How do you solve a problem like a poorly soluble API? Choosing an approach that maximises performance is critical. This means understanding the advantages and limitations of different process technologies. Bristol-Myers Squibb’s Ajit Narang takes us through the strategies worth considering for improved bioavailability.


When dealing with poorly soluble active pharmaceutical ingredients (APIs), it’s important to choose an approach that maximises formulation performance. Of the three key tenets of pharmaceutical development – stability, bioavailability, and manufacturability – poorly soluble APIs frequently encounter the problem of low and/or variable drug release leading to oral absorption and bioavailability concerns. Variable absorption of low-solubility drugs presents challenges to the accurate assessment of efficacy and the therapeutic window of new chemical entities (NCEs). In addition, high and unpredictable dosing might be required during dose escalation to achieve the maximum tolerated dose (MTD). A clear understanding of the rate-limiting step in oral drug absorption is crucial to the identification and selection of enabling technologies for improving oral bioavailability of poorly soluble APIs. This mechanistic understanding enables the selection of most appropriate excipients and process technologies.

Drug absorption can be either solubility or dissolution-rate limited. This is evident from the Noyes-Whitney equation. Most drugs are weak acids or weak bases, and follow a pH-solubility profile governed by the Henderson-Hasselbalch equation and the solubility of their ionised and unionised forms. The solubility and dissolution rate of weakly acidic drugs is low in acidic gastric fluids and high in basic intestinal milieu. The reverse is true for weakly basic compounds.

Dissolution-rate-limited drug absorption is characterised by slower in-vivo drug dissolution than the permeation rate through the intestinal membrane. In these cases, drug absorption increases with increase in dose and reduction of particle size. On the other hand, solubility-limited drug absorption is characterised by high dose:solubility ratio and high dissolution rate compared with the permeation rate. In these cases, drug absorption is not limited by the dissolution rate, but by the saturation solubility of the compound in the GI milieu at the site of drug absorption. Increase in dissolution rate (through decrease in particle size) or dose does not contribute to increased fraction dose absorbed (F). Non-linear dose exposure relationship caused by solubility-limited drug absorption is characterised by reduction in F with increase in the dose.

The acid test

Low or variable oral-drug absorption can also be associated with crashing out or precipitation of the drug upon transfer from acidic gastric pH to basic intestinal pH. This causes low drug solution concentration in the intestine, where features of dissolution-rate-limited drug absorption can then become evident. Redissolution of the precipitated drug in the intestinal milieu depends on the particle size, state of agglomeration of the precipitate, and its nature (amorphous vs crystalline). In addition to inter and intra-subject variability, drug precipitation upon intestinal transfer and pH shift can lead to food effect, gastric-pH interaction, presence of a second peak in drug absorption, and/or delayed/variable time to maximum drug concentration in the plasma.

"Drugs and compounds that are surface active and self-associate under basic pH conditions can lead to supersaturation of the drug at the site of absorption."

Since pH-dependent solubility of weakly basic drugs is a key contributor to drug precipitation, enabling technologies that lead to supersaturation of the drug upon pH shift can improve oral bioavailability. The rate and extent of oral drug absorption is then impacted by the proportion of drug and the duration a drug remains in the supersaturated solution state relative to the small intestinal transit time.

Drug precipitation upon pH shift takes place in two phases: nucleation and crystal growth. Certain phenomena contribute to inhibition or delayed kinetics of either of these phases, including self-association and complexation.

Complexation of poorly soluble drugs with soluble excipients, such as cyclodextrins, results in the complexed drug being the predominant molecular entity in the dosage form and drug solution – exhibiting its distinct physicochemical properties such as solubility, stability and diffusion coefficient. This approach has been used to alter the physicochemical and biopharmaceutical properties, and has been shown to improve the bioavailability of several APIs. Most complexes dissociate at the site of absorption or in vivo, leading to free drug being absorbed and present in the plasma.

Self-association of a drug in solution state can lead to the formation of dimers, trimers, multimers, or micelles. Micellisation of amphiphilic/surface active (surfactant) drugs is a well known phenomenon associated with a critical micelle concentration (CMC), above which interfacial occupation of the surfactants is complete and a significant change in solution properties is observable. In contrast, some drugs can self-associate to form stoichiometrically less-well-defined soluble structures in solution. These self-associated structures have low Gibbs free energy, compared with the monomeric form of the API, and provide a thermodynamic barrier to drug precipitation.

We have recently shown that drugs that self-associate under acidic conditions may exhibit prolonged and/or greater supersaturation upon transition of their aqueous solution into a basic pH environment, even though they may not be amphiphilic or surface active under basic pH conditions. Drugs and compounds that are surface-active and self-associate under basic pH conditions can lead to supersaturation of the drug at the site of absorption.

A helping hand

Certain formulation excipients have been identified that help create and maintain drug supersaturation at the site of absorption. Polymeric precipitation inhibitors that have been tested for this purpose include various grades and derivatives of polyvinyl pyrrolidone (PVP) and cellulosic polymers such as hydroxypropyl methyl cellulose (HPMC), methyl cellulose (MC), and carboxymethyl cellulose (CMC). Although the supersaturated state is thermodynamically unstable and would ultimately lead to drug precipitation, the kinetic delay of nucleation and crystal growth can be sufficient to allow prolonged maintenance of flux for passive diffusion across gastrointestinal epithelium. Extension of a drug’s crash resistance for a significant enough time that covers the absorption window of drug’s passage through the small intestine can improve a drug’s oral bioavailability.

