Reflection paper on the APV workshop on in vitro performance testing of topically applied and topically acting substances

Abstract

With great anticipation, the European Medicines Agency (EMA) guideline titled “Quality and Equivalence of Locally Applied, Locally Acting Cutaneous Products” officially came into effect on April 2, 2025. This regulatory document establishes the legal and scientific framework for the evaluation of generic topical medicinal products, particularly those for which systemic bioavailability is not a relevant endpoint. The guideline is designed to replace conventional clinical trials with scientifically justified alternative methodologies for demonstrating therapeutic equivalence to reference products in the context of generic marketing authorization. These methodologies include, most notably, in vitro release testing (IVRT), in vitro permeation testing (IVPT), stratum corneum sampling via tape stripping (TS), and the vasoconstriction assay for corticosteroids. Based on the draft guideline version released in 2018, preliminary experience has been gathered in recent years regarding the implementation and practical applicability of some testing parameters proposed. However, this early engagement also exposed several ambiguities and limitations in the draft guidance, prompting expectations that the finalized version would address these deficiencies and offer more comprehensive direction on the use of these methods. The present paper is intended to summarize these known limitations and critically examine selected aspects of the guideline. Thereby, it seeks to provide an informed perspective on the scope, robustness, and regulatory utility of the final guideline, and to facilitate a dialogue on its practical implementation in regulatory and industrial settings.

1. Introduction

The scope of the recently published European Medicines Agency (EMA) guideline ́Quality and equivalence of locally applied, locally acting cutaneous productś (No. EMA/CHMP/QWP/708282/2018) provides a central regulatory framework for the assessment of the quality and equivalence of medicinal products topically applied on humans. Thus, it covers locally applied, locally acting skin products, although the underlying principles are also transferable to other categories of topical medicinal products. In particular, the concepts set out in the guideline can be applied to ear and eye preparations as well as to locally acting products intended for use on the vaginal mucosa or the nail apparatus [1]. The quality requirements formulated can be applied for newly developed products and their marketing authorization as well as for post approval changes [1]. Thereby, a comprehensive understanding of the materials, formulation design, and manufacturing processes, which must ensure consistency, robustness, and reproducibility throughout the complete product life cycle, is necessary. Ideally, an appropriate physicochemical and biopharmaceutical characterization of representative product batches is carried out, which then serves as a reliable reference point for the evaluation of future changes after approval. Another central point of the guideline is the establishment of an equivalence concept that is tailored to new generic topical medicinal products (test product), whose application for approval is to be made possible by demonstrating equivalence with an approved topical product (originator, reference product). This is implemented through a step-by-step approach supported by clearly structured decision trees designed to help applicants and regulatory authorities to determine the most appropriate methodological approach. In a first step, different levels of similarity (Q1, Q2, Q3) are proclaimed, ranging from qualitative (Q1) and quantitative (Q2) composition to more detailed considerations of the physicochemical and structural properties of the formulations (Q3) to show pharmaceutical equivalence [1]. In the best case, the requirements for Q1, Q2 and Q3 are met to proceed with the approval procedure. For every different (and, in light of the variability of some originators and the strict equivalence criteria often more realistic) case, a series of test protocols is provided to support the presentation of (therapeutic) equivalence: (i) in vitro release tests (IVRT) using synthetic membranes to evaluate release kinetics, (ii) in vitro permeation tests (IVPT) and (iii) in vivo stratum corneum sampling (tape stripping, TS) to determine the pharmacokinetic profile of the active pharmaceutical ingredient (API) on human skin with regard to demonstrating equivalence between test and reference products, and (iv) pharmacodynamic vasoconstriction assay, which is specifically used for corticosteroids to enable sensitive measurement of local pharmacodynamic activity. Altogether, the explicit aim of the guideline is to reduce or even to replace time-consuming and cost-intensive clinical trials by implementing the proposed surrogate methods. However, due to the novelty of the guideline, there is a lack of experience among all parties involved and questions regarding scientifically sound and validated test parameters remain unanswered. In February 2025, a workshop on In vitro performance testing of topically applied formulations was held by the APV (Arbeitsgemeinschaft für Pharmazeutische Verfahrenstechnik e.V. (APV), International Association for Pharmaceutical Technology and Industrial Pharmacy) focused on the implications of the EMA guideline for manufacturers of topical dosage forms. This work covers aspects discussed during the workshop and is intended to give an overview of the advantages and disadvantages of key points of the guideline and provides a comprehensive overview of methodological, regulatory and scientific aspects. It also presents ideas that could be considered for future versions of the guideline.

