Marc Brown, PhD, is chief scientific officer and co-founder of MedPharm.

Marc Brown, PhD

	Scaling up the manufacture of a topical drug formulation to supply toxicology, clinical, and commercial batches requires a well thought-out strategy. Process development requires modest investment in time and money compared to the risk of batch failure and the consequent adverse effects on timings, costs, and project success. Furthermore, it is not only development programs for new chemical entities (NCEs) that can benefit from process development. It is a step that generic-drug companies omit at their peril, because originator products may not be based on optimized formulations and the correct process parameters needed for their manufacture may not be well defined. 

	A quality-by-design (QbD) approach, in combination with six sigma tools, provides a methodology from which a robust processing control strategy can be derived. The QbD strategy, given the appropriate scale-up considerations, lives with a product throughout its entire life, not just for the initial scale-up for toxicology or clinical batches but also when moving to commercial scale and future plant-to-plant transfers. The benefits of this approach are multiple. Above all, the building-in of quality as opposed to the determination of quality by testing clearly demonstrates to regulatory authorities that control over consistent manufacturing has been established. This increased level of assurance in manufacture reinforces the commercial viability of the product. In addition to experimental design to challenge critical process parameters (CPP), six sigma tools such as failure mode effect analysis (FMEA), input-output diagram, voice of the customer, benchmarking, and evaluation criteria add to the power of the QbD approach.

	Defining a control strategy around CPPs and critical material attributes (CMAs) is of particular importance with semi-solid products. The composition of such products is carefully derived by taking in every aspect of the target product profile while acknowledging the scientific and technical constraints imposed by the API, excipients alone, and in combination. The attributes of each material used can have a profound impact on the resultant formulation in terms of both safety and efficacy, but they can also dictate or heavily influence the manufacturing process especially in areas such as dissolution, mixing parameters, and heating or cooling processes. The safety and efficacy of topical pharmaceutical formulations, such as creams, gels, foams, and ointments, are clearly intimately related to the composition of the product; however, there is often a lack of appreciation for the relationship between clinical performance and the microstructure of the formulation, which is highly dependent upon manufacturing process parameters. 

	Any process development program aimed at product optimization must, therefore, take into account a broad range of processing parameters to ensure consistent manufacture of pharmaceutical product to the chosen specification. Creams tend to be one of the more complex topical pharmaceutical formulations (1), in which the combining of two immiscible phases requires a defined order and speed of addition of materials; defined speeds and times for the mixing and homogenization steps; and more often than not, a controlled heating and cooling rate during the process (2). The more complex the processing requirements, the more it is likely that the CPP will have an influence upon the control strategy and complexity of the scale-up program.

	A key tool from the start of any pharmaceutical development program is the quality target product profile (QTPP). The QTPP ensures that the broad objectives of the project are captured, including the patient and prescriber requirements and the attributes needed to ensure a safe, effective, and commercially viable treatment. The QTPP should always be referred back to when determining the impact of material attributes and process parameters on product quality. From the QTPP, the critical quality attributes (CQAs) of the product may be determined together with an estimation of the impact on product quality of individual raw material attributes and processing parameters (3). Benchmarking against competitor products, especially in the generic-drug market, is a useful guide to the QTPP where it can highlight key commercially relevant differentiators for the new product.

	Voice of the customer (VOC) is a six-sigma tool to aid the compilation of the QTPP. It allows the end user and the company sponsoring the development to provide clarity on what they need and want, or do not want in some cases. For a topical pharmaceutical product, the patient voice is paramount, and for orphan products, this is often efficiently captured through internet-connected patient groups. Other significant “voices” are the prescribers and investors. Simple surveys of patients or key opinion leaders (KOLs) are a highly effective means of acquiring valuable VOC data. VOC and the QTPP play a crucial role in the determination of CQAs. 

	An important element of any efficient process development program is the identification and mitigation of risks. It is widely recognized that the deployment of FMEA as a risk management tool, alongside an appropriate design of experiments, lead to control strategies that reliably produce drug products of the appropriate quality, whether they be based on NCEs or generic drugs. FMEA is a step-by-step approach for estimating the potential risk arising from all possible failures in the design or processing of the topical product. It also highlights where efforts should be deployed in risk mitigation-often defining the most suitable process controls. Many liquid and semi-solid topical pharmaceutical products are, by design or necessity, highly complex systems, often involving multiple phases (e.g., oil and water emulsions) with a defined range of droplet sizes. As such failures can arise in a number of different processing areas, including heterogeneity of drug content, or consistency and physical, chemical, or microbial instability. The efficacy, uniformity of dose, and safety of topical pharmaceuticals rely upon these formulations being homogenous, stable, safe, and easy to use. 

