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

Cynthia A. Challener

Advances in catalysis have had a tremendous impact on the pharmaceutical industry, enabling the synthesis of highly complex small-molecule intermediates and APIs in fewer steps, with increased atom economy, and often with high selectivities and yields. There are issues with many of the most widely used catalytic systems, however, including the cost, availability, and toxicity of many of the precious metals used to form the catalysts and the need for expensive and complex ligands to achieve the desired transformations. There has been much interest, therefore, in the development of alternative catalysts derived from abundant, nontoxic metals and simple ligand systems. New base metal catalyst platforms are also of interest because they may enable new chemistry not previously observed with traditional precious metal catalysts. Select examples with potential applications in pharmaceutical synthesis are discussed below.


  There is a finite supply of metals in the Earth’s crust, and humans are consuming many of them at an ever-increasing rate. As a result, the availability of certain transition metals for use in the production of catalysts is becoming increasingly limited due to growing demand for these materials in other applications, including the newer organic light emitting diode displays in smart phones and televisions and renewable energy systems, such as fuel cells, wind energy, and photovoltaics. 

In fact, in 2012, the British Geological Survey updated its risk list for the supply of chemical elements or element groups that are of importance to human society (1). The availability of each element was determined by considering its abundance in the Earth’s crust, the location of current production and reserves, the political stability of those locations, and the recycling rate and substitutability of the element. The rankings give an indication of which elements might be subject to supply disruption, most likely from human factors such as geopolitics or resource nationalism, along with events such as strikes and accidents. At the top of the list with the highest risk of supply disruption are the rare earths, platinum group metals, niobium, and tungsten. It should be noted that the demand for the elements (outside of their substitutability) was not considered in this analysis.

The platinum group metals are of particular interest to the pharmaceutical industry, because many important catalysts rely on platinum, palladium, ruthenium, rhodium, iridium, and osmium. These metals are expensive and experience high price volatility. The price of platinum, for example, ranged from $1400-$1500/ounce in March 2014, and has varied more than 5-fold in the last few years, according to Paul Chirik, the Edward S. Sanford Professor of Chemistry at Princeton University. With respect to platinum supply, he notes that the amount of platinum produced each year is only approximately 6% of the quantity of gold mined annually.


  Some researchers are raising the question of whether or not a reaction can truly be considered sustainable if such rare (and at times toxic) metals are required. According to Amir Hoveyda, Joseph T. and Patricia Vanderslice Millennium Professor of Chemistry at Boston College, even if a transformation is otherwise green (i.e. minimal/no solvent, high atom economy, no need for heating/cooling, little downstream processing requirements), it can’t be realistically considered as sustainable if a catalyst based on a rare transition metal is required. Hoveyda also notes that to be industrially practical, reactions must proceed in a reasonable period of time, ideally within six to eight hours, and catalysts must provide high selectivities and high yields.


  One reaction that Hoveyda has focused on is the copper-catalyzed synthesis of enantiomerically enriched homoallylic amines and alcohols, which are important intermediates for the preparation of many different biologically active molecules. As an alternative, his group has developed a transition metal-free route to enantiomerically pure amines and alcohols with inexpensive, easily prepared derivatives of the amino acid valine as catalysts for the reaction of a wide range of readily available unsaturated organoboron reagents with imines and carbonyls. In addition to not requiring any rare transition metal salts, the reactions are generally complete in six hours, require low catalyst loadings, and provide high yields and enantioselectivities. The effectiveness of this new set of catalytic reactions, according to Hoveyda, is due to the use of a proton and internal hydrogen bonding.


  Chirik, meanwhile, is developing alternative catalysts based on cobalt and iron, which are far more abundant than the precious metals. In addition, while most precious metal catalysts require two-point substrate binding for inducing high levels of enantioselectivity in asymmetric hydrogenation, because iron and cobalt are a new catalyst type, they have yet to exhibit this limitation, he points out. For example, Chirik’s cobalt asymmetric hydrogenation catalysts formed using simple and inexpensive chiral amine ligands are relevant to the pharmaceutical industry. In fact, Chirik developed the catalysts through a collaboration with Merck &  Co. that included high-throughput experiments. Notably, the cobalt catalysts are effective for the hydrogenation of both functionalized and unfunctionalized olefins in high yield and high enantiomeric excess. 

