Cynthia A. Challener is a contributing editor to Pharmaceutical Technology.

Cynthia A. Challener

Although drug repurposing has been receiving growing attention in recent years, drug companies have taken this approach, at least serendipitously, for decades. Viagra, originally developed as an angina treatment, is one of the most well-known examples. Today many groups are focused on identifying new indications for existing drugs including academic researchers; nonprofits such as Cures Within Reach; and government institutions including the National Center for Advancing Translational Sciences in the United States and the UK Medical Research Council. Other drug repurposing research ventures include public-private partnerships like the Center for Drug Repurposing (operated jointly by Ariel University [Israel] and Drug Rediscovery Ltd) and the Drug Repurposing Hub (a collaboration between the Broad Institute Cancer Program, the Center for the Development of Therapeutics, and the Connectivity Map group); and companies such as Biovista, GVK Bio, and NuMedii. In 2012, it was estimated that drug repurposing, repositioning, and rescue (DRPx) efforts accounted for 30% of FDA-approved new drug products (1). In addition, in 2015, Biovista President and Co-founder Aris Persidis reported that these drugs accounted for approximately 25% of pharmaceutical industry revenues (2).

	The known safety of existing APIs is one of the biggest benefits of using them for new indications discovery, according to Gini Deshpande, founder and CEO of NuMedii, a drug-discovery company that applies its proprietary big data platform based on integrative genomics, network-based methods, large-scale machine learning, and chemoinformatics to improve efficacy. One application of NuMedii’s technology is to identify new indications for existing APIs. “Safety data for these APIs can be leveraged to expedite the development process via the 505(b) 2 approach in the United States or the hybrid approach in the European Union,” she explains.

	For instance, a new chemical entity would normally take 10-15 years to get to the market, whereas an existing API being developed for a new indication can reach the market in five to eight years depending on the indication for which it is being developed. “This reduction in development time results in a significant cost and time savings for companies,” Deshpande says.

	Success rates in Phase III trials are also much higher for products developed via 505(b) 2-type approaches compared to traditional approaches. Evaluating new indications for existing APIs can also expedite the discovery and development of new therapies for rare diseases and for other conditions where existing APIs can be used to validate novel biology and thus speed up the discovery process, according to Deshpande. She notes that there are approximately 2800 drugs and more than 4000 compounds that have been discontinued at Phase II development, so there is a rich pool of existing APIs with which to work. 

	One of the biggest challenges to repurposing of existing drugs is in finding that novel indication. NuMedii’s big-data, artificial intelligence (AI)-driven methods provide a distinct advantage with such an effort, according to Deshpande. “Because our technology uses both data- and knowledge-driven approaches to find the broadest possible landscape of new indications, we often identify out-of-the-box new indications for existing APIs that have previously been undiscovered,” she explains. 

	Ensuring market exclusivity and reimbursement level are two other key challenges with this approach. These issues are further compounded by geographic-specific issues, because rules differ in the EU, the US, and the rest of the world. 

	Four major factors need to be considered when looking for new indications for existing APIs, according to Desphande: scientific rationale, clinical viability, commercial viability, and regulatory path. 

	Scientific rationale involves ensuring that the biology being modulated by an existing API is relevant and assessing whether formulation or delivery for the new indication is feasible. Clinical viability relates to whether there is a specific unmet need being solved by the API, the fit within a treatment paradigm, and the risk/benefit equation for using an existing API for the new indication. Commercial viability factors include the nature of the competitive landscape, reimbursement issues, and target pricing. Questions regarding the regulatory path are focused around the type and amount of available existing data that can be used for approval, and what additional studies need to be conducted and at what cost for the various phases of development.

	In the past, drug repurposing relied on serendipity, according to Deshpande. “Case reports of off-label uses and side effects of drugs reported (either approved or those in development) were often a good starting point to look for new indications/uses for these APIs,” she notes. Recent approaches have now made the process more systematic. She points to i

 approaches that span a multitude of both computational biology approaches such as the one NuMedii uses and computational chemistry methods, screening approaches using both cell lines and animal models and literature-based searches, although the latter can have potential issues due to prior art with regards to intellectual property.

	NuMedii began using genomic-data driven methods, but now leverages multiple types of ‘omic’ information, as well as information from the literature, clinical information, and other data types to ensure that the company captures the best possible indications for existing APIs, according to Deshpande.

	“Since 2010, NuMedii has demonstrated that its discovery technology and machine learning algorithms can dramatically impact the time, cost, and success rate of drug development in the fields of inflammation and oncology, thus revolutionizing the drug discovery process as we know it,” she attests. “We have demonstrated the predictive capability of NuMedii’s technology in the areas of inflammation and oncology with many pharma partnerships and projects,” Deshpande adds. One such example is NM-001, an undisclosed cardiovascular drug that has been in use for more than 30 years. NuMedii’s software uncovered wide-spectrum anti-inflammatory activity that was validated in an ulcerative colitis animal model. 

	Even with the successes being achieved with the information and approaches in use today, Deshpande believes that more data needs to be incorporated to ensure that the 

 approaches become more predictive. Policy changes are also necessary, in addition to advancements in the technology to identify new indications for existing APIs, to ensure appropriate incentives are in place to speed the use of existing drugs for new indications. “For instance,” says Deshpande, “the [Orphan Product Extensions Now] OPEN Act--a key part of the 21st Century Cures Act before it was removed from that bill just before being signed into law in December 2016--would have provided an additional six months of market exclusivity to drugs that are repurposed and approved by FDA to treat rare diseases,” she observes. 