	The CPPs are derived by establishing the relationship between the processing parameters, the CMAs of both the API and excipients, and the CQAs of the product. A concise way of expressing the process parameters is the six-sigma input–process–output (IPO) diagram (see example in 

). The IPO diagram highlights unit operations and which operational parameters should be investigated in an example manufacturing process.
	

	The first experimental design (DoE), in the form of screening study, is then derived from the outputs of the FMEA (see example in 

) and the IPO diagram. The objective of this pre-screen is to confirm the output from the FMEA, uncover any interactions between key parameters, and to determine processing parameters that are truly critical. Both the experience and expertise of the technical project team and efficient experimental design software are crucial for this approach. Pramod et al. noted that “though design of experiments is not a substitute for experience, expertise, or intelligence, it is a valuable tool for choosing experiments efficiently and systematically to give reliable and coherent information” (4). The FMEA allows the operator to capture the knowledge and decide which areas of the process are most critical and require experimental investigation; the DoE will then validate this thinking statistically and quantitatively.

	A crucial factor in the pre-screen and full experiment design for topical formulations is that experiments are conducted using equipment that is representative of larger-scale equipment in order to derive meaningful qualitative CPP data. At MedPharm, IKA LR1000 lab reactors are used, which allow for the control of all typical processing parameters. This approach is crucial to avoid generating “noise” and ensure the quality of the output and the robustness of the resultant control strategy. The understanding of the influence of scale on the CPP from high-quality experimental work conducted on small-scale forms the basis of any future scale-up work and technical-transfer activities. 

	For a complex cream, typically 12 experiments should be targeted to cover two to four CPPs in a pre-screening study in preparation for the manufacture of toxicity or clinical batches. The actual number of experiments may have to be expanded depending on the outputs and the associated risks.

	Another important factor is that experimental design in both the pre-screen and full study should attempt to push the product to failure to allow for the understanding of the design and control space boundaries. If the developer is conservative in experimentation, they can misunderstand the boundaries between success and failure, and the design space is limited to knowledge space (see 

	A third factor is the identification of what six sigma calls the key process output variable(s) (KPOV), the variables that determine success, and using evaluation criteria to establish whether the method employed to measure the KPOV will detect critical failure. If the answer is no, a mitigation plan is needed. The output of any experimental work will only be as good as the analytical method allows. Topical pharmaceutical products require an array of analytical techniques to evaluate their quality ranging from the commonplace, such as high-performance liquid chromatography (HPLC) and viscosity testing, to more sophisticated methodology, such as rheological evaluation, accelerated stressing to show the potential for separation (5), and 

release testing to check that there is no change in the release/thermodynamic activity of the drug from the formulation. 

	The design space is defined as the multidimensional combination and interaction of input variables and process parameters that have been demonstrated to provide assurance of quality (6). It is the boundary of the process parameters to which a product can be made and satisfies the QTPP and CQAs for the product. It is important to stress that there is not a hard border (7) with the knowledge space and that defined process inputs are as important as the measured output. Interpretation of the DoE should point to CPPs and any key interactions between these parameters. Analysis of variants (ANOVA) plots, which show both the mean and the distribution of date around each mean, are particularly useful in this respect as they will clearly show the significance of a parameter or interaction and associated variability. Many experimental design software packages can identify statistically significant effects across a range of parameters and present them in a concise graphic that decision makers find easy to understand. 

	An important point is that any process change made within the design space is not considered a regulatory change (8), and hence, has a direct bearing on the flexibility of any future manufacturing process. The more that is known about the boundaries of success and failure, the better the control strategy will be and the less impact that changes in excipient suppliers and specifications will have upon a product over its lifetime.  

	The control space represents a range of critical parameters within which the process will yield an assured output to meet the CQAs and target specification at all times. Getting from the design space to the control space can be achieved through the further use of experimental design; typically, two to five factors in full, or fractional factorial, or other surface response design informed by the data from previous experiment design work. A useful guide to DoE can be found online in the form of the engineering statistics handbook (8). 