In a second example, Chirik’s group has developed iron catalysts that mediate both inter- and intramolecular [2π + 2π] reactions of alkenes and alkynes to form cyclobutenes. What is interesting about this system is that it is effective for unactivated olefins and is also selective for 2+2 cycloaddition even in the presence of a diene that can undergo Diels-Alder (4+2) cycloadditions. According to Chirik, such selectivity is unique to iron and results from its ability to engage in radical chemistry with the supporting ligand.


  A third strategy for the development of alternative catalysts involves mimicking the behavior of enzymes. Such research is being performed by Robert Morris, a professor at the University of Toronto, who has developed new iron catalysts that perform what he calls “metal-ligand bifunctional catalysis,” in which the ligand and metal both play an active role in the catalytic cycle. Notably, he has shown that with the right choice of ligands, the electronegativity of the iron center can be adjusted such that it is close to that of carbon and thus able to form covalent bonds, particularly iron-hydrogen bonds. Surprisingly, carbon monoxide, which is often a poison to catalysts, is a required ligand as it is for nature’s hydrogenases.

Such catalysts are effective for the asymmetric transfer hydrogenation (ATH) of ketones and can be simply prepared from relatively inexpensive diamines and phosphine aldehydes, which self-assemble (promoted by the metal through a template effect) to form imines that bind to the metal. Morris notes that the catalyst is effective at low loadings and the iron can be readily oxidized upon exposure to air and then precipitated, allowing for simple purification. A third-generation family of these iron-based catalysts was reported in 2013 and should be available for purchase soon (2).  These catalysts are more active than any known catalyst for ATH and as selective as more expensive conventional ruthenium-based systems.

The catalyst may also be applicable for direct hydrogenation under a mild hydrogen pressure, and the reverse of the asymmetric transfer hydrogenation reaction may be useful for the kinetic resolution of racemic alcohols.


  Carbon-carbon coupling reactions have become widely used in organic synthesis, including the commercial manufacture of numerous pharmaceutical building blocks and key intermediates. Robin Bedford, professor of catalysis at the University of Bristol, has been investigating the potential for using iron catalysts to mediate these important transformations. In the past, coupling reactions involving iron have generally been Grignard reactions of “hard” nucleophiles. Bedford wanted to expand the repertoire to other “softer” nucleophiles (e.g., zinc) that are used in coupling reactions catalyzed by platinum and palladium.

What his group found was that the choice of ligand was crucial for creating iron catalysts that are effective for both the Negishi coupling of benzyl halides and phosphates with diarylzinc reagents and Suzuki couplings of arylboronates with alkyl and allyl halides and 2-halopyridines. Most importantly, the best ligand is the simple, inexpensive, bis(diphenylphosphino)ethane (dppe) ligand. The key, according to Bedford, is to prepare the catalyst in the correct manner. “One of the important lessons we learned in this work was that it is important to understand the identity and role of the active catalytic species and how ligands interact with iron in order to generate the desired species,” Bedford states.


  Iron is not generally thought of as a good metal for the preparation of catalysts because of its electronic structure. Platinum group metals, which are heavier metals with no unpaired electrons, have always been thought to form stronger bonds with ligands, according to Patrick L. Holland, a professor of chemistry at the University of Rochester. On the contrary, his group has shown that complexes based on metals with unpaired electrons can follow similar reaction mechanisms, and often do so more rapidly, as long as a balance is maintained between the weakness and activity of the ligand-metal bonds. 

Holland has developed special bidentate ligands that are designed to afford the high-spin electronic configuration in complexes with first-row metals. For example, his group has developed an iron(III) imido complex that oxidizes hydrocarbons and transfers a nitrene (NR) group to organic substrates.

“We believe that properly designed complexes of iron and other first-row metals can serve not only as alternatives to transition metal catalysts for the mediation of established reactions; given their unusual reactivity, they may also provide access to novel transformations,” observes Holland.

