Martin Koeberle, PhD, is senior manager, analytical development, at Hermes Pharma, Georg-Kalb-Strasse 5-8, 82049 Pullach, Germany.

Martin Koeberle

Quality has always been a key focus for the pharmaceutical industry, with drug developers traditionally using late-stage quality testing to ensure medicines are safe, reliable, and effective for patients. However, there is a growing appreciation within the industry that this approach may not be the most effective or efficient to safeguard quality. After all, quality cannot be “tested into” a product--it should be there by design from the very beginning of product development right up to the manufacturing process.
	
	The adoption of advanced process analytical technology (PAT) throughout the pharmaceutical value chain is now enabling forward-looking companies to move away from traditional quality testing methods and instead, use systematic, data-driven strategies to deliver quality outcomes. One such concept that is widely used in other industries and has been gaining traction within the pharmaceutical industry is quality by design (QbD). 

	In this article, the authors discuss the advantages of adopting a QbD approach, and how this concept can be applied throughout formulation and manufacturing workflows.

	Moving away from traditional product development

	QbD essentially involves designing quality into workflows upfront. The concept is aligned with the “Design for Six Sigma” paradigm, originally developed in the automotive industry. It represents a radical shift away from the empirical-based methods traditionally used in product development and manufacturing and has since been implemented across a variety of other manufacturing environments. 
	
	With a QbD approach, the product’s essential attributes are first defined. Risk and data analysis are then used to understand how the processes affect the product characteristics. QbD provides a robust framework for the design and implementation of processes that achieve a consistent level of quality and meet predefined standards. PAT plays a crucial role in the implementation of QbD--it allows the development and manufacturing processes to be monitored and optimized through real-time controls. Using this data-driven approach to improve manufacturing and development workflows can save time and money in the long run.

	The main issue with late-stage quality analysis is that it only detects and removes substandard products--it doesn’t prevent them from being created in the first place. As pharmaceuticals become increasingly complex, it’s more important than ever that quality is designed into the products from the initial concept to ensure patient safety.

	In recent years, key regulatory authorities, including FDA and the European Medicines Agency (EMA), have actively encouraged QbD and PAT through a number of initiatives. FDA recently began updating existing regulations to encourage the adoption of new analytical technologies that facilitate risk-based processes, and in the past few years, the agency has also released two QbD case studies for abbreviated new drug applications. Recognizing that QbD offers significant improvements in quality, regulatory authorities are now beginning to insist that pharmaceutical developers and manufacturers adopt QbD throughout the value chain, especially given that the technology needed to implement this approach is widely available.

	Stages of a QbD-based product development

	There are extra steps involved in the initial QbD-based product development. Additional upfront investment in terms of time, money, and resources will be required, but it protects against variability later on; it also minimizes risk, reduces waste, and saves time in the long run.
	
	Through detailed characterization of materials such as APIs and excipients, the critical quality attributes (CQA) are defined. CQAs are the chemical, physical, biological, and microbiological attributes that may significantly affect the quality of the finished product. This information enables the identification of critical process parameters (CPPs), which are the key variables that affect the production process. CPPs can be monitored to detect any deviation in manufacturing processes, and active control of these CPPs can counteract these deviations in order to ensure product quality and that CQAs are met.

	For example, Hermes Pharma was involved in developing an effervescent cough and cold formulation containing three APIs--acetaminophen, caffeine, and phenylephrine. Given the low concentration of phenylephrine, in the final product, a key challenge was to ensure content uniformity for all three APIs so that the product delivers reliable and reproducible efficacy when taken by patients. A QbD approach was adopted to improve the blend homogeneity of the APIs within the final product.

	The formulation involved three separate blending steps. Prior to adopting QbD, the blending time and speed would have been based on the formulator’s experience, rather than being verified by experimental investigation. With the QbD approach, monitoring of the mixture at various locations within the container was carried out using high performance liquid chromatography (HPLC) for the initial blending steps and near infrared (NIR) spectrometry for the final blending process (see 

). This data-driven approach meant that only when acceptable levels of blend homogeneity had been reached would the blending process continue to the subsequent step, resulting in more consistent levels of product quality.

