Formulation Strategies in Early-Stage Drug Development

Published on: 
Pharmaceutical Technology, Pharmaceutical Technology-10-02-2017, Volume 41, Issue 10
Pages: 20-27

Applying the right formulation strategies early in the drug development process can help avoid costly late-stage failures.

Improving R&D productivity continues to be a key challenge for the pharmaceutical industry. There is increasing pressure to speed up drug development and make it more cost-effective. Companies want to ensure that their most promising drug candidates eventually hit the market. But while speed to market is a crucial element, applying a structured approach starting at an early stage can help de-risk the drug development process and avoid costly late-stage failures. 

Priorities during early stage 

Efficacy and safety. During the early stages of a drug development program, time is of the essence, notes Torkel Gren, general manager at Recipharm Pharmaceutical Development. Establishing proof of efficacy and safety is the top priority, and formulation development is often left to later stages. Gren finds it strange that such an important activity as formulation development is often overlooked and started too late. “But I see it all the time,” he says. “Of course, there is no point starting formulation development before we have selected the drug candidate. The API has to be manufactured in sufficient quantities before formulation development can begin. However, to a certain extent, you can start formulation development with small quantities of non-GMP material. This approach is, no doubt, risky because solid-state properties of different batches can vary considerably at this stage. Nevertheless, especially with simpler formulations, useful information can often be collected from work with early batches.” 

The rate of attrition is high during clinical trials with only 10% of new drug candidates making it to market, according to Rob Harris, chief technical officer at Juniper Pharma Services. “Therefore, there is a natural reluctance within the industry to invest in the development of a robust dosage form until there is some confidence in the safety and efficacy of the drug.”   

Pharmacokinetics. Anil Kane, executive director and global head of Technical and Scientific Affairs at Patheon, observes that once a molecule has been selected as a potential candidate, all efforts tend to focus on proving that the drug substance gets absorbed in an animal species before a first-in-human Phase I study. “Typically, a formulation is prepared by simply suspending the drug in a hydroxypropyl methylcellulose suspension, with or without a surfactant, which is then dosed to an animal so that an absorption profile can be established,” he says. “In a similar manner, the first dose format prepared for a quick Phase I study is usually a neat API in a capsule.” 

Kane cautions that while these methodologies or procedures are simple and fast, hence, allowing pharmacokinetic assessments and safety studies to be performed, the results obtained are not always promising. “In many instances, it’s found that the drug was not properly suspended, due to hydrophobicity and poor wettability. In Phase I clinical programs, the results can show that the API was not absorbed sufficiently to obtain the desired plasma levels and there was not enough exposure,” he says. Gaining speed by deferring formulation development to a later stage has, in many instances, not met the goals of early development, Kane observes. 

Aaron Goodwin, principal investigator, Internal Research and Development, Capsugel, now a Lonza company, shares a similar view. He agrees that preclinical formulation development is generally focused on achieving a desired pharmacokinetic response in a translatable animal model while minimizing formulation development time and cost. Goodwin, nonetheless, points out that this can be challenging for molecules with poor bioavailability or a narrow therapeutic window given that in vitro–in vivo relationships are often not straightforward and, therefore, require additional time and cost to establish. “It’s tough to justify the additional cost and timing given the low probability of success for a preclinical API especially for a new mechanism of action,” he says. “However, if not addressed adequately, the drug product can be at risk of reformulation or even fail to demonstrate clinical efficacy in late stage when the stakes are much higher.”

Giving an example, Goodwin explains that adequate exposure at an initial dose may be achievable in preclinical studies with a crystalline or perhaps a micronized suspension; and progressing a simple, conventional formulation may be the quickest way to the clinic. “But if the compound properties suggest that oral absorption may be solubility limited at high doses, and this is supported by preclinical in-vivo data, advancing a simple, conventional formulation may risk not reaching a maximally effective exposure,” he says. “Advancing an amorphous form might take a bit longer to get to the clinic, but even if not optimized, it would likely circumvent this risk as well as reduce exposure variability.”

Kane emphasizes that formulation development, based on a systematic drug substance characterization and material properties, is required to avoid such failures or loss of time, investment, and opportunity in early development. 


Formulation challenges

Solubility. One of the main challenges in developing formulations for new drugs is the poor solubility of many of the compounds emerging from drug discovery, notes Harris. He estimates that 80% of new drug candidates are classified as poorly soluble, and points out that oral bioavailability will be compromised if the drug is not completely dissolved in the gastrointestinal tract. 