“Risk list 2012: An updated supply risk index for chemical elements or element groups which are of economic value,”

 Accessed Apr. 10, 2014.
  2.	W. Zuo et al., Science, 342 (6162), 1080-1083 (2013). 


Cynthia A. Challener is a contributing editor to 

Articles

Industry seeks alternatives to costly and toxic precious metals used to form the catalysts for organic synthesis.

Alternatives to Precious Metal Catalysts Sought

Transition metal-containing catalysts are widely used in organic synthesis, but many are based on expensive, rare metals. Researchers are seeking more sustainable alternatives.

Seeking Alternative Catalyst Solutions

Solid dosage forms are preferred for the delivery of APIs, but many therapeutic agents do not readily form stable crystalline materials or the crystals have poor solubility. Cocrystals alter the physical properties of APIs without chemical modification while maintaining the benefits of a crystalline solid form, and thus may provide a viable solution for some of these difficult APIs. Cocrystals can also be synthesized using numerous different methods, some of which use little solvent or are solvent-free and have the potential to provide real cost advantages. As a result, the potential for cocrystals is significant.

This potential is tempered, however, by the industry’s reaction to FDA’s classification of pharmaceutical cocrystals as “drug product intermediates” (DPIs) rather than APIs in its April 2013 regulatory guidance on cocrystals (1). This decision is generating significant discussion within the industry about the risks associated with the practical development of cocrystals. Based on the public comments on the draft guidance, manufacturers would prefer cocrystals to be considered drug substances rather than drug product intermediates. As the industry moves toward adopting the FDA guidance, the risk associated with developing cocrystal-based products appears to be increasing rather than decreasing. There are questions, for example, about expiration dates for drug products that incorporate cocrystals manufactured using processes that are consistent with API (drug substance) materials but regulated as DPIs. In addition, FDA is the only agency to date to formally define a regulatory scheme for cocrystals. If other agencies around the world take a different approach, it could become difficult for companies to develop cocrystal-based products that satisfy global regulatory requirements. The European Medicines Agency (EMA), for example, could consider cocrystals to be drug substances rather than DPIs.


“The selection of a specific crystal form for a given API is a profoundly important step from clinical, legal, and regulatory perspectives: the dissolution rate and intrinsic solubility of different crystal forms is variable and can strongly influence bioavailability; stability to temperature and humidity are critically dependent upon crystal packing; and the unpredictability (i.e., lack of obviousness) of crystal structures and physical properties means that there are legal issues and challenges in terms of obtaining and maintaining patent protection for an API,” says Ning Shan, director of preclinical development at Thar Pharmaceuticals. 

Highly lipophilic APIs have dissolution problems that may be overcome using a cocrystal because the new crystalline lattice being formed often has different dissolution properties compared to those of the crystals of the API alone, according to Chris Frampton, chief scientific officer of SAFC Pharmorphix. The formation of a new crystal lattice with a pharmaceutically acceptable coformer can overcome development hurdles, such as compaction problems and difficulties with hygroscopicity. In addition, for non-ionizable APIs, for which it is not possible to make a salt, the option to form a cocrystal can be advantageous over exploring exotic formulation methods or changing the API molecular structure to include the acid or base functionality required for salt formation, which can have an impact on the activity of the API. There is also the potential to use cocrystals as a lifecycle management tool, because their use can delay or divert generic competition.


Defining cocrystals as drug product intermediates, in which the coformer is treated as an excipient and the cocrystal is effectively considered to be a formulated material, has several consequences. A potential advantage could be in the area of cocrystal synthesis. Traditionally, APIs are crystallized from an organic solution, and such processes often require the use of large quantities of solvent. Utilization of solvent is common in drug substance manufacturing plants, but not in drug-product manufacturing facilities where DPIs are manufactured. Because cocrystals can be produced using solventless solid-to-solid or slurry conversion processes that are well documented in the literature, they are ideally suited for production in DPI facilities, according to Scott L. Childs, president and CSO of Renovo Research. “These novel synthesis techniques have been used to scale up cocrystals at gram scales and for screening purposes for many years. Now that cocrystals are classified as drug product intermediates, it is possible to consider these methods for cocrystal synthesis on the commercial scale,” he explains.  “The ability to prepare cocrystals using solventless synthesis methods in drug-product manufacturing facilities, which is where amorphous dispersions and other enabling formulations are manufactured, provides the opportunity to use ‘greener’ production methods and realize better manufacturing economics and reduced development times,” Childs adds.