	In most cases, transitioning from laboratory scale to mass production is rarely straightforward. However, scaling up is often significantly easier when QbD is employed, because the manufacturing process is better understood. It reduces the risk of encountering unexpected problems that negatively impact on production throughput and timescales.

	In the effervescent cough and cold remedy example mentioned earlier, the adoption of QbD significantly reduced the time required to optimize the scale-up process. By monitoring the CPPs throughout the multi-step process, the formulators were able to gain an understanding of the main contributions towards blend inhomogeneity. And by pinpointing issues specific to a particular step in the process, attention could be given to the factors involved, which helped overcome the challenges during scale-up. Without QbD, scaling up would have been based on a trial-and-error approach and would have likely taken considerably longer to optimize.

	Another benefit of having a good understanding of the CQAs and CPPs is the ability to easily establish a formulation and process design space that ensures compliance with the quality target product profile. Essentially, the need to register later adjustments with regulatory bodies after large-scale production has begun can be avoided, which is important as such revisions can lead to significant delays when it comes to bringing a product to market. Moreover, by using QbD, risk can be managed more effectively and it will be easier to determine exactly which product changes will require additional regulatory submissions.

	QbD also provides the flexibility to operate within a broader design space when changes to the manufacturing process occur that are outside one’s control. For example, at Hermes Pharma, hot melt coating is used to mask the unpleasant metallic taste of acetylcysteine in the production of a mucolytic product. However, it was discovered that the particle size distribution of the API varied between batches of the raw material--a situation that often occurs, because in many cases, suppliers do not fully specify the particle size distribution of their APIs. As a result, if the same amount of coating was applied to each particle, the coating thickness would vary and the unpleasant taste could come through. However, because QbD was applied together with PAT, the formulators were able to screen the particle size distribution of the incoming API and categorize these into distinct classes with an associated coating method. This approach provided a more consistent coating layer irrespective of particle size.

	If only one process was designed for a very narrow particle size range, the unsuitable particle size fractions would have to be discarded by sieving, which is of course not cost efficient. Alternatively, the formulator would have had to agree upon a very specific particle size with the supplier, which would incur additional costs. Working within a predefined and regulatory-approved design space avoids the need to apply for additional regulatory approval when changes in raw material specifications are encountered.

	QbD approaches based on PAT also facilitate real-time release testing. This approach gives manufacturers more timely information on product quality and means that any manufacturing problems can be dealt with faster, in a more informed manner.

	QbD also puts in place the quality assurance framework necessary for continuous process improvement. Compared to traditional batch testing, continuous process verification is much more able to push the processes towards optimization and maintain manufacturing parameters within acceptable limits. It also reduces the sampling effort, further boosting process efficiency.

	All of these process improvements mean that patients will benefit from an enhanced level of quality in their medicines, ensuring that the medicines are consistently safe and effective. As a result, customers benefit from a more reliable supply capability, preventing out of stock situations where product batches need to be destroyed due to poor quality.

	A stepwise approach to implementing QbD 

	There are three key stages in implementing a QbD approach to pharmaceutical development and production (see 

		Initial product profiling, risk analysis, and CQA/CPP determination during early development

		The design of experiments and the definition of possible and meaningful design spaces during the formulation/manufacturing development phase

		The establishment of a manufacturing and control strategy during scale-up and routine production. Manufacturing is also supported by ongoing process verification to ensure continued operation within the defined parameters.

	At Hermes Pharma, the first step in our development programs is to establish the quality target product profile (QTPP) for the product. The pharmacological and patient requirements as well as quality are considered in advance of manufacture, hence, providing a starting point for identifying CQAs in the next step. This is followed by risk-based characterization of APIs and excipients, including an assessment of factors such as particle size distribution, morphology, processability, and batch-to-batch or supplier-to-supplier variation. 

