AAPS PharmSciTech

, Volume 19, Issue 1, pp 1–10 | Cite as

Commentary: Why Pharmaceutical Scientists in Early Drug Discovery Are Critical for Influencing the Design and Selection of Optimal Drug Candidates

  • Margaret S. LandisEmail author
  • Shobha Bhattachar
  • Mehran Yazdanian
  • John Morrison
Open Access
White Paper


This commentary reflects the collective view of pharmaceutical scientists from four different organizations with extensive experience in the field of drug discovery support. Herein, engaging discussion is presented on the current and future approaches for the selection of the most optimal and developable drug candidates. Over the past two decades, developability assessment programs have been implemented with the intention of improving physicochemical and metabolic properties. However, the complexity of both new drug targets and non-traditional drug candidates provides continuing challenges for developing formulations for optimal drug delivery. The need for more enabled technologies to deliver drug candidates has necessitated an even more active role for pharmaceutical scientists to influence many key molecular parameters during compound optimization and selection. This enhanced role begins at the early in vitro screening stages, where key learnings regarding the interplay of molecular structure and pharmaceutical property relationships can be derived. Performance of the drug candidates in formulations intended to support key in vivo studies provides important information on chemotype-formulation compatibility relationships. Structure modifications to support the selection of the solid form are also important to consider, and predictive in silico models are being rapidly developed in this area. Ultimately, the role of pharmaceutical scientists in drug discovery now extends beyond rapid solubility screening, early form assessment, and data delivery. This multidisciplinary role has evolved to include the practice of proactively taking part in the molecular design to better align solid form and formulation requirements to enhance developability potential.


drug candidate selection preclinical formulation development drug candidate design pharmaceutical properties 


More complex biological targets and modes of drug action, such as the recent focus on the benefits of allosteric modulation [1] and polypharmacology [2], have shifted molecular parameters [3] out of the traditional “rule of five” chemical space. Unsurprisingly, many drug candidates from this non-traditional space demonstrate poor physicochemical properties such as aqueous solubility [4,5]. Over the past two decades, there has been an increasing effort to bring new drug solubilization and delivery technologies earlier in the development cycle and enable successful new molecular entity (NME) progression to become viable clinical and commercial therapies [6]. Advanced delivery technologies can also be successfully applied in the early discovery space for evaluation of critical “tool” molecules such as known competitor compounds, as well as positive or negative control compounds in exploratory preclinical proof of concept or safety studies. The need for enabled drug delivery measures to support a wide variety of preclinical studies has been greatly facilitated by the inclusion of formulation and pharmaceutical scientists in drug discovery teams.

Traditionally, the pharmaceutical scientist role has supported a multitude of activities in the early space including [7]
  • Evaluation of physiochemical properties

  • Biopharmaceutical evaluation

  • Formulation development for preclinical in vivo studies

In early physicochemical property evaluations, the focus is on rapid generation of solubility and stability as well as initial solid form and salt screening data (e.g., crystallinity and hygroscopicity). These evaluations typically occur in a moderate to high throughput fashion to keep pace with the rapid generation of candidate compounds to provide timely feedback to discovery teams. In addition to collecting and interpreting a multitude of physicochemical property data, pharmaceutical scientists highlight both the development opportunities and the risks of progressing particular drug candidates. Predictive solid-state characterization technologies for computing the crystal energy landscape show promise in guiding teams towards solid forms with ideal solid-state properties for optimum drug delivery [8].

Pharmaceutical scientists generally lead solid form selection and formulation strategy as part of the biopharmaceutical evaluation of lead compounds. As information about the chemical series develops, the limiting biopharmaceutical aspects of the molecules are often mitigated through formulation. Thus, optimal exposures and desired pharmacokinetic profiles can be realized. Occasionally, the designated delivery systems fail to achieve the desired exposure profiles and/or bioavailability, which leads to a situation where an otherwise promising candidate is not advanced. However, this failure often provides important and vital information regarding the limiting physicochemical properties of the molecule and/or compatibility with the chosen formulation. The learnings associated with these failures provide the opportunity to design more robust molecular scaffolds and subsequent drug candidates that work well with the intended formulation and delivery strategy.

At the compound optimization stage, formulation development for candidate molecules occurs rapidly to keep pace with the needs of the multitude of efficacy, pharmacokinetic, and toxicological studies. Solution, suspension, and other solubility-enabling formulations are developed to evaluate candidates and tool compounds in a number of different in vivo models. Often, delivery of high doses is necessary for non-orally optimized molecules with challenging physical chemical properties. The time investment to solve these delivery challenges is often worthwhile, because the information gained from these studies is invaluable in exploring structure-activity or structure-safety relationships. In the course of developing formulations for these studies, feedback on the delivery risks and development hurdles for the key lead proprietary candidates can be highlighted. Once the trends in the limiting or challenging physicochemical properties of the lead compounds are determined, the series can often be optimized for delivery, particularly when use of enabled dosage forms is required.

The shared experiences of the authors as pharmaceutical scientists supporting drug discovery teams in different companies indicate the value and necessity of a more active and expanded role for the discipline within discovery teams. This expanded role includes assessing and providing guidance on molecular factors and elements that limit enable formulations (e.g., solubility enhancement) for optimized drug delivery. This philosophy of “designing with delivery in mind,” combined with the knowledge and expertise of pharmaceutical scientists on the necessary structure and molecular attributes is envisioned to enable efficient design and selection of compounds, solid forms, and delivery formulations. This in turn will aid successful clinical translation, clinical development, and commercialization efforts. A more detailed assessment of the current and future directions for pharmaceutical scientists to support the selection and development of new chemical entities from in vitro studies at early stage drug discovery to the use of enabling technologies for development of clinical formulations is expressed in the following section.