The FDA guidance generally creates the opportunity for the application of non-traditional synthesis techniques to the preparation of cocrystals. There are processes being developed that have potential for application to the industrial-scale manufacture of cocrystals, such as methods using hot melt extrusion, continuous oscillatory baffled crystallizers, and resonant acoustic mixing techniques (2-4). Some solventless methods rely on various grinding techniques and melting approaches, according to Marcel Hoffmann, chief science officer with Crystallics. “Gaining an understanding of cocrystal formation and their interactions at the molecular, particle, powder, and product level is necessary in order to improve the manufacturing processes for this type of drug,” he adds. If the industry embraces the FDA guidance with respect to the manufacture of cocrystals in drug-product manufacturing facilities, there could be a significant shift in cocrystal manufacturing away from solvent-based processes and towards solventless processes, according to Childs.


Defining cocrystals as drug product intermediates has also created significant concerns. According to Hoffmann, depending only on differences in pKa, the same counter ion would form a salt with one API and a cocrystal with another API, and the two substances would be treated differently from a regulatory perspective. “In addition,” he notes, “this definition opens the door for generic companies to develop cocrystals and follow a 505(b)(2), or even bioequivalence route, rather than a more costly new chemical entity route.” 

Shan does point out, though, that a cocrystal still has the same active moiety as the original drug. “Even so, under the patent protection of the original drug, a cocrystal version of the same drug cannot circumvent the original patent,” he notes. “However, a cocrystal with desired properties can be patented, IP protected, and marketed, as mentioned previously in the context of life-cycle management.”

The FDA approach may not be consistent with the approach that other world regulatory agencies adopt for cocrystals, according to Childs. “If novel cocrystal manufacturing is enabled via the FDA guidance, but the same manufacturing approach causes a regulatory problem in other world markets, then that creates risk,” Childs observes. “There is currently no clear answer regarding how the FDA approach will fit in with the rest of the world, which is making it difficult to get buy-in and investment into this new technology from a very risk-averse industry,” he adds.

Shan further notes that the requirement in the FDA guidance that the developer of a cocrystal drug product must provide “assurance that complete dissociation of the API from its excipient occurs prior to reaching the site of action for pharmacological activity” could be a real challenge. 


At SAFC Pharmorphix, the ability to perform crystal structure analyses on really small crystals has been advantageous. “Knowing the crystal structure makes it possible to see how the two materials come together to form a supramolecular complex. Then we can go back to our list of potential coformers, tune them slightly, and follow an iterative process to develop the optimum cocrystal,” observes Frampton.

Renovo, meanwhile, has developed a new approach to formulating cocrystals that uses the improved cocrystal solubility more effectively. Traditionally, as is commonly done with salts, cocrystals are used in 

preclinical studies as “neat” material (pure crystalline powder), according to Childs. This approach is not optimal, because cocrystals are highly susceptible to solvent-mediated transformation into the poorly soluble form of the API when dosed as neat material.  

“We have developed a new approach, demonstrated with a danazol cocrystal, that utilizes solubilizing excipients and precipitation inhibitors to create an optimal supersaturation level and controlled dissolution event,” Childs explains.  The model system achieved a 10X increase in the area under the concentration time curve (AUC) in animal studies using the formulated cocrystal, compared to only a 1.7X increase in the AUC using the neat cocrystal solid.  “The results of this study suggest that the appropriate formulation of cocrystals can be an important part of the cocrystal development process,” he says. Childs believes that the concepts outlined in this study present a viable roadmap for moving cocrystals forward because it suggests that formulated cocrystals can directly compete with more established drug-delivery technologies such as amorphous dispersions and lipid-based solubilizing formulations.  