	Clearly, the control space for the CPP must sit well within the design space and sufficiently away from the edge of failure to ensure robustness. The interpretation and conclusions from an optimization design will show “best” settings to achieve a product that meets the QTPP and CQAs. A confirmation batch using the optimal settings will demonstrate that the response values from the DoE are close to their predicted values. 

	Using a stepwise and methodical QbD approach during the development and late-stage formulation of topical products provides a sound and robust platform in establishing the design space for process development and will ultimately enable the developer to provide a robust control strategy for manufacturing. Missing this step can lead to poor processing and physical instability in topical products, which directly impacts product performance and the patients who need them.

	The often complex liquid and semi-solid processing for topical products cannot be underestimated, and the marriage of experience, QbD, six-sigma tools, and experimental design ensure the manufacturing scale-up of complex topical products that can be conducted with minimum risk. Above all, experience has shown that a well thoughtout QbD-based process development strategy with in-depth knowledge of the products and processes can save not only time and money, but also embarrassment from having to explain the lack of understanding behind a critical batch failure that is holding up a development program for an important company asset.

Pharmaceutics: the Design and Manufacture of Medicines 

(Churchill Livingstone Elsevier, Amsterdam, Netherlands, 4th ed., 2013) p. 449. 
	2. M. Lowenburg, 

 36 (11) (2012).
	3. R.K. Chang et al., 

. 15 (3) 674–683 (2013). 
	4. K. Pramod et al., 

. 6 (3) 129–138 (2016). 
	5. D. Lerche and T. Sobisch, 

, Step 4 (August 2009).
	7. M. Adams, GSK, "Quality by Design & Design for Six Sigma," ISPE Breakfast Seminar Toronto (May 2009).
	8. NIST/SEMATECH, 

NIST/SEMATECH e-Handbook of Statistical Methods


	Vol. 43, No. 2
	February 2019
	Pages: 28–31

	When referring to this article, please cite it as C. Harrison, S. Pratt, and M. Brown, "Process Development Best Practices for Topical Drug Products," 

, is head of Process Development and Manufacture; 

is chief operating officer; and Professor 

, is chief scientific officer and co-founder, all at MedPharm.

Articles

Using a QbD approach in the development and formulation of topical products will enable the drug developer to provide a robust control strategy for manufacturing.

Process Development Best Practices for Topical Drug Products

	Once a formulation has been selected for a given API, its manufacture needs to be scaled up to supply toxicology, clinical, and commercial batches. Scaling up may appear simple, but most pharmaceutical companies have now learnt the value of a well thoughtout strategy for process development. With the ever-increasing costs and timelines associated with drug-development programs, risk mitigation invariably pays dividends. Process development requires modest investment in time and money compared to the risk of batch failure and the consequent adverse effects on timings, costs, and project success if they were to happen. Furthermore, it is not only development programs for new chemical entities (NCEs) that can benefit from process development. It is a step that generic-drug companies omit at their peril as it is often not realized that originator products may not be based on optimized formulations, and that the correct process parameters needed for their manufacture may not be well defined. 

	Quality by design (QbD) is an approach that aims to ensure the quality of medicines by employing statistical, analytical, and risk-management methodology in the design, development, and manufacturing of medicines (1). The practice has now been thoroughly embraced by the majority of the pharmaceutical industry as it has become clear that, far from being a hindrance (2), a QbD approach in combination with six sigma tools, provides a methodology from which a robust processing control strategy can be derived. The QbD strategy, given the appropriate scale-up considerations, lives with a product throughout its entire life, not just for the initial scale-up for toxicology or clinical batches but also when moving to commercial scale and future plant-to-plant transfers. The benefits of this approach are multiple. Above all, the building-in of quality as opposed to the determination of quality by testing clearly demonstrates to regulatory authorities that control over consistent manufacturing has been established. This increased level of assurance in manufacture reinforces the commercial viability of the product.

	In addition to experimental design to challenge critical process parameters (CPP), six sigma tools such as failure mode effect analysis (FMEA), input-output diagram, voice of the customer, benchmarking, and evaluation criteria add to the power of the QbD approach.