	There remains significant pressure on the pharmaceutical industry both from payers and patients to identify new therapies for diseases. “Given the total cost and time to develop new therapies from scratch, and the high failure rates involved, drug repurposing will provide an important avenue for the discovery of new treatments, particularly for rare diseases,” asserts Deshpande. She does not expect such an approach to completely replace traditional approaches, but rather supplement them. 

	“If one looks at developing countries like China and India, where the costs for new drug development will seem prohibitive, drug repurposing, repositioning, and rescue can rapidly create new treatments in these areas. That is precisely why we at NuMedii are proud to pioneer the use of big data and AI with systems biology to disrupt the traditional drug discovery paradigm and create a faster, highly predictive drug discovery engine to help more rapidly bring life-saving medicines to the people who need them most,” she states.

Drug Repositioning: Bringing New Life to Shelved Assets and Existing Drugs

 (John Wiley & Sons, Hoboken, NJ, 2012).
	2. A Persidis, “Myths and Realities of Repositioning,” Arrowhead 4th Annual Drug Repositioning, Repurposing & Rescue Conference (Chicago, IL, May 27-28, 2015).

Volume 41, Number 9
	September 2017
	Pages: 24–25

	When referring to this article, please cite it as C. Challener, “Expediting the Discovery and Development of Drugs," 

Articles

Drug repurposing, repositioning, and rescue can be faster, cheaper, and target more diseases than traditional drug discovery approaches.

Expediting the Discovery and Development of Drugs

Increased competition, the diversification of biologic drug substances, and globalization of manufacturing have accentuated the need for processes that are not only compatible with modular, flexible, smaller plants but also carry reduced total costs. Continuous and intensified processing is an attractive alternative that fits the needs of current and future manufacturing requirements.

	Cell-culture media directly impact the performance of upstream biomanufacturing processes; they dictate the potential productivity and reproducibility of specific processes while also influencing product quality attributes, according to Andrew Bulpin, head of process solutions for MilliporeSigma. “Cell-culture media must be optimized for each process, particularly perfusion processes and fed-batch systems that support higher profile fed-batch processes designed to minimize cost and speed development,” he observes.

	In addition to the cell line and cell-culture process parameters, cell-culture media is a key driver of bioproduction processes. Cell-culture media can have significant impacts on cell growth and viability and also drive protein production and protein quality, according to James Brooks, manager of worldwide media services and media development for BD.

	While significant time and effort are invested in the generation of an optimal cell line--from transfection to isolation and cloning, to full characterization--it is often the case that little thought is given to selection of the optimal medium for the clone, according to Brooks. He also notes that sub-optimal formulations have been shown to have deleterious effects on bioproduction processes and limit the potential of cell lines.

	Cell-culture media provide not only nutrients to the cell, but stimulatory and regulatory effects. “Media can impact cellular signaling pathways and be involved in the regulation of enzymatic activity within the cell,” Brooks explains. For example, metals contained within media formulations, such as manganese, copper, and iron, can serve as co-factors for enzyme activation and play a key role in driving protein titer and protein quality. In addition, sufficient levels of amino acids are required for normal cell growth and function, but also for the production of the correct target protein,” Brooks says. “Limitations in the amino acid compositions in some chemically defined (CD) media formulations have been shown to result in amino acid substitutions and the generation of sequence variants of the target proteins,” he says.

	The inherent diversity and complexity of animal cells limit the applicability of a universal cell culture media that can be used across all different production processes with optimal performance. Each cell line is unique, and to achieve its full potential, it is imperative to match an individual cell line with the optimal cell-culture medium that specifically meets its metabolic requirements.

	From the isolation of Chinese hamster ovary (CHO) cells into culture in the 1950s, Brooks notes, the cell line has diverged significantly, giving rise to a multitude of CHO subtypes and lines used for bioproduction today. “Each CHO subtype and each individual clone has its own specific nutritional and metabolic requirements. Through a medium optimization process, these requirements can be met, allowing the cell line to reach its full potential,” he explains.

	“Customization of a cell-culture medium requires the right tools and a specialized team of experts that can guide clients in the right direction and/or modify an existing formulation or design a new one if necessary in a minimum amount of time,” states Bulpin.

	Facing the challenges of chemically defined media

	Companies have been moving toward the use of chemically defined media to avoid the potential contamination issues associated with animal-derived ingredients. They have been challenged, however, to achieve the increasingly higher volumetric productivity levels now expected for fed-batch systems, according to Bulpin. “Chemically defined media are not necessarily the optimal process solution,” Brooks agrees.

	Nutritional limitations in chemically defined media formulations have been shown to impact protein titer and protein quality. Additionally, chemically defined does not mean chemically pure, according to Brooks. “Contaminants in CD media, such as trace elements present in vitamin stocks and some salts, have been repeatedly shown to impact cell-culture processes, particularly protein quality,” he says. Multiple companies have, in fact, published information where trace element contaminants in their CD formulations have impacted their monocolonal antibody (mAb) titers or glycosylation profiles, according to Brooks. “This issue is particularly pertinent in the development of biosimilar mAbs, where it is critical to match the quality parameters of the originator molecule,” he adds.