Lipid-based drug-delivery systems (DDS), including self-emulsifying DDS (SEDDS) and self-microemulsifying DDS (SMEDDS), have been shown to improve the bioavailability of several poorly soluble drugs. In these systems, the drug is solubilised in a mix of lipids and co-solvent(s) such that the DDS forms a stable opaque white emulsion (SEDDS) or a transparent nanoemulsion (SMEDDS) upon reconstitution with water with minimal agitation. Application of these systems is contingent on drug solubility in the selected lipids or mixture of lipids and/or co-solvents, which may be a limitation for highly crystalline materials. While the lipid-based systems are useful for providing the drug in the dissolved state at the site of drug absorption, their utility depends on the selection of the right lipid, co-lipid, and solvent among a relatively confusing array of choices. The lipids also tend to be expensive and processing technologies less commonly established in the commercial manufacturing environment of most organisations. Controlling and reducing the particle size of the drug can be accomplished by a variety of approaches, including milling, co-precipitation, formation of crystalline solid dispersions and controlled crystallisation. It is by far the most established and often-used approach in the industry for overcoming dissolution-rate-limited absorption challenges. Using APIs of well-defined low particle size, including nanoparticulate drugs, provides the advantage of rapid rate of drug dissolution. In certain cases, powder handling can become a concern from the manufacturability perspective due to the high surface energy, surface electrostatic charge, low bulk density, aggregation propensity, poor flow and potential for content uniformity issues, especially for low drug load formulations.

Scratch the surface

Co-processing of drugs and excipients is commonly used to modify the surface properties of the API by providing intimate contact with the excipient of choice. Co-processing of drugs with excipients can be accomplished by spray drying, co-precipitation, co-grinding, formation of crystalline solid dispersions, or co-milling during conventional manufacturing processes.

Adsorption of drugs on the surface of certain excipients can lead to increase in the surface area of drugs, thus increasing the rate drug release. For example, excipients such as kaolin or microcrystalline cellulose have been used as adsorbents in certain cases. Drug deposition in the pores of porous excipients such as silica can lead to a stable amorphous drug formulation, in addition to providing low particle size of the dissolving species. Adoption of these approaches requires careful consideration of process variables and control strategies to ensure reproducible manufacture.

"During early development, one is faced with the apparently conflicting choice of moving the compound forward in the pipeline through enablement or focusing instead on discovery efforts to look for an alternate compound that may not require an enabling technology."

Amorphous solid dispersions present the drug in a thermodynamically unstable amorphous state and a very small size of the dispersed state. These systems provide the advantage of increasing the saturation solubility of the drug as well as the rate of drug dissolution. Hydrophilic polymeric excipients, such as HPMC-acetate succinate (HPMC-AS), polyethylene glycol (PEG) and PVP, have been successfully used for the preparation of immediate release solid dispersions. The processes used for the preparation of solid dispersions, such as hot-melt extrusion or spray-dried dispersion, can result in significantly different powder properties of the resulting material in attributes such as density and porosity. The selection of these processes also depends on API characteristics such as miscibility with the polymeric excipients, solubility and stability in organic solvents, melting point and stability at elevated temperatures. The major challenge for amorphous systems is their relatively unpredictable physical stability, hygroscopicity and limited applicability to high-dose drugs given the need for high excipient:drug ratio.

Decisions, decisions

In light of extremely poorly soluble Biopharmaceutics Classification System (BCS) Class II (low solubility, high permeability) and BCS Class IV (low solubility, low permeability) NCEs dominating the pipeline, application of enabling technologies is increasingly being considered very early in drug discovery and development. The early stages of development are invariably rapid-paced with stringent timelines for preclinical and early clinical development. Making decisions on whether an enabling technology is needed and the selection of a solution for a particular drug invariably involves assumptions regarding the ability to develop assets, the enabling technology’s promise, and a potential path to commercial application. A general tenet in the selection and use of enabling technologies is to adopt the simplest and the most cost-effective of the available options, and to critically assess whether enablement is really needed.

Preliminary assessment of the need for enablement of an NCE and whether this need is met with the selected enabling technology is usually data-driven. These aspects raise questions about how much do we need to know before making such decisions? Where, when, and how do we use enabling technologies?

Perplexed with different choices of enabling technologies, different organisations usually resort to a cultural or historical preference of one over another. However, enabling technologies cannot generally be used in exchange for one another. Identifying the minimum amount of data that would be needed to make the decisions becomes intertwined with risk perception and risk management in drug-product development. Several organisations use a decision-tree approach to making formulation and technology decisions for NCEs at different stages of development. In addition, during early development, one is faced with the apparently conflicting choice of moving the compound forward in the pipeline through enablement or focusing instead on discovery efforts to look for an alternate compound that may not require an enabling technology.

These aspects make the decisions on the need and selection of enabling technology for a drug’s development challenging and interesting. Making these decisions in real time with limited data and within the timelines is usually done in a case-by-case basis with a combination of good science and good judgement.

An open forum for the formulation and drug development (FDD) and physical pharmacy and biopharmaceutics (PPB) sections of the American Association of Pharmaceutical Scientists (AAPS) will take place at the AAPS annual meeting this year, where these questions of risk and knowledge-based drug-product development can be debated further: where, when, and how do we use enabling technologies?