Involvement in early drug discovery provides pharmaceutical scientists the opportunity to influence structural design and ensure that clinical drug candidates have biopharmaceutically suitable or so-called “druggable” features. Thousands of compounds are screened with in silico tools and in vitro assays against predefined thresholds to yield a smaller yet still substantial subset of promising candidates for further in vivo pharmacodynamic and pharmacokinetic testing. Molecular knowledge gained at the early stages is broad but not deep, yielding a lot of data across several chemotypes or structurally similar compounds, but little specific data for any one compound. Emerging structure analysis tools, such as the Quantitative Estimate of Drug Likeness score (QED), are very helpful in quickly assessing a drug candidate for degree of drug-likeness and are useful for comparing large molecular data sets for clues about limiting pharmaceutical proprieties or molecular properties that are useful for efficient formulation [9]. Pharmaceutical scientists must therefore use prior experience, insight, new assessment tools, and judgment to interpret the available information and guide compound progression and selection decisions, an ability which can be as much art as science.

In Vitro Screening Support and Interpretation

In vitro assays are used to establish whether compounds possess the desired target affinity, off target toxicity liabilities, and biopharmaceutical properties such as permeability, metabolism, protein binding, and solubility [10]. These experiments are tiered to balance resource use, with the most critical and least resource intensive data collected first. As a result, greater amounts of information are collected on fewer compounds as they are successively filtered through each testing tier [11] (Fig. 1).
Fig. 1

Depth of knowledge gained during the drug discovery process (adapted from [11])

An unintended consequence of the initial focus on target affinity is that it often yields compounds with poor aqueous solubility [4,5]. This can be problematic for subsequent assays requiring higher concentrations. To overcome this challenge, compounds are predissolved in an organic solvent such as dimethyl sulfoxide (DMSO) prior to dilution into the aqueous assay media. This can create a thermodynamically unstable supersaturated state [12] that can lead to precipitation or container surface adsorption [13,14].

Nephelometry [15] is a turbidimetric assay commonly used to provide confidence that a negative assay readout is a true reflection of the compounds’ lack of activity rather than a solubility limitation. Essential structure-property relationship data can be collected for these poorly soluble compounds by incorporating an adsorption/precipitation inhibiting excipient into the assay media, albeit one which does not interfere with the assay readout [16]. Counterintuitively, false negative results due to precipitation or surface adsorption can also be beneficial for pharmaceutical scientists since compounds with poor physicochemical properties tend to be removed early from further consideration as drug candidates.

While solubility is a critical biopharmaceutical property impacting all areas of drug discovery, measuring equilibrium solubility values across large compound sets is not practical. Nephelometry and direct supernatant concentration measurements provide an early assessment of solubility. Unfortunately, these kinetic solubility values tend to overpredict equilibrium solubility, sometimes by orders of magnitude [15]. Despite this, the data can provide some insights. For instance, tracking kinetic solubility progression over time both across and between chemotypes provides an early warning of potential solubility risks for the ultimately selected drug candidates.

In Vivo Study Support and Interpretation

Promising drug candidates meeting the in vitro assay screening criteria are scaled up for further preclinical pharmacodynamic and pharmacokinetic profiling studies. At this stage, both the physicochemical nature of the compounds and the formulation strategy play crucial and sometimes underappreciated roles.

Material scale-up can result in a more thermodynamically stable polymorphic form, especially if the synthetic route or purification procedure has been further optimized. Such form changes may lower solubility, both in gastrointestinal fluids and in formulation vehicles. Early discovery formulations must overcome solubility limitations and satisfy several additional criteria:
  • Universality—effective delivery across a wide range of compounds

  • Capacity—effective delivery across a large range of doses

  • Throughput—efficient and consistent preparation across multiple dosing groups

  • Tolerability—safe to test animals and non-interfering with the assay readouts

Compatibility between the chemotype and the formulation is established early using a small subset of representative compounds, but the formulation may require modification as the chemotype progresses. The particular formulation strategy employed can also influence chemotype evolution, for example:
  • Emulsions favor more lipophilic compounds

  • Cyclodextrin solutions favor compounds with better complexing efficiency

  • Suspensions favor compounds with better dissolution and/or solubility characteristics

  • Supersaturated formulations favor compounds with greater inherent supersaturation tendency

The need for throughput also precludes the use of more advanced formulations such as amorphous solid dispersions. These early compounds may not be optimized for oral delivery and may be susceptible to high first-pass metabolism [17], necessitating delivery by parenteral routes such as intraperitoneal, subcutaneous, or intravenous administration in order to collect critical structure activity relationship information. In general, oral dosing is the ultimate and preferred route of administration and the metabolic liabilities are designed out of the molecule during further optimization iterations.

The potential for increased oral absorption from a supersaturated formulation can also be an important point of differentiation during compound optimization, especially for a poorly soluble compound [18]. Supersaturation is achieved by transferring a compound from a higher energy environment to a lower energy environment [19], such as
  • A higher energy solid form (amorphous, salt, or cocrystal)

  • Fast intestinal dilution of a cosolvent-containing vehicle

  • The gastric-to-intestinal pH shift for weakly basic compounds

The supersaturation advantage may be negated by rapid precipitation, and gastrointestinal differences in pH and bile content [20,21] between test animals can lead to different precipitation profiles and oral exposure variability. Precipitation inhibiting formulations and excipients is employed to reduce this variability, but care must be taken that the chemical potential driving force for intestinal increased absorption is not unintentionally lowered through oversolubilization that exceeds supersaturation limits [22].