	Bulpin notes that nutrients that were traditionally supplied with complex solutions such as hydrolysates need to be now supplied as individual chemically defined chemicals. Some of these nutrients, such as cysteine and tyrosine, have a maximum solubility and stability level below the concentration requirements for optimal performance. “In order to overcome this issue, companies choose different strategies; most commonly, feed solutions are divided in multiple parts, often with extreme pHs, and clients need to add them separately to the bioreactor, leading to potential unwanted pH shifts,” says Bulpin.

	MilliporeSigma developed a single-feed concept in which the nutrients are concentrated to more than 130g/L, allowing a reduction of the volume of feed added to the medium and thereby increasing the volumetric productivity. The feed is added at neutral pH using a simple reconstitution protocol and high solubility, according to Bulpin. It also contains cysteine and tyrosine derivatives that have been shown to release free cysteine and free tyrosine slowly throughout the culture. “This time-dependent release avoids tyrosine depletion, which can lead to sequence variants. It also results in free cysteine release while maintaining a reduced redox environment, which has been related to higher cell growth and productivity,” Bulpin says.

	Poloxamer-188, which has traditionally been available from only a single supplier, is a specific cell-culture ingredient that has caused difficulties in biopharmaceutical manufacturing in recent years. “Significant numbers of lots have resulted in toxic or low shear stress-protectant effects for cells,” Bulpin explains. MilliporeSigma developed an analytical test that allows the identification of the desire variety of poloxamer-188 and is employed by the company to validate each lot of poloxamer-188 to be used in cell-culture media. In addition, MilliporeSigma worked intimately with an alternative manufacturer and has validated this producer as a second supplier for poloxamer-188, according to Bulpin.

	The issues with chemically defined formulations are creating a renewed interest in peptone supplementation for cell-culture media across the biopharma industry, according to Brooks. “While chemically defined remains the ultimate desire, peptones continue to be used as cell-culture media supplements to enhance production and protein quality,” he says. “Peptones are well-characterized and long-established cell-culture supplements and feeds that can enhance cell growth and/or protein titer while helping maintain or achieve desired protein quality. The rich composition of peptones enables them to serve as an optimal nutritional source for cell culture. Peptones have also been shown to inhibit apoptosis and positively affect cell cycles,” Brooks adds.

	The continued move toward perfusion-based production has highlighted the need for more optimally designed formulations that can support the perfusion process. “It is a challenge to achieve steady-state, high-density cultures while maintaining consistent product quality. Most media formulations today are not designed for this application,” states Brooks. The overall cost of the base medium for perfusion processes is also a concern. With higher perfusion rates and medium exchange rates as frequent as 1-2 times per day, significant volumes of media can be consumed during a perfusion process. “It is therefore essential to have a medium that can achieve the desired growth, titer, and quality parameters within a cost-effective and manageable process,” Brooks asserts.

	In response to this need, MilliporeSigma developed and recently introduced a medium specifically for use in perfusion processes. It is designed to reduce the medium exchanges per day in a perfusion process while maintaining or even increasing specific productivity and product quality, according to Bulpin.

	MilliporeSigma expects biopharmaceutical industry needs for intensified fed-batch and perfusion processes and for high control over protein quality attributes to continue to increase, according to Bulpin.

	Brooks agrees that protein quality continues to be a major concern and focus in the industry--particularly given the increasing numbers of biosimilars and biobetters in development--and must be considered when developing cell-culture medium formulations. “The best way to make sure the formulation in use provides acceptable protein quality, such as N-glycan or charge-variant profiles, is to optimize the formulation specifically to match the cell line metabolic requirements,” he says.

	Protein quality supplements designed to modulate quality parameters following addition to existing cell-culture base media, are also beginning to emerge in the market, according to Brooks. “The most efficient way to achieve the desired protein quality parameters, however, is through the development and customization of the cell-culture medium to the specific cell line in use,” he stresses.

	When referring to this article, please cite it as C. Challener, “Optimizing Cell-Culture Media," 

Media manufacturers are focused on reducing risk, improving quality and consistency, and managing costs.

Optimizing Cell-Culture Media

Cannabinoids have great potential to treat a wide variety of diseases, leading to tremendous interest among pharmaceutical companies in accessing pharmaceutically pure cannabinoids for the development of novel therapies. Several companies are developing cannabinoid-based treatments, including Vitality Biopharma, GW Pharmaceuticals, Zynerba Pharmaceuticals, INSYS Therapeutics, Teewinot Life Sciences Corporation, and Nemus Bioscience.

	Some cannabinoids bind to the same receptors that are bound by endocannabinoids, which are naturally produced in human cells. The human endocannabinoid system (ECS) is a group of endogenous cannabinoid receptors located in the mammalian brain and throughout the central and peripheral nervous systems that are controlled by neuromodulatory lipids, according to Richard Peet, executive vice-president and research director at Teewinot Life Sciences Corporation. “The ECS is involved in a variety of physiological processes, including pain sensation, appetite, memory, and mood. It is for this reason that cannabinoids frequently influence various physiological functions when administered to humans,” he notes.

	Cannabinoids such as delta-9-tetrahydrocannabinol (THC), cannabidiol (CBD), cannabichromene (CBC), tetrahydrocannabivarin (THCV), and cannabidivarin (CBDV) have many potential uses, such as the treatment of nausea associated with chemotherapy, epilepsy, appetite control, diabetes, and inflammation, among many others.