Despite these significant regulatoryconcerns, interest remains in cocrystals. “Looking at the ratio between salt formation and cocrystal studies that Crystallics carries out on behalf of its customers, it is clear that the pharmaceutical industry is still hesitant to make cocrystals their final marketed products,” notes Hoffmann. He does, however, point to Esteve’s cocrystal of Tramadol and Celecoxib as one example of a cocrystal that is progressing towards a new drug approval. There is hope that a successful application of this cocrystal will pave the way for others. Crystallics is in fact seeing a renewed interest in the development of cocrystals, with most projects focused on tuning the physicochemical properties of the API, particularly solubility and dissolution rate. 

“I am convinced that pharmaceutical cocrystals are here to stay,” Hoffmann asserts. “With increasing research, our understanding will rapidly grow and cocrystals, unlike other bi-component crystals, offer the possibility of crystal engineering. Predicting the physicochemical properties up front is the holy grail of drug development,” he comments.  Frampton points out, though, that the current lack of predictability of cocrystal formation is necessary to some degree. “As soon as it is possible to predict cocrystal structures, we have a problem. Patent applications for salt forms are already being questioned on the grounds that they are predictable and obvious due to the pKa of the API and the limited list of pharmaceutically acceptable salt formers. The lure of the cocrystal remains the questionable and unexpected successful outcome.”

Childs adds that although additional advancements are needed to help the field mature, the science does not appear to be the limiting factor in the uptake of cocrystals by the pharmaceutical industry, but rather it is the perceived risk associated with cocrystals that seems to be slowing progress with respect to their industrial application. “With the appropriate time, effort, and investment, cocrystals can be used to consistently create improved pharmaceutical products. In specific situations, cocrystals should be able to compete directly with amorphous dispersions in terms of efficacy, but offer a superior product because of the lower risk and improved stability of the crystalline state compared to the amorphous state,” he concludes. 

Guidance for Industry: Classification of Pharmaceutical Co-Crystals

 99(4), 1693-1696 (2010).
  3. L. Zhao et al., 

, 2014 Advance Article, DOI: 10.1039/C4CE00154K.
  4. 	D.J. am Ende et al., 


  Cynthia A. Challener is a contributing editor at 

The perceived risks associated with cocrystals may be slowing progress of their industrial application.

Cocrystals: Industry Takes a Wait-and-See Approach

Highly potent active pharmaceutical ingredients (HPAPIs) make up the fastest growing segment of the worldwide API market, according to market research firm RNCOS, which predicts that the value of the global HPAPI market will reach $15.3 billion by 2017. Consultancy Transparency Market Research, meanwhile, estimates that the total HPAPI market will increase at a compound annual growth rate of 9.9% from $9.1 billion in 2011 to $17.5 billion in 2018.


  While the majority of HPAPIs have thus far been developed for the treatment of cancer (cytotoxics and cytostatics), particularly tyrosine kinase inhibitors and derivatives, newer hormone, narcotic, and retinoid-based drugs are also highly potent. In addition, HPAPIs are generally thought of as small-molecule drugs, but many biopharmaceuticals have occupational exposure limits (OELs) of 10 μg/m

 or less, a general guideline for classifying APIs as highly potent, according to Srinivas Achanta, regulatory affairs manager with Hetero. Of particular interest are antibody-drug conjugates (ADCs), which include a small-molecule, cytotoxic payload and an antibody connected with a linker and are designed to deliver the highly potent payload to targeted cells, reducing the likelihood of harm to normal cells. 


  Drivers of the HPAPI market, according to RNCOS, include a rising demand for cancer HPAPIs, increased private player participation, particularly in developed regions, and technological advances in process manufacturing of these challenging APIs. The number of ADCs in development has grown rapidly in recent years and is also a factor in the growth of the HPAPI market. The entrance of a growing number of competitors in the HPAPI market is, however, resulting in a highly fragmented market and may have a negative influence on growth, according to Transparency Market Research. The consulting firm also notes that a shortage of US FDA-approved manufacturing sites may act as a further restraint on growth for the HPAPI market. 