	Risk analyses are also performed to determine CQAs for the pharmaceutical product. The goal is to study and control any factor--be it physical, chemical, or microbiological--that could affect the quality of the product. It’s important to study APIs and excipients both in isolation and in combination to define their characteristics in as much detail as possible. Failure to keep these factors within the set boundaries creates the risk of manufacturing substandard products that can lead to manufacturing downtime or increased costs, and potentially harm supplier relationships. CPPs are variables contributed by the manufacturing process with the potential to impact the CQAs, resulting in a low-quality product. The greater the potential CQA impact, the more closely the CPP has to be controlled.

	Next, the design space is defined. For example, the design space could be the relationship between specific process inputs and the CQAs. If the design space is developed accurately and all essential aspects of the process are within the design space, then the CQAs must be acceptable. The design space enables manufacturers to demonstrate to regulatory bodies a comprehensive understanding of their process. The regulatory body then approves the design space as an operational range, giving manufacturers the flexibility to operate provided the workflow remains within that range (see 

) and avoid having to submit revised information to the regulatory authorities.

	Then, using the knowledge gained from the previous steps, together with the CQAs and CPPs, a control strategy is established. This control strategy must be defined in order to determine which analytical methods are required and where in the process they will be employed. PAT enables the control strategy to be further improved, as it facilitates the monitoring and control of the production processes in real-time.

	To maintain product quality across all batches, ongoing process verification should be performed. Unlike traditional verification approaches, such as the “three golden batches” method, which focuses on analyzing the first three batches only, continuous process verification does not treat validation as a discrete exercise. This way, the manufacturing processes are operating at the very highest levels of quality for the industry.

	QbD looks set to become increasingly important for the pharmaceutical industry as more and more companies are looking to boost production output, reduce throughput times, and lower costs by shifting from batch production to continuous manufacturing. However, for these “always on” processes, it just isn’t possible to perform quality testing within a series of continuous process steps in the same way as it is following each operation of a batch process.

	Additionally, as FDA, EMA, and other regulatory bodies further scrutinize production processes to ensure the highest levels of safety, QbD will make it easier for companies to demonstrate that they are operating within acceptable limits. As regulatory authorities start to insist that companies design quality into products at every stage of the pharmaceutical value chain, the use of QbD will become ever more important.


	Vol. 41, No. 9
	September 2017
	Pages: s20-s23

	When referring to this article, please cite it as M. Koeberle and W. Schiemenz," QbD: Improving Pharmaceutical Development and Manufacturing Workflows to Deliver Better Patient Outcomes," 

	Martin Koeberle, PhD, is head of analytical development & stability testing; and Wolfgang Schiemenz is senior manager of QbD & Drug Regulatory Affairs, both at Hermes Pharma.

Articles

Implementing quality by design in product design and formulation and manufacturing workflows can help improve efficiency and shorten development times.

QbD: Improving Pharmaceutical Development and Manufacturing Workflows to Deliver Better Patient Outcomes

Tablets and capsules are often coated for various reasons such as to improve their appearance or to make them easier to swallow. Coatings also play an important role in maintaining drug stability, activity, and effectiveness. This article reviews the benefits of solvent-free processes and techniques specifically optimized for coating new dosage forms such as orally disintegrating granules (ODGs). 


  Coating is usually achieved with one of a variety of techniques but regardless of the equipment and coating process used, there are several common challenges and considerations that will influence coating success. Firstly, the coating should be immediately stable (e.g., without the need for a curing step), reproducible, and cost efficient. Secondly, the coating process used during formulation development must be easy to transfer to manufacturing conditions and ready for scale-up, using excipients that can be easily and safely stored in bulk amounts. 

  In terms of the coating itself, some approaches can lead to structural polymorphism within the coating, which can render it difficult to consistently predict drug-release profiles, bioavailability, and efficacy while also affecting intellectual property (IP) protection and infringement. Such variations can arise during the coating process or develop over prolonged storage; hence, it is important to select a method where these factors have been studied and are well understood. 