	Why biocatalysis for cannabinoid synthesis?

 plant via biosynthetic pathways consisting of several enzyme-catalyzed steps. The 

plant produces only small quantities of many of these cannabinoids including the varin series of cannabinoids (THCV, CBDV, and cannabichromevarin [CBCV]), and cost-effective chemical synthesis of these molecules is not available, according to Peet.

	“Traditional chemical manufacture of these complex, chiral molecules typically requires several steps and elaborate purification from impurities,” observes James Lalonde, senior vice-president of research and development for Codexis. Chemical synthesis of cannabinoids such as THC also generates copious amounts of organic waste, takes many weeks to complete at the kilogram scale, and is very expensive, adds Peet. “Practical methods for chemical synthesis for many of the 111 known cannabinoids have not been developed,” he states. Enzymes that have been evolved in nature, in contrast, can synthesize these molecules cleanly with the exclusion of impurity-generating side reactions, according to Lalonde.

 plant is also not practical for production of large quantities of pharmaceutically pure cannabinoids. Extraction of targeted cannabinoids from the more than 100 related compounds in the plant is laborious and time consuming, according to Lalonde. “The 

 plant must be grown for three to four months and requires large quantities of water and energy inputs. Then obtaining pharmaceutically pure cannabinoids from plant tissue is very expensive, because many cannabinoids are produced in very small quantities in the 

 allows the synthesis of targeted single distinct compounds, rather than a complex mixture. “Production of these compounds can be controlled for purity and regulatory controls in the same manner as any active pharmaceutical. In addition, the biocatalytic process is independent from the plant growth cycle and the constraints of marijuana cultivation. Scale up of the synthesis of rarer, but desirable compounds would also be made possible in an enzymatic, 

 process in which enzymes that catalyze formation only of the desired compounds could be cloned and purified in high yield,” Lalonde explains.

	Teewinot has developed biosynthetic processes for the production of large quantities of cannabinoids including the varin series of cannabinoids. “Compared to chemical syntheses, these processes are less expensive, more efficient, and produce minimal amounts of chemical waste. As a result, biosynthetic processes make possible the production of pharmaceutically pure cannabinoids in much larger quantities than is practical through plant cultivation and cannabinoid extraction,” Peet asserts.

	The company uses bioinformatics and metabolomics to facilitate the development of synthetic biology and biocatalytic routes to cannabinoids. Genes in the 

 plant, as well as chemical compounds in plant tissues, are identified, characterized, and quantified if relevant. “By cloning specific gene sequences from the plant and incorporating them into microorganisms, we are able to develop new expression systems and processes for the production of naturally-occurring and specific cannabinoids,” Peet explains. Specifically, Teewinot has cloned genes from the Cannabis plant that code for cannabinoid biosynthetic enzymes and incorporated them into yeast and other microorganisms.

	Teewinot refers to its cannabinoid biosynthetic processes as CannSynthesis. Using its synthetic biology and biocatalytic processes, the company can now biosynthetically produce 18 different cannabinoids that have the identical chemical structures to those produced in the plant, including tetrahydrocannabinolic acid (THCA), THC, tetrahydrocannabivarinic acid (THCVA), THCV, cannabidiolic acid (CBDA), CBD, cannabidivarinic acid (CBDVA), CBDV, cannabichromenic acid (CBCA), CBC, cannabichromevarinic acid (CBCVA), cannabichromevarin (CBCV), cannabigerolic acid (CBGA), cannabigerol (CBG), cannabigerovarinic acid (CBGVA), cannabigerovarin (CBGV), cannabinol (CBN), and cannabicyclol (CBL), according to Peet.

 enzymes THCA synthase and CBDA synthase in a microorganism. THCA synthase converts the substrate CBGA into THCA and/or CBCA, while CBDA synthase converts the substrate CBGA into CBDA and/or CBCA. During biocatalysis, both reactions proceed in a pH-dependent manner in a bioreactor. The CBGA used in this biocatalytic process can be produced in any number of ways, including synthetic biology or chemical synthesis. Alternatively, the entire process can be done in a microorganism by means of synthetic biology.

	Teewinot can also produce next-generation, proprietary cannabinoid pro-drugs and cannabinoid analogs, many of which are novel and have never been produced in large quantities in pharmaceutically pure form. “Some of these compounds have improved efficacy and bioavailability and other beneficial characteristics that improve therapeutic outcomes,” Peet notes. He adds that it is now possible to test the efficacy of these compounds in the treatment of a wide variety of diseases, including inflammatory bowel disease, Crohn’s, epilepsy, cancer, migraine, fibromyalgia, and many others, according to Peet. For instance, in 2016, Teewinot licensed its patent-protected biosynthetic processes to Nemus Biosciences for use in the manufacture of its THC pro-drug for treatment of glaucoma.

	The company expects to undertake GMP production of cannabinoids using biocatalysis within the next three to six months, and is in conversations with a large number of pharmaceutical companies seeking to purchase cannabinoids produced by its biosynthetic processes for use in pre-clinical and clinical trials.

	In March 2017, Teewinot signed a letter of intent with Noramco, a producer of controlled-substance APIs, to commercially produce cannabinoids using its CannSynthesis technology. The agreement provides Noramco access to Teewinot’s patent-protected processes for the production of pure cannabinoids using biocatalysis and the company’s assistance with rapid implementation of the technology.

	Noramco will create 10-15 cannabinoid reference standards and evaluate the feasibility of using the technology to produce Dronabinol on a commercial scale. Dronabinol is known for its effectiveness in pain management, as an antiemetic, in the control of vomiting, as an appetite stimulant in the treatment of AIDS and for reducing the side effects of chemotherapy.