Karl Werner, senior director, Finished Product Development, Midas Pharma, concurs that the number of Biopharmaceutics Classification System class II and IV compounds has increased tremendously. “There are challenges when trying to formulate these drugs into an oral dosage form,” he says. “Poor solubility or bioavailability will result in incomplete or variable absorption, higher impact of pH and food on drug absorption, and poorly controlled pharmacokinetics.” In such cases, it is extremely important to develop a formulation that maximizes the chance of good exposure even if doing so requires additional time and cost, Gren stresses. But he also points out that in some situations, it is worth asking if formulation development is really the right solution to a solubility/bioavailability problem or if modifying the molecule may be a better way to develop a successful drug product.  


For weakly acidic or basic drugs, salt formation is often the preferred method for improving solubility because it is simple and effective. Other methods that have been widely used to address solubility issues in early development include particle size reduction by micronization or nanomilling to increase the dissolution rate, the use of surfactants or cosolvents, and complexation with cyclodextrins. But with the solubility of new molecules becoming more demanding, Harris notes that the application of more advanced solubilization technologies, such as lipid-based and self-emulsifying drug-delivery systems and amorphous solid dispersions, in early development is becoming more prominent. 

Polymorphism and stability. “Formulation strategies in early-stage drug development entirely depend on the complexity of the molecule and its critical nature and behavior,” says Kane, noting that some molecules may exhibit polymorphism, for example. “It is, therefore, important to understand the different polymorphic forms, their stability and properties, and potential to convert from one form to another,” he explains. “The formulation strategy would then be based on preventing the conversion of the selected polymorph and ensuring its stability throughout the clinical stability or shelf life of the product.” 

Dose range. Uncertainty over the dose range to be administered is another complication during early development, according to Harris. “A formulation strategy suitable for a 5–50 mg dose range is unlikely to be appropriate for a 100–1000 mg dose range and vice versa,” he says. “However, there are modeling software tools, such as GastroPlus, which can be used to predict an appropriate human dose based on physicochemical and animal pharmacokinetic data for the drug substance.”   

Limited amount of API. “Early-stage drug development is a balance between minimizing time to clinic and developing a progressable formulation that meets pharmacokinetic targets while at the same time has minimal pharmacokinetic variability,” says Goodwin. “This can be remarkably difficult when there is limited API for development.” In such cases, he recommends using API-sparing in-vitro tests to assess what types of formulations are likely to achieve clinical pharmacokinetics targets.

“Only minimal amounts of API are necessary to measure the neutral crystalline solubility as a function of pH, micelle partition coefficient, and amorphous solubility along with precipitation propensity. These values, along with estimates of permeability, can be input into simple models such as a maximum absorbable dose and dose number and dissolution rate models to get a basic understanding of absorption potential,” Goodwin explains. According to him, these basic models lack a high degree of accuracy, but they are useful for identifying the limiting factor for absorption and providing an initial estimate of an absorption dose response. “Ultimately, these estimates can be used to determine if a drug-delivery technology is needed to achieve the target pharmacokinetic profile,” Goodwin says. “And if an enabling technology is required, the information can indicate whether simple particle size reduction would work; or if the compound is ionizable, whether a high-solubility polymorph or salt form is likely to be adequate; or whether an amorphous form or lipid solution is necessary. This type of methodology decreases development time and API use by reducing the number of prototype formulations and in-vivo studies to nominate a proof-of-concept formulation for first-in-human studies.”

Understanding the drug molecule

Before selecting a formulation strategy, one needs to have a thorough knowledge of the physicochemical properties and biological attributes of the drug substance, in particular its solubility and intestinal permeability characteristics, Harris explains. Knowing what dose to administer is also important, he adds, because this information allows the formulator to select the most appropriate formulation strategy for the compound. 

Kane agrees that a systematic understanding of the molecule, its properties and challenges, and a sound, phase-appropriate formulation development strategy that addresses those challenges is key to success in early development. “For example, solid dispersions or lipid-based formulations can be used to address poor bioavailability challenges,” he says. “For molecules that are unstable, there are various ways of stabilizing the molecule through a detailed understanding of its forced degradation profile, ensuring early- and late-stage drug product stability.” 

Drug structure and physiochemical properties can also be used to assess formulation risks, according to Goodwin. “Melting point and glass transition temperature are examples of two physiochemical properties that are very informative for advanced drug-delivery technologies and easily measured with small amounts of API,” he says. “For example, an API with a low glass transition temperature would suggest that it may be difficult to achieve a stable amorphous form with high API loading in the drug product. A high melting point would suggest that it will likely be challenging to have a practical solubility in a lipid vehicle unless the dose is quite low. Although these are just a few examples, these physiochemical properties provide the basis for a preclinical risk assessment when evaluating enabling technologies for in-vivo performance, drug product stability, and manufacturability.”