It is also valuable for pharmaceutical scientists to look beyond standard pharmacokinetic readouts [C max, T max, and area under the curve (AUC)] and examine individual plasma profiles to observe the source of variability [23] (Fig. 2). In cases where alternative or enabling formulations do not improve exposure, physiological factors may be responsible. Codosing with an efflux inhibitor such as elacridar [24] can diagnose whether compounds and perhaps chemotype moieties are efflux substrates. Portal and jugular vein cannulated pharmacokinetic studies are useful in distinguishing poor intestinal absorption from hepatic first-pass effects [25,26], respectively. An improved formulation may overcome the former but not the latter.
Fig. 2

Plasma concentration versus dose relationship in toxicology testing (reproduced with permission from [54])


The crystallinity of solid forms and their stability are closely linked to the physicochemical and biopharmaceutical properties of new chemical entities. Many drug candidates can exist in more than one solid crystalline form, or polymorph. These forms are characterized by their unique X-ray powder diffraction, unit cell dimensions, and melting point. Changes in the crystalline structure such as defects and imperfections in the crystal lattice can change the apparent solubility and affect dissolution rates [27]. For example, amorphous solids lack crystallinity and are often more soluble than their crystalline counterparts. However, amorphous solids are thermodynamically unstable and pose a developability risk compared to the more physically and chemically stable crystalline solids. Thus, evaluating the potential for multiple solid forms of a drug candidate is an important aspect of the developability assessment as it may affect solubilization, compaction, and flow and impact manufacturability of dosage forms.

Forming salts or cocrystals represents alternative solid phases of a compound that can improve dissolution, solubility, hygroscopicity, and stability. Salts are formed by complete proton transfer to create ion pairs, whereas cocrystals are formed from an incomplete proton transfer or proton sharing. More specifically, cocrystals are defined as “crystalline structures made up of two or more components in a definite stoichiometric ratio where the arrangement is not based on ionic bonds” [28]. Cocrystal formation is mostly applicable to compounds with weakly ionizable groups that cannot form true salts.

Salts, cocrystals, and amorphous forms are also used in early development to improve dissolution which in turn can maximize exposure in preclinical pharmacokinetic and toxicology experiments. For developability assessments of these forms, the effect of temperature, humidity, compression, and grinding should be tested to determine if there are any solid form changes. Such form changes can, in turn, affect storage, stability, excipient compatibility, and manufacturing processes. Additionally, when salts and cocrystals are in solution or suspension, the physical instability of these forms due to disproportionation should be considered [29].

Understanding the relationship between the different solid forms and the most stable thermodynamic form is essential for ensuring a reproducible manufacturing process for both drug substance and drug product. Screening for possible crystalline forms of a drug candidate can be accomplished through either manual or moderate to high throughput procedures to assess for different potential polymorphs, salts, and/or cocrystals. Both methods typically involve a number of techniques to crystallize the drug candidate: use of various solvents, fast or slow evaporation, addition of anti-solvents, and possibly by mechanical stress [26]. Early development often involves testing a subset of these conditions to minimize material consumption when supplies are limited. More extensive screening can be pursued as more material becomes available. These screens are designed to maximize success rates in selecting a thermodynamically stable solid from that can be scaled up and reproduced successfully.

Yang et al. have recently shown the most stable polymorphs for benzene, a small, simple, symmetrical, and rigid organic molecule, can be predicted. However, predicting the most stable polymorphic forms of more complex and flexible small molecule drugs has remained elusive [30].


Absorption modeling of orally administered compounds is the pivotal component that connects the solid (or physical) form of the compound, the delivery system, the intended pharmacokinetic and toxicokinetic profiles, and the desired therapeutic outcome. A number of tools of varying levels of sophistication are available for this purpose and, when used appropriately, can provide valuable information with multifaceted applications. Absorption modeling can be used to guide formulation selection for pharmacology and toxicology studies, establish particle size parameters for clinical dosage forms, and enable formulation development and solid form (salt and cocrystal) selection for clinical formulations. In the hands of an experienced pharmaceutical scientist, the use of these tools can minimize the number of in vivo studies that would otherwise be necessary to drive these important project decisions.

The most basic modeling tools use the measured solubility, intestinal permeability, and particle size as inputs into simple tools to estimate the maximum absorbable dose as a function of particle size [31]. These tools generally provide reasonable guidance and are very convenient to use, especially for early assessments where there is limited knowledge of compound properties. However, for compounds with a significant pH-dependent solubility or a high supersaturation propensity, measured equilibrium solubility at neutral pH may not adequately model dissolved drug concentration throughout the entire absorption window. In these cases, the use of biorelevant dissolution tools along with more sophisticated absorption modeling tools such as MiMBa®, GastroPlus®, STELLA®, or Simcyp® might be better suited [32, 33, 34, 35, 36]. The solid form from either undissolved or precipitated material should also be assessed as a change in polymorphic form can affect solubility and dissolution [37]. A number of excellent reports discuss the comparative features and applications of the wide variety of biorelevant dissolution technologies available [38,39].

Figure 3 illustrates an absorption modeling flow scheme, incorporating the in vitro absorption parameters, the available non-clinical pharmacokinetic data, and an in silico absorption model to derive clinical absorption parameters in the context of the planned clinical dose range. The final outcome is an estimate of the amount absorbed as a function of administered dose for a specified particle size. Alternatively, plots can be generated to predict the relationship between amount absorbed and particle size at a specified dose.
Fig. 3

Illustration of the absorption modeling flow scheme

Absorption may also be influenced by gastric pH and feeding state leading to oral exposure variability, depending on the pH-dependent solubility profile and dissolution properties of the drug candidate. These factors should be characterized and appropriately applied in designing formulations as well as clinical studies. For Biopharmaceutics Classification System (BCS) Classes 2 and 4 compounds, particle size reduction alone may not provide adequate absorption over the desired clinical dose range. It may therefore be necessary to develop alternate options such as salts, cocrystals, amorphous solid dispersions, lipid-based, or nanocrystalline formulations to provide the desired exposure profile.

The in vivo dissolution behavior of crystalline solids can be relatively well characterized with standard biorelevant in vitro dissolution tools. However, the dissolution, supersaturation, and precipitation of enabled solids are more complex and the assessment tools are still evolving [40]. Consequently, judicious application of orthogonal dissolution screens and careful analysis and modeling of data can provide reasonably predictive results [41,42].

While the importance of modeling clinical absorption to inform drug product development and clinical research cannot be overstated, the same tools can be used to inform non-clinical in vivo studies in pharmacology and toxicology [43]. When appropriately applied, these tools can vastly influence compound progression in discovery lead optimization, increase the technical success of the studies, and minimize wasteful synthesis of material that might go unabsorbed despite the best enablement technologies.