	Although biosynthetic processes like those developed by Teewinot are new and improved methods of production that reduce cost and production time, increases purity, and allow for the synthesis of a much greater diversity of molecules that can be incorporated into drug formulations, the approach is different from traditional extraction or chemical synthesis. As a result, Teewinot has found that bringing this new technology to market has created the need to educate researchers and others in the pharmaceutical cannabinoid field. “Educating researchers includes describing the broad array of cannabinoids that can now reliably be manufactured, tested, and put into clinical trials for development of potential new therapies,” Peet comments.

	He does note, however, that because synthetic biology and biocatalysis are common pharmaceutical manufacturing processes, pharmaceutical companies and regulators are generally familiar with the technical aspects of these methods.

Cannabinoids have potential to treat a variety of diseases, but pharmaceutical companies must first develop cost-effective ways to access pharmaceutically pure materials without marijuana cultivation.

The Search for Efficient Synthesis of Complex Cannabinoids

Access to high-throughput synthesis and screening technologies has enabled the discovery of novel classes of small molecules that exhibit high potency. Unfortunately, many of these compounds suffer from poor solubility and bioavailability when administered in conventional solid-dosage forms, the preferred route of administration due to convenience and ease of use. Formulation as amorphous solid dispersions (ASDs)--most commonly via spray drying (SD) or hot-melt extrusion (HME) and more recently co-precipitation (CP)--is increasingly used to improve the performance of poorly soluble drugs. The choice of excipients for spray-dried formulations has a direct impact on the stability and efficacy of these ASDs.

	ASDs involve incorporation of the API in a metastable form within a polymer matrix. Spray-dried ASDs are prepared by dissolving the API with the polymer and other excipients in a solvent (or solvent mixture) and then removing the solvent through a rapid flash-drying process. “Metastable, amorphous forms have greater solubility and bioavailability, but because they are in a higher energy state, they desire, as do all things in nature, to move to a lower energy state, in this case via crystallization,” notes Márcio Temtem, associate director for particle design and formulation development at Hovione.

	The challenge is to avoid or at least delay for a sufficient length of time (the shelf life of the product) this change in the metastable state. The key excipient in ASDs--the polymer--forms a matrix that provides this needed stabilization. The matrix disperses the API molecules, preventing any interactions between them that could lead to crystallization. API molecules trapped within such a matrix also have reduced molecular mobilities, which further reduces their potential for crystallization.

	Several key factors must be considered when formulating spray-dried ASDs, such as the selection of the best ingredients to increase the “supersaturation effect” and the API/polymer ratio and the impact it may have on the stability and performance of the selected system. According to Temtem, “the API and polymer must form a true solid solution in which the polymer and API cannot be distinguished from one another and are mixed in such a way and in the right proportion that the mixture is thermodynamically and/or kinetically stable and the two compounds prefer to be together rather than apart.” Finding this “sweet spot” is achieved using various computational, high-throughput screening, and other experimental tools combined with knowledge and experience.

	Spray drying is just one method for preparing ASDs and may not be ideal for certain APIs or target formulations. Each method--spray drying, hot-melt extrusion, and co-precipitation--produces dispersions with different morphologies, surface areas, particles sizes, and other attributes that impact product release profiles. The manner in which the API is entrapped in the matrix is very different and thus also results in different levels of “disorder,” or in other words energy levels, according to Temtem.

	Process conditions during spray drying may also impact the chemical purity of some APIs. “Although the drying technique is gentle and takes place rapidly with the API droplets protected by evaporation of the solvent, it is necessary to assess an API for degradation during drying,” Temtem observes. He adds the use of solvents can also result in the plasticization of the amorphous solid dispersion, increasing the molecular mobility and thus the potential to crystallize.

	The first challenges to maintaining stability in ASDs, according to Meredith Perry, associate director of pharmaceutics with Pharmatek SD, Catalent, is the need to use a polymer that is chemically compatible with the API while also being miscible and improving the supersaturation of the API in aqueous media. “Although most polymers are relatively inert, some are hygroscopic or acidic and therefore inappropriate for compounds prone to hydrolysis or acid degradation,” she states.

	The second challenge is to select a solvent that provides sufficient solubility and chemical stability for both the active ingredient and the chosen polymer. “A wide range of organic solvents can be safely spray dried, so usually it’s possible to find a combination of solvents, such as polar protic and polar aprotic solvents, that meet these objectives,” Perry says.

	The next challenge with ASDs is excipient selection in the solid state. Any instability previously noted in the crystalline form of the API will likely be more pronounced in the amorphous state, according to Perry. “Specifically,” she observes, “hygroscopicity is typically worse in an amorphous form due to the hygroscopicity of some polymers and the high surface area of the ASD particles. Reactivity with acids, bases, and oxidizing agents is also generally worse due to the high energy state of the amorphous material.”

	The pharmaceutical industry favors the use of excipients that have been previously approved and have data supporting their use in humans. As a result, there are three main families of polymers used to form spray-dried ASDs, according to Temtem: cellulose-based polymers, polyvinylpyrrolidone-based polymers, and acrylate-based polymers.