	Defining a control strategy around CPPs and critical material attributes (CMAs) is of particular importance with semi-solid products. The composition of such products is carefully derived by taking in every aspect of the target product profile while acknowledging the scientific and technical constraints imposed by the API, excipients alone, and in combination. The attributes of each material used can have a profound impact on the resultant formulation in terms of both safety and efficacy, but they can also dictate or heavily influence the manufacturing process especially in areas such as dissolution, mixing parameters, and heating or cooling processes. The safety and efficacy of topical pharmaceutical formulations, such as creams, gels, foams, and ointments are clearly intimately related to the composition of the product; however, there is often a lack of appreciation for the relationship between clinical performance and the microstructure of the formulation, which is highly dependent upon manufacturing process parameters. 

	Any process development program aimed at product optimization must, therefore, take into account a broad range of processing parameters to ensure consistent manufacture of pharmaceutical product to the chosen specification. Creams tend to be one of the more complex topical pharmaceutical formulations (3) where the combining of two immiscible phases requires a defined order and speed of addition of materials; defined speeds and times for the mixing and homogenisation steps; and more often than not, a controlled heating and cooling rate needs to be applied during the process (4). The more complex the processing requirements, the more it is likely that the CPP will have an influence upon the control strategy and complexity of the scale-up program.

	A key tool from the start of any pharmaceutical development program is the quality target product profile (QTPP). The QTPP ensures that the broad objectives of the project are captured, including the patient and prescriber requirements and the attributes needed to ensure a safe, effective, and commercially viable treatment. The QTPP should always be referred back to when determining the impact of material attributes and process parameters on product quality. From the QTPP, the critical quality attributes (CQAs) of the product may be determined together with an estimation of the impact on product quality of individual raw material attributes and processing parameters (5). Benchmarking against competitor products, especially in the generic-drug market, is a useful guide to the QTPP where it can highlight key commercially relevant differentiators for the new product.

	An important element of any efficient process development program is the identification and mitigation of risks. It is widely recognized that the deployment of FMEA as a risk management tool, alongside an appropriate design of experiments, lead to control strategies that reliably produce drug products of the appropriate quality, whether they be based on NCEs or generic drugs. FMEA is a step-by-step approach for estimating the potential risk arising from all possible failures in the design or processing of the topical product. It also highlights where efforts should be deployed in risk mitigation-often defining the most suitable process controls. Many liquid and semi-solid topical pharmaceutical products are, by design or necessity, highly complex systems, often involving multiple phases (e.g., oil and water emulsions) with a defined range of droplet sizes. As such failures can arise in a number of different processing areas, including heterogeneity of drug content, or consistency and physical, chemical, or microbial instability. The efficacy, uniformity of dose, and safety of topical pharmaceuticals rely upon these formulations being homogenous, stable, safe, and easy to use.

Process–Output (IPO) diagram showing the unit operations of a sample manufacturing process.

) and the IPO diagram. The objective of this pre-screen is to confirm the output from the FMEA, uncover any interactions between key parameters, and to determine processing parameters that are truly critical. Both the experience and expertise of the technical project team and efficient experimental design software are crucial for this approach. Pramod 

. noted that “though design of experiments is not a substitute for experience, expertise, or intelligence, it is a valuable tool for choosing experiments efficiently and systematically to give reliable and coherent information” (6). The FMEA allows the operator to capture the knowledge and decide which areas of the process are most critical and require experimental investigation; the DoE will then validate this thinking statistically and quantitatively. 
	

	Key factors in the design of experiments 

	A crucial factor in the pre-screen and full experiment design for topical formulations is that experiments are conducted using equipment that is representative of larger-scale equipment in order to derive meaningful qualitative CPP data. At MedPharm, IKA LR1000 lab reactors are used, which allow for the control of all typical processing parameters. This approach is crucial to avoid generating “noise” and ensure the quality of the output and the robustness of the resultant control strategy. The understanding of the influence of scale on the CPP from high-quality experimental work conducted on small scale forms the basis of any future scale-up work and technical-transfer activities. 