In the course of a drug discovery program, many different compounds are evaluated during early stage formulation development. Typically, predicted and/or measured physicochemical properties are utilized to guide the initial formulation selection process. More recently, there is a high prevalence for utilizing enabled formulations early, when the biopharmaceutical properties of the chemical matter are unoptimized and often high doses as well as larger dose ranges are needed to fully explore efficacy and safety end points.

During the formulation screening and development process, “structure-formulation performance relationships” can be constructed with the understanding that these are subtle at such an early stage. Careful observations on which formulation approaches succeed or fail may be correlated to possible/probable specific structural attributes. Working hypotheses for structure-formulation performance relationships should also be developed for key lead series. From these relationships, the formulation scientist can recommend areas of the molecule to be optimized to increase the performance of the enabled formulation, thereby enabling a more robust drug delivery.

General descriptions of molecular properties that guide enabled formulation selection are outlined in Table I. This information includes the typical physicochemical space and preferred solid form of the drug candidate for each formulation approach. If a formulation platform appears promising for a certain series, the formulation scientist can guide the drug design teams to remain within a designated chemical space and incorporate any molecular changes determined from the structure-formulation performance observations which will improve or enhance the drug’s performance in that formulation.
Table I

Enabling Formulation Selection Criteria and Preferred Solid Form Recommendation

Enabled formulation approach

Typical physical chemical space

Preferred solid form

Non-aqueous pharmaceutical solvents (polyethylene glycols, propylene glycol, etc)

• Moderate lipophilicity (Log P), generally 1–3

• Utility increases with decreasing melting point (T m)

Salt a of base or free form of acidic or neutral molecules

Lipophilic self-emulsifying drug delivery systems (SEDDs): Type I, II, and IIIA [48]

• Log P >4.0

T m <225 °C,

• High solubility in triglycerides oils and lipid components

Free form (no salt forms)

Spray dried dispersions in cellulosic polymers (SDD)

• Methanol, acetone, ethyl acetate, or THF solubility >5 mg/mL

T m/T g (K) <1.4

T m <250 °C,

• Log P 1–7

• Chemically stable between pH 3–5 (for acid-containing polymers)

Free form (no salts)

Wet-milled nanosuspension

• Aqueous solubility <150 μg/mL, prefer in μg/mL range

• Log P 0–8

• Melting point 100–350 °C

• Crystalline

• Mechanical properties of brittleness

Salt or free form

Cyclodextrin-containing solutions

• Compatible aryl substitution and geometry required for complexation

Salt or free form

a Some “hard” acid salts such as hydrochloric, sulfate, and phosphate may be less soluble in pharmaceutical solvents than the free form

Non-aqueous pharmaceutical solvent solubilization (polyethylene glycols, propylene glycol, etc) is an early formulation approach that can be utilized for both oral and parenteral dosing [44]. This type of formulation platform can be translated into liquid filled capsules for drug product production to support clinical and commercial drug delivery. In most cases, non-ionized acidic or basic as well as neutral molecules are best to achieve optimal solubilization; however, salt forms of bases ameliorate potential oxidative chemical stability risks. General molecular and physical chemical property considerations for efficient formulation include a moderate lipophilicity (Log P, between1 and 3), the presence of several hydrogen bond acceptors, and a moderate melting point, T m (i.e., T m <200 °C) [45]. Relative to dose, the drug candidate should have sufficient solubility and long-term physical stability in these solvent systems.

An additional factor to consider is the high precipitation potential which solvent systems pose upon introduction to aqueous environments. This phenomenon is a major concern for bioperformance, and the risk increases with increasing dose. Precipitation inhibiting excipients is typically screened and utilized in these formulations. Screening for lipophilicity, crystallinity (melting point), solubility, and precipitation potential as well as chemical and physical stability in these solvent systems are key factors in guiding candidate selection. These factors should be considered early and throughout the design process. Alternatively, compounds may precipitate as an amorphous form through either liquid-liquid phase separation (LLPS) or liquid-solid phase separation (LSPS) [46,47]. In these cases, the supersaturation advantage is not completely lost as the amorphous form tends to be more soluble and faster dissolving than the original crystalline material.

Self-emulsifying formulations (SEDDs) of the lipophilic type, such as Type I, II, and IIIA [48], are often considered for hydrophobic compounds with moderate to high Log P (>3) and compounds with melting points less than 225 °C and preferably less than 150 °C. Formulation development factors include the solubility of the drug in the lipoid excipient components, the chemical and physical stability of the drug in the formulation, and the amphiphilicity of the drug, which can correspond to the degree of association and solubility of the drug in the surfactant excipients. Lipids can form peroxides over time and drug candidates with strongly basic sites (conjugate acid pKa >6) may be prone to oxidation. Consequently, recommendations to design teams may involve the decreasing of the basicity of free amine sites. Crystallization of the drug substance from the formed emulsion (i.e., after dispersal in water and/or following lipid digestion) is often a concern and molecular design factors which can decrease melting point, such as removal of hydrogen bond donors and acceptors, can alleviate emulsion crystallization problems.

Spray dried dispersions (SDD) to stabilize the amorphous form of a drug candidate in cellulosic polymers are a common enabled solid form alternative for insoluble crystalline materials. Key parameters that dictate successful SDD formulation include Log P, crystalline melting point (T m), and the amorphous glass transition temperature for the drug substance. Molecules with moderate Log P values [2, 3, 4] and fewer hydrogen bond donating or accepting heteroatoms are generally preferred. The amorphous materials should have relatively high T g values (i.e., >40 °C) for optimal physical stability to reduce the likelihood of crystallization. Molecular features that affect the glass transition temperature of small molecules include the molecular shape and volume, as well as the type and degree of shielding of hydrogen-bonding capable functional groups [49].