  Liquid coating is a reliable and well-established technology that is still the industry standard for coating many APIs but it has several disadvantages. Firstly, liquid coatings must be dried to form a solid final product, a time-consuming process that often requires an increase in temperature that incurs additional cost. The use of heat may also cause thermal stress, leading to API degradation. When using liquid coating, it may also be necessary to spray the tablet core or seed particle with several layers to create the final dosage form or intermediate with the  desired characteristics. This step adds time and complexity to the procedure and can have detrimental effects on the quality of the final product by causing variability in the crystalline form or an uneven, heterogeneous coating, thereby, leading to unwanted polymorphism or defects. Besides affecting stability, performance, and efficacy, such structural variations can also have important IP implications. For example, it may be that certain polymorphic forms of the API are protected by patents, and thus, must be avoided. For these reasons, solvent-free coating technologies are of interest.


Solvent-free coating offers several advantages over the use of solvents by reducing the time and costs associated with drying liquid coatings. It can also reduce the thermal stress imposed on the API and simplify the coating process by reducing the number of steps required to reach the final product. These methods include compression coating, electrostatic spray powder coating, supercritical fluid coating, photocuring, and hot-melt coating (HMC).


Generally, when coating tablets, the APIs are initially granulated before being blended with further excipients, compressed into tablets, and finally coated. It is also possible to use these uncoated tablet cores in a second compression step designed to surround them with an additional layer containing certain functional excipients, such as a second API or an additional dose of the same API. The method is well suited for producing multi-stage treatments that combine two or more APIs, particularly if they are incompatible and must be physically separated within the dosage form, or if their mode of action requires that they are administered to different parts of the digestive system.

Perhaps the biggest drawback of compression coating is that in the second compression step, it can be difficult to position the tablet core accurately in the center of the tablet, which is important for reproducible formulation and reliable pharmaceutical performance. Various efforts have been made to circumvent this problem, leading to the development of new approaches such as the one-step dry-coated (OSDrC) tablet manufacturing system, which can reduce the likelihood that the compaction process will negatively affect tablet performance (1). The excipients must also be carefully selected. Using appropriate fillers and/or binders is essential to successfully creating a solid dosage form by compression. Finally, because many APIs and excipients are simply not compressible in their crystalline or amorphous forms, an alternative solvent-free coating methodology must be selected. 


  An advantage of spraying liquid coatings onto tablet cores is that the final product is visually appealing, creating a clean, polished, and homogenous coating. To pseudo-replicate the process but without the need for solvents, researchers have developed ways to replace liquids with dry powders. One such approach, electrostatic spray powder coating, uses powdered raw materials that have been electrostatically charged through the application of high voltage. These materials are sprayed onto a neutrally charged seed pellet and will adhere to the surface by electrostatic forces. The coating is then fixed in place by temperature or exposure to infrared radiation for approximately 1 to 2 minutes (2, 3). The main advantage of the process is that the coating forms and hardens quickly while the final product can easily be customized in terms of properties and thickness by adjusting the composition and amount of mixture sprayed on the 
tablet core prior to fixation.

For the process to work, however, the coating and tablet core must possess certain conductive properties, or be modified using additional steps to introduce electric charge. To circumvent this issue, the conductivity of the core particle can by increased by wetting it with water to decrease resistance, although this method introduces moisture into the tablet and can lead to instability. Another way to improve conductivity is to modify the surface of the API substrate using polar groups. This method is usually achieved by dissolving the polar groups in a volatile solvent before spraying the solution onto the tablet core, thereby, reintroducing a solvent step into the process. Either way, the electrostatic attraction between the solid substances tends to be weak, hence, making it difficult to produce a thick coat without multiple rounds of spraying and fixing. The fixing step also requires moderate heating, hence, increasing the energy required to carry out the method, which potentially leads to API instability. Current research is focused on finding new ways to increase the conductivity of the tablet core and lower the glass transition temperature of the coating polymer (4).