2. Questions and comments on ‘Quality’

2.1. New development vs. generic products

The requirements of the guideline for demonstrating the quality of newly developed medicinal products, that fall within its remit, are extremely valuable. The explicit mention of all necessary information regarding the composition, indication and dosage as well as the primary packaging material is essential for the preparation of the documents for the authorization procedure of the new generic product. In particular, the request to name the respective function of each ingredient must be seen in the overall context of the formulation and is of great importance for a systematic (safety) assessment. Excipients play a particularly important role here, as they cannot be regarded as inert formulation components and can be decisive factors for product performance due to their influence on cutaneous absorption, e.g. by altering the thermodynamic activity of the API in the formulation or by altering the skin conditions and thus altering the APÍs diffusion coefficient. The awareness of the resulting need for regulatory control is also reflected in the guideline. But, despite these complex requirements, from an industrial point of view, the guideline does not clearly differentiate between the development of generics and innovative products. From a logical point of view, these quality requirements cannot be applied to generic products as they may contradict with the requirements of Q1/Q2/Q3 sameness. In this context the EMA guideline, Q1 requires the same qualitative composition, Q2 the same quantitative composition and Q3 requires similar physicochemical properties [1]. In fact, many of the originator products have been developed decades ago, when today’s quality standards and analytical capabilities to control them, did not yet exist. The phenomena observed in some originator semi-solids are high batch-to-batch variability in viscosity, syneresis, insufficient control of particle size (Ostwald ripening) of suspension-type formulations. The choice of excipients was often empirical and cannot be justified retrospectively based on the criteria proposed in the present guideline. From the authors’ perspective, a regulatory incentive for developing improved versions of originator products, which do not meet current quality standards, would be welcomed. This would allow progress in topical dermatological treatment options and patient benefits such as improved safety, usability, and convenience of application.
Paradoxically, originator products which do not meet today’s quality standards are best protected against generic competition and current quality requirement while at the same time sameness cannot be met which is not in interest of patients. It is expected that all stakeholders will gather experiences with the guideline and implementation within the upcoming years and it would be desirable that these feed into improvements, i.e. facilitated procedures for generics applications and post-approval changes.

2.2. Life cycle management

Overall, the requirements set out in the guideline present opportunities but also risks. Especially in the context of the life cycle management of a corresponding medicinal product and unplanned post-approval changes this might be relevant as changes in terms of composition, manufacture or packaging may have an impact on the quality, efficacy or safety of the product. Complex topical products do not only contain a variety of excipients that may have an impact on formulation performance but the microstructure of the formulation and the formulations impact on the skin condition may also have an impact on formulation performance. In this respect, there is a complex interplay between API, excipients, formulation microstructure and the skin which requires monitoring formulation characteristics at various levels. Apart from the more qualitative and quantitative composition, formulation microstructure as well as occlusion, emollience, and skin hydration properties are to be considered as potential critical factors for therapeutic efficacy as outline in the guideline [1]. Furthermore, changes due to formulation transformation after application, such as crystallization of the API [2], a change in pH-value or viscosity as well as absorption or loss of water (moisturizing/ drying out) or other evaporating components must also be comprehensively characterized and documented as part of the product performance characterization, and with regard to storage effects. In the event of deviations or changes, the quality of the product must then be verified again in accordance with the guideline.