	A third factor is the identification of what six sigma calls the key process output variable(s) (KPOV), the variables that determine success, and using evaluation criteria to establish whether the method employed to measure the KPOV will detect critical failure. If the answer is no, there needs to be a mitigation plan to change this. The output of any experimental work will only be as good as the analytical method allows. Topical pharmaceutical products require an array of analytical techniques to evaluate their quality ranging from the commonplace such as high-performance liquid chromatography (HPLC) and viscosity testing to more sophisticated methodology such as rheological evaluation, accelerated stressing to show the potential for separation (7), and in-vitro release testing to check that there is no change in the release/thermodynamic activity of the drug from the formulation. 

	The design space is defined as the multidimensional combination and interaction of input variables and process parameters that have been demonstrated to provide assurance of quality (8). It is the boundary of the process parameters to which a product can be made and satisfies the QTPP and CQAs for the product. It is important to stress that there is not a hard border (9) with the knowledge space and that defined process inputs are as important as the measured output. Interpretation of the DoE should point to CPPs and any key interactions between these parameters. Analysis of variants (ANOVA) plots, which show both the mean and the distribution of date around each mean, are particularly useful in this respect as they will clearly show the significance of a parameter or interaction and associated variability. Many experimental design software packages (e.g., SPSS, Minitab, JMP, and Design Expert) are now sophisticated enough to identify statistically significant effects across a range of parameters and present them in a concise graphic that decision makers find easy to understand. 

	An important point is that any process change made within the design space is not considered a regulatory change (10), and hence, has a direct bearing on the flexibility of any future manufacturing process. The more that is known about the boundaries of success and failure, the better the control strategy will be and the less impact that changes in excipient suppliers and specifications will have upon a product over its lifetime. 

	The control space represents a range of critical parameters within which the process will yield an assured output to meet the CQAs and target specification at all times. Getting from the design space to the control space can be achieved through the further use of experimental design; typically, two to five factors in full, or fractional factorial, or other surface response design informed by the data from previous experiment design work. A useful guide to DoE can be found online in the form of the engineering statistics handbook (10). 

, accessed June 26, 2018. 
	2. Eliot et al., 

. 1 (1) 105–122 (2013).
	3. M.E. Aulton and K.M. Taylor, 

(Churchill Livingstone Elsevier, Amsterdam, Netherlands, 4th ed., 2013) p. 449. 
	4. Editors of 

, “A Troubleshooting Guide for Topical Drug Manufacturing: Consider critical process parameters and strategies to optimize the manufacturing process,” Pharmaceutical Technology 36 (11) (2012).
	5. R.K. Chang et al., 

. 15 (3) 674–683 (2013). 
	6. K. Pramod et al., 

. 6 (3) 129–138 (2016). 
	7. D. Lerche and T. Sobisch, 

Journal of Dispersion Science and Technology 

, Step 4 (August 2009).
	9. Quality by Design & Design for Six Sigma ISPE Breakfast Seminar Toronto (May 2009) Murray Adams GSK.
	10. NIST/SEMATECH, 

QbD Strategies to Secure the Scale-Up of Semi-Solid Topical Formulations

The human skin protects the body from physical, mechanical, and chemical insults while preventing endogenous water loss. This function is predominantly achieved by a thin (10–30 µm) cornified outermost layer-the stratum corneum (SC)-generated through terminal differentiation of the basal epidermal keratinocytes. The stratum corneum protects the human body, but also severely limits drug delivery into and across the skin. The historical and theoretical understanding of the type of compound that will permeate the skin is based on the “500-dalton rule” (1, 2), where it is assumed that most compounds permeating the SC have a relatively small molecular weight (<500 Da) and are moderately lipophilic (log P 1–3.5) (3). Such an understanding was, however, based on the assumption of transdermal delivery across healthy skin with an intact barrier. Yet, according to the data from Citeline’s Pipeline database, 7% of topical medicines for the treatment of skin disease contain drugs with molecular weights from 600–1000 Da. Tacrolimus and pimecrolimus (804 Da and 810 Da, respectively), for example, are the two most well-known compounds that appear to contradict the “500-dalton rule.” Nevertheless, at present, there is no approved topical formulation containing a drug with a molecular weight of more than 1000 Da.