A fundamental stage in the coating process is the fixation of the coating to form a solid, protective layer. In many cases, this step involves the use of heat. This increase in temperature requires costly energy and many APIs become more susceptible to degradation as temperature increases. Photocuring attempts to avoid these problems by employing a method of fixation dependent on light energy rather than heat. As such, it does not require any heating or drying steps and is ideal for temperature-sensitive APIs. The process hinges on the use of light energy to trigger a polymerization reaction to convert a liquid or granulate surface coating into a solid material. The technique has been successfully used in pharmaceutical coating to create immediate and sustained-release dosage forms (5). 

While coating can proceed rapidly, oxygen in the ambient atmosphere can slow or halt the reaction and in some cases, must be purged from the formulation or manufacturing system using nitrogen, which adds complexity and cost to the process. While this technique has clear applications for heat-sensitive compounds, it is unsuitable for excipients or APIs that are photosensitive. Moreover, the process is expensive and is still in early phase development.


One way to avoid the downsides of traditional liquid coating is to use supercritical liquids, which have properties between liquids and gases along a continuum that can be manipulated. Like liquids, they can be used to dissolve the excipients and API, thereby facilitating the mixing and formation of a homogeneous coating. Carbon dioxide is most widely used because of its low critical temperature (31 °C) and pressure (74 bar), making it easy to manipulate under manufacturing conditions. The major benefit compared with standard liquid coating is that supercritical liquids can be rapidly transitioned to the gaseous phase by a reduction in vessel pressure, causing the excipient mixture to precipitate and form a solid. Unlike solvent-based methods, the technique leaves no residue and does not require any heating or drying steps. 

Supercritical liquid coating has so far proven to be effective for coating of fine particles. The supercritical liquid solvating power can be controlled by small changes in temperature or pressure, thereby, enabling the fine-tuning of coating properties (6). However, for the process to work effectively, the core particle onto which the coating forms must be insoluble in the supercritical fluid. In addition, high-pressure equipment is required to carry out the process, which adds further cost to an already expensive approach. 


  Despite the various methods available for coating traditional dosage forms that improve taste, mouth-feel, and ease of swallowing, many patients still complain that swallowing tablets and capsules is painful and unpleasant. According to a large pan-European study, more than 25% of a particular general practitioner’s patients complained of swallowing difficulties (7). 

Orally disintegrating granules (ODGs) are user-friendly dosage forms designed to meet patients’ needs, improve adherence, and increase treatment effectiveness. ODGs provide value for patients as they rapidly dissolve in the mouth, thereby circumventing any issues with swallowing. They can also be quickly and easily administered without the need for water and are much simpler to administer to children and the elderly. As they are not constrained by physical size, ODGs can also be formulated to contain larger amounts of API and/or multiple APIs in a single dose, simplifying treatment regimens. 

The effective coating of APIs for ODGs is dependent on many factors. Solvent-based coating is applicable for ODGs, but suffers from heating, drying, and potential stability complications while most organic solvents inherently pose various health and environmental risks. Although several solvent-free approaches have been used with some success for coating ODGs, especially electrostatic spray powder coating, most of them are time-consuming and expensive, often leading to a final product that does not adequately mask the bitter taste of most APIs. In addition, many approaches are not practical for coating large numbers of small granules. For example, compression coating is better suited for coating larger tablets. While new approaches such as supercritical liquid coating and photocuring may offer an effective way to coat ODGs, both methods are still experimental, expensive, and impractical for scale-up. 

In an effort to avoid the use of solvents and identify coating methods that might be suitable for ODGs, there has recently been significant research and development into novel methods of pharmaceutical coating. One such technology, hot-melt coating (HMC), is gaining popularity as it offers short processing times (typically under 1 h) and significantly lowers costs compared to other solvent-free methods. HMC avoids the use of solvents by instead utilizing molten mixtures, which are sprayed onto solid particles usually using a fluid bed coating setup. This mixture then solidifies upon cooling to form a protective coating around the seed particle containing the API. The excipients most suitable for HMC include lipids, waxes, fatty bases, and hydrogenated vegetable oils, as they are easy and quick to process, cost-effective, and have relatively low melting points (below 100 °C). 