	Common cellulose-based polymers used in spray-dried dispersions (SDDs) include hypromellose acetate succinate (HPMCAS) and hydroxypropylmethyl cellulose (HPMC). Widely used pyrrolidone polymers include polyvinylpyrrolidone (PVP, also known as polyvidone or povidone) and copolymers of PVP with vinyl acetate. A polyethylene glycol, polyvinyl acetate, polyvinylcaprolactam-based graft copolymer has also been used extensively in ASDs. Copoylmers of methacrylates with acrylic acid in different ratios make up the third family of polymers used to form the matrices within SDDs.

	“Polymer selection is the primary tool for chemically and physically stabilizing the API in an ASD because the two compounds are mixed at a molecular level,” notes Perry. For instance, HPMC capsules are often preferred over gelatin for hard shell capsules because HPMC has a neutral pH, low moisture content, and low hygroscopicity. Mannitol as a tablet excipient also meets these requirements.

	In third-generation SDDs, surfactants, typically d-α-Tocopheryl polyethylene glycol 1000 succinate (vitamin E TPGS) or sodium lauryl sulfate (SLS), are often added to prevent precipitation of the API during dissolution. “Surfactants help to maintain supersaturation of the API by forming micelles and/or other 3-D structures during drug product dissolution. The API is kept in solution by preventing interaction of API molecules with one another,” Temtem explains.

	Surfactant use is only effective at relatively low levels (< 5-10%); when used at higher levels, surfactants often undergo phase separation in the SDD. Due to their low melting points and sometimes low glass transition temperatures, when added to formulations in large quantities, may hinder product accumulation inside the units and result in physical and chemical stability challenges.

	For compounds that are pH-sensitive, excipients with pH buffering capacity, such as citric acid and sodium bicarbonate, can have a stronger stabilizing effect on ASDs than on crystalline APIs, according to Perry. In addition, solid-dosage forms can be coated for further moisture protection. Materials such as polyvinyl acetate-based (PVA-based) coatings provide a good barrier, and can be coated more thickly than film coats to provide further protection.

	Excipient suppliers are taking various approaches to the development of new products for use in SDDs. Most efforts are directed at developing improved versions of existing, approved excipients given that drug manufacturers are hesitant to risk attempting to get a new drug approved with a truly novel excipient; novel excipients do not have a separate approval pathway and are only approved when a drug-product formulation in which they are used receives approval.

	Thus, suppliers are looking to develop new grades of existing matrix polymers or to develop modifications to existing excipients. For instance, HPMCAS until recently was largely supplied by Shin-etsu. The company’s patent protection is expiring, and other suppliers such as Ashland and Dow are working to add this polymer to their portfolios. In addition, excipient suppliers are developing polymers with different combinations of acetate and succinate substituents to provide more options for drug developers. “These new excipients open new windows for formulators, particularly from a quality-by-design standpoint. Formulators now have many more polymer with a broad range of properties, allowing for the development of more effective products,” Temtem asserts.

	Another approach is the development of higher-performance versions of existing excipients that help manufacturers optimize their production processes. Temtem notes that Dow, for instance, is developing excipients for the pharmaceutical industry that have lower viscosities and thus allow for higher solids concentrations in SD formulations and/or the ability to process spray-dried systems under milder conditions.

	The development of novel excipients, or new chemical entities, is not as common but is ongoing despite the challenges in obtaining their approval. BASF, for example, is developing a new excipient with a good balance between the hydrophilic and hydrophobic moieties that provides good stabilization of APIs in solid dispersions while also maintaining supersaturation for an extended period of time, according to Temtem.

	Mesoporous silica is also a substance attracting significant interest from the pharmaceutical research community as a potential excipient for ASDs. Mesoporous silica has small pores and a high surface area and has been used for many years as a catalyst in chemical processing and for various applications in the food industry. “For ASDs,” observes Temtem, “the size of the pores in mesoporous silica is ideal for trapping API molecules and preventing them from interacting. The material is advantageous because silica powder has good characteristics in terms of its flowability, which leads to improved downstream processing. Silica is also an inert material, and thus, there is no potential for interactions with the API or the GI tract.”

	When referring to this article, please cite it as C. Challener, “Overcoming Excipient Challenges in Spray-Dried Dispersions,” 

Polymers and surfactants impact stability and long-term performance.

Overcoming Excipient Challenges in Spray-Dried Dispersions

While most basic research occurs in academia, contributions from industry are also undeniably significant and impactful, according to Jeff Song, director of development at Boehringer Ingelheim (BI) Pharmaceuticals. Several pharmaceutical companies have established strong R&D organizations that routinely discover innovative and practical chemistries, and technology companies such as Johnson Matthey and Codexis also engage in original research to provide new chemistries for API synthesis.

	In addition, both academia and industry are increasingly seeking opportunities for collaboration in precompetitive spaces, either through individual research agreements or multi-party consortia, according to Song. He notes that BI was one of the founders of the BI-Pfizer-AbbVie Consortium for Non-Precious Metal Catalysis, Flow Hydrogenation, and Computer Aided Synthetic Design. BI is also an industrial sponsor of the National Science Foundation’s Research Center, Center for Rational Catalyst Synthesis, and is intimately involved in its projects.

	Two areas of technology have emerged in recent years as particularly important for API manufacturing: flow chemistry and catalysis. “Both can be applied to a broad scope of APIs. Catalysis provides shorter/more atom-economical, more cost-effective, and greener processes, while also providing novel reactivities that were not accessible before. Flow chemistry is another green technology that expands the horizon of the types of chemistry that can be used for making bulk APIs,” states Chris Senanayake, US vice-president of chemical development for BI.