	The literature consists of publications claiming topical delivery of macromolecules. Why the discrepancy? Critical evaluation of this literature suggests that some of the methodologies can be misleading and may contain artifacts that make translating to a clinical significance difficult. One of the challenges when attempting to understand passive topical delivery of compounds is establishing a model or testing system that mimics the 

 condition. This task is even more challenging when the size of the compounds exceeds 1000 Da because many of the analytical tools used to detect and quantify small molecules are problematic for larger molecular weight compounds. Many researchers have used animal models, such as rodents and/or minipigs, either 

, to assess the passive delivery of large molecules. It can be argued that these models have yet to show good correlation to the clinical situation as proven by the lack of products on the market. In fact, many of these studies have been shown to be misleading as a result of inappropriate experimental design or a lack of contextualization of this limitation in the study conclusions.

	There have been, nonetheless, some glimmers of hope when polysaccharides, proteins, and oligonucleotides are explored. Hyaluronic acid (HA), a naturally occurring polyanionic polysaccharide up to 1000 kDa, is found in the skin and has been shown to be a key molecule involved in skin moisture due to its capacity to retain water (4). There have been many publications investigating the topical delivery of HA. One of the earliest and most comprehensive studies examined the topical delivery of HA following application to the human forearm 

 (5). Tritiated HA was detected in the dermis just below the epidermis, and the authors took many steps to eliminate artifacts to ensure the observations were real and valid. Since publication in 1999, the group did not report further findings on HA. However, there have been many other studies claiming to have shown HA permeation through human skin, yet the topical delivery HA continues to be taken with scepticism (6–11).

	In addition, latex proteins (14-52 kDa) have been proposed to cause hypersensitivity allergen-based reactions (12). Latex proteins are larger molecular weight compounds (14–52 kDa); hence, it is interesting how these proteins can produce allergic responses in the skin. It is presumed that the proteins are able to cross the skin barrier to elicit the response. In fact, there are studies that show latex proteins were able to penetrate excised human skin, and that exposure induced an IgG1 response 

	There have been several intriguing studies exploring the topical delivery of oligonucleotides within the past decade. Experimental studies using a nuclear factor kappa B decoy oligonucleotide (13 kDa) showed initial promise as a potential topical treatment for atopic dermatitis. The product progressed as Avrina into Phase1/2 (14) but despite the clinically positive data, it does not seem to have progressed further. Some of the constraints related to modifications of antisense molecules and their fixed structural nature may explain this lack of progression. More recently, topical application of a tumor necrosis factor (TNF)-α suppressing antisense spherical nucleic acid showed that treatment with the highest dose resulted in a statistically significant decrease in TNF messenger ribonucleic acid (mRNA) expression in psoriatic skin (15). This technology will be interesting to watch because it appears there are several similar compounds in early development.

	Aptamers are a subclass of large molecules that have been shown to have high binding affinity and selectivity with the ability to disrupt protein–protein interactions. Currently, it is not possible to disrupt these protein–protein interactions with traditional small molecules. Aptamers, therefore, represent a new class of molecules that could have antibody-like binding affinity with the possibility of topical delivery. Evidence from the literature supports their rapid clearance from the systemic circulation, thus limiting unwanted systemic side effects and restricting the biologic effects of topically administered aptamers to the skin (16). Interestingly, aptamers offer significant conformational plasticity and flexibility. Moreover, their structure can be modified without the loss of significant activity (17, 18).

	In a recent publication, researchers at GlaxoSmithKline, University of Reading, and MedPharm showed, for the first time, that a 62-nucleotide (20,395 Da) RNA-based aptamer, highly specific to the human interleukin (IL)-23 cytokine, permeated intact human skin to therapeutically relevant levels in both the epidermis and dermis (19). This observation was particularly surprising considering the compound was 40 times larger than what is commonly accepted as possible for passive topical delivery in the skin.

	In the study, the authors used multiple approaches to demonstrate the topical delivery of the aptamers, including fluorescently labeled aptamer, confocal microscopy, and a novel dual hybridization assay that used capture and detection probes with oligonucleotide precipitation to be able to quantify the aptamer at picomolar levels. They showed the IL-23 aptamer delivered into the skin was significantly above the cellular IC

 in the dermis) when treated topically using a simple cream formulation. This portion of the study used freshly excised human skin and a diffusion cell commonly referred to 

 penetration/permeation (IVPT) and is considered the “gold standard” for assessing topical delivery for both biopharmaceutical companies and regulatory agencies. To confirm the IVPT delivery and to help visualize this delivery, confocal microscopy was introduced on sections from the IVPT study, and this aptamer appeared to localize to the intracellular and extracellular compartments within the viable epidermis (see 

). It was interesting that this observation was noted to confirm a previous publication, showing the uptake of a different aptamer into primary human keratinocytes (20). From this independent observation, it could be extrapolated that intracellular and extracellular targets are possible with this technology.