A further challenge is the specification of rigid acceptance criteria, especially in cases where the variability of the performance parameters is considerable and strongly depending on the formulation. This is particularly evident in the case of substantial batch-to-batch variability. Even with the introduction of extended acceptance criteria, it would be questionable whether these adjustments would be sufficient to capture the full extent of the variability observed in practice for performance (IVRT) and rheological attributes as well as particle or droplet size [3]. A typical issue would be the reference product showing higher batch-to-batch variability in a quality aspect than allowed for the generic product. The following example may illustrate this issue: In hydrogels, the viscosity is typically determined by the polymeric thickener. For polymeric thickeners the European Pharmacopoeia (Ph. Eur.) typically allows a viscosity of 75 % to 140 % of the nominal value. Therefore, even from the same process, the same type and supplier of the thickening polymer, and exactly the same quantitative and qualitative composition, the batch-to-batch variability of such a formulation may be higher than the equivalence criteria of the current guideline allow. The generic product would thus be faced with the challenge to at the same time be equivalent and better (less variable in terms of rheological parameters) than the reference product. This also is particularly relevant from a life cycle management perspective, where consistent product performance across multiple production batches is critical to ensure both regulatory compliance and therapeutic reliability. It is emphasized that not only the inherent variability as a typical characteristic of topical formulations, but also methodological aspects can have an influence on batch variability, such as minimal deviations in product testing methods like rheological and performance tests. The handling of post-approval changes for products that were not developed in accordance with the current guideline requirements is also still questionable. Further challenges may be the management of the immense quantities of samples in practice, starting with their storage capacity under specified conditions, their management and analysis options, and the probable expansion of quality control units, both in terms of space and personnel. Last but not least, it may occur that the manufacturer of a drug product will have to react to a supply of excipients by using excipients from manufacturers different from those specified in the original marketing authorization. In such case, and if the authorities follow the guideline very strictly, equivalence of the drug product will have to be shown after incorporation of the excipient from the different manufacturer. This will lead to an enormous increase in costs related to Q3 and performance testing, possibly making it economically unfavorable to uphold the marketing authorization. The same applies to operational changes in the manufacturing process, an upscaling procedure or relocation of the manufacturing site.

Effective life cycle management on the part of manufacturers therefore requires proactive planning, robust analytical methods and careful risk assessment to ensure that product performance and regulatory compliance can be guaranteed throughout the product lifecycle, including any potential changes. However, a balanced regulatory approach is needed, which ultimately avoids the unintended exclusion of products based solely on intrinsic variability characteristics and an unnecessarily high consumption of time and money. It is reassuring that the guideline attempts to ensure that quality statements are not made in isolation but are the result of a systematic assessment of product-related evidence by embedding structured decision aids. Besides all questions remaining, this approach attempts to minimize subjectivity, to promote harmonization of regulatory applications and helps to improve predictability for industry and authorities.

3. Questions and comments on ‘Equivalence’

Equivalence can be interpreted as pharmaceutical (quality) or therapeutic equivalence. A stepwise procedure to present therapeutic equivalence (efficacy and safety) in lieu of clinical endpoint studies is proposed and decision trees for selection of test procedures to be applied are now implemented. These decision trees, which had been proposed earlier [4], provide a step-by-step and targeted path to demonstrate equivalence. By breaking them down into sequentially organized stages and dependencies, they enable individual decision-making for each development project, taking into account critical variables such as the type of formulation (simple vs. complex), the extent to which the criteria for pharmaceutical equivalence are met, and the availability of supporting in vitro data. In a first step, applicants are asked to decide, which decision tree is applicable. Depending on the type of formulation (solutions, simple or complex formulation) and if qualitative and quantitative (Q1 and Q2) equivalence and/or physicochemical properties (Q3) can be met, for concluding or demonstrating therapeutic equivalence, three different decision trees are available. They differ between Q1, Q2, Q3 comparability for solutions (Decision tree 1) and more complex, laborious and cost-intensive performance tests (IVRT, IVPT, TS and/or pharmacodynamic test) for all other kinds of formulations (Decision Trees 1,2,3). In this context, it should be noted that the guideline for a drug with a narrow therapeutic index only requires a study with a clinical endpoint. Nevertheless, the devil is in the detail and time will tell to what extent the guideline actually may needs to be adapted. The following problems and details are already relevant and worthy of consideration in the eyes of the authors.