Another objective of early-stage development is to ensure the formulation and process developed in Phase I can be transitioned into a scalable, manufacturable process in late-stage development, Kane points out. “The objective of late-stage development is to scale up the early-stage clinical formulation and process to a larger-scale product that can be commercialized, while ensuring and establishing its shelf life to be commercially viable. Keeping an eye on and envisioning a larger-scale process early on will help address challenges early in development. Designing robust processes with a thorough understanding of critical processing parameters that impact drug quality and defining the control strategies at a reasonable scale will ensure a robust manufacturing platform in late-stage development,” he says. 

Werner stresses that the risk of failure of a molecule is not only related to its pharmacological and pharmacokinetic properties or its toxicity, but also to its manufacturability. With reference to Midas Pharma’s systematic screening process for “difficult to formulate” drugs, he recommends including such technical considerations at an early development stage. He cites a project where solubility enhancement was achieved for a molecule, but pilot bioequivalence testing failed because it was found that higher scale production significantly changed the physicochemical behavior of the formulation. “This resulted in a substantial increase in development costs and an extended timeline, which likely could have been avoided by an earlier testing of the robustness of the formulation in terms of its scalability,” he says. 

Keeping in mind the required commercial viability streamlines the development of new drug products and significantly decreases the risk of failure at later stages, according to Werner. In another example involving a highly potent and poorly soluble compound, he explains how applying the right methodology can help avoid scale-up issues. “The bioavailability of the compound had to be improved and at the same time, become less variable with regard to food effects,” he says. “Several technologies were applied, including size-reduction (nanozation), spray-drying, and hot-melt extrusion. The nanoparticles agglomerated, and the first animal trial displayed highly variable adsorption rates. Application of amorphous suspensions, either prepared by solvent-based method (spray drying) or temperature-based method (hot-melt extrusion), showed more promising results. However, manufacturing trials revealed that the yield of the spray-drying process was substantially lower and the use of an uncommon organic solvent was required, which negatively affected the intended commercial viability. Hot-melt extrusion, on the other hand, not only enabled the desired positive effects on bioavailability and variability, but also helped to avoid the dust formation typical for a spray-drying process (dust formation is undesirable for highly potent drugs).” Hot-melt extrusion proved to be superior in this case.

The right approach 

Early-stage formulation strategies should not only focus on developing a dosage form that can be manufactured using simple and cost-effective processes but also one that offers a high probability of clinical success. “A sound formulation strategy is one that is ‘phase-appropriate’ and not over-engineered to a large commercial-scale in early development,” says Kane. “When developing a new drug, the chemistry, manufacturing, and control (CMC) strategy should involve a systematic drug substance characterization upfront to understand the molecule, its material properties, physical and chemical stability issues, and addressing these early in a phase-appropriate formulation development strategy.” Kane explains that this approach ensures a significant advantage by not needing to go back and address one or more challenges as they appear in the development process. 

According to Goodwin, efficient drug product development can be accomplished using a risk-based approach to formulation design based on fundamental- and mechanistic-based models. The identification of high-risk areas provides a basis for initial formulation selection and optimization. Harris also stresses that risk mitigation is crucial throughout the whole drug development process. “For early stages, it is vitally important to understand as much as possible about the attributes and behavior of the drug substance. This allows for selection of a formulation strategy that gives the compound the best chance of success in clinical trials,” Harris says. 

One future prospect that is already showing potential in dramatically reducing the amount of API and time necessary for drug product development is machine learning, Goodwin observes. “If trained on a wide diversity of data, these types of algorithms have the potential to predict drug-delivery technologies and processes that have the highest probability of successfully delivering an API-all based on a few simple API-sparing in-vitro measurements and calculated properties of the molecule,” he says. “In the past decade, this approach has gained a lot of recognition for predicting absorption, distribution, metabolism, elimination, toxicity attributes such as permeability, transporter and metabolism substrates, and estimates for clearance and volume of distribution.” Goodwin believes this methodology has huge potential in advancing drug development because of the high degree of prediction accuracy.

Article Details

Pharmaceutical Technology
Vol. 41, No. 10
Page: 20–27


When referring to this article, please cite it as A. Siew, “Formulation Strategies in Early-Stage Drug Development," Pharmaceutical Technology 41 (10) 20–27 (2017).