North America currently accounts for the greatest share of the market, followed by Europe. The greatest growth, however, is expected in Asia Pacific, with major activity taking place in India and China due to rising health awareness, continuing expansion of these economies, improving healthcare systems, and a rapid increase in the production of generics, according to Transparency Market Research.


  Because ADCs enable the targeted delivery of HPAPIs to specific cancer cells (or other problematic cells), they have attracted significant attention, and the number of ADCs in development has increased dramatically. “The toxin that is conjugated to the antibody in ADCs has been the fastest growing segment of the HPAPI market, with demand quickly increasing in recent years,” says Cynthia Wooge, global strategic marketing manager at SAFC. With three components to manufacture and combine together under containment conditions, the production of ADCs can be a complex process, adds Achanta. In addition, Wooge notes that certain new types of ADC compounds, which may be considered the next generation of ADCs, are considered to be even more potent than existing products and are pushing the limits on current capabilities with respect to the handling of low OEL materials.


  Single-use manufacturing technologies are on the rise for small-scale highly potent products such as ADCs, according to Jeff Marcoux, technical business development manager at Novasep. “Another important trend is the widespread acceptance of portable dedicated equipment, including both single-use and permanent systems for small-scale production of very highly potent compounds such as ADC payloads, which often have OELs lower than 0.1 µg/m

·8 h,” he notes. “Dedication of equipment and the adoption of single-use technologies can play an important role in controlling cross-contamination and maximum carry-over limits after cleaning (MACO) when implemented as part of a risk-based approach,” he adds. 

“All high potency APIs must be produced under conditions that not only protect the operators from exposure to the compounds, but also prevent contamination and the inadvertent carryover of a different product that was previously produced or is simultaneously being manufactured. While today’s modern HPAPI facilities are meeting today’s requirements, there is a continual need to improve these technologies going forward. To accomplish this task, we need to focus on the development of production and handling methods including single-use systems and/or new technologies that can provide increased protection of the API from cross-contamination. Such protection systems must focus on the cleaning and removal steps,” observes Achanta. “The development of such new technologies is of increasing importance as the number of multiproduct facilities, which post the greatest risk of contamination of HPAPIs, are growing around the world in order to meet the greater demand for these products,” he adds.


  Appropriate process design is in fact critical for the entire HPAPI production operation. “Most very highly potent APIs and ADC payloads require small clinical and commercial quantities, and the production of gram-scale GMP APIs and payloads can be challenging. The control of containment using flexible and small equipment, including glass equipment, is always a challenge and requires a tailored approach for each process and each unit operation,” Marcoux says. Appropriate process design at the development scale is also necessary to ensure that the process will fit the equipment and capabilities of the facility upon scale-up, according to Wooge. 

Achanta believes that one of the biggest challenges at large scale is the handling of powders and solids. “Up to a scale of a few kilos, it is relatively straightforward to manage solids, because bottles with alpha–beta split butterfly valve connectors can be used for charging reactors to maintain containment. On a larger scale, however, this approach is impractical,” he explains. 

Wooge also notes that appropriate cleaning methods must be implemented in order to achieve the required limits for residue levels in multi-use equipment. “The use of analytical methods that provide very low detection limits is necessary in order to confirm that the required residue levels have been achieved,” he says. He also adds that there is certainly an interest in continually achieving lower detection limits when handling HPAPIs.
  Finally, Achanta points to the ambiguity surrounding the classification of HPAPIs given that different pharmaceutical companies often have proprietary systems and the fact that often the classification of new APIs is unknown due to a lack of data as issues that must be managed with appropriate process design and containment controls.        

Appropriate process design and engineering are critical for the production of small-molecule and biologic HPAPIs.

HPAPIs: Fast-growing Segment Presents Challenges and Opportunities

Players across the biologics value chain are attracted by the advantages of continuous biopharmaceutical manufacturing.

Potential Benefits Drive Interest in Continuous Manufacturing

A global contract manufacturer of biologics discusses the increasing impact of health-based exposure limits in assessing cross-contamination risks.