Wet-milled nanosuspension (nanocrystal) formulations perform best when the drug candidates are crystalline and have a low aqueous solubility (i.e., <150 μg/mL) to prevent ripening and crystal regrowth. The bulk drugs should have high brittleness properties and low ductility and elasticity properties [50] to enable efficient size reduction through attrition milling. This deliver option applies to a broad range of drug molecules, with melting points generally 100–350 °C and Log P values of 0–5.

Aqueous cyclodextrin formulations are common for parenteral dosing of poorly soluble drugs. Solubility is improved via molecular complexation of the drug molecule within the more hydrophobic interior oligosaccharide “cage” of the cyclodextrin [51]. Generally, molecules demonstrate more efficient cyclodextrin complexation when they contain (i) monocyclic aromatic groups with little substitution or few ionizable centers at or near these aromatic rings, (ii) lipophilic groups, or (iii) low steric hindrance [52]. This information and the ability to predict cyclodextrin binding from a chemical structure can be utilized, but the pharmaceutical formulator to provide important guidance to medicinal chemists on what structural features will be beneficial for incorporation into the molecular framework.

Overall, there is a range of molecular features that pharmaceutical scientists can recommend to improve the performance of drugs candidates with enabled formulations. A critical understanding of the molecular features of lead compounds that affect formulation performance is gained through comprehensive early formulation screening and careful, thoughtful experimental observation.


Discovery and design efforts often extend into Phase 1 and Phase 2 clinical evaluation, with timely feedback informing optimization of backup candidates. This feedback may include the need for improved pharmacokinetic properties, physical or chemical stability, efficacy, safety, and selectivity.

The goal of the toxicology delivery strategy for any compound is to develop a safe formulation that provides sufficient in vivo exposures necessary to investigate the toxicology profile of the compound and establish a clear margin of safety. Determining the maximum tolerated dose (MTD) to demonstrate target organ or dose-limiting toxicity is a general expectation of regulatory authorities in support of clinical testing (Fig. 2). Ideally, the formulation provides a linear increase in exposure with dose. In practice, at least some dose-escalating exposure increase is often reasonable to achieve the desired exposure end points. Solubility limited molecules may exhibit some plateauing of exposure with dose, but ensuring that the formulation provides dose separation is critical to establishing safety margins.

In the absence of an MTD, other equally appropriate dose-limiting criteria may be considered, if met. For instance, establishing and characterizing either a maximum feasible dose (MFD) or an exposure-limiting dose (i.e., 1000 or 2000 mg/kg) which provides an acceptable exposure margin or a 50-fold exposure multiple relative to the clinical dose [53]. Several excellent publications on this topic describe the underlying principles of toxicology studies and formulation development in detail [54, 55, 56, 57].

In general, conventional aqueous suspensions or pH-adjusted solutions are preferred as they are fairly benign and can accommodate a broad dose range. These formulations are the safest and simplest of all options. However, when absorption is limited by low solubility and/or incomplete dissolution, alternate options are required to provide sufficient exposures and meet the needs of the toxicology studies. The various approaches described in Table I may be applied for developing these alternate formulations, with due consideration being given to the large doses that need to be delivered, the therapeutic targets under investigation, and the safety/acceptability of the vehicles for the durations of the studies.

Cosolvent and lipid-based vehicles offer simple alternatives to aqueous-based formulations; however, they may not be acceptable for long-term studies. Nanocrystalline suspensions and amorphous solid dispersions are often employed to support toxicology studies when conventional aqueous-based options are insufficient [58]. For compounds with the pH-dependent solubility, it is important to ensure that gastric pH variability does not impact the dissolution profile and thus the oral absorption of the compounds, especially in dogs [59].

The design and development of these enabling formulations are not an isolated activity. An integrated approach is necessary, incorporating a comprehensive understanding of the physicochemical and biopharmaceutical properties of the compound, the intended therapeutic target and dose, the desired safety margins, and the physiological aspects of the non-clinical species in which the studies will be conducted. It is also important to balance the benefits of the formulation options with the cost, potential risks, and technical complexities associated with their manufacture and handling.


Clinical formulations for Phase 1 and Phase 2a studies are generally designed to be as simple and inexpensive as possible to rapidly meet the clinical testing needs. The formulations should also offer flexibility for testing large dosing ranges, since doses are subject to change as additional information on non-clinical safety margins, clinical pharmacologic, and pharmacokinetic response are collected [60]. In developing the clinical formulations, it is extremely important for pharmaceutical scientists to have a comprehensive understanding of the planned clinical dose range, the dosing protocol (including in-clinic vs. take-home dosing and feeding state), the absorption parameters of the compound, the desired pharmacokinetic profile, and any disease/physiological parameters that might impact absorption in the clinical subjects. It is also important to ensure that the physical and chemical stability profile of the compound in question is well understood and will not impact delivery and bioavailability of the dose. Lastly, to the extent it is possible, clinical formulations should be developed with an appropriate line of sight to commercial development.

For immediate release oral delivery, the simplest clinical formulations would include solution or suspension formulations for the so-called “drug in bottle” (DIB) administration. Alternatively, neat drug candidate or simple dry blends can be supplied in a capsule or compounded into tablets for compression in a clinical pharmacy. Controlled release formulations can also be extemporaneously prepared in the clinical setting where necessary and can significantly speed up early clinical development [61].


Pharmaceutical scientists face a number of challenges and opportunities during the discovery and development of new medicines. In the discovery environment, they must adapt their functional expertise to the ever-changing needs of drug discovery teams. This requires employing specialized physicochemical, biopharmaceutical, and delivery strategy knowledge, as well as maintaining a general understanding of the efficacy, pharmacokinetic, and toxicological requirements of each particular program.