3.1. Challenges with Q3 equivalence

The idea behind the requirement of Q3 equivalence is that formulation microstructure may have an impact on API diffusion and release. This, in turn, may affect partitioning into the skin and thus penetration. This effect is especially valid for rheological aspects, where local viscosity will impact API diffusion coefficient in the formulation and for crystal size, habit and polymorph which may determine dissolution kinetics of the API in the formulation. Ensuring Q3 equivalence, before moving on to assess IVRT/IVPT thus makes sense not only from a rational point of view but also from an economical point of view. Though theoretical reasoning is valid, it will not necessarily turn out to be practically relevant to all formulations. For example, studies on semi-solid vehicle stabilized with glycerol monostearate showed that IVPT equivalence could be achieved despite missing Q3 equivalence in terms of rheological equivalence. This may be explained by the fact that local viscosity inside the formulation is more determining with respect to diffusion kinetics in the formulation than macroscopic viscosity assessed instrumentally [5]. A study on ibuprofene formulations revealed similar results. Differences in rheological parameters did not lead to alterations in IVRT. Interestingly, higher viscosity formulations did show lower flux values in IVPT in formulations with 1 % API but not in those containing 10 %. This shows that changes in rheological parameters may translate differently into IVRT/IVPT changes [6]. As another example, diclofenac diethylamine-loaded emulgels showed bioequivalence in an in vivo pharmacokinetic study in 32 healthy volunteers despite Q3 non-equivalence with respect to rheological parameters. This finding suggests that differences in rheological parameters beyond 10 % may be acceptable. Further studies would be needed to substantiate more rational limits across a broad range of formulations. In the meantime, Q3 equivalence should not be used as a knock-out criterion of bioequivalence in case that IVRT/IVPT equivalence is achieved. For more detailed reading, we would like to refer to the review by Ilic et al., which is still relevant even though based on the draft guideline [7].

API crystal size, habit and polymorph are also relevant to product performance as they guide dissolution kinetics. Therefore, these parameters should be the same in the test and reference product. Again, the requirement is sound from a theoretical point of view. In case dissolution is slower than diffusion in the formulation, partitioning into the skin and diffusion inside the skin, dissolution will determine the rate of API absorption. If the opposite is the case, dissolution will be fast enough to not affect API absorption and crystal size, habit and polymorph are of lower importance. In addition to this, formulation transformation after application to the skin needs to be taken into account. Evaporation of water and other solvents may lead to alteration of API solubility in the remaining formulation and may in the worst case lead to precipitation [2]. In this case, API absorption will be much limited. Thereby, it needs to be ensured that the test formulations show the same behavior. In case transformation of the formulation does not lead to API precipitation from the reference product, it must not occur in the test formulation either. If it occurs from the reference product, it must occur in a similar way also from the test formulation.

An aspect not covered by the guideline but known from a research perspective is alteration of the formulation due to shear forces during dispensing. Different dispensers may alter formulation microstructure to different extent which may then lead to differences in IVRT/IVPT. It remains to be assessed if this is an issue related to the majority of formulations or not, but it certainly should be kept in mind [8].

