunogenicity) and efficacy, yet there is no globally harmonized definition of acceptable aggregate species, size ranges, or levels, nor consensus on which analytical techniques should be primary for release versus characterization, especially given the differing sensitivities of the size exclusion chromatography, light scattering, analytical ultracentrifugation, and particle imaging methods. These challenges are further compounded at the release testing stage, where specifications must be sufficiently tight to ensure consistent product quality while remaining accommodative of analytical and manufacturing variability.

Potency testing is also a particular area of regulatory uncertainty, as biological activity is often measured using complex, variable cell-based assays that are sensitive to subtle changes in reagents, cell lines, and assay conditions. While regulators require validated assays and statistically justified acceptance criteria, there is limited guidance on how to establish clinically relevant potency ranges, manage assay drift over time, or interpret discrepancies between different bioassay formats, such as binding versus functional assays [214]. For parenteral products, these challenges are mitigated by decades of regulatory experience and precedents. However, they remain a source of uncertainty during lifecycle changes and comparability assessments, particularly when different regulatory agencies may place differing emphases on specific assays or attributes.

The regulatory gaps become more pronounced and consequential when biological proteins are formulated for inhaled delivery, a route that introduces additional layers of complexity not fully addressed by existing guidance documents, which were largely developed for small-molecule inhalation products. Inhaled biological proteins must meet traditional protein quality requirements and also satisfy device- and aerosol-related performance criteria, including delivered-dose uniformity, aerodynamic particle size distribution, and device robustness. However, there is limited regulatory direction as to how these aerosol metrics should be integrated with protein-specific critical quality attributes in a unified control strategy [28,29]. Importantly, aerosolization processes such as nebulization, soft mist generation, and dry powder dispersion expose proteins to shear stress, air–liquid interfaces, dehydration–rehydration cycles, and contact with device materials—all of which can induce aggregation, denaturation, or loss of biological activity. The current guidance does not clearly define expectations for characterizing protein integrity post-aerosolization, for establishing release specifications that account for aerosol-induced changes, or for linking in vitro aerosol stability data to in vivo pulmonary safety and efficacy. Furthermore, while inhalation guidelines specify aerodynamic performance tests, they provide little protein-specific perspective on the acceptable limits, test sensitivity, or clinical relevance, creating uncertainty around how much variability is tolerable and how changes in formulation or device components should be managed from a regulatory standpoint.

Viral safety and adventitious agent control represent another evolving area where regulatory expectations have not yet fully caught up with technological advances. Although ICH Q5A(R2) [215] reflects progress in recognizing modern detection methods, there remains a lack of harmonized standards for the validation, data interpretation, and routine application of these tools in characterization and release testing, particularly for inhaled products where additional exposure routes may raise distinct safety considerations.

Across both parenteral and inhaled biological proteins, the implementation of QbD principles highlights further regulatory gaps. Though sponsors are encouraged to define CQAs, link them to critical process parameters, and establish design spaces, regulators may vary in how they assess and accept QbD justifications. There is limited guidance on how to integrate advanced analytical data into regulatory decision making in a consistent and transparent manner (ICH Q8–Q11). For inhaled biologics in particular, constructing a robust QbD framework that connects manufacturing variability, formulation properties, device performance, aerosol characteristics, and clinical outcomes remains challenging due to the scarcity of regulatory precedents and the absence of clinically anchored acceptance criteria. Furthermore, different regional expectations—despite reliance on common ICH principles—can lead to divergent requirements for characterization depths, release specifications, and post-approval change management, increasing the development burden and potentially discouraging innovation in complex or nontraditional delivery approaches.

Characterization of subvisible particles is a critical component of the product development process for inhaled biological protein formulations, as SVPs can influence both safety and clinical performance. During product development, SVPs may originate from upstream manufacturing variability, formulation stresses, container–closure interactions, and stresses unique to aerosol delivery, including shear forces, air–liquid interfaces, and dehydration–rehydration cycles during nebulization or device actuation [110,172]. Effective SVP assessment informs formulation design, excipient selection, and device compatibility to minimize aggregation and preserve protein stability during storage and administration. Regulatory frameworks for parenteral products, such as pharmacopeial limits (USP <788>), provide a reference for acceptable particle levels, but these thresholds are not directly translatable to inhaled formulations, where the particle size, morphology, and composition affect the regional lung deposition, local immunogenicity, and tolerability [29]. Analytical characterization relies on complementary, orthogonal methods—including light obscuration, flow imaging microscopy, resonant mass measurement, and nanoparticle tracking analysis—to determine size distribution and morphology, and to distinguish proteinaceous particles from extrinsic contaminants [162]. Incorporating SVP data into product development enables risk-based specification setting, guides device and formulation selection, supports comparability assessments, and informs clinical strategy. Despite its importance, regulatory guidance specific to inhaled biological proteins is limited, highlighting the need for integrated approaches that link SVP characterization with aerosol performance, protein stability, and immunogenicity to support robust product development and regulatory submission strategies.