	To ensure the delivery observed with the other techniques was at therapeutic levels and the aptamer was bioavailable, the authors developed a Th17 mediated biological model using 

 human skin and showed the IL-23 aptamer was able to suppress IL-17 and IL-22 mRNA production (see 

) following topical application. Interestingly, there may be some structural commonalities between HAs, latex allergens, oligonucleotides, and aptamers, which potentially explain the positive observations for topical delivery; however, this hypothesis requires further investigation. Nevertheless, if this 

 work were to be confirmed clinically, this result could present a major breakthrough in dermatology and topical drug delivery as it could open new areas of research and potentially targets that are not accessible using traditional small molecules.

	A development strategy for macromolecules

	There are several critical experimental parameters required to ensure robust, artifact-free results to allow for improved translation to clinical situations. Some of these experimental parameters are the use of human skin with an intact or uncompromised barrier, clinically relevant dosing volumes, a validated highly sensitive analytical method for extraction and quantitation of the compound, and potentially, confirmation of biological activity and structure using human skin.

	The first step is to develop an analytical method to ensure the detection, analysis, and quantitation of the macromolecule is fit for purpose and free from interference. Having duplicate analytical methods using alternate techniques is an ideal approach to further confirm that the observations are valid. As with any topical program, it is imperative that proper preformulation studies occur to establish compound and formulation stability both from a chemical and biological activity sense. The next step is to screen compounds and formulations for passive topical delivery using human skin 

. The final and perhaps most crucial step is to ensure the macromolecule is tested for biological activity ideally using 

 skin to assess target engagement from a topical application. Given the counterintuitive challenge of proving topical delivery with macromolecules, study designs tend to include a majority of controls (both negative and positive) to ensure an unbiased, non-artifact, and robust result. 

Principles of Route-to-Route Extrapolation for Risk Assessment

 (Eds T. R. Gerrity, C.J. Henry) 93-127 (Elsevier, 1990).
	2. R.O. Potts and R.H. Guy, 

Transdermal and Topical Drug Delivery: from Theory to Clinical Practice

 (Pharmaceutical Press, 2003).
	4. E. Papakonstantinou, M. Roth, and G. Karakiulakis, 

 4 (3) 253-258 (2012).
	5. T.J. Brown, D. Alcorn, and J.R. Fraser, 

 113 (5) 740-746 (1999).
	6. A. Torrent et al., 

 18 (Supp 2) S229 (2010).
	7. B. Birkenfeld

 33 (25) 5947-5954 (2012).
	10. H.J. Lim et al., 

Colloids and Surfaces A: Physicochemical and Engineering Aspects

 94 (1) 634-641 (2013).
	12. E. Proksch, A. Schnuch, and W. Uter, 

 126 (8) 1792-1803 (2006).
	15. KT. Lewandowski 

 137 (9) 2027-2030 (2017).
	16. A.D. Keefe, S. Pai, and A. Ellington, Aptamers as therapeutics. 

 9, 537-550 (2010).
	17. Y.S. Kim and M.B. Gu, Advances in Aptamer Screening and Small Molecule Aptasensors. 

. 140, 29-67 (2014).
	18. Y.I.A. Nakamura and S. Miyakawa, 

, 138 (2) 282-290 (2017).
	20. R. Doble et al., 

	When referring to this article, please cite it as M. Brown and J. Lenn, "A Perspective on the Topical Delivery of Macromolecules,"

Pharmaceutical Technology APIs, Excipients, & Manufacturing 

 is chief scientific officer and co-founder of MedPharm; and 

is senior vice-president for MedPharm’s US operations. 

The human skin protects the body from physical, mechanical, and chemical insults while preventing endogenous water loss. This function is predominantly achieved by a thin (10–30 µm) cornified outermost layer-the stratum corneum (SC)-generated through terminal differentiation of the basal epidermal keratinocytes. The stratum corneum protects the human body, but also severely limits drug delivery into and across the skin.