	Chemistries routinely practiced in the pharmaceutical setting include amide bond formations, common acid/base promoted hydrolytic reactions to access amide and carboxyl moieties, carbon-carbon bond formations (Grignard, Suzuki-Miyaura, Negishi, Sonagishira, Heck), oxidations, reductions (classical and asymmetric), halogenations, esterifications, alkylations, aminations, nitrations, and classical and enzymatic resolutions, according to Marlon Lutz, a process development and flow chemist with Regis Technologies.

	 “Advances in catalysis, including both chemical and biocatalysis, have been the most impactful for API synthesis,” asserts Senanayake. Specifically, he points to the design and discovery of new (chiral) phosphine ligands that have allowed many efficient transformations such as cross couplings and the installation of chiral centers to be performed on large scales (1-3).

	Carbon-carbon bond forming organometallic chemistry is another chemistry worth highlighting, according to Stephen A. Munk, president of Ash Stevens, a division of Piramal Pharma Solutions. “Scalable palladium-catalyzed coupling reactions, such as Sonogashira and Suzuki reactions, allow for efficient carbon-carbon bond formation at the larger volumes required for API manufacturing,” he observes. (See an example in reference 4.)

	In biocatalysis, the use of directed evolution technology has resulted in the engineering of highly efficient enzymes for a wider range of chemical reactions than ever before, according to Song. Multiple examples of biocatalysis in API synthesis on the commercial scale have been reported (5-6).

	 “With lower processing times, solvent volumes, and reduced waste streams, flow chemistry (continuous processing) has enabled processes to be operated more efficiently, safer, cleaner, cheaper, and faster,” Lutz states. In addition, process analytical technologies (PATs) have enabled flow chemistry processes to be executed with higher confidence and reproducibility and generate products that are of high purity and yield. Furthermore, FDA supports the use of a continuous flow approach for producing APIs.

	 “Many pharmaceutical chemical processes are still performed in batch vessels, which are strictly limited by the process conditions (e.g., boiling points of solvents at atmospheric pressure) and often require long processing times for heating/cooling of large batch reactors. There has, however, been an increasing expectation for safer, more efficient, cost effective and greener synthetic routes. Flow chemistry has attracted attention as realization of its numerous benefits for achieving this goal has spread,” Lutz adds.

	In particular, flow chemistry is highly applicable for processes that require high mixing and heat/mass transfer, control of processes operating in the thermal runaway regime, and safely conducting hazardous chemistry that is difficult to handle in conventional batch reactors.

	At BI, flow chemistry has been employed for reactions that would not be possible using conventional batch protocols, according to Song. “Several of our programs involving high-energy reagents (azides), unstable intermediates (organolithiums), and extreme reaction temperatures (low or high) have benefited from flow technology,” he comments. In one example, approximately one metric ton of a propargyl borolane reagent was manufactured in good yield and high quality using flow chemistry (7).

	In addition, the ability to automate commercial flow reactors has impacted how process development is performed and translated into manufacturing, according to Lutz. “The use of flow chemistry enables process development and reaction optimization to be performed significantly more quickly than traditional approaches using batch vessels,” he explains. Employing PAT with flow reactor systems extensively reduces development times with elimination of the bottleneck of offline analysis (8).

	Furthermore, translating flow processes from research-scale models to commercial scale is relatively simple; if the geometries and design of the flow reactor are maintained, the number of flow reactors can be increased to achieve the desired product output (‘numbering up’), according to Lutz. He also notes that many of the commercial-scale flow reactors are designed to offer enhanced mixing and heat/mass transfer phenomena when increasing throughput.

	It is also worth noting, according to Lutz, that implementing flow chemistry can be beneficial for contract manufacturing organizations (CMOs) because they can easily use flow chemistry for problem steps that are low yielding, require harsh conditions, are medium- to high-risk due to the reaction conditions, or use excesses of expensive reagents.

	It isn’t only advances in chemistry and processing technology that have improved API manufacturing. The evolution of a contemporary data-rich laboratory environment with new equipment, software, process modeling techniques, quality-by-design methodology, multidisciplinary collaborations, more sensitive analytical instruments, enhanced safety analyses using reaction calorimetery, accelerating rate calorimetry (ARC), and thermal screening units (TSus) is providing a deeper understanding of chemical processes and their predictability at scale, according to Munk.

	“Today it is not only possible to improve yields and limit the formation of impurities; in-depth understanding of processes leads to better process control, safety, reliability, and robustness, along with improved product quality,” he explains. “It also allows for continuous improvement of the process and understanding where things fail and the consequences of the failure using statistical data without having to do hundreds of experiments,” Munk adds. New safety equipment/instrumentation and software allow for safer scale up of processes, while new laboratory reactor systems better model plant equipment and stirring conditions. A more multidisciplinary approach to process development looks at processes from more perspectives (modeling, statistically, engineering, safety, and environmental impact).

	Enhanced chromatography equipment and more

	While catalysis and flow chemistry are having the greatest impact on API manufacturing, there are many other advances providing measurable improvements. More efficient, environmentally compatible, and cost-effective production-scale chromatography equipment, including simulated moving bed (SMB) and supercritical fluid chromatography (SFC), is one example, according to Munk.

	Greater understanding of solid-state forms and genotoxic impurities are also important. Access to greater knowledge through rapid and effective solid form screening leads to the selection of more developable APIs (9). Identification of impurities and their sources makes it possible to develop more efficient, optimized processes with higher yields and purity profiles (10).