Compared to other solvent-free coating methods, HMC is particularly well suited for taste masking and careful manipulation of API release (8), making it ideal for formulating ODGs. By carefully selecting excipients based on molecular flexibility, hydrophobicity, melting point, molecular weight, and rigidity, as well as optimizing their relative concentrations, it is not only possible to fine-tune the release profile of the API to provide sustained or delayed release but also to provide immediate release products (see 

Although HMC can be carried out using a variety of methods and common manufacturing equipment including fluidized bed and spray coating, optimization of the protocol for each API/excipient mixture takes skill, knowledge, and experience. Air-flow, spray rate as well as spray and product temperature must be carefully controlled during the process to enable to successful melting and solidification of the excipient mixture to create a pleasant tasting, homogenous coating with the desired API release characteristics. These parameters are important and when the process is carried out correctly, dosage forms coated using HMC do not suffer from some of the polymorphic variations (and associated challenges) that other coating methods may exhibit. 

It is important to create formulation and manufacturing processes that embody quality-by-design (QbD) principles to enable a seamless transition during scale-up, and at the same time employ process analytical technology (PAT) that measure HMC parameters, such as coating thickness, to enable their accurate correlation with final product characteristics. Offline methods, such as scanning electron microscopy, can also be used to assess product quality, because they enable the surface and coating layer of ODGs to be visualized (see 


  While HMC is a suitable API-coating method for producing ODGs with delayed- or sustained-release profiles, optimizing the process for producing rapid-release medications such as analgesics has proven to be a challenge. A team at Hermes Pharma is collaborating with the Research Center Pharmaceutical Engineering GmbH (RCPE), INNOJET Herbert Hüttlin, CREMER Oleo GmbH & Co. KG, and the Karl Franzens University to identify and develop HMC parameters for producing such fast-acting medicines. The goal of the project is to develop new pharmaceutical formulations and manufacturing processes, and engineer new approaches and manufacturing controls for effective masking of the unpleasant tastes and smells associated with most APIs. 

Preliminary results from the ongoing project have highlighted that the production of fast-release medicines using HMC is feasible and amenable to scale-up. The transfer of the technology into production is underway. In several proof-of-principle studies, the parameters for formulating ODGs containing fast-acting APIs were successfully identified. The proof of concept studies included many different lipids, such as mono-, di- and triacylglycerides with varying carbon chain length (C14 to C22). In addition, more than 10 emulsifiers were tested. The resulting dosage forms provide rapid dissolution characteristics, similar to those of film-coated immediate-release tablets. Importantly, the granules are pleasant tasting and easy for patients to ingest. The HMC approach at Hermes is being extended to formulate other fast acting medicines such as ODGs, with the method expected to be fully validated and available for use by pharmaceutical developers in the near future.


  1.	Y. Ozeki, Y. Watanabe, H. Okamoto, and K. Danjo,

 21 (7) 1177-1183 (2004).
  2.	S. Bose and R.H. Bogner, 

. 12, 115-131 (2007).
  3.	M. Qiao et al., 

 83, 293-300 (2013).
  4.	M. Qiao et al., 

76 (2) 304-10 (2010).
  5.	S. Bose and R.H. Bogner, 

. 8 (3) E57 (2007).
  6.	Y. Wang, R. N. Dave, and R. Pfeffer,

 28, 85-99 (2004).
  7.	O. Andersen et al., 

. 115 (8) 947-9499 (1995).
  8.	S.G. Sudke and D.M. Sakarakar, 

3 (1) 16-26 (2013).
  9.	J.P. Kennedy and P.J. Niebergall, 

. 3 (1) 95-101 (1998).
  10.	V. Jannin et al., 


Detlev Haack, PhD, is director of R&D, and Martin Koeberle, PhD, is senior manager, analytical development, both at Hermes Pharma, Georg-Kalb-Strasse 5-8, 82049 Pullach, Germany.

The benefits of solvent-free processes and techniques specifically optimized for coating new dosage forms such as orally disintegrating granules are reviewed.