Based on the authors’ collective experience, pharmaceutical scientists can and should influence structure property relationships by “designing in” appropriate molecular features at the earliest stages of drug discovery. This can be done by ensuring compatibility with anticipated delivery strategies, such as solvent solubilization, lipidic emulsions, amorphous solid dispersion, particle size reduction, and salt or cocrystal formation. These strategies must maintain a “line of sight” to first in human (FIH) and later clinical studies and, ultimately, commercial formulation. This process, in turn, requires ensuring there are reasonable paths forward for both drug substance solid form, as well as the anticipated formulation and delivery strategies. Despite the complexities and challenges, the pharmaceutical scientists’ involvement during drug discovery and molecular design stages is crucial for the development of successful clinical candidates and ultimately new medicines for patients.


  1. 1.
    Conn P, Lindsley C, Meiler J, Niswender C. Opportunities and challenges in the discovery of allosteric modulators of GPCRs for treating CNS disorders. Nat Rev Drug Discov. 2014;13:692–708.CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    Peters J. Polypharmacology—foe or friend? J Med Chem. 2013;56(22):8955–71.CrossRefPubMedGoogle Scholar
  3. 3.
    Lipinski C. Lead- and drug-like compounds: the rule-of-five revolution. Drug Discov Today. 2004;14:337–41.CrossRefGoogle Scholar
  4. 4.
    Di L, Fish P, Mano T. Bridging solubility between drug discovery and development. Drug Discov Today. 2012;17(9–10):486–95.CrossRefPubMedGoogle Scholar
  5. 5.
    Augustijns P, Wuyts B, Hens B, Annaert P, Butler J, Brouwers J. A review of drug solubility in human intestinal fluids: implications for the prediction of oral absorption. European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences. 2014;57:322–32.CrossRefGoogle Scholar
  6. 6.
    Bhattachar SN, Bender DM, Sweetana SA, Wesley JA. Discovery formulations: approaches and practices in early preclinical development. In: Templeton A, Byrn S, Haskell R, Prisinzano T, editors. Discovering and developing molecules with optimal drug-like properties. AAPS Advances in the Pharmaceutical Sciences Series. 15. 1 ed: Springer-Verlag New York; 2015. p. 511.Google Scholar
  7. 7.
    Saxena V, Panicucci R, Joshi Y, Garad S. Developability assessment in pharmaceutical industry: an integrated group approach for selecting developable candidates. J Pharm Sci. 2009;98(6):1962–79.CrossRefPubMedGoogle Scholar
  8. 8.
    Price S, S. Reutzel-Edens. The potential of computed crystal energy landscapes to aid solid-form development. Drug Discovery Today. 2016(21):6.Google Scholar
  9. 9.
    Bickerton GR, Paolini GV, Besnard J, Muresan S, Hopkins AL. Quantifying the chemical beauty of drugs. Nat Chem. 2012;4(2):90–8.CrossRefPubMedPubMedCentralGoogle Scholar
  10. 10.
    Hughes J, Rees S, Kalindjian S, Philpott K. Principles of early drug discovery. Br J Pharmacol. 2011;162:1239–49.CrossRefPubMedPubMedCentralGoogle Scholar
  11. 11.
    Sinko C, editor How preclinical form and formulation optimization can contribute to improved R&D productivity Gordon Research Conference; 2013; Waterville Valley, NH.Google Scholar
  12. 12.
    Yamashita T, Ozaki S, Kushida I. Solvent shift method for anti-precipitant screening of poorly soluble drugs using biorelevant medium and dimethyl sulfoxide. Int J Pharm. 2011;419(1–2):170–4.CrossRefPubMedGoogle Scholar
  13. 13.
    Fukazawa T, Yamazaki Y, Miyamoto Y. Reduction of non-specific adsorption of drugs to plastic containers used in bioassays or analyses. J Pharmacol Toxicol Methods. 2010;61:329–33.CrossRefPubMedGoogle Scholar
  14. 14.
    Cai X, Walker A, Cheng C, Paiva A, Li Y, Kolb J, et al. Approach to improve compound recovery in a high-throughput Caco-2 permeability assay supported by liquid chromatography–tandem mass spectrometry. J Pharm Sci. 2012;101(8):2755–62.CrossRefPubMedGoogle Scholar
  15. 15.
    Hoelke B, Gieringer S, Arlt M, Saal C. Comparison of nephelometric, UV-spectroscopic, and HPLC methods for high-throughput determination of aqueous drug solubility in microtiter plates. Anal Chem. 2009;81:3165–72.CrossRefPubMedGoogle Scholar
  16. 16.
    Shah D, Paruchury S, Matta M, Chowan G, Subramanian M, Saxena A, et al. A systematic evaluation of solubility enhancing excipients to enable the generation of permeability data for poorly soluble compounds in Caco-2 model. Drug Metabolism Letters. 2014;8(2):109–18.CrossRefPubMedGoogle Scholar
  17. 17.
    Turner P, Brabb T, Pekow C, Vasbinder M. Administration of substances to laboratory animals: routes of administration and factors to consider. J Am Assoc Lab Anim Sci. 2011;50(5):600–13.PubMedPubMedCentralGoogle Scholar
  18. 18.
    Miller JM, Beig A, Carr RA, Spence JK, Dahan A. A win-win solution in oral delivery of lipophilic drugs: supersaturation via amorphous solid dispersions increases apparent solubility without sacrifice of intestinal membrane permeability. Mol Pharm. 2012;9(7):2009–16.CrossRefPubMedGoogle Scholar
  19. 19.
    Mathias NR, Xu Y, Patel D, Grass M, Caldwell B, Jager C, et al. Assessing the risk of pH-dependent absorption for new molecular entities: a novel in vitro dissolution test, physicochemical analysis, and risk assessment strategy. Mol Pharm. 2013;10(11):4063–73.CrossRefPubMedGoogle Scholar
  20. 20.