	Highly potent and complicated drugs such as antibody-drug conjugates and proteolysis targeting chimera are more prominent in the industry and are requiring significant containment infrastructure, purification equipment, and experienced chemists, according to Munk. The existence of different potency categorization schemes and the need to ensure operator and environmental safety pose significant challenges (11).

	Returning to specific chemistry examples, Song points to Knochel’s Turbo Grignard reagent as an important chemistry enjoying wide application in industry and academia. “This chemistry revolutionized the way in which Grignard reagents are generated. Due to its unique reactivity, Turbo Grignard offers benefits such as wider functional group compatibility and improved safety and cost-effectiveness for Grignard formation (12-13). Organozinc reagents have shown wider functional group compatibility as well, according to Marlon. (See reference 14 for an example.)

	While the advances discussed above are indeed significant, challenges in API synthesis and manufacturing remain. Fortunately, further developments are ongoing and provide significant potential for the future. For example, useful methods for fluorination and trifluoromethylation (15-16), which are needed given that approximately 20% of drugs contain at least one fluorine atom, are allowing medicinal chemists to access novel structures for structure-activity relationship studies and to secure intellectual property for new drug candidates, according to Senanayake. These methods are not ready for large-scale application, however. More work is required in both academia and industry to address substrate scope, catalyst loading, and general practicality issues. Lutz also points to greener metal-catalyzed fluorinations as showing great promise for the replacement of common and harsh fluorinating reagents (17-18).

	Direct functionalization of unactivated C-H bonds is another field showing promise for simplification of synthetic routes through reduction of the number of required steps, according to Lutz. He also points to palladium-catalyzed cyanations with potassium ferrocyanide (19) as providing convenient access to amides, amidines, esters, and carboxylic acids while eliminating the concerns associated with current cyanide-based reactions.

	Much work is also being directed at the development of non-precious metal catalysis, which can reduce costs and toxicity and enable greener processes, according to Senanayake. He notes that significant strides have been made in this area in the past five years, with the development of nickel (Ni)- and iron-catalyzed sp2-sp3 couplings of particular interest to the pharmaceutical industry (20-21).

	Microwave reactors, according to Munk, offer enhanced reaction times and enable transformations that will only work under microwave conditions. Photo-redox reactions, meanwhile, allow the use of the cheapest and cleanest energy source (i.e., solar power) for chemistry, according to Song. In addition, this technology can be combined with chemistry to deliver greener processes (22-23).

	 “API manufacturing requires not only effective chemical technologies, but seasoned, experienced process chemists who understand the necessary requirements for the development of GMP processes, including route selection, cost efficiencies, robustness, reaction mechanisms, environmental impact, and product quality, among other attributes,” asserts James M. Hamby, vice-president of business development for Ash Stevens. These process chemists many times develop and report interesting modifications, efficiency enhancements, and uses of established chemistries to effect large-scale transformations.

	 “Each process presents a unique set of challenges. No one chemistry or technology really stands out as above all the rest. That is why experienced chemists working in a multidisciplinary team deliver the best results,” Hamby concludes.

., 138 (47), 15473 (2016).
	2. C. H. Senanayake, et al., 

 358 (22), 3522-3527 (2016).
	3. C. H. Senanayake, et al., 

 18 (19), 4920 (2016).
	4. S. R. Gurung, et al., 

., 21 (1), 65-74 (2017).
	5. N.K. Modukuru, et al., 

., 18 (6), 810-815 (2014).
	6. Y.K. Bong, et al., US Patent US8895271 B2 Nov. 25, 2014 Codexis
	7. D.R. Fandrick, et al., 

., 16 (5), 1131 (2012).
	8. S.L. Buchwald, et al., 

., 2 (1), 145-154 (2007).
	10. J. H. Barkalow, et al., 

., 11 (4), 693-698 (2007).
	11. E. Dunny, et al., 

, Ahead of Print (2017).
	12. C. H. Senanayake, et al., 

 16 (16), 4090-4093 (2014).
	13. P. Knochel, et al., 

., 79 (10), 4253-4269 (2014).
	14. A. Dilman and V.V. Levin, 

., 57 (36), 3986-3992, 2016.
	15. E.N. Jacobsen, et al., 

., 138 (15), 5000 (2016).
	16. F.D. Toste, et al., 

., 137 (38), 12207 (2015).
	17. S. Buchwald, et al., 

, 325 (5948), 1661 (2009).
	18. T. Ritter, et al., 

., 132 (34), 12150 (2010).
	19. M. Utsugi, et al., 

. 18 (6), 693-698 (2014).
	20. C. H. Senanayake, et al. 

., 7, 5581-5586 (2016).
	21. J.N Desrosiers, Angew. 

., 55, 11921-11924 (2016).
	22. D.W.C. MacMillan, et al., 

., 138 (42), 13862 (2016).
	23. A.K. Vannucci, et al. 

Advances in chemical synthesis are enabling greener, more cost-efficient processes for API manufacturing.

Catalysis Advances Promote Greener API Manufacturing


	Vol. 41, No. 6
	June 2017
	Pages: 20–22

	When referring to this article, please cite it as C. Challener, “Accessing Cannabinoids Using Biocatalysis," 

Efficient synthesis of complex cannabinoids is possible while avoiding marijuana cultivation. 