3.2. Challenges with the release and permeation kinetics test protocols

3.2.1. Impact of composition of receptor medium on API solubility, formulation composition and skin integrity during IVRT and IVPT

The release of the API by diffusion is a passive process and in focus of the IVRT / IVPT experiment. In IVRT, synthetic membranes such as cellulose esters, polycarbonate, polyethersulfone or others are used, whereas in IVPT human skin from plastic surgery or body donors is used. The EMA does not require the use of a specific membrane but points out that it should ensure that the product and the receptor medium remain separate, should not be rate-limiting and should be compatible with the drug product formulation and not bind to the active substance. Although the focus of IVRT studies is to monitor API diffusion from the formulation into the receptor medium, all other molecules, which are sufficiently small to pass the membrane by diffusion will do so, too. These are e.g. excipients like citric acid used to adjust the pH of the formulation or organic solvents used for API solubilization in the formulation. Upon diffusion into the receptor medium, pH active substances may alter its pH of the receptor medium which in turn can have a significant influence on the ionization state of the API and therefore on its solubility in the formulation as well as in the receptor medium. This may result in altered release from the formulation. A buffered receptor medium ideally at the pH of the donor formulation is recommended to keep the pH constant. On the one hand, organic solvents diffusing from the formulation into the receptor medium may have a strong effect on API solubility in the formulation, which may affect API solubility and thus release. On the other hand, the amount of organic solvent present in the formulation is negligible with respect to the amount of receptor medium. Thus, its diffusion into the receptor medium will not impact API solubility therein.

But diffusion is not only limited to the direction from donor to receptor compartment and mentioned in the guideline as back diffusion of the receptor medium. This refers to the diffusion of water or, particularly important, of organic solvents from the receptor medium into the formulation. This process can significantly influence drug release because the influx of organic solvent may alter the structure of the formulation and the solubility of the API in it. Generally speaking, the higher the proportion of organic solvent, the higher the solubility of the API and the faster the release from a formulation. In contrast, a reduction of the organic solvent content in the receptor medium can change the solubility of the API in the receptor medium, thereby impeding release due to a reduction in the concentration gradient.

These aspects are particularly important in IVRT. Small organic molecules like methanol, 2-propanol etc. used to enhance the solubility of the API in the receptor medium can rather easily diffuse from receptor through the synthetic membrane into the donor compartment. This is often not taken into account when selecting a suitable receptor medium, where the main goal is to ensure sink conditions, resulting in unhindered diffusion and thus high release rates. An increase in the release rate is often accompanied by a greater change in the donor formulation used, so a balance must be found between sufficiently high API solubility in the receptor medium and adequately low alterations to the formulation during the experiment as some back diffusion of the medium cannot be ruled out completely.

In IVPT, diffusion of organic solvents and other small molecules will be limited as the skin is not as permeable as the synthetic membranes. As a result, alteration of the formulation is a minor risk. Nevertheless, organic solvents will diffuse into the skin and alter its characteristics. This may also lead to enhanced penetration and permeation which may overestimate skin absorption in vivo. Therefore, care must be taken when selecting receptor media other than aqueous buffer solutions. In IVPT, much lower permeated API concentrations are to be expected as the diffusional barrier if the skin is much higher than that of synthetic membranes. Therefore, one should evaluate whether solubility enhancement is needed at all. As an alternative to organic solvents, (bovine) serum albumin may be used. It enhances solubility of a range of APIs but does not alter the skin condition. Furthermore, it better reflects the in vivo situation [9].

3.2.2. Changes in formulation properties during the performance tests

The guideline requires that transformations of the formulation with respect to evaporation upon administration should be minimized in the experimental setup for IVRT [1]. IVRT under infinite dose conditions is robust in this respect but does not always reflect the clinical situation and is designed for quality purposes only. But for IVPT with dosing of realistic application quantities, such changes will occur probably in the majority of the formulations, especially if the donor compartment is un-occluded, as the guideline demands. Especially in cases where evaporation of volatile solvents or other phenomena might be relevant for API transfer to the site of action (e.g. due to changes in thermodynamic activity), further testing is needed as the guideline requests. Explicitly, the guideline states “residues should be equivalent with respect to quality, i.e., in terms of pharmaceutical equivalence”. A detailed analysis of Q1 and Q2 sameness of the residues will be elaborate but might be feasible, including sensitive analytical methods. It should be further detailed which physicochemical or microstructural properties (Q3) should be tested as most tests are certainly not feasible to be executed at the residues (e.g. rheological testing).