    Bergstrom CA, Holm R, Jorgensen SA, Andersson SB, Artursson P, Beato S, et al. Early pharmaceutical profiling to predict oral drug absorption: current status and unmet needs. European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences. 2014;57:173–99.CrossRefGoogle Scholar
  21. 21.
    Hatton GB, Yadav V, Basit AW, Merchant HA. Animal farm: considerations in animal gastrointestinal physiology and relevance to drug delivery in humans. J Pharm Sci. 2015;104(9):2747–76.CrossRefPubMedGoogle Scholar
  22. 22.
    Miller JM, Beig A, Krieg BJ, Carr RA, Borchardt TB, Amidon GE, et al. The solubility-permeability interplay: mechanistic modeling and predictive application of the impact of micellar solubilization on intestinal permeation. Mol Pharm. 2011;8(5):1848–56.CrossRefPubMedGoogle Scholar
  23. 23.
    Ding X, He M, Kulkarni R, Patel N, Zhang X. Investigation of clinical pharmacokinetic variability of an opioid antagonist through physiologically based absorption modeling. J Pharm Sci. 2013;102(8):2859–74.CrossRefPubMedGoogle Scholar
  24. 24.
    Levesque JF, Bleasby K, Chefson A, Chen A, Dube D, Ducharme Y, et al. Impact of passive permeability and gut efflux transport on the oral bioavailability of novel series of piperidine-based renin inhibitors in rodents. Bioorg Med Chem Lett. 2011;21(18):5547–51.CrossRefPubMedGoogle Scholar
  25. 25.
    Murakami T, Nakanishi M, Yoshimori T, Okamura N, Norikura R, Mizojiri K. Separate assessment of intestinal and hepatic first-pass effects using a rat model with double cannulation of the portal and jugular veins. Drug metabolism and pharmacokinetics. 2003;18(4):252–60.CrossRefPubMedGoogle Scholar
  26. 26.
    Matsuda Y, Konno Y, Hashimoto T, Nagai M, Taguchi T, Satsukawa M, et al. Quantitative assessment of intestinal first-pass metabolism of oral drugs using portal-vein cannulated rats. Pharm Res. 2015;32(2):604–16.CrossRefPubMedGoogle Scholar
  27. 27.
    Vippagunta S, Brittain H, Grant D. Crystalline solids. Adv Drug Deliv Rev. 2001;48(1):3–26.CrossRefPubMedGoogle Scholar
  28. 28.
    European Medicines Agency (EMA). Use of cocrystals and other solid state forms of active substances in medicinal products. Reflection Paper. 2014 Contract No.: CHMP/CVMP/QWP/136250/2014.Google Scholar
  29. 29.
    Lohani S, Cooper H, Jin X, Nissley BP, Manser K, Rakes LH, et al. Physicochemical properties, form, and formulation selection strategy for a biopharmaceutical classification system class II preclinical drug candidate. J Pharm Sci. 2014;103(10):3007–21.CrossRefPubMedGoogle Scholar
  30. 30.
    Yang J, Hu W, Usvyat D, Matthews D, Schutz M, Chan GK. Theoretical chemistry. Ab initio determination of the crystalline benzene lattice energy to sub-kilojoule/mole accuracy. Science (New York, NY). 2014;345(6197):640–3.CrossRefGoogle Scholar
  31. 31.
    Curatolo W. Physical chemical properties of oral drug candidates in the discovery and exploratory settings. Pharmaceutical Science & Technology Today. 1998;1(9):387–93.CrossRefGoogle Scholar
  32. 32.
    Ding X, Rose JP, Van Gelder J. Developability assessment of clinical drug products with maximum absorbable doses. Int J Pharm. 2012;427(2):260–9.CrossRefPubMedGoogle Scholar
  33. 33.
    Rohrs B. Biopharmaceutics modeling and the role of dose and formulation on oral exposure. In: Borchardt R, Kerns E, Hageman M, Thakker D, Stevens J, editors. Optimizing the “drug-like” properties of leads in drug discovery. Biotechnology: Pharmaceutical Aspects. IV. 1st ed. New York: Springer-Verlag; 2006. p. 512.Google Scholar
  34. 34.
    Kesisoglou F, Mitra A. Application of absorption modeling in rational design of drug product under quality-by-design paradigm. AAPS J. 2015;17(5):1224–36.CrossRefPubMedPubMedCentralGoogle Scholar
  35. 35.
    Huang W, Lee SL, Yu LX. Mechanistic approaches to predicting oral drug absorption. AAPS J. 2009;11(2):217–24.CrossRefPubMedPubMedCentralGoogle Scholar
  36. 36.
    Carino SR, Sperry DC, Hawley M. Relative bioavailability of three different solid forms of PNU-141659 as determined with the artificial stomach-duodenum model. J Pharm Sci. 2010;99(9):3923–30.CrossRefPubMedGoogle Scholar
  37. 37.
    Hawley M, Morozowich W. Modifying the diffusion layer of soluble salts of poorly soluble basic drugs to improve dissolution performance. Mol Pharm. 2010;7(5):1441–9.CrossRefPubMedGoogle Scholar
  38. 38.
    Reppas C, Friedel HD, Barker AR, Buhse LF, Cecil TL, Keitel S, et al. Biorelevant in vitro performance testing of orally administered dosage forms-workshop report. Pharm Res. 2014;31(7):1867–76.CrossRefPubMedGoogle Scholar
  39. 39.
    Brouwers J, Augustijns P. Resolving intraluminal drug and formulation behavior: gastrointestinal concentration profiling in humans. European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences. 2014;61:2–10.CrossRefGoogle Scholar
  40. 40.
    Kleppe MS, Forney-Stevens KM, Haskell RJ, Bogner RH. Mathematical models to explore potential effects of supersaturation and precipitation on oral bioavailability of poorly soluble drugs. AAPS J. 2015;17(4):902–17.CrossRefPubMedPubMedCentralGoogle Scholar
  41. 41.
    Chen Y, Liu C, Chen Z, Su C, Hageman M, Hussain M, et al. Drug-polymer-water interaction and its implication for the dissolution performance of amorphous solid dispersions. Mol Pharm. 2015;12(2):576–89.CrossRefPubMedGoogle Scholar
  42. 42.