Collectively, these regulatory gaps underscore a persistent disconnect between the scientific sophistication of modern protein analytics and the clarity of regulatory guidance, particularly as applied to release testing and inhaled biological proteins, where traditional parenteral paradigms are often insufficient. Addressing these gaps will require updated, modality-specific, and globally harmonized guidance that defines expectations for advanced analytical methods, establishes clinically meaningful acceptance criteria for key quality attributes, and integrates the aerosol and device performance with protein quality considerations, thereby enabling more predictable development, robust quality assurance, and timely patient access to safe and effective biological protein therapeutics.

8. Conclusions

Dossiers submitted to support the market authorization of drug products from regulatory agencies contain development and characterization data determinant of safety and efficacy. The forgoing treatise describes product characterization to support the safety, efficacy, stability, and usability of and (where applicable) post-approval changes in inhaled biological proteins. The robust characterization of the finished products based on in vitro tests and patient use studies determines the reliability of the drug products for their intended use. Regulatory authorities recommend designing characterization studies as well as the development and validation of the applicable analytical methods based on quality-assuring approaches, DOE, and QBD. Inhaled protein products contain drug formulations delivered by dedicated devices. For respiratory Drug–Device combination products, characterization studies may vary with the product designs. Nonetheless, determination of the single-dose content, delivered-dose uniformity, aerodynamic particle size distribution, and device robustness is common to all combination products. In addition, characterization of aerosol plumes may be required for products that emit doses in the form of plumes (e.g., SMIs). Product characterization also extends to safety-related testing, such as determination of degradation products and leachables. However, due to the fragile nature of protein molecules, safety testing for inhaled biological proteins includes the in vitro testing of the three-dimensional structural integrity and potency (biological activity) of the drug substance through development, manufacturing, handling, and patient use. Product characterization is submitted for fulfillment of a major part of the quality requirements. For inhaled biological proteins, all of the above is necessary to support product approval. The regulatory authorities require that each characterization study is based on the currently applicable quality standards while assuring the data integrity and robustness of the outcome.

Establishing characterization paradigms that reflect patient use conditions and the pulmonary microenvironment is fundamental to advancing the clinical relevance of in vitro testing for inhaled biologic protein products. While conventional in vitro assessments—such as emitted dose and APSD assessments—remain indispensable components of pharmaceutical quality evaluation, they are inherently limited in their ability to capture aerosol-induced structural perturbations, lung region-specific deposition, and preservation of biologic activity at the site of action. These limitations are particularly critical for locally acting inhaled biologics, for which the therapeutic performance is driven by local exposure, molecular integrity, and sustained functional activity within the respiratory tract. Therefore, next-generation in vitro characterization strategies must transcend traditional metrics and systematically incorporate clinically representative breathing profiles, disease-relevant airway geometries, and rigorous post-aerosolization physicochemical and functional assessments, including aggregation, oxidation, and potency. Advanced analytical methodologies with sufficient sensitivity to detect subtle but clinically meaningful changes in higher-order structures, conformation, and bioactivity are essential to establish meaningful links between in vitro measurements and in vivo performance. In parallel, lung mimetic platforms—such as air–liquid interface epithelial models, mucus surrogates, and ex vivo lung tissues—provide critical mechanistic insight into epithelial interaction, regional retention, and potential local immunogenic risks under conditions that closely approximate human pulmonary physiology.

Looking forward, the convergence of regulatory science and model-informed drug development is vital to the development of clinically relevant in vitro–in vivo relationships to support formulation optimization, device selection, and dose justification for inhaled biologic proteins. The adoption of clinically anchored, mechanism-based characterization frameworks will strengthen translational confidence, enhance regulatory decision making, and ultimately enable more predictable and robust clinical outcomes for inhaled protein therapies targeting respiratory diseases. It is evident that disruptive change is on the horizon for inhaled drug development driven by the demands of novel protein therapeutics.