3.2.3. Sites of action and which method to choose

Each sites of API actions will need individual test methods to be selected. This issue is mentioned by the guideline but not clearly solved. While IVPT is recommended for APIs with site of action in deeper skin layers, in the annex TS is recommended for products acting in the Stratum corneum (SC) but may also be suitable as a surrogate for products acting in deeper skin layers. At this point, the question arises, if the method should be selected on the basis of the site of action or with respect to the penetration depth of the API.

A more complex issue comes along with combination products that contain several APIs with different sites of action in the skin. They will require an adapted and differentiated approach. The heterogeneous spectra of the APIś physicochemical properties lead to distinct penetration depths, which is must also to be taken into account when selecting the permeation kinetics test. The chemical properties of the components to be examined thus require complex or multiple analytical methods for reliable quantification in IVRT, IVPT and also SC sampling-studies. With respect to the choice of method to investigate permeation kinetics, a precise proposal or commentary on this aspect in relation to formulations containing multiple APIs with different sites of action is not explicitly mentioned in the guideline. The usual procedure here would be a scientific advice procedure to clarify whether separate in vitro or in vivo models are necessary or whether a focused assessment on one active substance would be sufficient. Rapid access to the competent authority would be desirable here in order to speed up the approval procedure in the interests of a profitable exchange or to avoid unnecessary delays due to application and waiting times.

3.2.4. Alternative methods for performance testing

Besides IVPT using excised skin in diffusion cells and in vivo SC sampling, alternative methods have emerged over the past decade but are not accepted by authorities, yet. Namely, these are confocal Raman spectroscopy (CRS) and open flow microperfusion (OFM). While the former predominantly samples the SC, the latter evaluates drug levels in the dermis. They may thus be regarded as alternatives to SC-sampling and permeation experiments, respectively. In CRS laser light is focused on a sample and the inelastically backscattered light collected as Raman spectrum. This spectrum contains information on the nature and number of components in the sampled volume, i.e. the concentration of the API in the SC [10], [11]. It is non-invasive, provides high spatial resolution and is time efficient. A lot of research on the suitability of CRS to conduct skin penetration studies has been performed throughout the last decade, which was accelerated after a method to generate truly quantitative data from the spectra was published. First, studies showed that CRS yields similar values of penetrated API amount compared to TS [12], [13]. Then, a study applied the EMAs requirements for TS studies to CRS and showed that pivotal bioequivalence data can be acquired [14]. Another emerging method is OFM, which continuously samples the API that has permeated to the interstitial fluid of the dermis. Thereby, concentration–time profiles can be generated in a reliable way in in vivo studies. Information, that is not cannot be gathered with any method accepted by the guideline. Although extensive research on both methods has shown that they are capable of providing pivotal bioequivalence data [14], [15], the guideline still considers them not sufficiently established and calls for validation as part of a CHMP (Committee for Medicinal Products for Human Use) opinion. Unlike the United Stateś Food and Drug Administration (US FDA) that funds such research, the EMA does not and it will remain a (financial) challenge to the manufacturers of the instruments to provide the required data. At the same time, EMA is interested in the methods and open to them, which may enable close contact during validation and hopefully integration of the methods into the subsequent version of the guideline.

Download the full article as PDF here Reflection paper on the APV workshop on in vitro performance testing of topically applied and topically acting substances

or continue reading here

Dominique Lunter, Sascha Gorissen, Michael Herbig, Martin Hukauf, Adina Eichner, Reflection paper on the APV workshop on in vitro performance testing of topically applied and topically acting substances, European Journal of Pharmaceutics and Biopharmaceutics, Volume 220, 2026, 114971, ISSN 0939-6411, https://doi.org/10.1016/j.ejpb.2025.114971.