    Qian F, Wang J, Hartley R, Tao J, Haddadin R, Mathias N, et al. Solution behavior of PVP-VA and HPMC-AS-based amorphous solid dispersions and their bioavailability implications. Pharm Res. 2012;29(10):2765–76.PubMedGoogle Scholar
  43. 43.
    Wuelfing WP, Kwong E, Higgins J. Identification of suitable formulations for high dose oral studies in rats using in vitro solubility measurements, the maximum absorbable dose model, and historical data sets. Mol Pharm. 2012;9(5):1163–74.CrossRefPubMedGoogle Scholar
  44. 44.
    Strickley RG. Solubilizing excipients in oral and injectable formulations. Pharm Res. 2004;21(2):201–30.CrossRefPubMedGoogle Scholar
  45. 45.
    Jouyban A. Review of the cosolvency models for predicting solubility of drugs in water-cosolvent mixtures. Journal of pharmacy & pharmaceutical sciences : a publication of the Canadian Society for Pharmaceutical Sciences, Societe canadienne des sciences pharmaceutiques. 2008;11(1):32–58.CrossRefGoogle Scholar
  46. 46.
    Ilevbare G, Taylor L. Liquid–liquid phase separation in highly supersaturated aqueous solutions of poorly water-soluble drugs: implications for solubility enhancing formulations. Cryst Growth Des. 2013;13:1497–509.CrossRefGoogle Scholar
  47. 47.
    Mosquera-Giraldo LLT. Glass-liquid phase separation in highly supersaturated aqueous solutions of telaprevir. Mol Pharm. 2015;12(2):496–503.CrossRefPubMedGoogle Scholar
  48. 48.
    Singh B, Bandopadhyay S, Kapil R, Singh R, Katare O. Self-emulsifying drug delivery systems (SEDDS): formulation development, characterization, and applications. Crit Rev Ther Drug Carrier Syst. 2009;26(5):427–521.CrossRefPubMedGoogle Scholar
  49. 49.
    Kalogeras IM. A novel approach for analyzing glass-transition temperature vs. composition patterns: application to pharmaceutical compound + polymer systems. European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences. 2011;42(5):470–83.CrossRefGoogle Scholar
  50. 50.
    Hancock BC, Carlson GT, Ladipo DD, Langdon BA, Mullarney MP. Comparison of the mechanical properties of the crystalline and amorphous forms of a drug substance. Int J Pharm. 2002;241(1):73–85.CrossRefPubMedGoogle Scholar
  51. 51.
    Brewster M, Loftsson T. Cyclodextrins as pharmaceutical solubilizers. Adv Drug Del Rev. 2007;59:645–66.CrossRefGoogle Scholar
  52. 52.
    Chen W, Chang C-E, Gilson MK. Calculations for cyclodextrin binding affinities: energy, entropy, and implications for drug design. Biophys J. 2004;87:3005–49.Google Scholar
  53. 53.
    International Conference on Harmonisation (ICH). Guidance on Nonclinical Safety Studies for the Conduct of Human Clinical Trials and Marketing Authorization for Pharmaceuticals M3(R2). Google Scholar
  54. 54.
    Bhattachar SN, Bender, D. M., Sweetana, S. A and Wesley, J. A. Discovery formulations: approaches and practices in early preclinical development. Discovering and developing molecules with optimal drug-like properties: Springer; 2015.Google Scholar
  55. 55.
    nmHiggins J, Cartwright ME, Templeton AC. Progressing preclinical drug candidates: strategies on preclinical safety studies and the quest for adequate exposure. Drug Discov Today. 2012;17(15–16):828–36.CrossRefGoogle Scholar
  56. 56.
    Neervannan S. Preclinical formulations for discovery and toxicology: physicochemical challenges. Expert Opin Drug Metab Toxicol. 2006;2(5):715–31.CrossRefPubMedGoogle Scholar
  57. 57.
    Palucki M, Higgins JD, Kwong E, Templeton AC. Strategies at the interface of drug discovery and development: early optimization of the solid state phase and preclinical toxicology formulation for potential drug candidates. J Med Chem. 2010;53(16):5897–905.CrossRefPubMedGoogle Scholar
  58. 58.
    Kesisoglou F, Mitra A. Crystalline nanosuspensions as potential toxicology and clinical oral formulations for BCS II/IV compounds. AAPS J. 2012;14(4):677–87.CrossRefPubMedPubMedCentralGoogle Scholar
  59. 59.
    Bhattachar SN, Perkins EJ, Tan JS, Burns LJ. Effect of gastric pH on the pharmacokinetics of a BCS class II compound in dogs: utilization of an artificial stomach and duodenum dissolution model and GastroPlus, simulations to predict absorption. J Pharm Sci. 2011;100(11):4756–65.CrossRefPubMedGoogle Scholar
  60. 60.
    Ku MS, Dulin W. A biopharmaceutical classification-based Right-First-Time formulation approach to reduce human pharmacokinetic variability and project cycle time from First-In-Human to clinical Proof-Of-Concept. Pharm Dev Technol. 2012;17(3):285–302.CrossRefPubMedGoogle Scholar
  61. 61.
    Thombre AG, Berchielli A, Rogers JF. Extemporaneously prepared controlled release formulations for accelerating the early phase development of drug candidates. Drug Discov Today. 2014;19(5):694–700.CrossRefPubMedGoogle Scholar

Copyright information

© American Association of Pharmaceutical Scientists 2017

Open Access This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (, which permits any noncommercial use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors and Affiliations

  • Margaret S. Landis
    • 1
    Email author
  • Shobha Bhattachar
    • 2
  • Mehran Yazdanian
    • 3
  • John Morrison
    • 4
  1. 1.Pfizer Global Research and Development Research FormulationsCambridgeUSA
  2. 2.Lilly Research LaboratoriesA Division of Eli Lilly and CompanyIndianapolisUSA
  3. 3.Pharmaceutics DepartmentTeva Branded Pharmaceutical R&D IncWest ChesterUSA
  4. 4.Discovery Pharmaceutics, Bristol-Myers SquibbWallingfordUSA

Personalised recommendations